Chapter V: Introduction (2)

Northwest African Sector.--The continental margin from northwest Morocco to Dakar is remarkably uniform and rather closely resembles the northeastern United States sector. The continental shelf and slope are well developed (Fig. 19). The shelf is 15 to 70 miles in width and thus is somewhat narrower than either the North American shelf or the Anglo-French shelf. The shelf break ranges from 50 to 80 fathoms. The continental-slope gradients range from 1:15 to 1:40 and are thus somewhat less steep than in the American sectors. Prominent benches are common at 300, 600, 850, 1200, and 1600 fathoms. The continental rise is well developed and is compound. The main contrast between the North African and American sectors is the greater width of the African continental rise. Off northeastern United States a line of isolated volcanic peaks cuts across the continental rise and abyssal plain. In the North African continental margin volcanic peaks are larger, more numerous, and lie in coalescing lines or along ridges. The Cape Verde and Canary groups lie in the continental rise near the outer edge of the upper continental rise. All provinces except the continental shelf widen from Gibraltar southward toward Cape Verde.

On profile E-11 off Casablanca the distance from the shelf break to the lower continental rise is only 50 miles as compared with a similar measurement of 500 miles at Cape Verde. Off Casablanca the continental slope extends to 1400 fathoms where the gradient drops to less than 1:40 from 1:10-1:20 on the continental slope. The upper continental rise which widens to more than 100 miles farther south is only poorly developed off Morocco. No other deep-sea echo-sounding profiles are available for the Moroccan continental margin. Surveys of the continental slope made by the French Hydrographic Service during the past few years will, when published in full, undoubtedly provide much valuable information on the topographic benches in this important area (Grousson, 1957).

HEEZEN _et al._, PL. 8

FIGURE 1. CONTINENTAL RISE WEST OF ST. NAZAIRE, FRANCE

FIGURE 2. BISCAY ABYSSAL PLAIN. NOTE SMALL MID-OCEAN CANYON

FIGURE 3. BISCAY ABYSSAL PLAIN

PDR RECORDS EUROPEAN CONTINENTAL RISE AND BISCAY ABYSSAL PLAIN

Depth in fathoms.]

Heezen _et al._, PL. 9

FIGURE 1. SMALL-SCALE ROUGHNESS, UPPER CONTINENTAL RISE

FIGURE 2. ROLLING TOPOGRAPHY, UPPER CONTINENTAL RISE

FIGURE 3. CAPE VERDE ABYSSAL PLAIN

FIGURE 4. ABYSSAL HILLS]

Heezen _et al._, PL. 10

PDR RECORD OF ABYSSAL HILLS, SOUTHEAST OF BERMUDA RISE

Note sub-bottom echos from beneath intermontane basin floor. Depth in fathoms.]

Heezen _et al._, PL. 11

Area of each photograph is about 6 by 8 feet.

PLATE 11.--_OCEAN-BOTTOM PHOTOGRAPHS ON THE CONTINENTAL MARGIN AND OCEAN-BASIN FLOOR_

FIGURE 1. (Station T1-3, photo 27) Depth 260 fathoms, location 47° 42´N., 07° 34´W., just below shelf break west of St. Nazaire, France. Note small ripples which appear to be superimposed on larger ripples.

FIGURE 2. (Station T1-3, photo 7) Depth 285 fathoms, location several hundred feet from photograph in Figure 1. Note holothurians and solitary coral attached to rocks. Ripple marks are less prominent than in Figure 1.

FIGURE 3. (Stations T1-16, photo 28) Depth 2600 fathoms, location 46° 50´N., 11° 25´W., northern part of Biscay Abyssal Plain. Note tracks of bottom crawlers, and the prominent conical mounds, each with a central hole.

FIGURE 4. (Station T1-18, photo 29) Depth 2650 fathoms, location 43° 56´N., 11° 12´W., southern part of Biscay Abyssal Plain. Note meandering ridge made by subsurface burrower. Note also starfish, upper left, and quantity of fecal pellets and holes in the bottom.

FIGURE 5. (Station T1-20, photo 53) Depth 2850 fathoms, location 42° 18´N., 14° 47´W., northeastern part of Iberia Abyssal Plain. Note holes with converging tracks and large mound in upper right.

FIGURE 6. (Station T1-58, photo 14) Depth 3072 fathoms, location 29° 17´N., 57° 23´W., Abyssal Hills southeast of Bermuda Rise. The round objects are manganese nodules. Note shark's tooth in lower right. Note also small holes indicating bottom dwellers, and meandering raised ridge of sub-bottom burrower. Of particular interest are the small moats surrounding many of the nodules; they are probably scour marks caused by bottom currents. These currents must be very gentle since none of the nodules seems to show evidence of recent rolling.

Positions of stations shown on Plate 30.]

Profile E-12 passes from the African coast between Fuerteventura and Gran Canary toward the northeast. Here the continental slope extends only to 1000 fathoms. The Canary Islands rise abruptly from the continental rise. Except for gradients of the order of 1:15 on the steep slopes of these volcanic islands the gradients of the continental rise are 1:300-1:1000.

Profiles E-13 and E-14 end on the east near Gran Canary and thus do not show most of the upper continental rise. They do show the remarkably wide and nearly level lower continental rise which reaches a width of more than 500 miles.

Profiles E-15 and E-16 lie off Spanish Sahara. In both profiles the gradient is 1:10 to 1:20 between the 50-fathom shelf break and a bench at 300-500 fathoms. In both profiles the gradient drops below 1:40 at about 1200 fathoms. Both profiles show numerous prominent benches on the continental slope. The upper continental rise with gradients of 1:350-1:1200 extends to the western limit of the profiles.

In profiles E-17 and E-18 the continental slope becomes gentler, and only in the upper 500 fathoms of E-18 does the gradient exceed 1:25. The upper continental rise is about 60 miles wide with depths predominantly about 1600 fathoms, and the lower continental rise lies at about 2100 fathoms and is very smooth.

