Origin of the sediments and submarine geomorphology of the inner continental shelf off Choctawhatchee bay, Florida

Origin of the sediments and submarine geomorphology of the inner continental shelf off Choctawhatchee bay, Florida

Marine Geology - ElsevierPublishingCompany, Amsterdam - Printed in The Netherlands ORIGIN OF THE SEDIMENTS AND SUBMARINE GEOMORPHOLOGY OF THE I N N E...

1MB Sizes 1 Downloads 31 Views

Marine Geology - ElsevierPublishingCompany, Amsterdam - Printed in The Netherlands


(Received March 10, 1966) (Resubmitted May 8, 1967)

SUMMARY A portion of the inner continental shelf in the northern Gulf of Mexico exhibits sinusoidal, submarine ridges and troughs oriented roughly 70 ° to the strand line. Almost perpendicular to the ridges and troughs are two linear sand bodies lying at 70 and 90 ft. (21 and 27 m) depths. A thin lamina of clayey silt covers the coarse, clean sands forming this topography. Sediments comprising the sand bodies are distinguishable from surrounding sediments by their larger grain size and higher heavy mineral content. The ridges display coarser, better sorted sediments with higher heavy mineral percentages than the troughs. It is proposed that the sand bodies originated as barrier islands and/or spits during the Late Wisconsin regression. Fluvial action cut the ridge and trough topography into the terrace and the recent, Wisconsin sea level rise modified the topography to its present form. Radiocarbon dates the youngest aspects of the 70 ft. deep sand body as at least 5,000 years B.P.

INTRODUCTION The northeast coast of the Gulf of Mexico is concave seaward with a series of lagoons, bays and estuaries separated from the Gulf by a barrier island complex along its western extremes. Almost centered along this coast is Choctawhatchee Bay in northwest Florida (Fig.l). The Choctawhatchee River, heading in southern Alabama, debauches into the eastern Choctawhatchee Bay where it is forming a delta. The spit separating the bay from the Gulf is situated on truncated Tertiary beds about 100 ft. below sea level and displays sand ridges of over 70 ft. maximum elevation above sea level (TANNER,1964). 1 Present address: Department of Geology, University of Southern California, Los Angeles, Calif. (U.S.A.). Marine Geol., 5 (1967) 299-313





:ig.l. L o c a t i o n m a p .




.- z~24




I 86"[20"






,o,,~. ,




' ie



The area has a humid, semi-tropical climate with an annual rainfall of 125 cm. Surface currents are generally eastward in the summer and westward during the winter (LEIPPER, 1954). Overall longshore drift is to the west (PRICE, 1954) with average breaker height at 35 cm (GoRSLINE, 1966). Marine Geol., 5 (1967) 299-313



Submarine topography The --6t3 and --120 ft. (18 and 36 m) contours roughly parallel the coastline in this area whereas the --90 ft. (27 m) contour forms a broad V whose apex is located just east of the Bay entrance (Fig.2). Between water depths of 60 and 120 ft. the bottom topography consists of ridges and troughs orientated approximately perpendicular to the trend of the --90 ft. contour. These are best defined in deeper water where they are spaced 2,000 ft. (600 m) apart and exhibit a maximum relief in excess of 30 ft. (9 m). Precision depth recorder traces show the ridges to be asymmetrically higher on their southwest facing slopes. To the east the ridges and troughs terminate shoreward against two sand bodies roughly outlined by the --70 and --90 ft. contours respectively (Fig.2). Although almost perpendicular to the shore in the east, the 70 ft. deep sand body abruptly turns westward and subparallels the trend of the continental shelf for 11 nautical miles until directly offshore from the entrance to Choctawhatchee Bay. This deposit, over 1 nautical mile wide, rises almost 20 ft. (6 m) above the normal slope of the continental shelf. The 90 ft. deep sand body, subparalleling the 70 ft. deep sand body, is of lesser aerial extent and height and is more completely dissected.

