California Division of Mines and Geology Bulletin 202 Geology of the Point Reyes Peninsula, Marin County, California

San Andreas Fault

The San Andreas fault zone, forming the long straight inlet of Tomales Bay and the long straight valley between Olema and Bolinas Lagoon, is overwhelmingly the most important structural and physiographic feature of the Point Reyes Peninsula.

In order to avoid confusion, in this report the words "San Andreas fault zone" are used in the sense employed by Noble (1926) to designate the band of roughly parallel fractures which accompany the "fault trace," or master-fault. The words "1906 fault trace" are used to designate the actual break in the ground which occurred at the time of the San Francisco earthquake of April 18, 1906. This 1906 fault trace lies wholly within the "San Andreas fault zone." The term "rift," used by Lawson (1908) to designate the belt of topographic features along the fault zone, will not be used in this report.

Quoting Crowell (1962), it can be truly said that, in the section between Bolinas Lagoon and Tomales Bay, the fault zone "consists almost everywhere of a broad shallow trough filled with fault land forms, such as scarps, slice ridges, sag ponds, shutterridges and offset streams." Driving along State Highway 1, these topographic features are visible from the road. In particular the passer-by will notice the slice ridges near the Hagmaier Ranch, the big sag pond east of State Highway 1 about a mile north-northwest of Five Brooks and the shutterridges along the 1906 fault trace near the National Park Headquarters. At Bolinas Lagoon and Tomales Bay, the trough of the fault zone has been invaded by the ocean. Sharp (1954) described seventeen primary fault features and seven secondary fault features in southern California, and most of these physiographic features also can be seen here.

The fault zone is distinguished not only by the land forms but also by the fact that the Franciscan rocks, which to the east consist almost entirely of graywacke, within the fault zone consist mostly of serpentine, pyroxenite, decomposed ultrabasic rocks, and other rock slices and boulders, termed "tectonic inclusions" by Schlocker and Bonilla (1962), which are demonstrably not in place and which have presumably been moved along the fault zone. The Calera limestone block near Five Brooks is an example. Many of these stand as isolated boulders in the grassy meadows.

The Franciscan of the fault zone is overlain by the marine Merced Formation at the southeast end of the zone and by the fresh-water Olema Creek Formation at the northwest end. In the vicinity of Five Brooks this serpentine-bearing phase of the Franciscan Formation is exposed at the surface. The fault zone in this area seems to be bounded by two large fault traces, one on each margin, termed the east and west boundary faults for convenience. The nearly parallel sides of the long narrow inlet of Tomales Bay suggest that it too is bounded by two parallel faults.

The east boundary fault trace is distinguished by many sag ponds and strong topographic evidence of a large fault east of State Highway 1, northwest from Five Brooks. The topographic expression of this fault here is greater than along the 1906 fault trace, suggesting that movements on the east boundary fault have been large and frequent in the not too distant past. This fault seems to disappear under terrace deposits to the northwest and the southeast, but no doubt it is represented by the east side of Tomales Bay.

To the south, it seems likely that the continuation of this fault determines the position of the east side of Bolinas Lagoon. The occurrence of hot springs on the beach just north of Rocky Point suggests that the faulting may extend as far south as that locality. Tectonic inclusions can be seen beside State Highway 1 near here.

Another large fault seems to form the west boundary of the San Andreas fault zone. This fault, also marked by striking topographic features, has a more pronounced effect on the geology since it separates Monterey Shale and granitic basement in the north from the Franciscan of the fault zone. The west boundary fault trace is occupied by Paradise Valley near Bolinas; further northwest, it is occupied by the upper portion of Pine Gulch Creek.