Profiles E-19, E-20, and E-21 cross the continental margin off Dakar and the Cape Verde Plateau which rises from the lower continental rise. The Cape Verde Plateau consists largely of the coalescing bases of the volcanic Cape Verde Islands. The lower continental rise and the abyssal plain reach their maximum width at this latitude. The width of the ocean, the width of the Mid-Atlantic Ridge, the width of the abyssal hills, and the depth at the axis of maximum depth all reach their maximum values for the North Atlantic at this point. The characteristics of the continental rise in this sector are listed in Table 2. The reliability of these figures is much poorer than those given for northeastern United States, owing to the smaller number of profiles in this sector.

TABLE 2.--_General characteristics of the continental rise
northwest Africa Sector_

Values measured from Profiles E-11 to E-21

====================================================================
Depth
Segment Upper edge Lower edge Gradient Width
====================================================================

Upper continental rise
1 1200 ± 200 1500 ± 200 1:90 ± 30 30 ± 10?
2 1500 ± 200 1600 ± 200 1:200 ± 100 30 ± 15?
3 1600 ± 200 1800 ± 100 1:100 ± 50 25 ± 15?

Lower continental rise
1 1800 ± 100 2000 ± 100 1:400 ± 200 75 ± 50
2 2000 ± 100 2000 ± 100 1:1500 ± 500 150 ± 50
3 2000 ± 100 2700 ± 100 1:500 ± 200 200 ± 50

Abyssal plain 2700 ± 100 3000 ± 75 1:1250 ± 250 200 ± 50
--------------------------------------------------------------------

A famous submarine canyon, the Fosse de Cayar, lies just north of Cape Verde. Other submarine canyons are certainly present in the sector since any profile parallel to the strike of the topography reveals large irregularities probably related to canyons. Echograms (Pls. 9, 13) taken in the continental rise in this sector show distinct contrasts in the topographic detail of the sea floor. The rugged topography of the abyssal hills (Pl. 9, fig. 4) contrasts sharply with the nearly flat, extremely smooth abyssal plain. (Pl. 9, fig. 3). The continental rise is nowhere so smooth nor so flat as the abyssal plain. The continental rise ranges from 10- to 20-fathom rolling hills 5-10 miles in width to 2- to 5-fathom hills a few hundred feet across (Fig. 19). At the seaward edge of the abyssal plain the echo sounder penetrates the bottom to reveal interfaces 5-20 fathoms below (Pl. 13, Fig. 4). Sub-bottom penetration of 5-15 fathoms is occasionally encountered on the continental rise on local topographic highs. Lower continental-rise hills of the type observed off eastern United States have not been observed off Africa.

BENCHES AND TERRACES OF THE CONTINENTAL MARGIN

The topography of the continental margin provinces is divided into a series of benches or terraces. The largest are the continental shelf and slope (continental terrace) and the upper and lower continental rise. Superimposed on each of these major features is a series of smaller benches and terraces which range from features a few miles wide to simple breaks in the gradient of the continental slope.

Many of these features can be traced for hundreds of miles (Heezen _et al._, in press); some are intermittent, others change in depth with distance along the shelf; still others are only locally developed. We can propose at least four possible origins for terraces or benches: (1) ancient shore features; (2) structural (or rock) benching; (3) block faulting; and (4) landslide or slump scars.

The submerged terraces within a few hundred feet below present sea level can probably best be explained as ancient beaches formed during the lower sea levels of the Pleistocene. The fact that the same levels are found along coasts of diverse geology and tectonic development supports the eustatic origin of terraces between sea level and 70-100 fathoms. The gradients of the continental shelf are so low and the benches are so persistent that block faulting and slump scars are excluded as general explanations. The benches of the continental slope extend to depths of 1500 fathoms and vary in depth from point to point along the continental slope. These cannot be Pleistocene eustatic levels unless we consider that they were formed prior to recent large crustal deformations. Again the persistence of the benches for many miles argues against a fault-scarp or slump-scar hypothesis. Thus, while the benches of the continental shelf are probably ancient beaches, particularly those traced at the same depth for thousands of miles, the benches of the continental slope are probably rock benches, while some may represent step faulting.

SUBMERGED BEACHES ON THE CONTINENTAL SHELF: In Table 3 the depths of terraces or persistent levels of the continental shelf are listed for selected points in the North Atlantic. There is a remarkable uniformity in these data; the same levels are found near Newfoundland, in Florida, and on oceanic islands far from the glaciated areas.

On the basis of data obtained in the North Atlantic it is not possible to date the different terraces, but probably most were formed in the period between 12,000 and 5,000 years B.P. when the sea rose in consequence of the melting of the Wisconsin glaciers. Coring and dredging on these submerged ancient beaches could probably produce material datable by the radiocarbon method.

TABLE 3.--_Depth (in fathoms) of prominent continental-shelf
terraces_

Each column based on only one nonprecision echogram

=============================================================================
|St. | |Charles-| | |Bar |St. | |
Placentia|John's,|Norfolk,|ton, |Bimini,|Miami,|bados,|Vincent,|Dakar|Dakar
Bay, Nfd.|Nfd. |Va. |S. C. |B.W.I. |Fla. |B.W.I.|C.V.I. |North|South
=============================================================================
10 12 10 8 10 8 10
15 15 15
20 18 20 20 18 20 20
25 24
30 30 28 30 28 28
35 38 35 35 32
40 38 38 38
42 45 42 45 42 45
50 50
55 55 58 56 54 55 55
68 65 68 65 60
72 76 72
80 80 80 80 85 81 80 78
-----------------------------------------------------------------------------

CONTINENTAL MARGIN BENCHES: On each profile across the continental margin is a series of benches and changes in gradient which range from the shelf break to slight changes in gradient on the continental slope.

If a field geologist enters a new area of sedimentary rocks where road cuts do not exist he invariably goes to the stream valleys, and here he gets his first and best view of the geologic section. The stream's gradient is adjusted to the resistance of the rocks over which it cuts, and the form of the valley-side slopes reveals the nature of the underlying rocks even if they are grassed over.

This obvious field method had never been fully applied to the continental margin. Stetson (1936) dredged in the canyons of Georges Bank, and his hauls included Cretaceous sandstones and Tertiary marls and green sands. He concluded that the canyons had been cut deep into the continental margin to expose the underlying Cretaceous rocks, but he considered the continental slope the product of depositional processes.