Previous work Most of the sedimentary investigations in the northeastern Gulf of Mexico have been concerned with deeper water sediments. DIETZ (1963) interpreted the area offshore from Panama City, 45 nautical miles east of Choctawhatchee Bay, as a rough relic shelf on the basis of a submerged forest of partially fossilized stumps in 40 ft. (12 m) of water. LUDWlCK(1964) includes the nearshore sediments o iTChoctawhatchee Bay as part of the Cape San Bias sand facies composed of 90 ~ terrigenous and 10~o carbonate sands whose thickness is estimated to be less than 50 It. (15 m). An alternation is noted in the distribution of the fine- and medium-grained sands until the water depth reaches 120 ft. along his line of samples. GRIFFIN (1962) identifies two major clay mineral suites in the northeastern Gulf of Mexico: (1) Apalachicola River type clay which is dominantly kaolinite, and (2) Mississippi River type clay which is dominantly montmorillonite. Choctawhatchee Bay contains 80-100~ kaolinite whereas the montmorillonite increases to 30-40~o farther offshore.

Field and laboratory investigation Bathymetric, geophysical and sedimentary investigations were carried out in order to study the nature and origin of the topography in this area and to relate these if possible to contemporary sedimentary environments. Samples were collected in October, 1963 from the United States Coast and Geodetic Survey ship "Hydrographer" using a coring device. Twenty-five stations were located by Raydist in an area

Marine Geol., 5 (1967) 299-313



from 1 to 14 nautical miles offshore and in depths ranging from 50 to 123 ft. Samples from these stations are believed to represent all sedimentary environments as interpreted from submarine topography (Fig. 1). The base and data for constructing the topographic map were obtained from the United States Coast and Geodetic Survey Chart 1264 and the precision depth recorder profiles made during the cruise. In conformity with Coast and Geodetic Survey charts, topographic data were measured in feet. The cores were split, photographed, and samples were obtained from the first 2 cm, intervals of 20 cm, and each lithologic change with depth. Silt and clay particles were sized by Coulter Counter; the sands were dry sieved. The statistical parameters of the size distribution were computed by digital computer. Oriented clay slides were made and relative clay mineral abundances were estimated using the ratios of the areas of triangles best fitted to X-ray diffraction peaks (GRIFFIN, 1962). The heavy minerals from the 1.0~0-1.5~ and 3.0~0-3.5~o size fractions were separated from the samples using tetrabromethane and centrifugation. Both heavy and light mineral constituents of the 1.0~-1.5~0 size fraction were identified. The carbonate was analyzed by EDTA titration and organic carbon and nitrogen determined by Coleman analyzers. Three samples of calcareous material were age-dated by 14C at the Radiocarbon Dating Laboratory of the Department of Geology at the University.


The lithologies of the 24 cores, averaging 57 cm in length, show several similarities. All are predominantly mottled in color and consist of light brown-to-gray quartz sand with varying amounts of mollusk and echinoid fragments, except core 23 which has a large silt and clay content and core 19 which displays a large range in texture. A surface layer of silt and clay varying from 1 mm to 3 cm thick is present in a majority of the cores and was included in the top 2 cm textural samples. A layer of reddish brown sand a few centimeters thick was found in tburteen cores. Its elevation in each core ranges from near the top to 44 cm below the surface of the cores. Core 19 (28 cm long) grades downward from a brown, clayey silt at the top to dark, shelly, coarse sand with granule- and pebble-sized limestone rock fragments. A delicate, recent alcyonarian stem was attached to one of the large rock fragments. The rock fragments are well indurated and contain angular quartz grains, wellpreserved small mollusk shells and fragments and other skeletal debris. Calcite is the cementing agent. The sediments are predominantly sand-sized; the mean grain sizes ranging from --1.20~ to 4.8(p and averages 0.9(p (coarse sand by the Wentworth classification). The silt and clay layer located at the surface of a majority of the cores is readily apparent in the size analysis with only core 23 becoming noticeably finer toward the base (Fig.3). Phi standard deviations of the samples range from 3.2 to 1.1 with an Marine Geol., 5 (1967) 299-313


N . J . H Y N E A N t ) H. G . G O O D E L L

!oI ° Z


~-S g o


(.9 Z



'!! (2_




Marine Geol., 5 (1967) 2 9 9 - 3 1 3








Fig.4. Organic carbon and nitrogen distribution.

