Thinner granite tongues penetrate the gabbro in irregular fashion , while marginal shattering of the gabbro is abundantly evidenced by the xenoliths of all shapes and sizes included in the outer parts of the granite .sx Both acid and basic rock are penetrated by sheets of red granite-aplite , for the main part inclined ( south ) at low angles .sx These same features are repeated again and again along the north coast of Jersey .sx The more intimate relations between the two rocks are most favourably exposed some 300 yards farther west , on and near .sx the walls of a sea-cave .sx In this neighbourhood the basic rock is penetrated in intricate fashion by a maze of veins of granite ranging in width from two feet to a mere fraction of an inch .sx The veins follow no definite direction , but , judging from the sharply angular forms of many of the xenoliths , the gabbro must have been in the solid state when invaded by the acid magma .sx The exposures hereabouts illustrate in spectacular fashion the several stages in the stoping process ; indeed it is difficult to envisage a more convincing example .sx In the case of the rock-mass illustrated in fig. 15 it is obvious that , were the granite-member of the complex to be removed by some process , the isolated blocks of gabbro could be replaced .sx in their original positions with little loss .sx In this , the first stage of the stoping process , it is clear that the amount of xenolithic material assimilated by the advancing magma is negligible :sx the mechanical effect , which is a mere shattering due to unequal expansions , is very striking , however , and was definitely not restricted to the roof , but at least in the earlier stages of intrusion , affected also the wall-rock .sx That the granite magma , though doubtless highly viscous , possessed remarkable penetrative power is shown by the thinness of some of the veins ( see Plate 6A) .sx The advance of a tongue of magma must have been instantaneous with the opening of a fissure as a consequence of the high hydraulic pressure to which it was subjected .sx Large blocks of gabbro with twenty-foot sides are commonly seen completely insulated from the parent mass by sheets of granite sometimes only an inch or two in thickness .sx Although one is at first tempted to regard the granite veins as having wedged off these xenoliths , there is here no evidence that the granite acted in other than a purely passive capacity , providing the substratum of lower specific gravity into which the shattered blocks could subside through the operation of gravity , and also the necessary lubricant to facilitate the sinking .sx A continuation of the same process on a more limited scale led to the brecciation of these major xenoliths , producing a peripheral shattering at first , but ultimately reducing them to a size suitable for chemical reaction with the magma .sx In a number of instances we have noticed that the fractures bounding the xenoliths may be strongly curved ( Plate 7B ) , and it is clear that all rounded enclaves do not owe their form to magmatic corrosion or assimilation , as is perhaps too commonly assumed .sx No doubt the manifestation of curvilinear joints , occult in the rock , is stimulated by the unequal heating effects consequent upon immersion in the granite-magma .sx In other cases , however , rounding has doubtless been aided by a process akin to exfoliation .sx Opponents of the stoping hypothesis have more than once maintained that blocks broken off from the roof of an intrusion would be prevented from sinking by the hydrostatic pressure and the viscosity of the magma .sx In the present case the blocks have demonstrably sunk , some to considerable distances , and it is clear that a continuation of the process must have resulted in a replacive relationship being established .sx As bearing on this matter it may be recorded that at one point west of Ronez two rounded xenoliths , each some six feet in diameter , are embedded in a grey-coloured , streaky hybrid granite rich in porphyritic felspars .sx The dark schlieren in the .sx neighbourhood of the upper basic xenolith have been deflected downwards into approximate parallelism with its periphery .sx Where the two xenoliths are nearest together , the schlieren are squeezed into close proximity , and at the same time the felspar phenocrysts lose their idiomorphism , passing first into aligned almond-shaped augen , and finally into mere light-coloured streaks .sx Obviously these features have been produced by the pressure of the upper xenolith upon the viscous fluid separating it from the lower one .sx Fig. 16 .sx In addition to the mechanical disintegration described above ; however , the evidence that xenoliths engulfed at an earlier stage have suffered profound modification is so clear that chemical analyses and study of thin sections would be necessary only to provide quantitative data .sx In the field the evidence is conclusive .sx Xenoliths representing all stages in the process by which they lost their identity and became completely merged in the invading magma are abundant .sx The one extreme , the first stage , is represented by fragments frozen in immediately after separation from the parent magma ; at the other extreme we have the " ghosts " whose original xenolithic character is inferred by reason of a slight darkening of colour and a richness in coloured minerals in an otherwise normal granite .sx As the result of continued submergence in the magma the xenoliths become progressively paler in tint and changed in mineral composition .sx Illustrations of interaction of another sort are to be seen in both the north-western and the south-eastern granites .sx Large xenoliths , rudely spherical in form , have apparently suffered much marginal corrosion , judging from their fretted outlines .sx A curious feature is a variation in grain size in the xenoliths , the texture becoming progressively finer towards the periphery , and the impression is conveyed that the xenoliths have been actually chilled against the invading magma , which is of course absurd .sx The granite veins are only an inch or two in thickness , and the xenoliths are sometimes some yards in diameter , and it is clear that another explanation must be sought .sx Generally , the darker and finer-textured the marginal facies , the clearer and more distinct is the junction between the