However Upham (1894) had suggested that the continental slope formed a continuous outcrop of Tertiary and Cretaceous sediments from Newfoundland to Florida, a suggestion the writers consider quite probable. That is to say, an analogy can be made between the continental slope and one face of the Grand Canyon or to an erosional escarpment bounding a high mesa or plateau like the Book Cliffs of Utah and Colorado.

Only a few areas of the world are sufficiently well sounded to provide data for a study of structural benches. One cannot expect to see identical structural benches in each profile even across a slope composed of a laterally uniform series of horizontal beds of contrasting lithologies. The exact mode of erosion, the local system of jointing, and chance variations in a number of other variables make it necessary to have a large number of closely spaced, accurately located profiles. We are fortunate that the Coast and Geodetic Survey has surveyed virtually the entire continental slope from Georges Bank to Norfolk, Virginia. Almost all these sounding lines are run at right angles to the strike of the topography and are thus suitable for analysis of structural benches. In this same area the dredgings of Stetson (1936) on Georges Bank and the Esso Hatteras Light test provide us with information on the stratigraphy of the sediments which form the continental shelf and slope. The seismic work of Ewing and collaborators (1937 _et seq._) provides us with further information on the dips and on the depths of a number of sedimentary rock series of contrasting lithology.

Fishermen began finding fossiliferous rocks on Georges Bank well over a century ago. They were not particularly pleased to obtain rocks instead of fish and generally threw the accursed rocks back into the sea. Some curious fishermen brought a few of the rocks to shore, however, and in time some of these were received by the museums (Upham, 1894; Dahl, 1925). These rocks contain Tertiary and Cretaceous fossils. The depths and positions of recovery of the rocks were generally unknown to the museums, and no clear idea could be gained of the exact occurrence of this material. Stetson (1936; 1949) conducted a series of scientific dredging operations in the Georges Bank area. His aim was to recover more of these older rocks from known depth ranges and positions.

He concluded that the older rocks outcrop only in the submarine canyons. He found no consistent depth ranges for the series of Miocene and Upper Cretaceous rocks obtained.

Chart shows position of sounding lines and dredge hauls used to construct projected profile and inferred geologic section shown in Figure 21 (a). Sounding lines from Coast and Geodetic Survey Chart 1313.]

In the Georges Bank area a series of prominent benches continues along the continental slope. If these are structural benches we should be able to trace them up the canyons and thus determine the dip of the formations. If the benches are the result of step faulting or landslide scars they would not extend up the canyons. In order to test these alternatives, a series of profiles has been plotted from the surveys of the Coast and Geodetic Survey. A line was drawn which paralleled most of the contours of the continental slope for 20 miles or more (Fig. 20). A second line was drawn at right angles to this first strike line. All sounding lines in the area were projected to this second dip line along lines parallel to the strike line, and plotted as a composite projected profile (Fig. 21). If the first line was essentially the strike of rock layers then we should be able to determine the dip of the beds by picking the successive benches as they occur on successive profiles across the canyon. In Figure 21 the results of the analysis of Oceanographer and Hydrographer canyons are presented. The dredge hauls by Stetson (1949) have also been projected on this profile. We see that a major bench occurs at _A_ which passes below the hauls in which fossiliferous Navarro (Upper Cretaceous) was obtained and through the upper limit of the hauls where Matawan (Upper Cretaceous) was obtained. Thus if this horizon or one closely parallel to it is the Navarro-Matawan contract, the apparent discrepancies of the depth ranges of Stetson's dredge hauls are explained. Much more important to the present study is the dating of a prominent structural bench.

Location of soundings for profile (a) shown in Figure 20. Soundings projected along strike to construct profile. Soundings for both profiles taken from Coast and Geodetic Survey Chart 1313.]

Well logs from Swain (1947) and Spangler (1950). Four sounding profiles made by R. V. ATLANTIS are projected to profile. Note that resistant formations form prominent structural benches on continental slope.]

The Esso Hatteras Light No. 1 test encountered crystalline rock at 1640 fathoms depth, beneath Lower Cretaceous strata (Spangler, 1950). The several holes drilled in the vicinity revealed remarkably constant dips over a wide area. This is in fact true of the whole coastal plain. Since the Hatteras well is only 17 miles from the continental slope, it seems reasonable to project the dips to the continental slope. We can then observe whether prominent benches on the continental slope correlate with resistant strata in the well. We find (Fig. 22) that they do. In the area between Cape Hatteras and Nova Scotia several cores have revealed reworked Eocene, Miocene, and Cretaceous Foraminifera. In 1947 Northrop and Heezen (1951) obtained a photograph and a core at 500 fathoms on the continental slope. The core contained Eocene (Jackson) sediments, and the photograph showed a rock ledge below the marl sampled. Although sediment cores, particularly those of reworked material, do not provide as reliable information as dredge hauls, this outcrop of un-reworked Eocene may also be used in dating the structural benches of the continental slope.

The structural benches between Cape Hatteras and Cape May are remarkably uniform and persistent (Fig. 23). On the basis of the extrapolation shown in Figure 22, the structural benches in this area have been correlated with the formations encountered in the Hatteras well. North of Cape May a major angular unconformity separates the late Tertiary and Cretaceous formations. Eocene has not been found north of Nantucket. Upper Cretaceous has been found at about 400 fathoms off Georges Bank (Fig. 21), and Lower Cretaceous has been dredged at 200 fathoms off Banquereau Bank, Nova Scotia.

Between 1945 and 1950 workers on the ATLANTIS made several sounding profiles east of Georgia, North Carolina, and South Carolina. Each crossed the precipitous Blake Escarpment. It was quite clear that no sediment could be accumulating on such a steep escarpment and that beds of ancient sediments and perhaps crystalline rocks must outcrop on the escarpment. In 1949 and 1950 a few cores were taken on the escarpment which encountered Miocene and Eocene sediments in depths of 500-800 fathoms (Ericson, Ewing, and Heezen, 1952). The marked similarity of all topographic profiles further supported the view that the escarpment was formed by the outcrop of an orderly sequence of horizontal sedimentary rock layers. With this specific problem in mind a cruise was made to the Blake-Bahama area on the research vessel ATLANTIS, in 1951. More than 50 cores were obtained. Sediments of Recent to Upper Cretaceous age were obtained on the Blake Escarpment and from the steep walls of the Bahama Channels (Ericson, Ewing, and Heezen, 1952). Seismic-refraction work by Katz and Ewing (1955) and Nafe _et al._ (unpublished) and reflection work by Ewing and Landisman (unpublished) have revealed that distinct seismic interfaces can be traced into the structural benches on the Blake Escarpment. The ancient sediments from the Blake Escarpment and the log of the Andros well allow the dating of some of these formation contacts. At present the most prominent bench at 1200-1500 fathoms appears to mark the base of the Upper Cretaceous. Dredging and further coring on the Blake Escarpment below 1400 fathoms is one of the most promising projects of its type despite the great difficulties involved.

