GULF RESEARCH REPORTS Published by the GULF COAST RESEARCH LABORATORY Ocean Springs, Mississippi November 1973 Gulf Research Reports Volume 4 Issue 2 January 1973 A Quarter Century of Geology at the Gulf Coast Research Laboratory (1948-1973) Ervin G. Otvos Jr. Gulf Coast Research Laboratory DOI: 10.18785/grr.0402.01 Follow this and additional works at: http:// aquila.usm.edu/ gcr Part of the Geology Commons, and the Marine Biology Commons Recommended Citation OtvoS; E. G. Jr. 1973. A Quarter Century of Geology at the Gulf Coast Research Laboratory (1948-1973). Gulf Research Reports 4 (2); 151-165. Retrieved from http:/ / aquila.usm.edu/ gcr/vol4/iss2/ 1 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. A QUARTER CENTURY OF GEOLOGY AT THE GULF COAST RESEARCH LABORATORY (1948-1 973) by ERVIN G. OTVOS, JR. Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 After its founding at the Mississippi Academy of Sciences meet- ing on May 3, 1947 in the Buena Vista Hotel, Biloxi, Mississippi, the Laboratory offered two biological courses during its first summer ses- sion but no geology. Geology has been an integral part of Gulf Coast Research Laboratory since its second year of existence. During the first few years the abandoned Civilian Conservation Corps camp in Magnolia State Park, east of Ocean Springs, was used for housing the Laboratory, The activity was directed by Dr. R. L. Caylor, Chairman of the Science Division, Delta State Teachers College in Cleveland, Mississippi. An assortment of temporary buildings served as resi- dence, office, library and dining hall. From surplus army equipment additional dormitory buildings were constructed and vessels of the Mississippi Sea Food Commission were borrowed at Ocean Springs Harbor for field work. The Mi.ssissippi Board of Trustees of Institu- tions of Higher Learning was induced to assume responsibility for the Laboratory and the State Legislature established it as a state institu- tion in 1948 and appropriated ten thousand dollars for its operation in 1948-50. Doctor Caylor .served for several years as director of the Labora- tory. He was only on the grounds permanently during the summer. He invited Dr. Richard R. Priddy, Chairman of Geology at Millsaps College in Jackson (1948-1972) to help develop a geology teaching program and join the Executive Board of the Laboratory, Priddy’s participation had a profoundly beneficial effect on the geological work for years to come. With youthful enthusiasm and interest in his sub- ject he gradually overcame the problems created by the lack of equip- ment, primitive, inconvenient facilities and his initial unfamiliarity with coastal geology. Makeshift instruments (sediment corers, oxy- gen and pH-meters, etc.) had to be manufactured. In the coming de- cades practically all Millsaps geology majors and several chemistry majors attended geology courses offered during the summers. These courses, in fact, were made a recommended summer field camp sub- stitute for the Millsaps geology majors. While in 1947 only botany and marine biology were taught, in August of 1948 geology became the third established course and a two-week marine sedimentation problems class was taught by Priddy with the assistance of C. P. Marion, a graduate student at Mississippi 151 152 Otvos State College. Between 1949 and 1955 the course was called Marine Sedimentation. After 1948 all geology courses lasted three weeks and carried three hours credit. The four students in 1948 (C. A. Barton, E. R. Campbell, F. G. Clark and W. E. Cook) were all from Millsaps and students from this college were always very well represented in the geology classes of the following summers (Table 1). In the early years sediment sampling from the Biloxi Bay bot- toms and the surrounding beaches occupied the course, along with the description and grain size analysis of the samples. Later the class projects became more varied and carried students to many different locations along the Mississippi shore. Marion taught the geology course during the summers of 1949-50. In 1952 Arthur T. Allen of Emory University was the instructor and during the following three summers Dr. Olin T. Brown of Mississippi Southern College (later University of Southern Mississippi) taught the course. Priddy taught in 1951 and again from 1955-58. In those formative years all profes- sors practically donated their efforts and time. In the beginning the teaching and research work at the Laboratory was largely restricted to the summer months. As late as 1954 the year-round staff consisted of three part-time clerical and maintenance personnel and only two full-time .scientific staff members, including Dr. A. E. Hopkins, the director who had become the first full-time director in 1952, It was not until 1961 that the first full-time geologist assumed his duties. The physical expansion of the Laboratory and its geological facil- ities proceeded slowly but steadily. With the help of the Executive Secretary of the Board of Trustees of Institutions of Higher Learn- ing, the Laboratory acquired the Smart property on Davis Bayou, a mile and a half south of the facilities in Magnolia Park. The purchase price was $35,000. A 2V2-story main building (Fig. 1) that be- came known as the “Big House,” was a former summer home that had been constructed in 1900. The surrounding 49 acres of land was a beautiful .setting on a small peninsula bounded by Halstead, Stark and Davis bayous. In 1950, the “Big House” was occupied by Labora- tory personnel and a fii'e in the Magnolia Park facilities during Au- gust of that year hastened the completion of the move to the Smart property in 1951. One of the auxiliary structures behind the site of the future Hop- kins Building, by the boat harbor, became the “mud shack”, a one- room geological classroom-laboratory building. This had been a form- er servant’s quarters. Between 1961 and 1965 the small kitchen and pantry of the “Big House” were used for the geology office and stor- age room. With the exception of two wooden dormitories, Laboratory buildings were clustered on low ground along Davis Bayou. The dor- mitories were reassembled on the slopes of the “Hill”, the densely wooded, flat-topped “hinterland” which formed the center of the small 3 a iM oi Ci a u 0 j:} ct X ij u Vi a Cfl a> 01 cc 03 O U 3 o Cfl V GQ cm « 1 3 a> C g g s c« IV u c 03 ■c c V CM ■<3- CM o VO I-H CO LO I-H i-H cn i-H i-H i-H vO X cn 00 LO LO CO CM LO CM ON i-H CM O O CO CO LO »-H CM I-H ''O' t-H O CM Ht Ht Ov CM X CM vD CM VO o i-H i-H CO I-H VO CO LO o LO CM CM o 00 o CO 00 cn CM »— 1 C30 CTN 1— H vD 00 CM CM 00 CM 1—1 o 1-H CM •JC ■-H i-H o 'cr o o CO VO CM CO vD •—1 CM LO Hj- o CM X •K CM CO /-s X X Pi o • Pi O I — 1 X o 3 p 3 o 3 3 3 • 'H 3 X P 3 3 X 3 X 3 X X 3 3 3 P •3 3 •H 3 X X S 0 Du 5C 50 3 cn CO Pi I-H X 3 X C X 1 — \ 3 X CO 3 ‘H 3 O X 34 3 Pi 3 B o •H X P •H O X CJ) Pu ‘H X X 3 3 o P. 3 P 3 3 X 3 •H 3 o 3 •K 3 3 •H CO X s«/ 3^ 3 3 3 3 3 •H 3 X 3 X 3 P X X 3 •H o X o 3 < X X 3 a s X X o X •H V-*' • 'W' • X o C-J) S p p O H CO CU CO u 3 o a >> 50 o o 0) c •H }-i r-( C« U •H CO X 0-1 CO CO X Qi ^ H CO X X d § 0 1 — I OJ [i4 0) CO * CO cd 3 0) •-* O 4J O CO CO O 0 o 0) !