Northwest of Pine Gulch Creek, the west boundary fault appears to control Olema Creek on the west side of the peculiar rhomboidal detour it makes in its course just south of Five Brooks. Three thousand feet northwest of Five Brooks, it seems to coincide with the 1906 trace and then diverges at a small angle to the west of the 1906 trace until at Olema it is nearly 2000 feet west of it. Northwest of this point, its course is obscured by terrace deposits, but it is no doubt represented in a general way by the granitic margin of the west side of Tomales Bay. This west boundary fault was observed in the Bolinas sea cliff in 1913 (Bradley, 1915, p. 243) as a vertical fault. Projecting it southeastward, it is intriguing to imagine it connecting with the Pilarcitos fault (San Mateo county), which lies in a similar geological environment with the granitic Montara Mountain to the west, and the Pliocene Merced Formation to the east, underlain by Franciscan rock. It might also be speculated that the San Bruno fault corresponds to the east boundary fault of the Olema Valley (see San Franciscan sheet, Geologic Map of California).

Many additional fault traces are found in the fault zone and show topographic expression. The most clearly marked is, of course, the most recent—the rupture caused by the 1906 earthquake. This is described in fascinating detail by Gilbert (1908), and students of the area can be grateful that this eminent and thorough geologist was available to describe the Point Reyes Peninsula section of the rupture in the classic report of the California Earthquake Investigation Commission. Gilbert's descriptions are so clear that it is easily possible over sixty years later to retrace his footsteps in spite of the growth of vegetation and changes in cultivation and houses.

No attempt will be made here to repeat Gilbert's work, since it could hardly be improved upon, although he had no reliable topographic map of the area. He described the surface expression of the rupture as taking three forms: the ridge phase, the trench phase, and the echelon phase; traces of the first two can still be seen. The minor surface features of the 1906 rupture are surprisingly persistent.

The 1906 rupture at Bolinas, which cut the tip of Stinson Beach sandspit, was about 3500 feet from the west boundary of the fault zone as expressed by the Merced—Monterey contact. Northward at Five Brooks the fault zone is only about 1500 feet wide, and the 1906 trace is over against the west side of the zone. Farther northward, the zone widens and forms the submerged depression of Tomales Bay. The 1906 trace is established by lying more or less centrally between the two edges of this inlet (Lawson and others, 1908). The 1906 fault trace cut Tom's Point and Sand Point (near Dillon's Beach) but apparently did not cut Millerton Point. At Bodega the 1906 trace was close to the east side of the fault zone.

Many other shorter, more discontinuous, fault traces are to be seen in the fault zone, and some are as impressive as the 1906 trace. It seems clear that the fault zone has been the locus of a very large number of earth-ruptures (accompanied no doubt by large earthquakes) over a long period of geologic time.

WIDTH AND LENGTH

Tomales Bay is about 7000 feet wide at its northern end, suggesting that the width of the fault zone there is of that order. Southward, the fault zone narrows to 1500 feet at Five Brooks, widening again southward to about 4000 feet at the head of Bolinas Lagoon, and is about 8000 feet wide at the latitude of Bolinas (assuming the east shore of Bolinas Lagoon is the east boundary of the fault zone). This width is narrow compared with the 6 mile maximum width of the fault zone in southeastern California (Noble, 1926). The total length of the segment of the San Andreas fault zone in the Point Reyes Peninsula, measured from Tomales Bluff to Bolinas, is about 30 miles.

ASSOCIATED FAULTS AND FRACTURE TRACES

It is mentioned above that many fault traces are to be recognized by their topographic expression in the fault zone itself. Outside the fault zone there are a number of dislocations that can be similarly recognized; these can be divided arbitrarily into: (1) associated faults and (2) fracture traces.

Gilbert (1908) reported a rupture of the surface which he attributed to "a divergent branch of the fault" in the north part of Bolinas, trending approximately north-south and fading out in both directions near the present site of Bolinas School (Lawson and others, 1908, p. 81). Since this rupture was in the alluvial plain, it seems equally possible, in hindsight, that it was caused by lurching of the incompetent alluvium. There seems to be no other evidence of a north-south fault in this vicinity. Several topographic expressions of fault traces traverse the Merced terrane to the north of the school, but these are all aligned in a northwesterly direction. This branch of the fault described by Gilbert would lie entirely within the fault-zone as defined above.

