University of Washington Publications in Geology The Geology of the San Juan Islands

HARO FORMATION

Principal Features. The peninsula known as Davidson Head, located at the northern extremity of San Juan Island, is composed of conglomerate, shale, slate, sandstone, graywacke, grit, and limestone of upper Triassic age. These rocks, which occupy an area of only 48 acres, make up what is here called the Haro formation. So far as it is definitely known, no other rock outcrops of this age occur in the San Juan Island region.

Lithology and Structure. The lowermost strata exposed on Davidson Head are made up chiefly of conglomerate with occasional thin interbeds of reddish colored sandstone and shale. The sandstone layers pinch out along the strike and conglomerates take their place. The shales frequently show spheroidal weathering. The conglomerates generally have a coarse greenish-colored matrix, although in places the matrix is calcareous and stained with ferric oxide. In many cases the boulders of the conglomerate are not well sorted, for large subangular fragments occur along with smaller pebbles of various sizes. The boulders are composed of fragments of the Orcas chert and Leech River graywacke, granite, dacite porphyry with virtreous phenocrysts of sanidine up to half an inch in diameter, dark greenish altered andesite, and fine-grained bluish-gray limestone. The thickness of the conglomerate member as exposed on Davidson Head is 920 feet.

The conglomerates are overlain by thin-bedded carbonaceous shale, slate, graywacke, grit, and limestone. The layers of limestone are interbedded with carbonaceous shale and the individual strata do not exceed four feet in thickness. The uppermost strata of the Haro formation are largely concealed by glacial drift, but the location of the fault line between the rocks of the Haro formation and the Orcas cherts is evident from the surface features.

The rocks occurring on Davidson Head are somewhat shattered and dislocated so that the dip and strike vary considerably from place to place. However, the average strike is nearly east and west and the dip is moderately steep towards the south. The sediments belonging to the Haro formation have not been intruded by igneous rocks within the region in which these rocks are exposed. The total thickness of the Haro formation occurring in the vicinity of Davidson Head is 1250 feet. Aside from a moderate amount of induration these rocks have suffered from a relatively slight amount of metamorphism.

Age and Correlation. Nearly all of the strata which overlie the conglomerates are abundantly fossiliferous, but the fossils occur only as impressions. All of the fossils collected belong to the genus Halobia, which is restricted in its occurrence to the upper Triassic.

The limestones are metamorphosed to such a degree that only the distorted outlines of Halobia can be distinguished. The fossils are best preserved in the carbonaceous shales and slates which occur just below the limestones. In many cases the shells have been replaced by pyrite and marcasite.

The conglomerates occurring on Upright Head, Humphreys Head, north end of Blakeley Island, Decatur Island, James Island, southeast portion of Orcas Island, Obstruction Island, Peapod Rocks, and Sinclair Island, all of which being located in the upper part of the Leech River group, contain dacite porphyry and andesites along with fragments of chert and graywacke. On first examination the writer considered these rocks to be Jurassic or even of later age, because it was supposed that the andesites and dacite porphyry belonged to the Vancouver volcanics. The conglomerates of the Haro formation also contain similar andesites and dacite porphyry. Dikes of similar material actually cut the Orcas cherts on Turtleback Mountain Range, but their relationship to the other igneous intrusions could not be determined. Although the rocks at the above mentioned localities have been considered with the Leech River group, it is possible that they should be correlated with the Haro formation.

LOWER CRETACEOUS SYSTEM OR SERIES

SPIEDEN FORMATION

Principal Features. Isolated outcrops of conglomerate, sandstone, and shale, called here the Spieden formation, occur in the northwest portion of the map-area. The outcrops of this formation are apparently confined to Spieden Island, Sentinel Island, and Sentinel Rock. The latter two islands are composed chiefly of conglomerate, while Spieden Island is also composed of conglomerate except along its northern margin. The rocks of the Spieden formation are evidently separated from those of the Nanaimo series to the north ward by a normal fault trending in an east and west direction. Spieden Island is located on the upthrow side of this fault. Apparently there is another normal fault which follows the bed of the channel to the south of Spieden Island. In this case Spieden Island is located on the down-throw side of the fault.

Lithology and Structure. The Spieden formation is composed of conglomerate, breccia, sandstone, shale, and argillaceous limestone. Within the map-area the conglomerates make up about 85 per cent of the rocks. They contain fragments up to a foot in diameter although they are usually not larger than one inch. The fragments are composed of andesite, diorite, granodiorite porphyry, milky quartz, jasper, graywacke, chert, argillite, and limestone. Boulders of granodiorite occur very sparingly and these are generally fine grained. A calcareous sandstone forms the matrix in most cases, although in some beds a ferruginous matrix prevails. The conglomerates on Sentinel Island are identical in lithology with those on Spieden Island.

SPIEDEN ISLAND

Along the northern margin of Spieden Island there is a group of thin bedded and somewhat carbonaceous shales. These continue at least as far as low tide, but their thickness is unknown. Although they are soft and unmetamorphosed they have been badly contorted by folding. They grade upward into sandy shales and sandstones, with an occasional bed of argillaceous limestone. These beds are usually olive-gray in color, and the individual strata average six inches in thickness. Throughout a thickness of about 35 feet these beds are richly fossiliferous.