two rocks .sx This contrasts strongly with other cases in which obvious admixture has taken place , giving virtually a passage from gabbro to granite in a few inches .sx It appears at first sight as if quite thin veins of extremely hot and fluid granite were capable of causing actual peripheral fusion of the basic rock with which they were in contact , the fused zone then cooling down more rapidly than during the initial consolidation of the gabbro , with the consequent development of a finer grained texture .sx This must , however , be ruled out as highly improbable .sx Even if it be admitted that the magma was superheated ( which few would admit as a possibility at the present time ) , in this case thin veins are called upon to supply the heat necessary to fuse basic rock .sx The observed effects may be explained adequately in terms of Bowen's reaction principle :sx for reaction allows for the diffusion of alkalies , silica and fluxes , including water , into the basic xenoliths , from the magma .sx This diffusion modifies the composition of the invaded rock in such a way that its .sx melting point approximates more and more closely to that of the invading magma .sx In this modified mass crystal phases will arise , which , because they are forming in merely softened rock and not in fluid like the surrounding granite , will .sx be small and irregular , thus giving the appearance of a " chilled margin .sx " ( This rock is described on p. 205) .sx In a particularly informative exposure a short distance to .sx the west of Ronez , several stages in the assimilation process may be studied in contiguous xenoliths ( fig. 16 ) ; Inclusion " A " shows the common marginal shattering , resulting from the sudden .sx immersion of cold xenolith in hot magma .sx Inclusion " B , " engulfed at an earlier period .sx shows some brecciation , but in .sx addition , a softening and streaking out of the fragments , which , by admixture with the invading magma has produced a streaky hybrid rock of gneissose appearance , while this in turn gradually merges into normal granite .sx Finally , inclusion " C " shows the peripheral `fusion' followed by the development of a chilled .sx margin , described above .sx It must be emphasised that these and kindred changes .sx were impressed upon the xenoliths when the magma had become practically immobile :sx it had lost its penetrative power .sx Further , the xenoliths are separated from the parent rock by comparatively short distances only .sx We should be doing little justice to the evidence exhibited in such spectacular fashion if we did .sx not , with Daly , assert that the xenoliths still recognisable as .sx such are merely the " rearguards of a host , " and that the processes .sx so obviously operative during the last stage of uprise must have been active at greater depths and for a long period .sx It is perfectly evident that this was not accomplished without profound .sx modification of the invading magma .sx Discussion of this point .sx would , however , carry us far outside the scope of this paper .sx The Form of the Complex .sx In the preceding paragraphs the nature of the junction of the granite with its roof and walls has been detailed , and emphasis .sx has been laid upon the fact that , wherever segments of the roof are visible , the granite is seen in intimate association with gabbroid rock .sx It is an association that is common elsewhere , and records of composite intrusions of granite and gabbro , or ( in the case of minor intrusions ) , of their hypabyssal equivalents , are frequent in petrological writings .sx The Carrock Fell gabbrogranophyre complex , Carlingford , and the several Tertiary examples from the Western Isles are well known cases in point .sx The two contrasted rock types in these examples are regarded as co-magmatic .sx There is no reason to doubt the consanguinity of the jersey granites with the gabbroic roof-rock ; but as has been shown , the interval of time between the arrival of the basic magma and the acid material which followed it was sufficiently long to allow the former to have become solidified and well jointed .sx At Ronez and Sorel Point the gabbro reaches sea level and below ; but elsewhere along the north coast it occurs as small isolated patches capping the cliff-tops , the lower parts consisting of granite .sx In the extreme north-west the gabbro is represented only by scattered xenoliths in a grey , contaminated granite .sx In the south-east of the island the granite is again capped by the isolated remains of a basic intrusion , more extensive than in the north-west complex ; while the south-western granite bears evident traces of basic rock as xenoliths , and is clearly contaminated .sx There is thus strong presumptive evidence that the basic rocks form fragments of a sheet-like intrusion that was originally at least coextensive with the granites .sx The base of the sheet is now far from regular :sx it appears to have been tilted slightly from west to east , so that its presence in the west is inferred only from the petrographic character of the granite and of its xenoliths .sx Eastward it intersects the cliff-tops and descending to lower levels , appears to thin out in the complicated melange at the western end of Les Rouaux beach , between the Jersey Shales and the unconformably overlying volcanic agglomerates of Belle Hougue ( Fig. 13) .sx Thus it appears here to have been in the nature of an intraformational sheet , but probably did not keep to this level .sx In addition to erosion from above , the sheet has been subjected to active stoping from beneath ; while collapse along lubricated planes of weakness has lowered great roof pendants and rafts to levels far beneath the original base of the sheet .sx The alternative to supposing the basic rock to have been sheet-like in form has little to recommend it .sx It seems much more likely that one gabbro and one granite are represented in these three complexes , rather than three granites and three gabbros showing exactly the same relationships .sx The sheet hypothesis is supported by the occurrence , in the Ronez quarries , of a series of different basic rock types , rudely stratiform ( p. 198 ) , and by the quasi-horizontal sills of aplitic granite that intrude the gabbro , presumably along pseudo-bedding joints , in the neighbourhood of the Hermitage in the south of Jersey .sx