Soundings by Coast and Geodetic Survey; 35° 30´N.-38° 30´N.]

Lee (1951), who made a topographic study of Exuma Sound, Bahamas, traced several prominent benches through 51 cross sections of the sound.

Data from Day et al. (1956) and Hill (1957). Geologic ages are those assigned by Day on the basis of velocity; they are not based on dredging or drilling.]

Seismic-refraction profiles have been made across the continental slope southwest from the English Channel. These studies were initiated by Bullard and Gaskell (1941) and have been most recently reported on by Day _et al._ (1956). The seismic section of Day _et al._ (1956) (Fig. 24) suggests that the prominent bench at 1600 fathoms and the short but steep scarp just below represent the outcrop of the metamorphic basement on the continental slope. It is postulated that the prominent bench at 900 fathoms may represent the base of the Mesozoic, and the smaller bench at 300 fathoms the base of the Miocene. Tertiary sediments have been obtained from the walls of canyons in the Bay of Biscay in depths down to about 1500 fathoms (Bourcart and Marie, 1951). The age assignments in Figure 24 are taken directly from Day _et al._ (1956) and have not been confirmed by dredging on the continental slope.

The writers conclude that the majority of the topographic benches of the continental slope and other category II provinces are structural benches which reflect the outcrop of resistant rock layers. This of course implies that the continental slope is not a simple depositional feature but a structural or erosional one. Since the structural benches are present both in the canyons and on the un-dissected slope, the occurrence of Tertiary and Cretaceous rocks on the continental slope cannot be explained by erosion of submarine canyons into an otherwise depositional terrace in the manner implied by Stetson (1949).

The existence of such persistent benches implies that the entire width of the category II provinces must be at most only thinly covered by recent sediments. Since the discovery of the great importance of turbidity currents and the relatively low slopes necessary for their occurrence, it has been a great puzzle to the writers how sediments could be permanently deposited on the present continental slope. The answer is simply that they are not. In addition to the turbidity currents which provide a mechanism for the seaward transport of sediment down the continental margin, deep-ocean currents probably sort and transport much sediment along a course parallel to the continental slope. It has recently been demonstrated (Swallow and Worthington, 1957) that velocities of 10-20 cm/sec are attained by ocean currents which flow parallel to the continental slope. The particular measurement referred to was made at 1600 fathoms on the continental slope south of Cape Hatteras where a 17 cm/sec southward-moving current was observed. The strong current is not a local phenomenon since it was found in the South Atlantic by Wüst (1935) and is predicted by theories of ocean circulation (Stommel, 1957). Photographs of ripple marks on the continental slope (See for instance Fig. 13 in Elmendorf and Heezen, 1957) had indicated high velocities, but it was not possible to distinguish between a current and an oscillatory origin. The total effect of slides, slumps, turbidity currents, and strong ocean-bottom currents is the removal of most of the unconsolidated Recent sediments from the continental slope.

The deposition of a series of Mesozoic and Tertiary sediments on the subsiding margin of the continental block has produced a wedge of sedimentary rock largely of shallow marine facies. Each successive strata laid down on the shelf was abruptly terminated at the shelf break by the processes of erosion which continuously or periodically clear the unconsolidated sediment from the continental slope. Deposition on the shelf was interrupted by several marine regressions which produced unconformities in the stratigraphic sequence. Nafe and Drake (1957) observed that the increase of seismic velocity with depth and therefore the increase in compaction with depth is more rapid on the continental shelf than in the deep sea. This is probably in part the result of erosion of previously deposited sediments and sedimentary rocks along the unconformities and in part the result of ground-water cementation during periods of emergence.

Each unconformity should mark a lithologic change and consequently a change in the resistance to erosion of the rock series. Many structural benches may indicate surfaces of unconformity. The most recent unconformity in the sequence lies between the surface of the emerged Wisconsin continental shelf and the overlying post-glacial shelf sediments.

The shelf break is defined as the most prominent break in slope between the continental shelf and continental slope. The most prominent break may locally be a Pliocene or Miocene structural bench, but elsewhere late Pleistocene or Recent strata may form the shelf break. Rates of subsidence, erosion, and sediment supply vary from place to place along the continental shelf, and the lack of conformity either in depth or in age of the shelf break is thus easily explained.

The deeper structure of the continental margin indicates a fundamental structural discontinuity at the base of the continental slope (category II provinces). It would seem a small extrapolation to attribute a fault origin to the continental slope. Although faulting may have played a large part in the earliest history of the category II provinces, alternate periods of sedimentation and marine planation on the continental shelf and long-continued erosion by slumps, turbidity currents, and deep-sea currents on the continental slope, together with a general subsidence of the area, could have alone produced the characteristic form of the continental terrace.

Further work on structural benches co-ordinated with a study of ancient sediments from dredges and cores should enable us to draw a geologic map of the continental slope of eastern United States and Europe (Heezen _et al._, in preparation).

GEOLOGY AND GEOPHYSICS OF CONTINENTAL MARGIN PHYSIOGRAPHIC PROVINCES

_Seismicity of the continental-margin provinces._--Plate 29 shows the distribution of epicenters in the North Atlantic. Except in the Puerto Rico Trench and the adjacent Antilles Arc, earthquakes are exceedingly rare in the continental-margin provinces of the North Atlantic. From the Bahamas through the Grand Banks the only earthquakes reported since 1910 are two near the Laurentian Channel. In the eastern Atlantic an earthquake belt crosses the continental margin near Gibraltar but does not seem to be directly associated with the Atlantic continental margins. If the continental slope in the area covered by Plate 29 is a fault scarp, we must infer that the motion has now been arrested.