The fossiliferous layers are overlain by fine-grained conglomerate and breccia. The arenaceous or calcareous matrix of the conglomerate greatly predominates over the boulders. Here and there, thin and well stratified sandstones occur throughout the conglomerate beds, but the sandstones pinch out along the strike and do not form definite horizons. The conglomerates have a thickness exceeding 2,000 feet.

The conglomerate beds of the Spieden formation have an average strike of N 70°-75° W, and they dip to the southward at angles of 45-60 degrees. Along the north shore, at the foot of Spieden Bluff, the fossiliferous shales and sandstones strike N 65° W, and dip 65° SW.

The conglomerates of the Spieden formation differ from those of the Nanaimo series, in the scarcity of boulders of granodiorite and other batholithic rocks. The late Jurassic batholiths apparently had not been deroofed in this region by upper Knoxville time.

SENTINEL ISLAND

Sentinel Island is composed of conglomerates identical in lithology with those occurring on Spieden Island. The alternating calcareous and ferruginous matrix of the conglomerates is peculiar to the Spieden formation. At the north edge of Sentinel Island the beds strike N 80° W and dip 45° SW. At the south edge of the island the beds strike N 82° W and dip 55° S.W. The thickness of the strata outcropping on Sentinel Island is about 800 feet.

The outcrops occurring on Sentinel Rock are the equivalent of some of the strata on Sentinel Island.

Age and Correlation. Along the north shore of Spieden Island there is a belt of fossil-bearing sandstone and shale. The fauna was examined by Dr. T. W. Stanton, who determined its age to be lower Cretaceous and equivalent to the upper part of the Knoxville formation of California.

Aucella crassicollis Keyserling is by far the most abundant fossil and it makes up fully ninety-five per-cent of the fauna. The identical fossils are found on the Nooksak River, north of Mount Baker. One of these fossil beds is located in a road-cut about two miles east of the village of Glacier, in section 5, T 39 N, R 7 E.

The fossils collected on Spieden Island include the following;—

The Spieden formation is to be correlated with the Pasayten formation which outcrops in the Hozomeen Range and other localities in north central Washington.¹⁶

---

¹⁶Smith, G. O., and Calkins, F. C., A Geological Reconnaissance Across the Cascade Range near the Forty-ninth Parallel: U.S. Geol. Survey Bull. 235, pp. 28-30, 1904.

---

DESCRIPTION OF NEW SPECIES

Order Ammonoidea
Family SILESITIDAE

GENUS HOLCODISCUS UHLIG
HOLCODISCUS?? STANTONI n. sp.

Plate XII. Figs. 3, 4, and 5.

Shell compressed, convex, and narrowly umbilicated; the umbilicus somewhat rounded with steep inner wall, occupying about one-fifth of the entire diameter; volutions closely involute, the inner ones being almost covered by those which succeed them; aperture higher than wide, sub-elliptical, but deeply emarginate by the encroachment of the preceding volution.

Surface marked with numerous slightly elevated, flexuous, transverse ribs, which are always somewhat narrower than the shallow concave spaces between them; the ribs are always bifurcate, and usually one or both branches are again divided so that near their summits they are generally trifurcate or tetrafurcate. There are approximately 25 major or parent ribs to each volution. The major ribs, though less strongly developed, continue to the inner wall of the umbilicus.

Sutural line not well seen, there being only two immature specimens collected that show a sutural line.

Dimensions.

The largest specimen known to the writer, and preserved only as a cast, measured about 125 mm. in diameter. The largest specimen actually collected has the dimensions: maximum diameter, 65 mm.; maximum thickness, 17 mm.; diameter of the umbilicus, 12 mm.

Locality. The type specimen was found on the north shore of Spieden Island at the foot of Spieden Bluff. The paratype was found at the same locality.

The same species is found in a road-cut along the south bank of the Nooksak River, about two miles east of the village of Glacier, in section 5, T 39 N, R 7 E.

Disposal of Type. University of Washington Paleontological Collection.

H. stantoni differs from Holcodiscus cumshewaensis, which species it most closely resembles, in the size of the umbilicus. H. cumshewaensis, in the type specimen at least, is much more loosely coiled than H. stantoni.

Named in honor of Dr. T. W. Stanton of the U. S. National Museum at Washington.

Family PHYLLOCERATIDAE

GENUS PHYLLOCERAS SUESS
PHYLLOCERAS SPIEDENENSIS n. sp.

Plate XII. Figs. 1 and 2.

Shell smooth and moderately inflated; the umbilical margin rounded and indistinctly defined; volutions increase rapidly in size, are closely convolute, the inner ones being completely covered by each succeeding volution; aperture higher than wide, nearly circular in the type specimen, but more elliptical in some of the other specimens collected; aperture, emarginate because of the encroachment of the preceding volutions, is somewhat pointed in the direction of the umbilicus, the specimen being very thin at this point.

Sutural line.

Dimensions.

The thickness at the axis of the umbilicus approaches zero.

Locality. The type specimen was obtained on the north shore of Spieden Island at the foot of Spieden Bluff.

Disposal of Type. University of Washington Paleontological Collection.

P. spiedenensis, in the type specimen, is more inflated than is normal in the genus Phylloceras. In some of the specimens this is reduced more nearly to the normal inflation for the genus. The type specimen does not seem to be distorted, for several other specimens were collected possessing a similar shape.

PELECYPODA
Superfamily ANATINACEA Dall

GENUS PLEUROMYA AGASSIZ
PLEUROMYA THOR n. sp.

Plate XII. Figs. 8 and 9.