Gilbert also described a "bedrock crack" on Mt. Wittenburg, crossing the "northeastern spur of the peak near its junction with the main crest," and trending northwest-southeast. Aerial photographs show a sharp straight ravine, very suggestive of a fault, starting close to the National Park Headquarters in Bear Valley and heading toward the northeastern spur of Mt. Wittenburg. However, efforts to find remaining traces of this bedrock crack on the ground were unsuccessful, so this particular topographic indication should probably be referred to as a fracture trace (see below).

In addition to the above, Gilbert mentioned a "branch fault trace" about 1/2 mile long, traversing the ridge between the Second Valley and Third Valley at Inverness. Gilbert's report shows a photograph of this fault trace (plate 47A) and indistinct relics of it are to be seen on the ground today. Its geologic setting suggests that it was indeed a genuine bedrock crack, but no unequivocal indication of it is to be seen in the aerial photographs, perhaps on account of the subsequent growth of scrub and trees.

Aerial photographs disclose a band of apparent "fracture traces" (Lattman, 1958) west of the fault zone in the vicinity of Mud Lake. The fracture traces form a lineament which leaves the fault zone in a northwesterly direction and then curves concave to the fault zone until it becomes almost parallel to it. In the center of this lineament lies Mud Lake, an anomalous pond resembling a sag pond, on the top of Inverness Ridge. The lineament becomes indistinct and dies out not far north of Mud Lake.

Examination on the ground has failed to disclose the features responsible for this lineament; indications of fault topography are abundant but indistinct. The lineation could be due to bedding in the Monterey Shales; however the thickly forested area yields few measurable attitudes, and the general impression is that dips and strikes in the shales are not as regular as would be required for this explanation.

That this photogeologic lineament is a fracture trace is supported by an observation of Gilbert (1908. p. 76):

"A pond known as Mud Lake, on the divide at the head of Pine Gulch Creek and about a mile from the fault-trace, suddenly and permanently lost its water at the time of the earthquake. At the same time a small spring on the east side of the ridge and about 0.75 mile in a direct line from the pond, was suddenly enlarged, a torrent of water gushing from it for several hours and then gradually diminishing".

This occurrence could well have been due to the opening by the earthquake of some of the fractures in the vicinity of Mud Lake. Mud Lake today contains water, the 1906 crack having no doubt been filled with sediment.

Clague (1969) considers it probable that the formation of Mud Lake is involved in older landslides. This however is not incompatible with the fracture trace explanation of the photogeologic lineament described. The long lineaments on the aerial photographs are reminiscent of the faint surface lines reported by Warne (1953) in the vicinity of Bakersfield and attributed by him to minor faulting, although the forest cover on Inverness Ridge makes the features there more discontinuous to the eye than those in the San Joaquin Valley.

A drainage anomaly at the point where the lineament from Mud Lake intersects the fault-zone trough also suggests that the lineament is due to a fault or fracture trace. It is concluded that these faint lineations seen in the forest cover of the south part of Inverness Ridge are expressions of minor faulting or cracking related to movement on the San Andreas fault zone. The more prominent "fracture traces" are shown on plate 1. Evidence of faulting is sparse on the Franciscan block east of the fault zone, suggesting that these rocks may have yielded by folding instead of by faulting.

DIRECTION OF THE MOVEMENT

There is no question about the direction of the movement on the San Andreas fault in 1906. Lawson and others (1908) reported that practically all the movement in 1906 was right-lateral; that is, the ground on the southwest side of the fault moved horizontally northwest relative to that on the northeast side. Table 10 shows the 1906 displacements.

Table 10. Horizontal surface displacements in the earthquake of April 18, 1906.