-• OJ o OJ o 3 C Q) 0^ O •H 50 O rH o s CU a •H 5-1 CO O 0) 4-1 C 3 - •H CU X M TD 5-1 £0 3 0 S 4-1 J*-* CO ate Pleistocene (Sangamon) Waveland-Gulfport- Belle Fontaine mainland barrier ridge system is now being investi- gated in great detail, Other held studies revealed that there is no geo- logical proof for the existence of earlier Pleistocene, higher coastal A Quarter Century of Geology 161 Figure 7. Physical Marine Geology class in the Sedimentation Laboratory, Oceanography Building, 1971. Mary Whitten (left) and Norma Weaver, University of Mississippi students with author. (Photo by Catherine Campbell) ridges, previously widely reported from several areas of the north- eastern Gulf coastal region. An extensive core-drilling program, so e.ssential in coastal re- gions with few natural outcrops, was initiated in October 1971 to solve the numerous stratigraphic-geomorphological problems. The most important result of this project to date has been the recognition of a widespread, continuous, shallow stratigraphic unit under the Sangamon-agc barrier ridge zone and under certain areas of the Late Pleistocene coastwise alluvial plain. This fossiliferou.s unit (named Biloxi Formation) appears to represent the marine transgression phase of the Sangamon Interglacial and is correlative probably along the whole northern coast of the Gulf of Mexico. Plans also call for stratigraphic exploration drilling in Mississippi Sound and on the off- shore barrier islands in order to learn more about the pre-Pleistocene, Pleistocene and Holocene history of the north-central Gulf coastal- nearshore area and its relationship with the .subsiding zone of the Mississippi River Delta region. Geological field work including future detailed boring in Missis- sippi offshore barrier islands and the south Hancock County marsh- beach ridge complex should clarify the Holocene history of the present 162 Otvos offshore-inshore area and the changes (crosional destruction, island migration, accretion) which profoundly affect them. Long-range ef- fects of natural (hurricanes) and man-made changes on these fea- tures and the resulting effects on the mainland coast and the Missis- sippi Sound are not yet adequately knowui. Such geological studies would contribute not only to the understanding of the formation and destruction of barrier islands and mainland barrier beaches during the Late Quaternary in general, but also to the clarification and possi- ble solution of serious environmental problems along the Mississippi Coast. Results of geological research of the past few years were demon- strated in May 1973 during a two-day field trip of the New Orleans Geological Society. The trip covered a wide area between New Or- leans East and Dauphin Island, Alabama. A guidebook published for the occasion summarized the present knowledge about the Miocene- Recent coastal formations and the presently active geological pro- cesses, At the request of the Commission on Shorelines of the Interna- tional Association for Quaternary Research, we are also collaborating in supplying regional information for the preparation of a set of world shoreline maps, covering the Late Pleistocene, Late Holocene and Recent shore positions. ACKNOWLEDGEMENTS Dr. Gordon Gunter, Director Emeritus and Professor of Zoology at the Laboratory, encouraged the writing of this account and re- viewed the manuscript. Doctor Gunter and Doctor Snowden supplied data used by the author as did Doctor Priddy, whose mimeographed descriptions of the geological activities at Gulf Coast Research Lab- oratory were invaluable in the preparation of this article. Michael Bograd greatly assisted in the tabulation of the student attendance data. BIBLIOGRAPHY (Geological publications and the.ses associated with the Laboratory) (1) Papers, books and reports Clark, S, F. 1960, Behavior of some Gulf Coa>;t muds as compared to some high- way clays‘ Jour. Miss. Acad. Sci., Vol. 6 (1954—60), pp. 197-201. Foxworth, R. D. and B. Z. Ellis. 1960. Preliminary study of heavy minerals of Mississippi Sound; Jour. Miss. Acad. Sci., Vol. 6 (1954-00), pp. 217-220. Fo.xworth, K. D., R. R. Priddy, W. B. Johnson and W. S. Moore. 1962. Heavy min- eral.s of sand from recent beaches of the Gulf Coast of Mi.ssissippi and asso- ciated islands: Miss. State Geol. Survey Bull., No. 93. A Quarter Century of Geology 163 Franks, J. S., J. Y. Christmas, W. h. Silur, R, Combs, R. Waller and C. Burns. 1972. A study of nektonic and benthic faunas of the shallow Gulf of Mexico off the State of Mississippi as related to some physical, qhcmical and geologi- cal factors 1 Gulf Research Reports, Vol. 4, No. 1. Holladay, C. O., J. H, Foster, J, H. Hetrick, Jr. and R. D. Foxworth. 1957. A study of the shifting of Marsh Point Sand Spit: Jour. Miss. Acad. Sci., Vol. 6 (1954-60), pp. 203-206. Otvos, E. G., Jr. 1970. Development and migration of barrier islands, northern Gulf of Mexico: Reply; Geol. Soc. Amcr. Bull., Vol. 81, No. 12, pp. 3783 88. 1971, Relict eolian dunes and the age of the “Prairie” coastwise terrace, southeastern Louisiana: Geol. Soc. Amer. Bull., Vol. 82. No. 6, pp. 1753-58. 1971. (with J. Pinson, B. W. Brown, C, M. Hoskin, T. J. Laswell, et al.) , Geology section, p, 20, in Earth resources data and technological studies relevant to the cultural, physical and economic development of the State of Mississippi, Task 1 Contract NAS9-11607, Final Report, (also individual geology report). _. 1972. Pre-Sangainon beach ridges along the northeastern Gulf Coast- fact or fiction?: Gulf Coast Assoc. Geol. Soc. Trans., Vol. 22, pp. 223-228. 1972. Mississippi Gulf Coast Pleistocene beach barriers and the age problem of the Atlancic-Gulf Coast “Pamlico” — “Ingleside” beach ridge sys- tem : Southeastern Geology, Vol. 14, No. 4, pp. 241-250. 1973. Genetic and age problems of the Moreau-Caminada Holocene coastal ridge complex: Southeastern Geology, Vol. 15, No. 1, pp. 37-43. 1973. Geology of the Mississippi-Alabama coastal area and near-shore zone: Field Trip Guidebook: New Orleans Geol, Society, New Orleans. 1973. Inverse beach sand texture^ — coastal energy relationship on Missis- sippi-Alabama Coast barrier islands: Jour. Miss. Acad. Science, Vol. 18 (in press) . 1973. Phase 111; Scdimcntology, in Cooperative Gulf of Mexico Estua- rine Inventory and Study, Mississippi; State of Mississippi Marine Conserva- tion Commission and Gulf Coast Research Laboratory. Ed.i J. Y. Christmas, pp. 123-137. Price, J. B. and G. Gunter. 1964. Studies of the chemistry of fresh and low'-salinity waters in Mississippi and the boundary between fresh and brackish watgr: Int. Revue ges, Hydrobiol., Vol. 49, No. 4, pp. 629-636. Price, J. B. and R. R. Priddy. 1959. Colorimetric determinations of nitrite and nitrate nitrogen in brackish coastal waters: Bull. Mar. Sci. Gulf and Carib- bean, Vol. 9, No. 3, pp. 310-314. 