Shell compressed, being most convex near the anterior margin; valves closed in front but slightly open behind; anterior end short and sloping nearly in a line from the beaks to the ventral margin; posterior end much longer, and somewhat pointed at its junction with the ventral margin; cardinal margin excavated and gently curved; umbones large, broad, and prominent; beaks small, anterior, and curved forward and downward; umbonal ridges nearly obsolete.

Surface marked by irregular, moderately fine concentric striations. Hinge teeth and muscular impressions unknown.

Dimensions.

Locality. The type specimen was collected on Spieden Island, at the foot of Spieden Bluff.

Disposal of Type. University of Washington Paleontological Collection.

P. thor is distinguished from the typical Pleuromya subcompressus Meek, by the fact that its beaks are less prominent, its surface markings finer and more irregular, and its posterior end is more pointed.

PLEUROMYA TYPA n. sp.

Plate XII. Fig. 7

Shell moderately convex, rounded in outline, the height being but little less than the length; umbones broad and somewhat flattened; beaks small, elevated, slightly anterior, and curved forward and downward; valves closed in front, but apparently open behind; anterior end sloping rapidly to the ventral margin; posterior end rounded, though somewhat longer than the anterior end; anterior umbonal ridges well developed; posterior umbonal ridges obsolete; ventral margin convex and strongly curved.

Surface marked by deep, irregularly disposed, concentric striations. Hinge teeth and muscular impressions unknown.

Dimensions.

Locality. Collected on Spieden Island at the foot of Spieden Bluff. Poorly preserved casts were seen at the Nooksak River locality, two miles east of the village of Glacier.

Disposal of Type. University of Washington Paleontological Collection.

The ventral margin of P. typa is much more strongly curved than that of any other Pleuromya known to the writer.

Superfamily PECTINACEA Reeve

GENUS LIMA BRUG
LIMA SPIEDENENSIS n. sp.

Plate XII. Fig. 6.

Shell small, moderately convex, obliquely subovate, posterior side produced below; beaks sharply incurved; ears small; surface markings consisting of twelve narrow, radiating ribs; spaces between the ribs smooth and much wider than the ribs themselves; characters of the interiors of the valves unknown.

Dimensions. A single fragmental specimen has the following approximate dimensions,—length, 12+mm.; height, 15+mm.; thickness, 7 mm.

Locality. Collected on Spieden Island, at the foot of Spieden Bluff.

Disposal of Type. University of Washington Paleontological Collection.

L. spiedenensis differs from Lima suciensis in the fact that it contains no minor radiating ribs, and by the lack of concentric lines of growth.

UPPER CRETACEOUS SYSTEM OR SERIES

NANAIMO SERIES

Principal Features. The rocks of the Nanaimo series^{17, 18, 19} outcrop on the small islands which fringe the northern margin of the San Juan Island group. The rocks are composed of unmetamorphosed conglomerates, grits, arkosic sandstones, and shales. They appear along the north shore of Orcas Island, the outcrops beginning at the base of Buck Mountain and extending westward around Point Doughty and continuing southward as far as Point Kimple. Rocks belonging to the Nanaimo series occur on Stuart Island, Satellite Island, Johns Island, Ripple Island, Cactus Islands, Flattop Island, Gull Rock, White Rocks, Waldron Island, Bare Island, Skipjack Island, Parker Reef, Patos Islands, Sucia Islands, Clements Reef, Matia Islands, and on the Barnes and Clark groups of islands.

---

¹⁷Richardson, James, Report on the Coal Fields of Nanaimo, Comox, Cowichan, Burrard Inlet, and Sooke, British Columbia: Geol. Survey Canada, Report of Progress, 1876-77, pp. 160-192, 1878.

¹⁸Dawson, G. M., The Nanaimo Group: Amer. Jour Sci., vol. 39, pp. 180-183, 1890.

¹⁹Clapp, C. H., Geology of the Nanaimo Map-Area: Geol. Survey Canada, Mem. 51, pp. 1-127, 1914.

---

The outcrops of the upper Cretaceous rocks are usually small and isolated and the geological record is fragmental. These rocks at one time covered the whole map-area with the possible exception of the extreme eastern or southeastern portion. The sediments belonging to the Nanaimo series in this region have not been intruded by any igneous rocks.

On the basis of determinations made on the fossil beds occurring on Sucia Waldron, and Skipjack islands, as well as on Vancouver Island and vicinity, the rocks of the Nanaimo series have been placed in the upper Cretaceous and are essentially equivalent to the Chico Cretaceous of California.

PLATE XV. Above: Upper Cretaceous rocks along the north shore of Orcas Island. Below: Limestone ledge on the shore of East Sound, at the foot of Mount Entrance.

Lithology and Structure. In the type locality the Nanaimo series has been divided into a number of formations mainly on the basis of their lithology^{20, 21} The series is composed entirely of conglomerate, grit, arkosic sandstone, shale, and coal. Though some of the formations are fossiliferous they are noticeably lacking in true limestones.

---

²⁰Clapp, C. H., Geology of the Nanaimo Map-Area: Geol. Survey Canada, Mem. 51, pp. 44-80, 1914.

²¹Clapp, C. H., Sooke and Duncan Map-Areas; Vancouver Island: Geol. Survey Canada, Mem. 96, pp. 224-227, 1917.