The possibility of vertical movement in the Marin County segment of the fault trace was carefully examined by Gilbert, and he concluded that it could be demonstrated only that the part of Bolinas Lagoon east of the fault had probably subsided 12 inches relative to the part west of the fault. His photographs show that in many places the 1906 trace is bordered by what appears to be a fault scarp, but these scarps were due in part to landsliding and in part to lateral movement of slices along the fault, bringing higher ground in juxtaposition with lower ground (Gilbert's "fault ridges".) Lawson and others (1908, pl. 52A) illustrate a scarp which is actually the head of a Lake Ranch landslide. The fault ridges face indiscriminately southwest or northeast, making their origin from vertical movement unlikely. Gilbert himself saw the possibility of the "appearance of vertical displacement largely due to combination of horizontal displacement with slope of ground." Gilbert saw little convincing evidence of vertical movement on the 1906 fault trace; most California geologists today probably concur that horizontal movement is the dominant feature of the San Andreas fault zone (Dickinson and Grantz, 1968).

However, Oakeshott (1966) makes a plea for retaining a flexible position with respect to the possibility of vertical movement on the fault, and it is worthwhile looking at the regional geology with this in mind. The highest point of Inverness Ridge, lying between the fault zone and Drakes Bay, is Mt. Wittenburg (elevation 1407 feet). Mt. Wittenburg is overlain by Miocene marine sediments so Inverness Ridge has been uplifted more than 1400 feet since the Miocene. The beds on Inverness Ridge dip southwestward into the Drakes Bay syncline.

On the northeastern side of the fault zone in Marin County lies the "San Francisco—Marin fault block" (Lawson, 1914), composed mostly of Franciscan rocks. This fault block dips northeastward, in the opposite direction, and its eastern margin has been invaded by the waters of San Francisco Bay, just as the center of the Drakes Bay syncline has been invaded by the waters of the Pacific Ocean. The highest point on the San Francisco—Marin block along Bolinas Ridge bordering the Olema Valley is about 1440 feet. This means that the rocks on both sides of the San Andreas fault zone have been uplifted, forming now a kind of roof-ridge structure along the fault zone. This is compatible with the generally accepted conclusion that the San Andreas fault zone is an area of compression. Whether the uplift has taken place as vertical faulting, or otherwise, is not disclosed by the evidence. But uplift there has been in the geologic past, and enough subsequent erosion to expose the granitic pluton of the Point Reyes Peninsula at the surface.

Nevertheless, all the topographic evidence within the zone itself points to long-continued lateral (horizontal) movement. The rocks on the two sides of the zone are quite different so there is no opportunity in the area to measure the total displacement on the fault zone by comparing the two sides. However, an interesting geological detail bearing on the nature of the movement came to light during the field work for this report. The lake beds of the Olema Creek Formation contain abundant granitic material almost certainly derived from the granitic rocks of Inverness Ridge. This granitic sedimentary material extends south in Olema Creek almost as far as the Boyd Stewart Ranch. The nearest granitic outcrop today is the small inlier in Bear Valley near the old Country Club. If this inlier provided the material, it must have been transported northward to the fault zone via Bear Valley Creek, and hence will have reached the fault zone approximately at the Vedanta Retreat (formerly Shafter Ranch), about 12,500 feet along the fault zone, from the Boyd Stewart Ranch. This detrital granitic material has been displaced this distance laterally southeastward along the fault zone since it was deposited. Since long-term average annual rates of movement along the fault zone are variously estimated as between 1/2 inch and 2 inches per year, these numbers are compatible with the conclusion that the Olema Creek beds are Pleistocene in age. The Olema Creek beds lie on the northeast side of the 1906 trace, so the right-lateral movement could have taken place on this trace.

Thus, there is some evidence that right-lateral movement has been taking place here since Pleistocene time. What movement took place before that time has to be determined by correlations across the fault outside the Point Reyes Peninsula. At the present time, most students of the question feel that right-lateral movements at average rates between 1/2 and 2 inches per year have persisted since Cretaceous times (Dickinson and Grantz, 1968).

The fact that the Bolinas end of the valley, in which the San Andreas fault lies, was filled with Merced marine sediments during late Pliocene time indicates that the Olema—Bolinas valley was already eroded at that time, suggesting that the San Andreas fault zone is at least older than the Merced Formation. The remnants of terrace beds containing Franciscan material, lying on the Monterey Shale, suggest that the cutting of the Olema Valley was post Miocene. Thus the erosion of the Olema—Bolinas valley took place between late Miocene and late Pliocene time.