1961. Rapid volumetric determination of sulphate, calcium and magne- sium in high lime-magnesian brackish coastal waters: Bull. Mar. Sci. Gulf and Caribbean, Vol. 11, No. 2, pp. 198-206. Priddy, R. R. 1954. Recent Mississippi Sound sediments compared with some Up- per Cretaceous sediments: Gulf Coast Assoc. Geol. Soc, Trans., Vol. 4, pp. 159-167. Priddy, R, R, and R. M, Crisler, Jr. 1955. Preliminary survey of sediments in parts of the Mississippi Sound: Jour. Miss. Acad. Sci., Vol, 5 (1951-53), pp. 226- 230. Priddy, R. R., R. M. Crisler, Jr., C, P, Sebren, J, D. Powell and H. Bui-ford, 1955. Sediments of Mj.ssissippi Sound and inshore waters: Miss, State Geol. Surv. Bull., xNo, 82. Priddy, R. R. and B. L. Smith. 1960. Recent sedimentation on Horn Island, Missis- sippi: Gulf Coast Assoc. Geol. Soc. Field Trip Guidebook, pp, 5-8. Snowden, J. 0., Jr. 1965, Settling velocities of selected clay minerals in brackish and normal sea water; Jour. Miss. Acad. ,Sci., Vol. 11, pp. 123-126. Upshaw, C. F,, W. B. Creaih, and F. L. Brooks. 1966. Sediments and microfauna off the coasts of Mississippi and adjacent states: Miss. State Geol. Survey Bull., No. 106. 164 Otvos (2) Published lecture abstracts DeVries, D. A. 1969. Grass-balls formed on a Mississippi beach during Hurricane Betsy: Geol. Soc. Amer. Abstracts with Programs, 1969, Pt. 6, North-central Section, pp. 10-11. DeVries, D. A. and R. R. Priddy. 1964. An eight-year study of Marsh Point Sand SpiL Jour. Miss. Acad. Sci., Vol. 10, p. 182. Moore, W. S. 1962. Eh-pH relations in sulfide deposition in Mississippi Sound: Jour. Miss. Acad. Sci„ Vol. 8, p. 100, Moore, W. S. and J. O. Snowden, Jr. 1967. Sulfates in Mississippi coastal waters: Jour. Mhss, Acad. Sci., Vol. 13, pp. 59-60. Otvos, E. G., Jr. 1972. Pre-Sangamon beach ridges along the northeastern Gulf Coast-fact or fiction?: Bull. Amer. Assoc. Petr. Geologists, Vol. 66, No. 9, p. 1901. Price, J. B. and R. R. Priddy. 1960. Colorimetric determination of nitrite and nitrate nitrogen in brackish coastal waters: Jour. Miss. Acad. Sci., Vol. 6 (1954-60), p. .349. 1960. Rapid volumetric determination of calcium and magnesium in brackish coastal waters: Jour. Miss. Acad. Sci., Vol, 6 (1954-60), p. 351, 1960. Rapid volumetric determination of sulfate, calcium and magne- sium in high lime-magnesium brackish coastal waters: Jour. Miss. Acad. Sci-, Vol. 6 (1954-60), p. 399. Price, J. B., R. R. Priddy, W. S. Moore and W. 1. Luke. 1962. Investigation of flocculent materials in Mississippi Sound: Jour. Miss, Acad, Sci., Vol. 8, p. 111. Priddy, R, R. 1955. Humu.s of Mississippi Sound: Geol. Soc. Amer. Bull., Vol. 66, No. 12, Pt. 2, p. 1717, 1960. Additional studies of Gulf Coast sediments: Jour. Miss. Acad. Sci., Vol. 6 (19.54-60), p. 58. 1962. Mississippi Sound heavy minerals compared to other “heavies” of the Gulf Coast and South Atlantic Coast: Jour. Miss. Acad. Sci., Vol. 8, pp. 102-103. Priddy, R. R,, R. M. Crisler and H. Burford. 1954, Sediments of parts of Missis- sippi Sound: Geol. Soc. Amer. Bull., Vol. 65, No 12, Pt. 2, pp. 1366-67. Priddy, R. R., R. J, Gentile and R H. Lyons. 1964. Observations on the sulfur cycle in some Mississippi Sound muds: Jour. Miss, Acad. Sci., Vol. 10, pp. 49- .50. Priddy, R. R, and E. J. Johnson. 1966, Sulfur and sulfur bacteria in some Missis- sippi Sound muds; Geol. Soc. Amer. Special Paper, No. 87, p. 260. Snowden, J. 0., Jr. 1961. Recent sedimentation of Biloxi Bay, Mississippi: Progr. Ann. Mtiig. Geol. Soc. America, Cincinnati. Snowden, J. O., Jr. and E, Otvosi Jr. 1971, Chemical water quality and sedi- ment-water relations in Louisiana and Mississippi estuaries: Abstl*. Second Coastal and Shallow Water Research Conference (Univ. of Southern Cali- fornia, University Press), p. 215. 1972. (In Press). Chemical changes in interstitial sediment water and clay mineralogy in Louisiana-Mississippi estuaries: Clays and the Marine En- vironment, Proc. of the 21st Conf. Clay Minerals Soc. (3) Masters Theses Barton, C. A. 1952. The sediments of Biloxi Bay, Mississippi: M. A. Thesis Uni- versity of Illinois. Foxworth, R. D. 1958. Heavy minerals of sand from recent beaches of the Gulf Coast of Mississippi and associated islands: M. A. Thesis, University of Mis- souri. Lindsey, R, H., Jr. 1962. A study of recent forminifera along a salinity gradient A Quarter Century of Geology 165 through the Mississippi Sound, Mississippi: M. S. Thesis, University of Mis- sissippi. Marion, C. P. 1951. A study of recent marine sediments in the Biloxi-Ocean Springs area of the Mississippi Gulf Coast: M, A. Thesis, Mississippi State College. Robbins, W. H. 1961. Ecology of foraminifera of Biloxi Bay, Mississippi: M. S. Thesis, University of Missouri. Snowden, J. O., Jr. 1961. Geological and chemical environment of Biloxi Bay, Mis- sissippi: M. A, Thesis, University of Missouri. (4) Papers in Preparation Otvos, E. G., Jr. Inverse beach sand texture — wave energy relationship, north- east Gulf Coast barrier islands. A Holocene beach ridge system (“Hancock County Chenier”), South- western Mississippi Coast. Sediments and sediment chemistry of Holocene St. Louis Bay deposits. A late Pleistocene transgressive unit, northern Gulf Coast (Biloxi For- mation) . Holocene history of the Mississippi-Alabama coastal area. Development and recent history of Mi.ssissippi-Alabama coastal barrier islands. The late Holocene Hancock-New Orleans coastal-submarine ridge sys- tem. Otvos, E. G., Jr. and J. O. Snowden, Jr. Geology of the Back Bay of Biloxi. Snowden, J. O. and E. G. Otvos, Jr. Chemical water quality and sediment-water reactions in Louisiana and Mississippi estuaries: Louisiana Water Resources Research Institute Techn. Rprt. No. 5. Louisiana State Univ,, Baton Rouge, La. Gulf Research Reports Volume 4 Issue 2 January 1973 The Occurrence of the Remarkable Scyphozoan, Deepstaria enigmatica, in the Gulf of Mexico and Some Observations on Cnidarian Symbionts Philip J. Phillips Texas A6’M University DOI: 10.18785/grr.0402.02 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Phillips, P. J. 1973. The Occurrence of the Remarkable Scyphozoan, Deepstaria enigmatica, in the Gulf ofMexico and Some Observations on Cnidarian Symbionts. GulfResearch Reports 4 (2): 166-168. Retrieved from http:// aquila.usm.edu/gcr/vol4/iss2/2 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. THE OCCURRENCE OF THE REMARKABLE SCYPHOZOAN, Deepstaria enigmatica, IN THE GULF OF MEXICO AND SOME OBSERVATIONS ON CNIDARIAN SYMBIONTS by PHILIP J. PHILLIPS^ Department of Biology Texas A&M University College Station, Texas 77801 On 8 July 1965 one damaged specimen of the remarkable scypho- zoan, Deepstaria enigmatica Russell 1967, was taken in the course of a sampling program conducted by the Department of Oceanography, Texas A&M University in a ten-foot Issacs-Kidd midwater trawl in the Yucatan Basin. The trawl was put into water at Lat. 19° 58' N, Long. 85° 14' W. This is the first report of Deepstaria in the Atlantic region. This medusa is otherwise known only from the Pacific Ocean where it has been captured by slurp gun from the submersible Deep- star over the San Diego Trench and has been taken in midwater trawls by the Scripps Institution of Oceanography Mid-Pacific Ex- pedition (Barham and Pickwell 1969 and Russell 1967). The present specimen, although badly torn, is clearly recogniz- able as D. enigmatica. The anastomosing canal lattice has the charac- teristic pattern described by Russell. The medusa is a deep