---

BARNES AND CLARK ISLANDS

The rocks on the Barnes and Clark group are chiefly coarse conglomerates but they contain interbeds of sandstone and some shale. The pebbles of the conglomerate are composed of andesite, granodiorite, diorite, chert, argillite, graywacke, and specimens of all of the known older formations. The typical matrix is a grayish-brown arkosic sandstone.

The rocks on Barnes Island strike about N 12° E and dip 80° SE. On the north end of Clark Island the rocks strike N 8° W and dip 50° SW.

The southeast end of Clark Island shows a strike of N 30° E, and a dip of 30° SE. The islands and reefs making up the Sisters group, have a similar trend and dip.

Barnes and Clark Islands have been formed by the two limbs of the same syncline and evidently they belong to the same horizon. The north part of Clark Island also forms the limb of an anticline, with the south end and the Sisters group forming the other limb. These folds plunge to the southward at a moderate angle.

Figure 3. Outline map of the Barnes and Clark group of islands.

MATIA ISLANDS

Matia Islands are formed by a fragmental portion of a monoclinal fold which has a persistent strike of N 67° W, and a dip of 68° NE. The islands are composed of three parallel resistant formations separated by two less resistant ones. The latter are formed largely from shale and sandstone. The resistant formations are composed of medium-sized conglomerate and coarse buff-colored arkosic sandstone, with occasional scattered pebbles. In places there are irregular patches of coarse conglomerate. A generalized section across Matia Islands shows:

The sandstones are usually crossbedded and are always arkosic. Occasionally there are fragments of Cretaceous trees which are now turned to coal or partly silicified. The sandstones and conglomerates are evidently delta deposits.

SUCIA ISLANDS

Sucia Islands are formed by the more resistant strata of a plunging syncline, the less resistant ones being covered by sea water. The syncline plunges to the eastward and the individual strata consequently outcrop in the form of a horseshoe, with the open side toward the east.

At the extreme southern edge of Sucia Islands there is a coarse conglomerate with fragments composed almost entirely of the Leech River schists, and fragments of the milky quartz veins that commonly cut these schists. The bluish-gray pulverized schists serve as the matrix of the conglomerate. The larger boulders are composed chiefly of white milky quartz and they stand out in strong contrast with the bluish-colored matrix. Because the conglomerate contains fairly large angular fragments of the fragile schist, and because it contains little besides the fragments of the Leech River schists, the source of the material must have been close at hand. In all probability the conglomerate is the basal member of the Nanaimo series at this locality.

The conglomerate grades upward into a bluish sandy shale composed of pulverized fragments of the schists. This in turn is overlain by light to dark gray sandy shales. The shales are fossiliferous and they contain calcareous and concretionary beds of three inches or so in thickness, at intervals of about 15 feet. The shale appears to be identical with the Haslam formation on Vancouver Island.^{22, 23}

---

²²Clapp, C. H., Geology of the Nanaimo Map-Area: Geol. Survey Canada, Mem. 51, pp. 53-56, 1914.

²³Clapp, C. H., Sooke and Duncan Map-Areas, Vancouver Island: Geol. Survey Canada, Mem. 96, pp. 224-227, 1917.

---

The shale is overlain by a coarse to medium-grained buff-colored sand stone. The sandstone is even-grained and possesses a parting normal to the bedding-plane. For this reason it has been used for the manufacture of paving blocks.

Above the sandstone there is a less resistant formation, presumably a shale, which is entirely covered by soil or by tide-water.

The generalized section as exposed on Sucia Islands, is as follows:

By differential chemical action of the salt water on the sandstones of the Nanaimo series curious erosion surfaces have resulted. Sometimes hollow caverns have been produced, and more commonly the whole surface resembles a honeycomb. (See Plate XIV,B).

The northern limb of the synclinal fold on Sucia Island is also the southern limb of an anticline, with Clements Reef representing the northern dip.

PATOS ISLANDS

Patos Islands are composed entirely of cross-bedded sandstone and conglomerate. The individual strata pinch out rapidly along the strike and no division of the rocks into formations is possible. These cross-bedded sediments were laid down as delta deposits, and excellent examples of top-set, fore-set, and bottom-set beds are now exposed. From the nature of the sediments accurate measurements of the strike and dip are not possible. However, the strike follows parallel to the south shore-line, and the beds always dip to the northward at angles ranging from 45 to 65 degrees. The thickness of the rocks exposed on Patos Islands is about 1450 feet.

PLATE XVI. Above: Point Disney, Waldron Island, showing the immense conglomerate bed overlain by fossiliferous shaly sandstone. Below: The Point Disney conglomerate.

ORCAS ISLAND

The outcrops of the upper Cretaceous rocks occurring along the north and northwest shores of Orcas Island are composed chiefly of thin alternating beds of sandy shale and shaly sandstone. These beds change in lithology rapidly along the trike. The sandstones are generally well cemented, and like all of the rocks of the Nanaimo series in this map-area, they are always arkosic. They contain fragments of volcanic rocks, together with relatively undecomposed fragments of acid plutonic rocks. Quartz, in some instances, is only a subordinate constituent. Silica is the usual cementing material.

The massive conglomerate that forms Point Doughty, besides containing fragments of all of the older rocks exposed on the San Juan Islands, includes boulders of coarse basic plutonic rocks of several varieties not seen in this region. To the southward, and overlying the Point Doughty conglomerate, there are several thinner beds of conglomerate, sandstone, and lignitic shale. The latter are rich in fossil leaf impressions.