One argument in favor of long-continued lateral movementon the San Andreas fault can be based on the complete difference between the rocks on the two sides of the fault zone—Franciscan on the east and granitic plus Tertiary on the west. To account for this geological setting, Lawson (1914) invokes tremendous oscillating vertical movements on the San Andreas fault to ac count for the absence of the Franciscan Formation on the west side, in spite of the example of the convulsive right-lateral movement of 1906 which he himself observed.

Application of the principle of simplicity (Anderson, 1963) would lead one to prefer the hypothesis of a long-continued lateral movement, similar to that actually observed, to the hypothesis of the large vertical movements necessary to strip rocks of the Franciscan Formation from the west side and the Tertiary rocks from the east side of the fault.

SHUTTERRIDGES

The fault zone between Inverness Park and Five Brooks is distinguished by a succession of elongated ridges parallel to the 1906 trace on the northeast side of the fault zone. Gilbert (1908) called these ridges "peculiar in that their western, or more strictly southwestern, base being determined by faulting, is nearly straight; while their northeastern base, modified by the erosive action of Olema Creek, is scalloped." The ridges, as far southeast as the Truttman Ranch, are composed of granitic and Monterey Shale debris, clearly derived from Inverness Ridge; they correspond, therefore, to Buwalda's (1936) definition of shutterridges as "ridges which by fault shift, horizontally or vertically or in both components, tend to move across, block and shut in the ravines or canyons of streams crossing the fault." Bear Valley Creek and Cold Creek (5000 feet northwest of Five Brooks) have certainly been blocked by the movement of these ridges, so that the term shutterridge seems appropriate. No doubt those ridges, composed of granitic and Monterey Shale debris, were originally portions of outwash fans of streams coming from Inverness Ridge, which originally flowed across the 1906 trace from west to east. The ridges were shifted to the southeast by right-lateral movement on the northeast side of the 1906 trace.

EARTHQUAKES AND CREEP

The San Andreas fault in California is associated historically with several large earthquakes and many small ones. The Point Reyes Peninsula segment of the fault has been relatively quiet seismically since the tremendous 1906 earthquake. On segments of the fault near Hollister and Parkfield—Cholame, creeping displacement is taking place at the surface, and shallow small earthquakes are frequent.

Inspection of the 1906 trace in Marin County fails to show any tectonic creep, and if movement is taking place along some other segment in the fault zone, it has passed unobserved. This lack of creep seems to be consistent with the absence of small earthquakes along this stretch. Special efforts to locate evidence of recent fault traces have been unsuccessful.

Although progress has been made in earthquake prediction, at this time no certain method of forecasting the occurrence of earthquakes has been developed. A study of the San Andreas fault zone in Marin County convinces one that a great deal of earth movement, presumably accompanied by earthquakes, has taken place in the past. There is no reason to expect any future change in this situation, but whether the next earthquake on this segment of the fault will occur today, tomorrow, next year, or 100 years hence cannot now be determined.

STREAM DIVERSION

At the north end of Olema Valley, Bear Valley Creek emerges from the hills west of Olema and flows northeastward until it reaches the 1906 trace, where it makes an abrupt right-angle turn to the northwest. It does not cross the 1906 fault trace but continues northwestward into Lagunitas Creek at the head of Tomales Bay. It is prevented from joining Olema Creek by the shutterridges which have moved across its course.

The next stream to the south (unnamed) flows northeasterly from the hills until it reaches the fault trace of 1906, where it turns northwest for 1000 feet, after which it resumes its northeasterly course and flows into Olema Creek. The sharp turn of this stream to the northwest is apparently due to its northeasterly course having been blocked by a shutterridge.