purple- blue in color and appears to have had a diameter in excess of 70 cm. The mesogloea is 18 mm. thick near the center of the disc. Oral arms are not discernible and portions of the disc are considerably distorted by local extreme contraction. Fragments of female gonadal material are attached to parts of the subumbrellar surface. These gonadal frag- ments, too small to permit any observations on gross gonad morphol- ogy! contain zygotes in various stages of early cleavage. Ova and zygotes are small (c. 100 microns diameter) and have very little, if any, yolk. The depth at which this jellyfish was taken cannot be stated with any certainty since the collecting device fi.shed from the surface to a depth of 2400 m., no '‘at depth" closing device having been used. The occurrence of this unique scyphomedusan in the Gulf of Mex- ico constitutes a considerable range extension. Closer examination of midwater trawl samples from other regions and more extensive sam- pling will probably lead to the discovery of more specimens of Deep- staria from other areas, I strongly suspect that D. enigmatica may have a circumglobal di.stribution in tropical and subtropical waters. As is the case with many oceanic organisms, the zoogeographic dis- 1 Present address; North Atlantic Division, U.S. Army Corps oX Engineers, 90 Church Street, New York, N. Y. 10007. 166 Deepstaria enigmatica in the Gulf 167 tribution of Deepstaria can probably be correlated with intensity of sampling and workers who can identify it. Barham and Pickwell discuss a commensal or symbiotic associa- tion between the giant parasitic isopod Anuropus and Deepstaria, Photographs taken by Deepstwr show one of these isopods clinging to the subumbrellar surface at the time of capture. Barham and Pick- well state that theTnedusa ‘Mid not pulsate or show normal swimming movements. Instead the medusa appeared flaccid and seemed to be floating passively.” Ru.ssell noted the absence of stomach, epithelial lining and parts of the coronal muscle, and suggested the possibility the medusa was moribund at the time of capture. Taking into account the occurrence of nematocysts in the stomachs of some anuropids (as reported by Menzies and Dow 1958 and cited by Barham and Pick- well) Barham and Pickwell suggested that the isopod feeds on the jellyfish, incapacitates medusa movement and creates a “floating pro- tective environment,” which they state is a process that stops “some- what short of the reduction of salps and pyrosomes into thin-walled houses by the well known amphipod, Phronma sedenteria'' It should be noted that Barham and Pickwell did not demonstrate nematocysts in the stomach contents of the Anuropvs in association with Deep- staria, While the isopod may be traveling on an atenic host or is ac- tually a true parasite of Deepstaria there are no criteria for establish- ing just what can be considered normal swimming movement for this medusa, if indeed it noi*mally does generate swimming movements, I find it difficult to believe that Anuropus incapacitates the medusa since my own personal and published observations (Phillips, Burke and Keener 1969) indicate that medusae can serve as hosts for a wide variety of crustaceans and other metazoans without incapacitation. In actual fact the moribund state will render an organism prey for a wide range of animals which would not normally attack it, as may possibly be the case for Deepstaria, Additionally, it is exceedingly dubious that Phronhna invades salps or pyrosomes and converts them into thin-walled houses. The structure of the amphipod house bears little semblance of that of any tunicate and the resemblance is a superficial one at best according to Leo Berner, Jr., (personal communication). It is much more likely that a Phronima secretes its own domicile. Examination of Phronima houses evinces no evidence of zooids of any type ever having been em- bedded in the gelatinous matrix. There is also a very striking correla- tion between house proportions and amphipod size, a situation that would not be so well defined if the Phronima did not secrete it. Symbiotic relationships between Scyphozoa and other metazoans are very common. Pickwell and Barham cite seven such relationships and their list can be considerably lengthened. Fish-jellyfish associa- tions are probably the best known (Marisueti 1963 and Phillips, Burke and Keener 1969) and many are known that involve Crustacea and 168 Phillips medusae (Outsell 1928 and Phillips, Burke and Keener 1969). It should be noted that there are trematode and cestode parasites of medusae (Dollfus 1931 and Stunkard 1969) as well as vermiform parasites of unknown phyletic aflfinilies which are commonly found in some rhizostomes (Moestafa and McConnauj^hey 1966 and Phillips and Levin, in prep.), I have also found on very rare occasions hyperiid amphipods embedded in the mesogloea of two species of calycophoran siphonophores (Diphyes dispar and Chelophyes appendiculata) . Me- tazoan parasites or symbionts of Cnidaria are not at all rare; they are merely seldom looked for and even less seldom investigated. It remains to be seen whether or not life cycle completion in Anuropm is dependent on cnidarian hosts. Data presented by Barham and Pickwell regarding the association between Annropus and Deep- stama are insufficient for drawing conclusions. ACKNOWLEDGEMENTS I would like to thank Dr. Leo Berner, Jr. (Department of Ocean- ography, Texas A&M University) and Dr. Sewell H. Hopkins (De- partment of Biology, Texas A&M University) for critical review of the manuscript. This study was supported in part by NSF Grant GP 3555. The Gulf of Mexico Deepstaria enigmatica specimen was made available to me through the courtesy of Dr. W. E. Pequenat and Dr. L. Berner, Jr. and has been placed in the U. S. National Museum, Washington, D. C., U.S.A. This paper in modified form was part of a dissertation submitted in partial fulfdlment of the requirements for the Ph.D. degree in biology at Texas A&M University. LITERATURE CITED Barham, E. G. and G. V. Pickwell. 1969. The giant isopod, Anuropus: a scypho- zoan symbiont. Deep Sea Research. 16:525-9. Dollfus, R. P. 1931. Nouvel addendum a mon “Enumeration des cestodes du planc- ton et des invertebres marins” (1) Annales de Parasitologie Humainc et Com- paree. 9(6) :552-60. Gutsell, J. S. 1928. The spider crab, Libinia dubia, and the jellyfish, Stomolophiis meleayris found associated at Beaufort, North Carolina. Ecology. 9(3) :358-9. Mansueti, R. 1963. Symbiotic behavior between small fishes and jellyfishes with new data on that between the stromateid, Pepriius alepidotus and the Scy- phomedusa, Chrysaora quinquecin ha, Gopeia. (1) ;40-80. Moestafa, S. H. and B. H. McConnaughey. 1966. Catostyliis ouwensi (Rhizostomae, Catostylidae) , a new jellyfish from Irian (New Guinea) and Ouwensia cato- styli n. gen., n. sp., parasitic in C. otiwcnsi, Treubia. 27(1) :l-9. Phillips, P. J., W, D, Burke and E. J. Keener, 1969, Observations on the trophic significance of jellyfishes in Mississippi Sound with quantitative data on the associative behavior of small fi.shes with medusae. Transactions American Fisheries Society. 