Where the beds are not crumpled by thrusting, the rocks of the Nanaimo series on Orcas Island have a persistent strike of N 65° W, and they invariably dip to the southward. It is not probable, however, that the outcrops all belong to a single monoclinal fold, for the section includes several unexposed horizons of great thickness, and the adjoining areas to the west are faulted and broken.

The section, starting from Point Thompson and proceeding south-westward, is as follows:

The lignitic shale containing the fossil plants is crumpled and broken.

Still farther southward the rocks are concealed for a distance of half a mile. At this point a coarse conglomerate bed about 35 feet thick outcrops at the water's edge. The conglomerate bed which forms the greater part of Freeman Island evidently belongs to the same horizon. The sandstone and shale strata on Freeman Island strike N 55° W, dip 65°-75° SW, and overlie the conglomerate. The corresponding strata on Orcas Island are crumpled and broken.

Point Kimple, which is located about three-quarters of a mile south of Freeman Island, is composed of medium-textured conglomerate with thin interbeds of sandstone and shale. There are five horizons of this conglomerate and their general strike is N 80° W. They dip to the southward at angles of 20-30 degrees.

The rocks of the Nanaimo series form a submarine shelf or platform that extends from Orcas Island to Parker Reef. The sandstones and sandy shales which form Parker Reef strike N 65° W and dip to the northward at an angle of 60 degrees. Apparently they represent part of the northern limb of the anticlinal fold, the southern limb of which is exposed on the north shore of Orcas Island.

WALDRON ISLAND

The upper Cretaceous rock exposures on Waldron Island are largely confined to the higher southeast side, and to scattered points along the northern margin. The elevated region extending northeastward from Point Disney is composed of bluish-gray sandstone, coarse conglomerate, and fossiliferous shaly sandstone. The boulders of the conglomerate frequently attain a diameter of several feet. Altered andesite, granodiorite, and chert are the most abundant constituents of the conglomerate boulders, although the older rocks of the region are well represented. In addition, the conglomerate contains many boulders of coarse-textured basic plutonic rocks, and also nephelite and cancrinite syenites, all of which are foreign to this locality.

The rocks composing the southeast side of Waldron Island are folded into a basin-shaped fold. The structure has been complicated by the fact that the formations have slipped on each other with a sort of rotational motion.

Figure 4. Outline map of Waldron and Skipjack Islands.

Scattered outcrops occur along the north shore of Waldron Island, but they cannot be followed inland because of the thick covering of glacial drift. These outcrops are composed of buff-colored sandstone with a minor amount of conglomerate. The sandstones contain abundant remains of fossil Ostrea.

There is an anticlinal fold between Waldron and Bare islands, while between Point Hammond and Fishery Point, there is a synclinal fold. The average strike is about N 65° W. On the east side of the island, immediately north of the elevated region, there appears to be an anticline with an axis that would intersect the opposite side of the island to the south of Fishery Point. Just how the fold was produced in the elevated portion of the island is not clear.

Several years ago a hole was drilled in the east central part of Waldron Island, to a depth of nearly 1500 feet. The writer has not been able to secure the log of the drill-hole, but the greater part of the core is still on the island. The location of the hole was unfortunate, for the drill penetrated through several hundred feet of conglomerate and sandstone the section of which was already well exposed along the shore-line.

BARE ISLAND

Bare Island is composed of alternating beds of conglomerate, grit and buff-colored sandstone. The rocks are cross-bedded, with a strike of N 70°-80° W and dip about 80° NE. Fossil Ostrea are found on this island.

SKIPJACK ISLAND

Skipjack Island is composed of alternating beds of coarse and fine conglomerate, grit, and shaly sandstone. The sandstone is relatively soft, and it is eroded with sufficient rapidity to form embayments with parallel sides between the conglomerate strata. The section on Skipjack Island is about 500 feet thick. The beds strike from N 85° E to nearly east and west, and dip to the northward at an angle of about 70 degrees.

WHITE ROCKS

White Rocks, which are located about a mile to the south of Waldron Island, are composed of grit and conglomerate with some interbeds of coarse sandstone. The beds strike N 30° W and dip 41° NE.

GULL ROCK

Gull Rock is composed of coarse conglomerate with an interbed of less resistant sandstone, the latter being largely eroded away. The beds strike N 65° E. and dip 65° SE.

FLATTOP ISLAND

Flattop Island is composed of about 250 feet of coarse conglomerate, which is underlain by at least 35 feet of thin-bedded dark-gray shale and shaly sandstone. The beds strike N 65°-70° E and dip 25°-26° SE.

RIPPLE ISLAND

The formations exposed on Ripple Island are evidently equivalent to those occurring across the channel on Johns Island. A formation consisting of thick and thin-bedded buff-colored sandstone is overlain by about 125 feet of conglomerate. This in turn is followed by thin-bedded shale and sandstone. The beds strike N 80° W and dip 45° SW.

PLATE XVII. Figs. 1, 2, and 3. Condonella suciensis n.sp. Type specimen. Figs. 4, 5, & 6. Type specimen. Figs. 7 & 8. Cucullaea suciensis n.sp. Type specimen. Fig. 9. Fusulina sp. Enlarged specimen from the limestones of the Leech River group, Orcas Island.