30 •






[ 0.04 0.08 |40 DEPTH [ 6 0 C M IN CORE NAUTICAL MILES ~'~'','"~-3 4











0 :>



;> Z


N. .I. HYNE A N D H. G. G O O D E L L

average phi standard deviation of 1.8. Phi skewness ranges from t?.8 to --0.6 and averages 0.1 and phi kurtosis ranges from 9.2 to --1.4 and averages 0.9. These sediments are relatively free of heavy minerals, with the 1.0~0-1.5~0 size fraction ranging from 0.0 ~ to 0.4 7/ooheavy minerals and averaging 0. I ~ (predominantly kyanite, magnetite, ilmenite and rutile). The 3.0g,-3.5q) size fraction ranges from 0.1 ~o to 17.5~ and averages 3.4~,,;. Pyritized foraminifera occur in the finegrained core 23. Leucoxene covers magnetite and ilmenite grains in some cores. The light materials consist of quartz, clastic limestone, foraminifera and other biological materials with no appreciable plagioclase, orthoclase, glauconite, muscovite or biotite. Foraminifera are concentrated in the surface samples of most of the cores but are found throughout core 23. Organic carbon ranges from 0.18~ to 1.92~ and averages 0.48~o whereas organic nitrogen averages 0.02~ and ranges from 0.01 ~ to 0.08~. The average carbon-nitrogen ratio is 21.9. Organic carbon and nitrogen appear to be randomly distributed with depth in most of the cores (Fig.4). The fine-grained sediments of core 23 contain less organic carbon and nitrogen than is expected but loss of protein with time is reflected in the increasing carbon-nitrogen ratios with depth in the core. X-ray analysis shows the existance of two clay mineral suites. The first, typified by the surface silt--clay layer, is a kaolinite suite; the second a montmorillonite and illite suite with no measurable amount of kaolinite. Nine cores change from a kaolinite to the montmorillonite-illite suite with depth in the core. In no instance does kaolinite reappear in any measurable quantity after this change occurs.

MacroJbssil ident(fication Although no core contains a complete assemblage, almost all macrofossils belong to surf zone, lagoonal, or nearshore environmental assemblages of PARKER (1960). The only fossil identified directly from the clastic limestone fragments is Abra aequalis SAY. If other species found in core 19 were derived from the clastic limestone, then the assemblage shows many species of similar ecologic affinities which do not appear in the other cores.

14C D E T E R M I N A T I O N S

Three 14C dates were obtained. A fragment of the clastic limestone from 20 to 28 cm in core 19, dated 22,042 years B.P. Two samples were obtained from 33 to 39 cm below the top of core 10 representing the 70 ft. deep sand body. One sample which consists of several well-preserved shells, mainly Macrocallista maculata LINNE, produced a date of 4,918 years B.P. whereas the other composed of fragmented calcareous skeletal material, dated 6,000 years B.P.

Marine Geol., 5 (1967) 299-313








Fig.5. Geological echo profiles.




~ 1 1 0


















.n ©



Y. J. H Y N E A N D H. G . G O O D E L L

CONTINUOUSSEISMICPROFILES The seismic profiles reflect a horizontal subbottom strata beneath the ridge and trough and sand body topography. A buried channel in the eastern part of the area was intersected by four of the profiles (Fig.5). Although the channel is approximately 1,200 ft. (365 m) wide, no indication of it exists on the present bottom topography except where it passes under the partially dissected center of the 70 ft. deep sand body.