98(4) :703-l2. Russell, F. S. 1967. On a remarkable new seyphomedusan. Journal of the Marine Biological Association of the United Kingdom, 47(3) ;469-73. Stunkard, H. W. 1969. The morphology and life history of Neopechona pyriforme (Linton 1900) n. gen., n. comb. (Trematoda: Lepocreadiidae) . Tke Biologi- cal Bulletin. 136:96-113. Gulf Research Reports Volume 4 Issue 2 January 1973 Stranding Records of a Finback Whale^ Balaenoptera physalus, from Mississippi and the Goose-Beaked Whale^ Ziphius cavirostris, from Louisiana Gordon Gunter Gulf Coast Research Laboratory J.Y. Ghristmas Gulf Coast Research Laboratory DOI: 10.18785/grr.0402.03 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Gunter; G. andj. Christmas. 1973. Stranding Records of a Finback Whale^ Balaenoptera physalus, from Mississippi and the Goose- Beaked Whale, Ziphius cavirostris, from Louisiana. Gulf Research Reports 4 (2); 169-173. Retrieved from http;// aquila.usm.edu/gcr/vol4/iss2/3 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. STRAJNDING RECORDS OF A FINBACK WHALE, Balaenopiera physalm,, FROM MISSISSIPPI AND THE GOOSE-BEAKED WHALE, Ziphius eavirostris, FROM LOUISIANA by GORDON GUNTER and J. Y. CHRISTMAS Gulf Coast Research Laboratory Ocean Springs, Mississippi INTRODUCTION Stranding records are sometimes the only source of data con- cerning some species of cetaceans and it is appropriate that these in- cidents be reported in some detail. This is particularly true of rare or little known species. Bottle-nosed dolphin (Tursiops truncahis) car- casses stranded on the Mississippi and Louisiana coasts are so com- mon that they receive little attention. Other species occur and have, no doubt, been overlooked or confused by laymen with I'nrsiops. Even the rare stranding of large whales in this area may go unreported ex- cept for news stories that fail to identify the animal properly and which are not readily available to cetologists. The stranding in 1967 of one of the first definitely reported fin- backs, Dalaerioptera physahis, in the Gulf of Mexico is given here. It is the first Mississippi record, Moore (1953) noted that “only six specimens of the goose-beaked whale have been reported on the eastern coast of North America” and provided four new records from the Florida west coast. The present report includes a Louisiana stranding in 1969. SPECIES ACCOUNT Balaenopiera physalus The Gulfport Ship Channel extends from the western end of Ship Island, a barrier island off the Mississippi coast, northward across Mississippi Sound for 11,5 nautical miles to the Gulfport Harbor. It has a limiting depth of 40 feet. On 7 April 1967 a baleen whale in- vaded Mississippi Sound by the ship channel for the first time that this event has ever been reported. This is also probably the first time a whale has entered the bay during the past several hundred years because the waters were too shallow until the ship channel was dug. The general events have been described by Heiliger (1967) as fol- lows : On the late afternoon of April 7, Captain Peter Skrmetta and Peter B. Lasanen were returning on the Pan-American Clipper, an excursion boat which operates between Gulfport and Ship Island, when they saw a whale spouting near the mouth of the Gulfport Har- 169 170 Gunter and Christmas Stranding Records of Whales 171 bor. Needless to say, this was most surprising. The next day this whale was found to have run ashore and died inside of Ship Island. It was later towed closer to Fort Massachusetts, the landing point on the island, as an attraction to tourists and excursionists aboard the Pan-American Clipper. This attraction lasted until the odor of the decaying animal became overwhelming, when Captain Skrmetta moved the carcass to the Cfulf beach and partially buried it with the help of a dragline dredge mounted on a barge. A group from the Gulf Coast Research Laboratory, including the senior author, visited Ship Island on April 11 and inspected this whale by going around it in a skiff. Heiliger (1967) had previously estimated the length as 45 feet and the weight at about 12 tons. The animal was a maturing male with a protuberant penis well over a meter long. There w'as a good bit of blotchy white on the right side of the anterior belly and the lower side of the flukes was white. The in- ner flippers were also white. Some of this can be seen in the tw^o ven- tral view photographs (Plate I, a & b). These and all other aspects of the specimen corresponded to the finback whale, Balaenoptera phy- salus. After this whale had been buried for a few months its head was dug up and exhibited at Fort Massachusetts for the delectation of the excursionists to Ship Island. The actions of thieves and the Hurricane Camille on August 17-18, 1969 have reduced the skull (25 June 1972) to a remnant of one jawbone. Records of Balaenoptera sp. from Louisiana have been summa- rized by Lowery (1943), The Louisiana State University Museum had a piece of baleen from a probably Louisiana or Mississippi specimen of B. physalns taken about 1928. Several more probable and a couple of definite records from Louisiana will be summarized by Lowery (in press) . Zi phiiis cu inrostris We first received the report of a whale stranded on the Chande- leur Islands of Louisiana from Mr. Bob Stevens who saw it on a rou- tine inspection flight of the Gulf Islands National Wildlife Refuge in April 1969. The carcass was located later on 15 May 1969 by J. Y. Christmas, Tom Mcllwain and Lionel Eleuterius. It was partially buried in the berm about five miles south of Chandeleur Lighthouse on the Gulf beach (approx, lat. 29° 58' N. long. 88° 49' W.), Heavy surf prevented landing on the Gulf beach, but Captain Kenneth Mel- vin w'as able to put the party ashore from Chandeleur Sound, across the island. The skin had sloughed off the carcass so that muscle tissue was exposed on some areas of the body. It lay on the right side with dorsal fin and flukes buried. The beak was about half buried with jaws open at 10-15°. The whole body was blackened so that natural color could 172 Gunter and Christmas Plate 11. Upper view of skull (a) and mandibles (b) of a specimen of the goose-beaked whale found on the Chandeleur Islands, Louisiana in April 1969. Courtesy of Dr. G. H. Lowery, Jr. Stranding Records of Whales 173 not be ascertained. The viscera were pushed into the mouth and an area about 46 cm in diameter around the anus was distended. When dug out, the dorsal fin and flukes were essentially intact but their shape was not clearly discernible. Teeth could not be seen. Somewhat crude measurements made with a meter stick were as follows: Total length — 559 cm. (18.1ft.) Maximum body depth — 106 cm. Tip of beak to dorsal base — 336 cm. Tip of beak to eye — 71 cm. Tip of beak to pectoral base- 132 cm. Lower jaw length — 38 cm. Upper jaw length — 35.6 cm. Flukes, Maximum width — 112 cm. Blowhole (located behind eye) — 7.6 cm. The carcass was left intact because any parts that could have been removed could not be carried across the island to the boat. About a month later, J. Y. Christmas and Richard Waller re- turned and landed on the Gulf beach. Decomposition was considerably advanced. The