CACTUS ISLANDS

Cactus Islands are composed of conglomerate, cross-bedded sandstone, and shale. The generalised section exposed on East Cactus Island is as follows:

At the east end of Cactus Islands the average strike of the rocks is N 73° W and the dip is 55°-63° SW. At the west end of the group the average strike is about N 65° W and the dip is 60°-66° SW.

JOHNS ISLAND

The generalised section exposed on Johns Island is as follows:

At the east end of Johns Island the beds strike N 65° W and dip about 35° SW. At the west end of the island the beds have an average strike of N 60° W and a dip of 50°-55° SW.

STUART ISLAND

The rocks on Stuart Island have been closely folded into an anticline and a syncline whose general trend is N 70° W (See Fig 4.) About 1500 feet of conglomerate with irregular patches of cross-bedded sandstone are exposed on the north limb of the anticline. Underlying the conglomerate there is a great thickness of alternating layers of dark carbonaceous shale and light gray sandstone. The shale strata usually have a thickness of one to six inches, while the sandstone layers are generally somewhat thicker.

The south limb of the anticline, which is at the same time the north limb of the syncline, is represented most prominently by the conglomerate formation mentioned above. This conglomerate forms a ridge that extends from one end of the island to the other, and it connects the two main parts of the island. The same conglomerate formation expresses itself on the south limb of the syncline, where it forms Tiptop Mountain, the highest elevation on the island. Reid Harbor is located in the axis of an elongated structural basin. Along the shore to the southwest of Tiptop Mountain, the rocks are compressed into chevron folds.

Figure 5. Outline map of Stuart, Satellite, and Johns Island.

SATELLITE ISLAND

Satellite Island is composed of the same formations as those outcropping on Stuart Island. The anticline occurring on Stuart Island extends across Prevost Harbor and embraces the rocks on Satellite Island. The rocks have an average strike of N 80° W.

Age and Correlation. The lignitic shales occurring south of Point Doughty on Orcas Island have been described by Newberry.²⁴ He considered the fossil plants to be identical with those occurring in the sandstones in the vicinity of Bellingham, and all of these formations were referred to the Cretaceous. Some of these lignitic beds contain a large thick-shelled species of Ostrea, which also occurs on Waldron Island in association with known marine upper Cretaceous fossils.

---

²⁴Newberry, J. S., Description of the Fossil Plants Collected by George Gibbs, Geologist to the United States Northwest Boundary Commission under A. Campbell: Jour. Boston Soc. Nat. Hist., vol. 7, pp. 506-525, 1863.

---

The best known fossil beds of this region are those occurring on the Sucia Islands. Species from this locality were described and figured by Meek²⁵ in 1876, by White²⁶ in 1884, and by Whiteaves²⁷ in five volumes, 1876-1903.

---

²⁵Meek, F. B., Descriptions and Illustrations of Fossils from Vancouver and Sucia Islands, and other Northwestern Localities: U. S. Geol. and Geogr. Survey of the Territories, Bull., vol. 2, pp. 351-376, 1876.

²⁶White, C. A., Cretaceous Fossils from Vancouver Island Region: U. S. Geol. Survey, Bull., pp. 33-48, 1884.

²⁷Whiteaves, J. F., Geol. Survey Canada, Mesozoic Fossils, vol. 1-5, 1876, 1903.

---

The fossiliferous formation on the Sucia Islands is 700 feet thick and the fauna contains a great variety of species. The formation, as a whole, may be characterized by the abundance of the following species:

Juoceramus vancouverensis Shumard
Cinula obliqua Gabb
Baculites chicoensis Trask
Crassatellites conradiana Gabb
Trigonia evansana Meek
Margarita ornatissima Gabb
Cyprimeria lens Gabb
Glycimeris suciensis n. sp.

The fossiliferous horizon on Skipjack Island is not more than a foot thick. It is especially characterized by the following:

Perna excavata White
Trigonia evansana Meek
Cinula obliqua Gabb
Glycimeris suciensis n. sp.

At Point Hammond on Waldron Island, a large thick-shelled species of Ostrea occurs in association with Trigonia evansana. The same species of Ostrea occurs near Fishery Point on Waldron Island, on Bare Island, and at various points on Orcas Island. The richest fossil-bearing horizon on Waldron Island is located just above the Point Disney conglomerate. Although a large variety of species is present, it is characterized by Cucullaea ponderosa Whiteaves, Cucullaea truncata Gabb, Glycimeris suciensis n.sp, Cinula obliqua Gabb, and Trigonia evansana Meek.

DESCRIPTION OF NEW GENUS AND NEW SPECIES

GASTROPODA

GENUS CONDONELLA, NEW GENUS

Shell small, discoidal, each whorl being coiled upon the preceding one; umbilicus broadly conical, converging towards the posterior side; anterior side convex; posterior side flat or somewhat concave; each whorl slightly and obliquely emarginate upon the preceding one; aperture sub-ovate to crescent-shaped.

Named in honor of Herbert T. Condon, comptroller of the University of Washington.

CONDONELLA SUCIENSIS n. sp.

Plate XVII. Figs, 1, 2, 3.

Shell consisting of six whorls, increasing rather slowly in size; test thin, sub-nacreous, and crossed obliquely by numerous fine transverse lines.

Dimensions.

Locality. The only specimen known was collected on Sucia Island at a point about 300 feet above the base of the fossil-bearing shales.

Disposal of Type. Paleontological collection of the U. S. National Museum, Washington, D. C.