* * * * *

_Magnetic anomalies and continental-margin provinces._--The first continuously recorded total-intensity magnetic data for the Atlantic were obtained in 1946 on a flight from Long Island, New York, to Bermuda, and return (Keller _et al._, 1954). The first data from a ship-towed magnetometer were obtained in 1948 (Heezen, Ewing, and Miller, 1953).

The initial measurements showed (1) several large (> 300γ) anomalies on the continental shelf, (2) a large (> 500γ) anomaly over the continental slope, (3) an exceptionally smooth field over the category III provinces, (4) rough fields with 5- to 15-mile wide 100 to 200γ anomalies over oceanic ridges and rises, (5) enormous anomalies (> 1000γ) over seamounts and islands, and (6) large anomalies (> 500γ) over the crest of the Mid-Atlantic Ridge.

The large anomalies on the continental shelf were considered to indicate large volcanic cones buried by the sediments of the continental shelf (Miller and Ewing, 1956). The anomaly over the continental slope was considered to be the magnetic-edge effect associated with the abrupt edge of the continental block. The rough fields over the oceanic rises and ridges were attributed to variations in the susceptibility of the crustal rocks, probably related to volcanic extrusions and intrusions. The anomaly over the crest of the Mid-Atlantic Ridge was not recognized as a general pattern until later when many measurements were made. The major puzzle for which no explanation was readily obtained was the origin of the smooth field over the continental rise. The smooth or quiet field has been observed on each crossing of the category III provinces, and even such a sharp-sided feature as the Puerto Rico Trench failed to show a magnetic anomaly (Davidson and Miller, 1956). We can thus state that all category III provinces are characterized by a smooth magnetic field.

The significance of this point is not yet clear, but it must be of major importance to the full understanding of the continental margin.

_Crustal structure and continental-margin provinces._--Maurice Ewing and his associates have made more than 1000 seismic-refraction measurements to determine the crustal structure of continental margins. Most of the published results pertain to the eastern continental margin of the United States (Drake et al., in press), but much material will soon be published on the continental margin of Sahara, Sierra Leone, Brazil, Argentina, Guiana, Spain, Morocco, the Gulf of Mexico, and the Caribbean Sea. In addition extensive work has been accomplished by Bullard, Gaskell, and Hill in the western approaches to the English Channel. In some areas seismic, gravity, and magnetic measurements were made along the same lines and have been subjected to an integrated analysis (Worzel and Shurbet, 1955c).

In Plate 26 four profiles replotted from published sections are shown. Each profile is plotted at 40:1 exaggeration, and depths are indicated in fathoms and kilometers to facilitate comparison with the profiles and topographic descriptions given elsewhere in this paper. Some stations which lie up to 100 miles from a profile have been projected along the strike of the topography. In general each profile indicates (1) a thick lens of sediment and sedimentary rock beneath the continental rise; (2) a major discontinuity in crustal structure at the base of the continental slope; and (3) a wedge of low-velocity sedimentary rocks ("unconsolidated") overlying a lens of higher-velocity ("semi-consolidated and consolidated") sedimentary rocks beneath the continental shelf. The basin, filled with higher-velocity sedimentary rocks, is formed by an upturned lip of crystalline basement rocks near the continental slope. Local variations in thickness and velocity give rise to interesting minor configurations, but the basic structure is nearly identical in all the sections. The main contrast is that the category III provinces are underlain by thick sedimentary rocks in the case of the continental rise, and successively thinner accumulations of sedimentary rock are associated with the marginal basins and marginal trenches.

Since the main difference in the continental-margin sections seems to be a greatly enhanced continental-rise sedimentary section to the north as compared with the Blake Plateau--Bahama--Puerto Rico region, it seems reasonable to conclude that the main difference in the topography is the result of different sedimentation rates or a different time of origin, rather than of a vastly different structural origin. The sedimentary rock column beneath the continental rise is truly geosynclinal in thickness. Whether this great lens of sediment and sedimentary rock will eventually be deformed into a mountain range to form a new addition to the continent as advocated by the accretion school or whether it will lie forever on the sea floor is one of the great speculations of geology.

* * * * *

_Sediment distribution and physiographic provinces of the continental margin._ The recent sediments of the category I provinces are largely sands, silts, silty lutites, and carbonates (Stetson, 1938; Shepard and Cohee, 1936; Newell, 1955). All these sediments can be assigned to the orthoquartzite suite of Pettijohn (1957). Judging from the ancient sediments dredged from the continental slope and obtained by drilling, the category I provinces have been receiving orthoquartzite sediments at least since the Cretaceous. In category II provinces Recent sediments are either lacking or are very thin. Those Recent sediments which temporarily remain in the province range from foraminiferal lutites, found in such areas as the Bahamas, to silty lutites relatively poor in pelagic fossils, found in the higher latitudes. In general category II provinces are characterized by nondeposition or erosion.

The sediments of the category III provinces include both pelagic and clastic terrigenous sediments. The outer ridge is covered by _Globigerina_ ooze in depths shallower than about 2200 fathoms and by red clay in deeper areas. The continental rise is generally covered by silty foraminiferal lutites, but, where submarine canyon deltas spread out from the mouths of submarine canyons (Ericson _et al._, 1951, 1952), beds of graded sand and silts from a few centimeters to a few meters in thickness are interbedded with lutites. In the marginal trenches and marginal basins graded calcareous sands alternate with low-carbonate lutites. The category III provinces of northwest Africa are dominated by two groups of volcanic islands and seamounts which contribute flows and volcanic detritus to the continental-rise sediments. In general the category III provinces are dominated by sediments ascribable to the graywacke suite of Pettijohn (1957).

* * * * *

_Past, present, and future of continental-margin physiographic provinces._--It takes only a little imagination to see a historical or genetic sequence in the four profiles of Plate 26. The Puerto Rico Trench--outer-ridge profiles thus may represent a continental margin in youth, the Blake--Bahama--outer-ridge profile a margin in late youth or early maturity, and Newfoundland and northeast United States profiles a margin in late maturity or old age.