head, tail and left pectoral were removed, placed on a sheet of polyethylene and dragged into the water. A piece of nylon webbing was secured around the whole bundle and it was lifted to the trawl deck of the boat. The return trip stank, At the Gulf Coast Research Laboratory the bundle was buried above normal tides. A few days after Hurricane Camille had inun- dated the whole area we found that only shallow erosion had occurred at the burial site, exposing the end of the beak. One tooth was found on the ground. The head was still putrid. It was covered and not dis- turbed again until Dr. George H. Lowery, Jr. removed the material to Louisiana State University for museum preparation. He has kindly furnished photographs of the cleaned specimen which is now LSUMZ15609. Plate 11 is an upper view of the .skull (a) and the mandibles (b) of this specimen. Doctor Lowery (personal communication) was able to confirm our tentative identification as Ziphi'US cavirostris. lie later carried photographs to London where F. C. Fraser at the British Museum concurred in this identification. LITERATURE CITED Heiliger, Dudley P. 1967. Outdoors along the coast. (Newspaper column). The Dixie Guide, May 1967, p. 16, (A monthly newspaper published in Gulfport, Mississippi, now defunct.) Lowery, George H., Jr, 1943. Check-list of the mammals of Louisiana and adjacent waters. Occasional Papers of the Museum of Zoology (Louisiana State Uni- versity), Number 13:213-57. (In Press). The mammals of Louisiana and adjacent waters. Louisiana State University Press. Moore, Joseph Curtis. 1953. Distribution of marine mammals to Florida waters. The American Midland Naturalist, 49(1) : 117-58. Gulf Research Reports Volume 4 Issue 2 January 1973 Some Effects of Hurricanes on the Terrestrial Biota^ With Special Reference to Camille (Reprint) Gordon Gunter Gulf Coast Research Laboratory Lionel N. Eleuterius Gulf Coast Research Laboratory DOI: 10.18785/grr.0402.04 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Gunter^ G. and L. N. Eleuterius. 1973. Some Effects of Hurricanes on the Terrestrial Biota^ With Special Reference to Camille (Reprint). Gulf Research Reports 4 (2): 174-185. Retrieved from http:// aquila.usm.edu/ gcr/vol4/iss2/ 4 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. SOME EFFECTS OF HURRICANES ON THE TERRESTRIAL BIOTA, WITH SPECIAL REFERENCE TO CAMILLE' by GORDON GUNTER and LIONEL N. ELEUTERIUS Gulf Coast Research Laboratory Ocean Springs, Mississippi INTRODUCTION There have been very few articles concerning the effects of hur- ricanes upon marine and shore organisms. Some effects on fishes have been described by Hubbs (1962) and in that paper he reviewed some of the previous references. Information on animals killed or injured by hurricanes is scarce because potential observers in areas where they strike are generally more concerned with practical personal matters than biological stud- ies right after a bad storm. The senior author has been in or very close to seven West India hurricanes as they came ashore. Each time he was somewhat forewarned and had determined to make some type of quantitative appraisal of killed animals following these storms. However, on no occasion has this been done. Nevertheless, the tw^o waiters have collected some fragmentary information worth record- ing. Some Damages to the Fauna on the Mississippi Coast Hurricane Betsy — 9-10 September 1965 When Betsy passed the Mississippi Coast on its way to devastating areas of New Orleans, the center was approximately 50 miles south of Horn Island. Thus the wind blew more or less from east to west along the Ocean Springs^Pascagoula coastal area. Following this storm hun- dreds of sea balls mostly of the marsh grass Spartina were found on the mainland beach of Mississippi Sound near Gulfport. The water rose to a height of about six feet on the Laboratory grounds and left a strand of debris along the beach. In areas where there was marsh grass, thousands of little drowmed mice, the Eastern Harvest Mouse, Reithrodontomys humuliSy lined the shore in a little browm wundrow which w'as sometimes 100 yards long without a break. Also lying on the beach every ten yards or so w^as a dead raccoon, Pro- cyon loto7\ There were so many on the Laboratory grounds that they ^ This is a verbatim reprinting of the paper with the same title published in Volume 3, Number 2 of this journal, which was so improperly laid out that re- prints with the plates could not be made and the legends were missing. 174 Hurricane Effects on Biota 175 had to be hauled away. Raccoons are good swimmers and they cer- tainly did not come from the surrounding nearby marsh-land, Horn Island, which lies eight miles offshore, was either completely or nearly submerged during this storm. The most reasonable assumption is that these coons were drowned in Mississippi Sound, after being washed off of Horn Island and their bodies were finally blown onto the main- land shore. Hurricane Camille — 17-18 August 1969 This has been publicized variously as the worst storm that ever struck North America, or as the worst that has come ashore in this country in 143 years. Old, indefinite accounts indicate that something like this struck Florida in the 1700s. It is quite certain that Camille was the most powerful hurricane that has struck a w’ell-populated shore of the United States. The weather planes flying through it clocked the winds at 218 mph and recorded the lowe.st natural baro- metric pressure that has ever been read (26,01 inches) . The ''Big House,” an old landmark of the Laboratory and of the coast, was splintered and even the brick pillars upon which it stood were washed aw'ay. We have no way of quantifying the destruction of animals, ex- cept to say that the clean-up agency, the 43rd Battalion, Corps of Engineers, U. S. Army, reported removing 28 tons of animals from the beach between Biloxi and Uulfport on August 22-24, Most of these animals were dogs and cats, but some horses and cattle were mixed in. After the storm many dogs were homeless and many were systematically shot because they were starving. Following the storm, the writer was waked up by a bird singing lustily just outside his window; this was the only bird seen for about a w^eek. An unknown number of wild animals, birds, dogs and other life including human beings lost their lives in the storm; after about three days, the odor of decaying animals was noticed in the atmos- phere and lasted about a week before it gradually went away. The bird population and the squirrel population virtually disap- peared but both came back after a few weeks time, most noticeably the jaybirds and a few gray squirrels {Sciurus carolinensis) . The birds disappeared again, probably because they could find nothing to eat. This was certainly true of the squirrels and they w^ere reduced to gnawing the bark off of felled water oaks (Gunter and Eleuterius 1971). When the storm struck, the seeds of various nut trees — chiefly the hickory, black walnut, and thousands of pecans which are planted in this area^ — were just beginning to mature. Many of these were blown 176 Gunter and Eleuterius down and approximately half of the