The specimen was examined by Dr. Stanton and Dr. Dall, who said that it resembled the fresh-water genus, Planorbis, more than any marine form they had ever seen. Since the specimen was found in strata that were rich in strictly marine fossils it cannot be considered as a variety of that genus.

Superfamily ARCACEA Deshayes

GENUS GLYCIMERIS DACOSTA
GLYCIMERIS SUCIENSIS n. sp.

Plate XVII. Figs, 4, 5, and 6.

Shell small, moderately compressed convex, equilateral, and almost round in outline; beaks small, nearly central and incurved, projecting but little above the superior border; outer surface marked by a close, regular net-work of radiating and concentric raised lines.

Dimensions.

Locality. The type specimen was collected on the Sucia Islands where it is fairly abundant. It is also found on Skipjack and Waldron Islands.

Disposal of Type. University of Washington Paleontological Collection.

G. suciensis was classified by Whiteaves as Glycimeris veatchii. The writer examined a large number of specimens from the Sucia Islands, Skipjack Island, and Waldron Island, and none were seen that exceeded the type C. suciensis in size. The type G. veatchii is much higher in proportion to its length, than any of the specimens observed.

GENUS CUCULLAEA LAM.
CUCULLAEA SUCIENSIS n. sp.

Plate XVII. Figs. 7 and 8.

Shell moderately large, ventricose, rounded and equilateral; anterior and posterior ends rounded without any pronounced shoulder; beaks prominent, broad, fairly close together, curved inward and a little forward, placed nearly central; cardinal area moderately large, broad, and marked with well defined, divergent, ligamentary grooves; shell thick, but tapering rapidly at the ventral margin; surface marked by moderately coarse, well defined radiating lines and by irregular and less strongly defined coarse concentric lines.

Dimensions.

Locality. Collected on the Sucia Islands at the south margin of Fossil Bay.

Disposal of Type. University of Washington Paleontological Collection.

C. suciensis is distinguished from C. ponderosa, and from C. fruncata, by its rounded, equilateral shape, and by the character of its surface markings.

EOCENE SYSTEM OR SERIES

CHUCKANUT FORMATION

Principal Features. The coal-bearing sandstones and conglomerates occurring on the northern part of Lummi Island and in the vicinity of Bellingham Bay to the eastward are here referred to as the Chuckanut formation. These rocks form the lower part of White's Puget group.²⁸

---

²⁸White, C. A., On the Puget Group of Washington: Amer. Jour. Sci., 3rd ser., pp. 443-450, 1888.

---

The sandstones of the Chuckanut formation are generally cross-bedded and somewhat arkosic. They are usually cemented less firmly than the sandstones belonging to the Nanaimo series and they contain more interbedded lignitic material. As a general rule the Chuckanut sandstones are lighter in color than those of the Nanaimo series. The conglomerate strata occurring in the Eocene and Upper Cretaceous formations appear to be identical with regard to the nature of the materials composing their boulders and pebbles.

The Chuckanut formation was evidently laid down in brackish or even fresh water, and fossil leaves and plants are very abundant in many of the strata.

Lithology and Structure. The rocks of the Chuckanut formation exposed on Lummi Island consist of cross-bedded and poorly consolidated arkosic sandstones and conglomerates. The conglomerates are indistinguishable from those occurring in the Nanaimo series. The sandstones are composed of undecomposed fragments of granitoid and volcanic rocks mixed with fragments of chert and argillite. Quartz is frequently a subordinate constituent only, although in some horizons it is very abundant. Lignitic material derived from fossil palm trees is found interbedded with and scattered through the sandstone strata.

On Lummi Island the rocks belonging to the Chuckanut formation trend northwesterly. They occupy the bottom of a distorted syncline, the greater part of which has been eroded away. The Chuckanut sediments have been laid down upon the eroded surfaces of the Eagle Cliff porphyrites.

PLATE XVIII. Above: The sandspit and lagoon at Argyle, San Juan Island. Below: The Upper Cretaceous rocks at Point Thompson, Orcas Island.

Age and Correlation. The sedimentary strata occurring on the northern part of Lummi Island are identical in lithology and plant remains with those outcropping on the mainland to the eastward. The rocks of the Chuckanut formation are well exposed along the Chuckanut Drive on the Pacific Highway.

It is interesting to note that the rocks of this district contain several plant species which are also found in the upper Cretaceous rocks to the south of Point Doughty on Orcas Island. Newberry²⁹ compared the fossil plants from both localities and decided that they were identical and of upper Cretaceous age.

---

²⁹Newberry, J. S., Description of the Fossil Plants Collected by George Gibbs, Geologist to the United States Northwest Boundary Commission under A. Campbell: Jour. Boston Soc. Nat. Hist., vol. 7, pp. 506-525, 1863.

---

Recent work on these fossil plants by Dr. Knowlton has placed the rock formations in the vicinity of Bellingham Bay in the lower Eocene. According to Dr. Knowlton the beds do not represent the lowermost portion of the Eocene and their flora is much different from that occurring on Orcas Island. The writer found that the fossil leaves occurring on Orcas Island are interbedded with strata which contain a marine upper Cretaceous fauna.

The Chuckanut flora is very different from the middle or upper Eocene flora occurring in the vicinity of Seattle.

QUATERNARY SYSTEM

In many places the islands of the San Juan group are covered with a deep mantle of glacial till and sediments. The glacial geology of Puget Sound and vicinity has been discussed at some length by Bretz³⁰ and a bibliography of previous writers on the subject may be found in his report.³¹

---

³⁰Bretz, J. Harlen, Glaciation of the Puget Sound Region: Wash. Geol. Survey Bull., 8, 1913.