The lens of sediments that has filled the marginal trench off northeastern United States is truly geosynclinal in thickness (Drake _et al._, in press). The sediments of the shelf lens are similar in lithology to the orthoquartzite suite of a mio-geosyncline (cf. Kay, 1951), and the thicker continental-rise lens is probably similar in lithology to the graywacke-volcanic suite of a eugeosyncline. (Kay, 1951). It seems virtually certain that ancient orthogeosynclines were, before deformation, closely analogous to the continental margins. The major problem is completing the sequence from filled geosyncline to folded mountains is the mechanism by which the earth's crust beneath the geosyncline thickens by 20-30 km. Although we cannot predict the future of the present continental margin with any assurance until the mechanism of thickening is understood, it seems probable that before deformation the older orthogeosynclines were similar in most details to the modern continental margins.

It seems possible that the present continental margins will in some future geological period be uplifted into folded mountains.

OCEAN-BASIN FLOOR

GENERAL DESCRIPTION

The second of the three basic topographic divisions of the oceanic depression is the ocean-basin floor. Included in this division are those provinces of the oceanic depression that are not included in the continental margin or the mid-oceanic ridge.

The ocean-basin floor is divided into three categories: (1) abyssal floor, (2) oceanic rises, and (3) seamounts and seamount groups. The first category includes two types of provinces, abyssal plains and abyssal hills, which occupy the deepest portion of the ocean-basin floor. Included in these provinces are such features as abyssal gaps and mid-ocean canyons. The second category includes the larger positive features of the ocean-basin floor, and the large seamounts and seamount groups fall in the third category. The landward limit of the ocean-basin floor is the 1:1000 gradient isopleth along the continental margin. Along the mid-oceanic ridge the boundary is taken as that scarp or scarp zone where the average level rises appreciably above the axis of maximum depth of the basin floor. Broad elevations which rise above the basin floor as isolated rises are termed oceanic rises and are included in this discussion even though they may be structurally more closely related to the mid-oceanic ridge or the continental margin.

ABYSSAL PLAINS

_General discussion._--An abyssal plain has been defined (Heezen, Ewing, and Ericson, 1954) as "an area of the deep-ocean floor in which the ocean bottom is flat and the slope of the bottom is less than 1:1000".

Abyssal plains have been found in all oceans, characteristically at the foot of the continental rise. Koczy (1954; 1956) and Gaskell and Ashton (1954) have described the abyssal plain south of the Bay of Bengal. Koczy (1956) has described plains in the equatorial Atlantic on either side of the Mid-Atlantic Ridge. On Expedition Deep Freeze workers on the USS GLACIER, using a PDR, discovered an abyssal plain in the Weddell Sea. Workers on the R. V. VEMA in 1957 confirmed the existence of abyssal plains in the Argentine and in the Cape and Angola basins of the South Atlantic. Menard (1955) has described abyssal plains off the California and Alaska coasts. Abyssal plains are really important and are present off all coasts except those having a long, continuous, unfilled marginal trench.

The abyssal plains shown on the physiographic diagram have been named in order to facilitate referencing. The abyssal plains of the western Atlantic between Newfoundland and the West Indies are best known. The "abyssal plain south of Newfoundland" (Heezen _et al._, 1954) has been named the Sohm Abyssal Plain following the usage of Murray (1912) who referred to this area as "Sohm Deep" in several publications. The abyssal plain south of the Bermuda Rise referred to as the "abyssal plain in the Nares Basin" (Heezen, Ewing, and Menzies, 1955; Luskin et al., 1954) is named on the province chart (Pl. 20) the Nares Abyssal Plain, also following the usage of Murray (1912), who named this area the Nares Deep. The abyssal plain southwest of the Bermuda Rise was included in the Nares Abyssal Plain until the discovery in 1955 that the two parts were separated by a steep abyssal gap at 24° N. 68° W. necessitated the splitting off of this plain from the Nares Abyssal Plain. The name Hatteras Abyssal Plain used on the map was suggested by the close proximity of the very prominent Cape Hatteras.

An abyssal plain is probably present in the Newfoundland basin, but no PDR soundings have been obtained in that area. The name Newfoundland Abyssal Plain is suggested subject to confirmation by a precision survey.

The outer ridge which runs south from Cape Hatteras paralleling the coast lines of the Bahamas encloses the Blake-Bahama Basin (Ericson, Ewing, and Heezen, 1952); a narrow abyssal plain is found along its western margin. This plain is named the Blake-Bahama Abyssal Plain.

Between the Greater Antilles and the Bahama Banks is a depression called the Old Bahama Channel in the north; farther south it is known as the Hispaniola-Caicos Channel. The Hispaniola-Caicos Channel in particular contains an abyssal plain which, although much smaller than the large, deep-sea abyssal plains, has all the characteristics of slope flatness and sediment composition. This is called the Hispaniola-Caicos Abyssal Plain.

The smallest, yet perhaps the most striking, Atlantic abyssal plains are those at the bottom of the Puerto Rico Trench at depths of about 4358 fathoms (4585 corrected). They exhibit the proper degree of flatness, low gradient (Ewing and Heezen, 1955), and the shallow-water clastic sediments typical of abyssal plains.

South of Cuba in the southwest corner of the diagram lies a major deep-sea trench. Only the eastern end of this Cayman Trench is shown in the diagram. On the floor of this steep-walled trench lies a narrow trench plain or system of semi-connected trench plains which lie at depths of about 3000 fathoms south of Guantanamo, Cuba, but deepen to 3692 fathoms south of the Cayman Islands. These are known as the Cayman Trench Plains. This area will be described in more detail in a later publication devoted to the topography of the Caribbean.

In the eastern Atlantic each abyssal plain has been crossed at least three times by ships employing a PDR, a coverage which, although sufficient to establish their existence, is vastly inferior to the coverage of the western Atlantic abyssal plains. The Biscay Abyssal Plain occupies a large portion of the Bay of Biscay. It is connected by an abyssal gap at 43° 30´N., 12° 00´W., to the Iberia Abyssal Plain which lies west of the northern two-thirds of the Iberian Peninsula. A small abyssal plain west of the Tagus River takes its name from this river. The great abyssal plain which skirts Madeira on the east and south and extends over a vast area to the west is referred to as the Madeira Abyssal Plain. The Canary Islands and associated banks form the boundaries of a small oval depression known as the Canary Abyssal Plain. The vast abyssal plain which skirts the African continent west of Cape Verde is named the Cape Verde Abyssal Plain.