DISTINGUISHINGSEDIMENTARYCHARACTERISTICS The characteristics of four sedimentary occurrences were compared statistically using a " t " test with 95 ~ confidence limits used for decision. Twenty-six sand body samples were compared with 45 nearshore sediment samples (Table I) and ten ridge samples with nineteen trough samples (Table II). The sand bodies have significantly larger mean grain sizes and higher heavy mineral percentages than do the nearshore sediments. The ridges display larger mean grain sizes, smaller phi standard deviations, lower silt percentages and higher heavy mineral percentages than do the troughs.

PROPOSED ORIGIN OF THE SUBMARINETOPOGRAPHY The layer of fine sediment cevering the coarse, shelly sands is composed of kaolinite in agreement with the present patterns of clay mineral sedimentation in this area (GRIFFIN, 1962). The kaolinite contributed mainly by the Apalachicola River, to the east of Choctawhatchee Bay, is derived from lateritic soils of the drainage basin. This easily eroded layer overlying the relatively clean sands which form the topography, suggests that the topography is relic and that the agents which formed



Sand bodies

Surrounding marine sediments

Phi mean grain sizO

0.381 --0.57 - 1.44 1.75 1.49 - 1.94 5.490/01 2.37%-15.56%

0,681 --0.07 - 2.33 1.58 1.06 - 2.67 3.69%1 0.61%-17.48%

Phi standard deviation Heavy mineral percentagO 3.00 ~ - 3.50 ~0

average range average range average range

1 Significantdifferenceat 95% confidencelimits. Marine Geol., 5 (1967) 299-313







Phi mean grain size1

0.291 --0.14- 0.39 1.491 1.21- 1.81 0.01 --0.12- 0.22 0.17 %1 0.04- 0.59% 6.95%1 2.06--15.56% 13.90% 1.64-53.38 %

1.161 0.52- 2.33 1.871 1.60- 2.67 0.05 --0.63- 0.53 2.57 %1 0.58- 9.49% 1.40%1 0.49- 3.51% 15.01% 1.97-36.13 %

Phi standard deviation 1 Skewness Silt percentage1 Heavy mineral percentage1 3.00¢p-3.50to Carbonate percentage

average range average range average range aveCage range average range average range

x Significant difference at 95% confidence limits. the ridges, troughs, and sand bodies were effective during a radical environmental change from present conditions. PARKER and CURRAY (1956) observed that while the best-developed submarine terraces and platforms in the northwest Gulf of Mexico occur at 54-60 ft., 186-204 ft., and 270-288 ft. below sea level, a minor grouping of platform tops occurs at about 90 ft. below sea level. EMERY (1958) reports a shallow terrace off southern California at about - - 8 0 ft. and FAIRBRIDGE(1961) states that - - 4 2 to --98 ft. platforms have a world-wide occurrence. A satisfactory geometric and sedimentary model of the sand bodies are barrier islands and spits. Typical of these in this part of the G u l f coast are St. George Island and St. Joseph Spit located 100 miles to the southeast of Choctawhatchee Bay. These barriers are concave landward, over 16 nautical miles long, 1-43.1 nautical mile wide and reach a maximum elevation of 50 ft. (15 m) above sea level. The sediments of St. George Island and St. Joseph Spit (BRENNEMAN and TANNER, 1958) are finer and better sorted than those of the sand bodies off Choctawatchee Bay. Differences in stream competency, wave energy and duration of sea level stands during the Pleistocene may have contributed to this. A comparison between modern environments and the sedimentary parameters of the sand bodies supports a barrier island origin. SHEPARD (1960) found the sediments of Texas barriers to be coarser than the nearshore Gulf deposits and BRADLEY (1957) distinguished between the sub-aerial environment of Mustang Island, Texas, and the marine environment of the lagoon and nearshore Gulf on the basis of greater heavy mineral concentrations on the barrier. Both these criteria are statistically significant on the - - 7 0 and - - 9 0 ft. sand bodies (Table I). However, the sedimentary parameters established for distinguishing barrier islands by SHEPARD (1960) and BRADLEY (1957) would be more definitive with a close association of lagoonal deposits. Seismic Marine Geol., 5 (1967) 299-313