³¹Ibid., pp. 10-12.

---

Although at least four major periods of glaciation are recognized in many parts of North America, but two have been established as occurring in the Puget Sound region. The sediments and till derived from these glacial invasions have been grouped as follows:

(1) The Colwood sediments, deposited since the retreat of the last glacier.

(2) The Vashon till and sediments, deposited by the last glacier.

(3) The Puyallup sediments, deposited during the interval of time between the Admiralty and Vashon glacial periods.

(4) The Admiralty till and sediments, deposited during the next preceding glacial period or epoch.

A number of small remnants of strongly indurated tillite occur on Burrows Island and also on Allan Island. These remnants of tillite do not occur in contact with the more recent glacial till or sediments. They are found only on the south or protected slopes of the hills, and they occur as pockets on the underlying rocks of the Fidalgo formation. The fragments of the tillite are as strongly cemented as those of the conglomerates occurring in the Nanaimo series. The cementing material or matrix of the tillite is light gray in color and somewhat calcareous. Whether these remnants of tillite belong to the Admiralty glacial period, or to an earlier glacial period, is not known.

The glaciers performed a large amount of erosion in the general region of the San Juan Islands, overriding the highest mountain tops and greatly modifying the earlier topography. The glacial erosion was noticeably more intense at the southern ends of the islands, while the northern ends are often covered with glacial drift. The great bulk of these sediments was deposited during the Puyallup interglacial epoch, and they are usually covered with a thin layer of Vashon till. In places the Vashon till and sediments contain huge erratic boulders.

The direction followed by the upper portions of the glacial ice in this region was not controlled by the underlying topography. The courses of the deepest glacial erosion were partly determined by the presence of fault or fracture zones or by previously existing valleys or channels. Near sea-level the glacial striations usually follow parallel to the courses of the present water channels, while at higher elevations the striations trend nearly north and south. In the northwest portion of the map-area the course of the glacier was slightly east of south. In the southwest portion the striations trend almost exactly north and south, and do not give any indication that part of the ice sheet turned southwestward through the Strait of Juan de Fuca.

Glacial grooves with a depth of several feet are encountered in many parts of the region. Perhaps the most remarkable display of glacial grooving is found on the rocky slopes to the east of Iceberg Point on Lopez Island. Here the glacial striations trend almost at right angles to the strike of the upturned strata of the Leech River group.

On the southeast portion of Orcas Island, to the north of Obstruction Pass, the rocky hills are drumlinoidal in shape and polished by glacial action.

On San Juan Island Bretz recognized two localities, Cattle Point Hill and Bald Hill, as the best examples of recessional moraines occurring in the Puget Sound region.

The sediments occurring on Waldron Island, Lopez Island, Guemes Island, Sinclair Island, Portage Island, and Decatur Island belong largely to the Puyallup inter-glacial epoch. In most cases these sediments are overlain by a layer of till deposited as recessional morainal material by the Vashon glacier.

Since the last glacial period there has been a general uplift throughout all of the area previously covered by the ice. At several localities high above the present sea-level on Waldron Island, the Sucia islands, San Juan Island, Lopez Island and Orcas Island, there are abundant marine fossils. Of these, Pecten hastatus, Cardium corbis, Paphia staminia, and Saxidomus giganteous are the most abundant. At one locality near Deer Harbor Bretz found marine shells in a well at an elevation of 290 feet.

The San Juan Islands exhibit abundant and excellent examples of recently upraised beaches. These usually occur at elevations of 15 to 25 feet above present high tide.

Figure 6. The Argyle Lagoon and sandspit.

At the present time, some interesting sandspits and lagoon lakes are being formed in the San Juan Islands. Among these, the Argyle Sand Spit and Lagoon on San Juan Island, and Fisherman Bay on Lopez Island, are the most interesting.

Figures 7, 8, 9. Three stages in the evolution of the Argyle Lagoon. The arrows show the direction of the prevailing currents.

The Argyle sand spit would have closed the entrance to the bay had it not been for the presence of Little Island which is composed of solid rock. Practically all of the materials composing the sand spit and sandy hook were derived from Bald Hill.

Figure 10 The lagoon and sandspit at the north end of Blakeley Island.

At the north end of Blakeley Island a small rock mass has been tied to the main island by means of sand spits. A similar physiographic feature is encountered on the south shore of Orcas Island about three-quarters of a mile to the east of the village of Orcas.

Double sand spits converging to a point and containing a shallow lagoon between them, are of very common occurrence in the San Juan Island area. In many cases the lagoons have been filled up with sediment so that they no longer contain any water.

At several places along the shores of the San Juan Islands, clam beds uplifted above the present high tide-level are seen to grade downward into beds that now contain living clams of the same species. Care must be used that such deposits are not confused with the so-called "kitchen middens," or shells left by the Indians who formerly camped along the shores.

At the eastern margin of the map-area the deltas of the Samish and Nooksak rivers are at the present time encroaching on some of the San Juan Islands. The Samish delta has already encroached to such an extent that shallow water and luxuriant growths of eel-grass extend almost as far west as Guemes Island. Nooksak River, with its distributary, Lummi River, are supplying so much sediment that a submerged sand bar connecting Lummi Island with the mainland is now in the process of formation.