The eastern margin of the map just reaches the abyssal plain in the western Mediterranean which Heezen and Ewing (1955) have named the Balearic Abyssal Plain.

Abyssal plains were not discovered until the use of continuously recording echo sounders was extended to abyssal depths following World War II (Tolstoy and Ewing, 1949). While the early nonprecision echo sounders were sometimes adequate to distinguish the limits of particular plains, real advances in their study required the development of precision sounders and the extensive use thereof (Heezen, Ewing, and Ericson, 1954). Since the very existence of abyssal plains has been known for less than 10 years and only recently have any of them been adequately delineated, the problem of their nomenclature has never before arisen.

In the cases of the Nares and Sohm abyssal plains we have taken the name Murray has given to the "deeps" in which the plains occur. For Cape Verde, Iberia, and Newfoundland abyssal plains we have followed the name Wüst (1940b) proposed for the basin within which the plains lie. The Biscay, Tagus, Hatteras, Blake-Bahama, Hispaniola-Caicos, and Balearic have been given the name of a prominent near-by land area, in accordance with recognized practice. The Horseshoe Abyssal Plain takes its name from the horseshoe-shaped line of seamounts which surrounds it except on the eastern side.

* * * * *

_Regional description._--The relief of the abyssal plains is so low that ordinary small-scale exaggerated profiles (Pls. 22, 27) reveal nothing of its detailed character. In a few areas exceptionally good PDR records have been obtained which show relief of a few feet (Pl. 12).

SOHM ABYSSAL PLAIN: The best-studied abyssal plain is the Sohm Abyssal Plain south of Newfoundland. In addition to the 1953 VEMA PDR survey (Heezen _et al._, 1954) the area was studied extensively in 1955 and crossed by the trans-Atlantic cruises of 1954 and 1956. The plain is T-shaped and generally about 200 miles wide; depths range from 2700 to 3000 fathoms (uncorrected). The depth increases in each arm of the T toward the stem where the slope changes to south. The greatest depth is found at the south boundary of the plain near 29° N. In the northern part of the plain small topographic irregularities are unknown, but toward the south peaks 50 to 500 fathoms high increase in number until they finally replace the plain. North of the Bermuda Rise the plain surrounds a line of huge seamounts. However, even here no small hills are found. In the strip extending between 30° and 37° N., the east and west boundaries of the plain are formed by scarps 200 to 800 fathoms high. The boundary on the north side of the Bermuda Rise is formed by a distinct shallowing, but no scarp is seen.

There is some justification for separating the northwest arm of the Sohm Abyssal Plain north of the Bermuda Rise from the rest of the Sohm Abyssal Plain since a mid-ocean canyon runs from the northwest to the central sections of the plain. A sill might lie northeast of the northeast tip of the Bermuda Rise. Here the bottom gradient changes slightly, and an extremely large magnetic anomaly is associated with the region. However, the plain narrows only slightly at this point, and, since we cannot tell how many individual basins were filled to form an abyssal plain, we will consider that the Sohm Plain now includes the whole area, keeping in mind that the northwest arm may have only recently merged with the rest.

South of Newfoundland the continental rise merges with the abyssal plain with no scarp or line of hills. Here the boundary is taken at the point where the gradient is gentler than 1:1000. This is not an entirely arbitrary definition since an appreciable change in gradient generally occurs near this point. South of the boundary the gradient gradually decreases to 1:3000 at 37° N., beyond which it decreases even more gradually and reaches 1:5000 at the southern edge of the plain.

HATTERAS ABYSSAL PLAIN: Northeast of the Bermuda Rise on a line between Long Island (N. Y.) and Bermuda a small area of abyssal hills lies between the continental rise on the northwest, the Bermuda Rise on the southeast, the Sohm Abyssal Plain on the northeast, and the Hatteras Abyssal Plain on the southwest. Although this area has been traversed by more tracks than any area of similar size in the Atlantic, the relationships here are still somewhat obscure. This area of some 6000 square miles is one of irregular relief, but it is not known if it is of the character of an irregular sill or a complex abyssal gap, although the latter seems more probable. The northwest margin of the Hatteras Abyssal Plain is abruptly formed by the lower continental rise hills. The eastern boundary is generally distinct and in places is formed by a 50- or 100-fathom scarp, but generally it is not so impressive as the east and west boundaries of the Sohm Abyssal Plain. South of about 34° N., the western boundary of the Hatteras Abyssal Plain is formed by the outer ridge. In some sectors the eastern flank of the outer ridge is rather gentle, and the abyssal plain gently laps the ridge without a sharp break. This boundary is generally sharp north of 27° N. but to the south is less definite. The Hatteras Plain slopes southeast to about 31° N. where the direction of slope changes to south. A gradient of about 1:3000 and depths of about 2900 fathoms are reached at 25° N. Within about 60 miles of Vema Gap the plain is broken by numerous mid-ocean canyons which converge on the gap (Pl. 14, fig. 3). The Hatteras Abyssal Plain has no known seamounts or sea knolls, in contrast to the Sohm Abyssal Plain and Nares Abyssal Plain.

The sediments of the Hatteras Abyssal Plain resemble the northern abyssal plains in having many beds of quartz silts (Pl. 28) and contrast sharply with the Blake-Bahama Abyssal Plain where calcareous sands are abundant. On the west the outer ridge forms a continuous barrier to turbidity-current sediments coming from the near-by Bahamas so that the only source of turbidity-current sediments is from the Cape Hatteras region to the north or possibly from the Hudson Canyon via the suspected abyssal gap mentioned above. The type of sediment is entirely consistent with the gradient of the plain which slopes southward from the Hatteras region.