y , J. HYNE AND tl. (i. G O O D E L I ,

profiles from this study indicate an ancient channel behind the - -70 and ....90 ft. sand bodies in a position that would be a logical extension of the Choctawatchee River during a lower sea stand. Lagoonal deposits partially derived from the river should be associated with the sand body barriers. Recent work on both relic and modern barrier islands (reviewed by SANDERS, 1962) supports in tile main GILBERT'S model (1885, 1890) in which barrier islands, formed by littoral drift, do not migrate with rising sea level. In accordance with the Gilbert model, a rise in sea level raises the zone of wave action on a barrier and the barrier responds by accreting and thickening with longshore drift supplying the sediments. When longshore drift is no longer able to maintain the barrier island, it becomes submerged and a new barrier island may form landward. A transition time will exist when the focus of wave action and longshore drift is gradually transferred toward the shallower barrier island. During this time, the old barrier becomes in effect an offshore bar on which shoaling waves may mask or obliterate all surface traces of beach, dune and inner flat environments while retaining the sediment mass. Lagoonal sediments behind the drowned barrier island are covered with the new barrier island sediments, sediments washed over from the old barrier island, and normal nearshore sediments, but are not subjected to excessive erosion. The success of locating and sampling relic lagoonal deposits depends upon the thickness of this sediment cover which is only slightly penetrated by gravity cores because of its compact nature due to relatively large grain size and good sorting. Core 23, from a deep trough dissecting the --90 ft. sand body, consists of 82 cm of dominantly silt and clay with both shallow water and bay species macrofossils. Increasing carbon nitrogen ratios with depth in the core, pyritized foraminifera and the fine-grained lithology indicates that the sediments were deposited in a low energy reducing environment. The clay mineralogy averages 65% kaolinite and 35 ~ montmorillonite, significantly more montmorillonite than reported for the present day Choctawhatchee River and this area of the continental shelf (GmFHN, 1962). The thickness of this deposit and the clay mineralogy makes its deposition on the open continental shelf in the last 5,000 years improbable. This lithology is similar to typical lagoonal sediments (SHEPARD, 1960) indicating that the --90 ft. sand body was washed over its lagoonal deposit during Late Wisconsin sea level rise. The clastic limestone, dated at 22,042 years B.P., occurs as large fragments at the bottom of core 19 and fits the descriptions of beach rock (G~NSBURG, 1953) although the calcite cement is probably due to solution and recementation. The core texture suggests that it sampled a slump deposit from a beach rock ledge underlying the --70 ft. sand body. During sea level rise in the Late Wisconsin, remnants of an ancient barrier or marine terrace supported by the beach rock would act as a focus for wave attack and littoral drift resulting in the subsequent draping of new barrier sediments over the beach rock core. 14C age dates of 6,000 and 4,918 years B.P. on shell material from the near surface of the --70 ft. sand body probably represents the drowning stage of the sand bodies when they were assuming the form of offshore bars and receiving younger, marine sediments. Marine Geol., 5 (! 967) 299-313