foliage of those remaining was denuded by breakage of the limbs. The same thing was true of oaks and acorns. Additionally^ the nuts themselves were beaten off the trees that remained standing. Presumably for that reason the Eastern Gray Squirrel, which was quite common, had not returned in its for- mer numbers by April 1970. Before the storm it was quite common to see as many as eight of these at one time in a relatively .small area of trees in the senior author’s front yard. After the storm, he saw none for one week and then he saw a lone squirrel. The squirrel population apparently increased in about three weeks to a month after the storm, then declined again. This observation would bear out the supposition that squirrels moving in from other areas could not find sufficient food and moved out again. The same thing apparently was true of the jaybirds. There was a decided diminution in the number of birds which came to feeder stations during the following winter. For instance, dozens of birds and sometimes a few' hundred in one afternoon for- merly fed at a home facing the beach just to the side of the Labora- tory grounds. The most numerous species, sometimes present in the dozens at a time, was the Savannah Sparrow. During the past winter only three or four have appeared at a time. The owner, before Camille, had to keep watch on the Starlings and jaybirds because they dis- turbed and ran off the others, but has had no trouble since the storm. General observations show' that the Brown Thrashers, the jaybirds and the Cardinals are present in very diminished numbers even today (April 1970). These facts have been noticed by other people and recorded, espe- cially in The Dixie Guide by Mr. Clayton Rand who has gone through three bad hurricanes at his home in Gulfport. Mr. Rand has men- tioned in his paper several times, the last being February and March 1970, that during former hurricanes there were many snakes and frogs everywhere in the area and that the mosquitoes were quite bad. He has remarked three times in his monthly newspaper that there was a great absence of life following Camille, even of the birds. To the senior writer, however, the most amazing thing has been the disappearance of the ants up until this time (April 1970). The black carpenter ant and the Argentine fire ant and several other smaller species were quite common in his yard. Apparently they all succumbed to the storm, e.\cept for a minute yellow species that goes by the name of sugar ant, which has been seen one time. Bread and other foods set out for dogs and cats w'ere formerly covered with ants in a matter of minutes; but, even this long after the storm, they may remain untouched by ants for days. We do not know the extent of de- struction of the Argentine fire ant, but locally they are gone. It is to be expected that termites and termite feeding animals and Hurricane Effects on Biota 177 possibly woodpeckers would increase greatly in numbers due to the thousands and thousands of felled trees and rotting timber, a good bit of which, after having had the top broken off, is still upright. Damages to the Flora on the Mississippi Coast There are very few reports of the effects of hurricanes, typhoons, or cyclones (tornadoes) on coastal vegetation. Sauer (1962) reported the effects of cyclones on the coastal vegetation of a tropical island (Mauritius) in the Indian Ocean. Chamberlain (1959) and the U. S. Department of Agriculture (1960) reported some of the effects of Hurricane Audrey on the vegetation of south Louisiana. Previous hur- ricanes which struck the Mississippi coast inflicted minor damage to the vegetation; one of the worst of these storms known to the junior author occurred in 1947. The “eye” or center of Hurricane Camille came ashore in the Pass Christian-Bay St. Louis area and the path was well marked by the effects of the storm on vegetation. The most apparent and obvious effect was the destruction of the ti’ees. In Jackson County most of the trees blown down were oriented with the tops pointing toward the northwest. In Harrison County near Gulfport, the trees became ori- ented with the tops pointing toward the west-northwest and in the Pass Christian-Bay St. Louis area, they were oi’iented in an east-west direction, but some tree tops pointed eastward and some pointed to the west and the trees were nearly parallel in alignment (Figs. 1 and 2). The paradoxical alignment was apparently a result of the initial winds from the east, followed after the “eye” passed over the area, by winds from the west. Trees west of Bay St. T^ouis near Pearl River were oriented with the tops toward the east-northeast and near Sli- dell, Louisiana, they were down in a northeast direction. The intensity of winds from Hurricane Camille could be seen in the number of trees felled, the number increasing as the wind velocity increased toward the path of the “eye.” In fact, without referring to other data, one could determine the storm’s path by observing the east-west direction in which the trees were blown down and by the gradual increase in the numbers of trees destroyed as the center of the path was approached. Tornadoes or extremely turbulent winds ripped through many areas on the periphery of the hurricane and the paths of their “touch downs” were well documented in the vegetation. In Magnolia State Park, which almost adjoins the Laboratory propei*ty, there is one area 50 feet wide and 17 tree lengths long, which the second author at- tributed to these tornadic gusts. The junior author conducted two vegetatioiial surveys to compare the intensity of damage to areas on the periphery of Hurricane Ca- 178 Gunter and Eleuterius mille with areas nearer the center. In Jackson County, these surveys showed that in one tract, 4% of the trees were blown down and 10% were damaged to the point that survival was in question. The plant community was dominated by Quercus yiigra (water oak) with Piniis elliotii (slash pine), Carya glabra (hickory) and Qmrcus rubra (red oak) being the subdominant species. This 40-acre tract in Magnolia State Park was approximately 22 feet above sea level. Destroyed trees in decreasing order were: red oak, slash pine, water oak, and hickory. It was noted that the heart wood (xylem) of the red oaks had been weakened by pathogenic attack and were rotted. Dess than 10% of the pines destroyed were uprooted; they were twisted or broken off at heights ranging from 5 to 20 feet above the ground. The large tap- root characteristic of the pines apparently held the trees up; they were not blown down easily, but could be broken. Other trees blown down in adjacent plant communities were Magnolia grandiflora (mag- nolia), Nyssa biflora (black gum), Liquidamhar styraciflora (sweet gum) , and Lireodendrum tulipera (tulip tree or yellow poplar) . Another survey was conducted on 87 acres of forested land north of Pass Christian in Harrison County, bordering the Wolf River and Red Creek Road. Approximately 10 acres here was bottomland forest along the river and adjacent low-lying drainage areas. The rest of the land was approximately 25 feel above .sea level and covered with Piniis elliotii (slash) , Pinus