Heezen _et al._, PL. 14

FIGURE 1. SUB-BOTTOM ECHOES ON RECORD FROM SOUTHERN BERMUDA RISE

FIGURE 2. MID-OCEAN CANYON NO. 2

FIGURE 3. MID-OCEAN CANYON IN HATTERAS ABYSSAL PLAIN, WEST OF VEMA GAP

PDR RECORDS OF MID-OCEAN CANYON NO. 2 AND CANYONS AT VEMA GAP

Depth in fathoms.]

Heezen _et al._, PL. 15

Area of each photograph is about 6 by 8 feet.

PLATE 15.--OCEAN-BOTTOM PHOTOGRAPHS OF SEAMOUNTS

FIGURE 1. (Station V4-12, photo 16) Depth 700 fathoms, location 35° 12´N., 15° 18´W., on seamount of the Horseshoe Group. Note bioclastic debris lodged in depressions on the rocky bottom. Photographs shown in Figures 1, 2, and 3 are no more than a few hundred feet apart.

FIGURE 2. (Station V4-12, photo 10) Depth 700 fathoms, location 35° 12´N., 15° 18´W. Rippled sandy bottom.

FIGURE 3. (Station V4-12, photo 7) Depth 710 fathoms, location same as Figures 1 and 2. Note current ripple marks. The photograph indicates an appreciable current from left to right.

FIGURE 4. (Station V4-16, photo 19) Depth 1100 fathoms, location 35° 07´N., 13° 04´W., on the side of Ampere Seamount. Note winnow row of dark and light gravel-sized detritus which resembles similar material on the Rift Mountains shown in Figure 1 of Plate 19.

FIGURE 5. (Station V4-18, photo 26) Depth 75 fathoms, location 35° 10´N., 12° 55´W., near the top of Ampere Seamount.

FIGURE 6. (Station V4-14, photo 47) Depth 1100 fathoms, location 35° 12´N., 15° 22´W. on same seamount as Figures 1-3. Bottom appears to consist of light-colored sediment thinly veiled by dark coarser sediment. The white polka dots appear to represent piles of the underlying light sediment brought to the surface by burrowers.

Positions of stations shown on PLATE 30]

BLAKE-BAHAMA ABYSSAL PLAIN: The Blake-Bahama Basin is a long narrow basin between the great Blake Escarpment and calcareous Bahama Island banks on the west and the outer ridge on the east. A narrow abyssal plain 20 to 70 miles wide spreads out from the base of the escarpment (Pl. 5, fig. 3). The plain is shallowest and widest just east of the Northeast Providence Channel and deepens and narrows toward its northern and southern ends. Sediment cores taken 60 miles off the mouth of Northeast Providence Channel contained graded beds of calcareous sand a few meters thick (Ericson _et al._, 1952). The material was obviously derived from the Bahamas, presumably through the action of turbidity currents running out through the Providence Channels, the southern entrance of Exuma Sound, and the numerous submarine canyons which dissect the slopes off the islands. The depth of the plain ranges from 2600 to 2750 fathoms, and its area is about 7000 square miles.

NARES ABYSSAL PLAIN: South and southeast of the Bermuda Rise lies the Nares Plain, a 37,000-square-mile abyssal plain that slopes eastward from Vema Gap toward the Mid-Atlantic Ridge. From 62° to 64° W. its northern and southern boundaries are abrupt, and only three or four small abyssal hills have been observed in the plain. East of 64° W. the gradient of the plain exceeds 1:2500, and the frequency of hills increases until at 61° W. the plain consists of a series of fingers which extend into the abyssal hills. The sediment cores obtained in the Nares Abyssal Plain consist of alternating red clays and quartz silts (Ericson _et al._, 1952; in press). The Nares Abyssal Plain is the deepest of the broad abyssal plains of the ocean-basin floor. It is also the farthest from the land. If the turbidity currents smoothed the Nares Plain they would all have had to flow through Vema Gap since the outer ridge-trench complex to the south prevents currents from this area from reaching the Nares Abyssal Plain. The idea of a route through Vema Gap is supported by the fact that the plain slopes eastward from the gap.

HISPANIOLA-CAICOS ABYSSAL PLAIN: The depression between the Bahamas and the coasts of Cuba and Hispaniola deepens east of Cay Lobos and reaches its maximum depth of 2220 fathoms northwest of Cape Francis Viejo of Hispaniola. At Cay Lobos the channel is a steep V-shaped depression with a maximum depth of 400 fathoms, but to the east this V-shaped channel opens out to an abyssal plain about 1500 fathoms deep. Between this point and Great Inagua Island the flat floor slopes east reaching a depth of 1575 fathoms just southwest of Inagua. This portion is called the Old Bahama Abyssal Plain. South of Great Inagua is another V-shaped channel in which the depth drops from 1700 fathoms on the west to 2200 fathoms on the east. From this point to Cape Francis Viejo the Hispaniola-Caicos Abyssal Plain is nearly flat. Sediments collected from this plain confirmed the turbidity-current origin of the smooth topography. The cores contained plant debris, shallow-water fossils, and a variety of reworked older material (Ericson, Ewing, and Heezen, 1952; Ewing and Heezen, 1955; Ericson _et al._, 1955).

PUERTO RICO TRENCH (ABYSSAL) PLAINS: A trench plain is an abyssal plain in the bottom of a deep-sea trench. Since the discovery of the Puerto Rico Trench (Abyssal) Plains similar features have been reported for the Middle America Trench (Fisher, 1954), the Kurile-Kamchatka Trench (Udintsev, 1955), and the Peru-Chile Trench (Zeigler, 1958). Two trench plains are known from the Puerto Rico Trench. The smaller one occupies a basin south of a median ridge, and the larger one lies along the deep axis of the trench. They range in width from 1-2 miles to about 12 miles. Sediment cores taken in the trench plain contained beds of graded calcareous sand containing fragments of _Halimeda_ and shallow benthic Foraminifera (Ericson, Ewing, and Heezen, 1952). The depth of the larger trench plain ranges only slightly from 4358 fathoms (4585 fathoms corrected). The trench shallows eastward, and the plains disappear.

CAYMAN TRENCH (ABYSSAL) PLAINS: The Cayman Trench contains trench plains very similar to those in the Puerto Rico Trench. The limits of the Cayman Trench Plains shown on the physiographic diagram may not be too reliable; it has not been redrafted since the most recent tracks shown on the track chart were obtained.