31 1

Of the 24 cores, nine change from a kaolinite to a montmorillonite-illite suite with depth; four have only a montmorillonite-illite suite; and three contain more montmorillonite-illite than the kaolinite suites of the present Choctawhatchee and Apalachicola Rivers. West of the study area, the Alabama River system contributes a clay mineral suite of approximately equal quantities of kaolinite and montmorillonite (GRtFFIN, 1962). The inclusion of western clay mineral suites and the coarser, less sorted sediments of the sand bodies compared to those of St. George Island and St. Joseph Spit, substantiates a possible western source for some of the relic barrier sediments. The --70 and --90 ft. sand bodies, implanted on the western flank of the northeast trending --100 to --70 ft. contours as they are indented coastward (Fig.2), may represent the termination of an eastern paleodrift system resulting from the refraction of east-west wave fronts. Although a sand wave origin for the ridge and trough topography was considered, lack of modification of the sand body crests into ridges and troughs, dissimilarities between the sedimentary characteristics of this area and those established for the sand waves on Georges Shoal (STEWARTand JORDAN, 1964) and lack of sufficient tidal amplitudes or orientation led us to alternative mechanisms for their origin. The general aspect of the ridge and trough topography appears strikingly similar to the sub-aerial, sub-parallel stream erosion patterns developed on the Pleistocene and Recent marine terraces of the coastal plain. GREMILLmNet al. (1964) described the drainage pattern on the high terraces of Florida, where many small streams flowing perpendicular to the present shoreline intersect larger streams flowing parallel to the coastline following the deepest parts of ancient lagoons. VERNON(1942) also described the trellis-like stream drainage on the terraces just landward of the study area. Here small consequent streams flow down the terrace slopes perpendicular to the coastline. Unable to breach the relic beach ridges, they coalesce forming larger streams which parallel the coastline until eventually crossing the beach ridges at low points. If the ridge and trough topography is of fluvial origin, then the problem of the preservation of these features during Wisconsin sea level arises. Wave attack would be concentrated on the southern end of the southeast oriented ridges and symmetrical ridges would be eroded on their western flanks with longshore currents depositing sediments on the eastern sides of the ridges producing an asymmetry similar to that noted in the submarine topography of this area. Ridges so modified would exhibit the sedimentary characteristics noted in this investigation: higher concentrations of heavy minerals, larger mean grain sizes, smaller phi standard deviations and lower silt contents than the troughs. With this proposed model a sequence of events may be postulated: (1) During the Late Wisconsin regression barrier islands and beach rock were deposited on a marine terrace which at present is less than 100 ft. (30 In) deep. Some of these sediments may have been derived from the west. (2) Sub-parallel, fluvial action cut a ridge and trough topography on the terrace and partially dissected the barriers. (3) During the Late Wisconsin transgression both barriers were modified and Marine Geol., 5 (1967) 299-313



subsequently drowned. A large sediment mass of the deepest barrier ( -90 ft. or - - 2 7 m sand body) was washed over lagoonal deposits, which were partially derived from the Choctawhatchee River, and over the ridge and trough t o p o g r a p h y immediately shoreward. The shallower barrier ( - - 7 0 ft. or --21 m) was less modified possibly due to its protected position and/or beach rock core and finally drowned 5,000 years ago. (4) The ridge and trough t o p o g r a p h y was subjected to wave action resulting in the asymmetrical shapes and sedimentary characteristics which it now displays. (5) A fine layer of silt and clay from the Apalachicola and Choctawhatchee Rivers was deposited over the topography.

ACKNOWLEDGEMENTS The authors are indebted to the United States Coast and Geodetic Survey for their sponsorship of this study, especially Mr. J. W. Kofoed. C o m m a n d e r W. R. R a n dall and the officers and crew of the United States Coast and Geodetic Survey ship " H y d r o g r a p h e r " are due special thanks for their patience, efficiency, and professionalism displayed during the coring operations. Mr. A. Weeks and Mr. R. Harbinson made the seismic profiles. This paper is part of a Master's thesis submitted to the G r a d u a t e School of the Florida State University by Mr. N o r m a n J. Hyne.