Further southeast and still on the southwest side of the 1906 fault trace, a stream, locally known as Cold Creek, joins Olema Creek close to bench mark 167, about 3500 feet northwest of Five Brooks. The upper part of its course is northeast, but on approaching the 1906 trace it describes a wide curve to the southeast, flows southeast along the fault trace for 2000 feet, and then joins Olema Creek to flow northwest. It is as though its junction with Olema Creek had been pushed southeastward by the long shutterridge whose highest point is marked as 257 feet elevation on the topographic map (Inverness quadrangle). Other streams to the southeast of Cold Creek also show a southerly bend.

At Five Brooks, as the name implies, the drainage is very complex and is undoubtedly affected by fault movements. South of Five Brooks, the streams coming from the west do not flow into Olema Creek, which is separated from them by a series of fault-slices; instead they form the headwaters of Pine Gulch Creek which in the upper part of its course runs along the west boundary fault of the San Andreas fault zone. In this area, Olema Creek and Pine Gulch Creek run parallel to one another in opposite directions, separated only by about 1500 feet, with Olema Creek flowing into Tomales Bay to the north and Pine Gulch Creek into Bolinas Lagoon to the south.

Pine Gulch Creek, instead of going directly southeast into Bolinas Lagoon as one would expect, turns southwest through Paradise Valley and eventually empties into Bolinas Lagoon near Union School.

The writer has considered various possible causes for this behavior. The stream could perhaps have ponded and diverted through Paradise Valley by a landslide in the neighborhood of Woodville; but there is no obvious evidence of this. The same effect could have resulted from tectonic uplift in the vicinity of Woodville. No doubt the stream has been affected by fault movements, but it is difficult to see exactly what type of movement could have caused this anomaly. Higgins (1961) suggests that its anomalous course might be due to superposition from a blanket of terrace deposits.

A more likely immediate cause would seem to be stream-capture by a tributary of the actively eroding McCormick Creek. This creek presumably originally flowed from the south side of Pablo Point through Paradise Valley into Bolinas Lagoon; after the rejuvenation by the recession of the ocean which formed Bolinas Mesa, a northeasterly tributary may have cut back and captured the waters of Pine Gulch Creek, diverting them away from their course which originally led to the end of Bolinas Lagoon.

On the northeast side of the fault zone in the Olema Valley, there is little evidence of stream diversion due to fault movement.

Regional Structure

The Point Reyes Peninsula is synclinal in general form. The oldest rocks crop out on Inverness Ridge and at Point Reyes, on the northeast and southwest margins of the territory; the central synclinal part is occupied mainly by the Pliocene Drakes Bay Formation. The axis of the central syncline, termed the Point Reyes Peninsula syncline by Weaver (1949), is occupied by the drowned valley of Drakes Estero, which suggests recent sinking of this area. The Bolinas terrace extends discontinuously northwest as far as Drakes Bay, where it dips under the present beach on the east side of the syncline. The present position of the older beach deposits along Point Reyes Beach suggests recent uplift at Point Reyes and Tomales Point, with an intervening syncline.

Cores from oil well tests suggest that Point Reyes syncline thickens to the south under the ocean. This is also suggested by regional gravity anomalies (Chapman and Bishop, 1968).

Local Structure

FOLDS

An easily accessible anticlinal fold is exposed in the sea cliff about 6000 feet southwest from Drakes Beach County Park. This fold is strongly asymmetric, with southwest dips of about 40 degrees on the southwest flank and northeast dips of about 5 degrees on the northeast side. The axis trends northwest across the peninsula. Southwest of the axis, there is a shallow complementary syncline, the east end of which is clearly outlined in the offshore kelp beds (photo 26). These features are referred to in this report as the Mendoza anticline and syncline after the ranch on which they are situated.

Photo 26. The Mendoza syncline. underwater in Drakes Bay, is outlined in kelp. Photo by Towill, Incorporated, San Francisco, and U. S. Army Corps of Engineers.

An unsuccessful exploratory oil well drilled on the axis of the Mendoza anticline encountered granitic basement at a subsea depth of about 600 feet. It seems likely that the anticline expressed in the Drakes Bay Formation reflects vertical faulting in the basement complex.

A small steep anticline involving the Monterey Shale as well as the Drakes Bay Formation extends northward from the head of Schooner Bay at the north end of Drakes Estero. At the latitude of the south end of Abbotts Lagoon, it heads into a strong anticlinal fold trending approximately east-west. The hill which divides Abbotts Lagoon into two parts seems to be a topographic expression of this east-west uplift. To the east the uplift appears to extend back into the granitic basement, causing a swing in the granitic outcrop at the Sir Francis Drake Highway.

Further to the southwest another line of folding, running more or less northwest-southeast, is expressed by inliers of Monterey Shale which appear in the outcrop of the Drakes Bay Formation at and northwest of the Home Ranch. This fold extends southeast as a steep zone as far as Glenbrook Creek, where an inlier of granitic rock is exposed in the creek bed. This or a similar steep zone extends southeast through Muddy Hollow and the Laguna Ranch, nearly to the Y—Ranch, southeast of which it disappears in the complexities of the Wildcat Ranch landslides.

A syncline in Monterey Shale is exposed on the beach about 4000 feet northwest of the mouth of Arroya Honda. The "Religious Colony slump" of Clague (1969) can be seen in the cliffs above the syncline. Northeast of the Religious Colony (Palomainn) site is a steep zone of vertical and overturned beds in the Monterey Shale, which dies out to the southeast and disappears into the Double Point landslide to the northwest.

Previous writers (Lawson, 1914; Leck, 1921; Douglas, 1943) have described an anticline parallel to the west side of Bolinas Lagoon and about 1/2 mile to the west. The present mapping did not confirm the existence of this fold. The attitudes of the incompetent Monterey Shale exposed in the cliff between Bolinas and Duxbury Point seem to be essentially vertical (judging by the layers of concretions). Near Duxbury Point the beds dip generally about 45 degrees to the southwest.

At Duxbury Point some subsurface structural complexity at depth is indicated by east-west striking beds which dip southward and by oil and gas seeps in the vicinity.

Numerous other minor folds exist, particularly in the vicinity of U—Ranch and Home Ranch. The long sinuous folds shown by Weaver (1949) are not confirmed by the present mapping, with the exception of his "Country Club anticline," which causes the small inlier of granitic basement in Bear Valley. The incompetent Monterey Shale is often so contorted, possibly by sliding down the continental slope during deposition, that small exposures are unreliable indicators of structure.

FAULTS

A textbook example of a nearly vertical fault can be seen in the cliff section due west of the Kehoe Ranch (photo 27). This reverse fault, apparently related to the uplift of Tomales Point, strikes east-west and has displaced Laird Sandstone against granitic basement. Other minor faults, too small to depict on plate 1, extend east-west across the point here, the largest lying completely in granitic rock.

Photo 27. Reverse fault in the sea cliff near Kehoe Ranch. Laird Sandstone has been displaced against granitic basement rock.

Weaver (1949) shows a northwest-southeast fault immediately south of White Gulch on Tomales Bay near its northern end. Laird Sandstone is in place on the northeast side of this fault, although eroded from most of the surrounding granitic rock. It seems very likely that such a fault is present here, but since no evidence of it was seen, other than the topography and the presence of the Laird Sandstone, the fault is not shown on plate 1. Both the topography and the presence of the Laird Sandstone could be due to an originally irregular surface of the basement rock.

The anomalous east-west alignment of the Point Reyes Ridge, with its nearly straight south-facing cliff, can best be explained by postulating the existence of an east-west fault, down-thrown to the south, under the sea just off shore. If this is correct, the Point Reyes Ridge south face is a fault-line scarp. The gravity field seems to confirm this (see "Geophysical Surveys").

The Point Reyes Ridge is broken by a number of small faults in a complex pattern. More faults exist here than are shown on the map. Some of them seem to be pre-Drakes Bay Formation in age.

Numerous small faults can be seen in the Drakes Bay Formation that forms the white cliffs of Drakes Bay; these are insignificant in displacement and are not shown on the geologic map.