BRADLEY,J. S., 1957. Differentiation of marine and subaerial environments by volume percentage of heavy minerals, Mustang Island, Texas. J. Sediment. PetroL, 27 : 116-125. BRENNEMAN,L. and TANNER,W. F., 1958. Possible abandoned barrier islands in Panhandle, Florida. J. Sediment. Petrol., 28 : 342-344. DETZ, R. S., 1963. Wave-base, marine profile of equilibrium and wave-built terraces: a critical appraisal. Bull. Geol. Soc. Am., 74 : 971-990. EMERY,K. O., 1958. Shallow submerged marine terraces of southern California. Bull. GeoL Soc. Am., 69 : 39-60. FAIRBRmGE,R. W., 1961. Eustatic changes in sea level. Phys. Chem. Earth, 4 : 99-185. GmBERT, G. D., 1885. The topographic features of lake shores. U. S., Geol. Surv., 5th Ann. Rept., pp.69-123. GmBERT,G. D., 1890. Lake Bonneville. U. S., Geol. Surv., Monograph, 1 : 438 pp. GINSBURG,R. N., 1953. Beach rock in South Florida. J. Sediment. Petrol., 23 : 85-92. GORSLINE,D. S., 1966. Dynamic characteristics of West Florida Gulf Coast beaches. Marine Geol., 4 : 187-206. GREMILUON,L. R., TANNER,W. F. and HUDDLFSTUN,P., 1964. Barrier-and-lagoon sets on high terraces in the Florida panhandle. Southeastern Geol., 6 : 31-36. GRIFF1N, G. M., 1962. Regional clay mineral facies--products of weathering intensity and current distribution in the northeastern Gulf of Mexico. Bull. Geol. Soe. Am., 73 : 737-768. LEn'PER,D. F., 1954. Physical oceanography of the Gulf of Mexico. U. S. Fish Wildlife Serv., Fishery Bull., 89 (55) : 89-100. LtroW~CK, J. C., 1964. Sediments in northeastern Gulf of Mexico. In: R. L. MILLER(Editor), Papers in Marine Geology, Shepard Commemorative Volume. Macmillan, New York, N.Y., pp.204-235. Marine Geol., 5 (1967) 299-313



PARKER,R. H., 1960. Ecology and distributional patterns of marine macroinvertebrates, northern Gulf of Mexico. In: F. P. SrmPARO, F. B Pr|LEOER and TJ. H. VAN ANDEL (Editors), Recent Sediments, Northwest Gulf of Mexico. Am. Assoc. Petrol. Geologists, Tulsa, Okla., pp.302-327. PARKER, R. H. and CURRAY,J. R., 1956. Fauna and bathymetry of banks on the continental shelf, northwest Gulf of Mexico. Bull. Am. Assoc. Petrol Geologists, 40 : 2428-2439. PRICE, W. A., 1954. Shorelines and coasts of the Gulf of Mexico. U. S. Fish Wildlife Serv., Fishery Bull., 89 (55) : 38-62. SANDERS, J. E., 1962. Effects of rise of sea level on established barriers. Meeting Geol. Soc. Am., Houston, Texas, 12 Nov., 1962. SHEPARO,F. P., 1960. Gulf coast barriers. In: F. P. SHEPARD,F. B PHLEGER and TJ. H. VAN ANDEL (Editors), Recent Sediments, Northwestern Gulf of Mexico. Am. Assoc. Petrol. Geologists, Tulsa, Okla., pp.197-220. STEWART, H. B. and Jordan, G. F., 1964. Underwater sand ridges on Georges Shoal. In: R. L. MILLER (Editor), Papers in Marine Geology, Shepard Commemorative Volume. Macmillan, New York, N.Y., pp. 102-116. TANNER,W. F., 1964. Nearly ideal drift system along the Florida panhandle coast. Z. Geomorphologie, 8 : 334-342. VERNON, R. O., 1942. Geology of Holmes and Washington Counties, Florida. Florida, GeoL Surv., GeoL Bull., 21 : 161 pp.

Marine Geol., 5 (1967) 299-313