taeda (loblolly) and Pimis palustris (longleaf) in various stages of growth. The owner considered the area a game reserve and left it undisturbed. Results of a sample showed that ap- proximately 70% of the bottomland species were blown down. The species were Magnolia virginiana (sweet bay) Liquidamhar styraai- flora (sw'eet gum), Taxodium distichum (bald cypress), Acer rubrum (red maple), and the area was dominated by Quercus nigra (water oak) . Ninety per cent of the trees in the low-lying area had diameters greater than 24 inches at breast height and there were between 100 and 150 trees per acre. An estimated total of 201,000 board feet of hardwood timber was lost. Approximately 10,000 slash, loblolly, and longleaf pine trees with diameters greater than 10 inches were present on the higher sites and there were only 300 of these trees that were not damaged, i.e. 97 % were destroyed. Many of those standing were not expected to survive due to lack of limbs, missing tops or split trunks. A total of 607,600 board feet of pine was estimated as lost. Many young trees were crushed by the falling trees, and other understory plants and habitats for wildlife were destroyed. At the time of the survey (March 1970), beetles, especially Ips avuhu^, bps grandicoUis, and Ips calligraphns, had infested many of the downed trees and rot had begun. The specific names of the beetles were furnished by Dr. Virgil Smith, entomologist, U. S. Forest Service, Gulfport, Mississippi. Twisted and split saw logs could not be salvaged for use. Paper wood operations were expected to Hurricane Effects on Biota 179 be hindered by the tangled mass of trees. Practically all of the pine trees were second growth and ranged from 16 to 68 years old. The water oaks and other hardwoods were much older, ranging from 100 to 125 years. These two tracts simply show by comparison that the most dam- age to the vegetation was caused by winds occurring near the center of Hurricane Camille’s path. Another observation was the destruction of Querciis virginiana (live oak) along the Peach front from Biloxi to Pass Christian. Ap- proximately 25,000 live oaks were growing along the beach before Camille and one-fourth were destroyed by wind and water and one- half were damaged. Those trees nearest the beach were partially in- undated and the roots eroded by wave action. The immediate beating action of wind and the physiological “drought” resulting from the salt spray reduced these evergreens to bare branches (Figs. 3 and 4) . Many slash and longleaf pines may have been killed as a result of the inundation of low-lying areas near the mouth of the Wolf and Jourdan Rivers. The trees are dead but standing; however, this could be the result of other, internal damage since many trees on the barrier islands were covered by salt w'ater and survived. This observation needs further study. The Corps of Engineers, U. S. Army, estimated that a total of 1.2 million board feet of saw timber and one million cords of pulp- wood in MissLssippi w'ere lost. On the Mi.ssissippi Test Facility in Han- cock County, an estimated 6,000 cords of pulpwood were damaged and only 60 of the downed trees could be salvaged for lumber. It has been reported that a total of 290 million cubic feet of pine alone was lost in South Mississippi (Van Hooser and Hedlund 1969). The barrier islands presented a pattern of destruction similar to that on the mainland. Petit Bois Island was affected relatively little but there was a gradual increase in damage on the islands to the west. Horn Island was heavily eroded on the outside beaches. The marsh vegetation was pushed dowm and pressed to the soil .surface by the w’ater as it passed over the island (Figs. 5 and 6). Ship Island was cut into three pieces and more than one-third of the vegetation, most of which was herbaceous, was removed. Cat Tsland was heavily dam- aged. Large oaks were uprooted by wave action and many pines were broken by the wind. Large sand dunes were leveled, the sand redis- tributed over much of the adjacent low-lying marsh. Tons of plant materials .swept from the Louisiana marshes and the barrier islands were deposited on the mainland in large windrows. Marshlands were affected insignificantly because the water cov- ered them early in the hurricane and they were not exposed to the terrific beating of wind and wave that occurred later. Spartina alter- 180 Gunter and Eleuterius niflora (smooth cord grass) flowered on schedule (September through November). Shrubs found along the periphery of marshes, where they formed thickets, acted as baffles and protected trees and, in some cases, homes. Many upland understory areas were denuded of her- baceous and woody shrubs where they were located near water. The botanical regime of South Mississippi was disturbed by Hur- ricane Camille of August 1969, probably to a greater extent than by any other hurricane in the history of Mississippi, and the greatest in- fluence on the terrestrial vegetation was the destruction of the trees. Hurricane Effects on Biota 181 Figure 1. Heavily damaged pine stand in Hancock County show- ing parallel but opposite direction alignment of fallen trees. Note direction in which standing trees are leaning. Figure 2. Heavily damaged pine stand illustrating parallel but opposite direction alignment of fallen trees. This effect was caused by passage of the “eye” of Hurricane Camille through Bay St. Louis-Pass Christian area. 182 Gunter and Eleuterius Figure 3. Damaged live oaks (Quercus virginiana) along Highway 90 near Long Beach, Mississippi. Hurricane Effects on Biota 183 Figure 4. Damaged home and live oak (Quercus virginiana) along Highway 90 and open waters of Mississippi Sound at Long Beach, Mississippi. 184 Gunter and Eleuterius Figure 5. Marsh near the south beach of Horn Island. Altitude approximately 1,500 feet. Figure 6. Low altitude view (600 feet) of same marsh shown in Figure 5. Note flattened plants of J uncus roemeriamis and Spartina alterniflora as a result of wave action across island. Hurricane Effects on Biota 185 LITERATURE CITED Chamberlain, J. L. 1959. Influence of Hurricane Audrey on the coastal marsh of southwestern Louisiana. Coastal Studies Institute, Louisiana State Univer- sity, Technical Report, lOB, ONR 35608. Gunter, G. and L. Eleuterius. 1971. Bark eating^ by the common gray squirrel fol- lowing a hurricane. Amer. Mdl. Nat. 85(1) :235. Hubbs, Clark. 1962. Effects of a hurricane on the fish fauna of a coastal pool and drainage ditch. Tex. Jour. Sci. 14(3) :289-96. Sauer, J. D. 1962. Effects of recent tropical cyclones on the coastal vegetation of Mauritius. J. Ecol, 50:275-90. U. S. Dept, of Agr. Soil Conservation Service. 1960. Effects of saline water from Hurricane Audrey on soils and vegetation. Alexandria, La., Special Rept. (Minco). Van Hoqser, Dwane D, and Arnold Hedlund. 1969. Timber damaged by Hurricane Camille in Mississippi. U. S. Forest Service Res. Note. 80-96:1-5. Southern Forest Experiment Sta., New Orleans, La. Gulf Research Reports Volume 4 Issue 2 January 1973 Protozoan Symbionts from the Anemone Bunodosoma cavernata from Galveston Island^ Texas Philip J. Phillips Texas A6’M University DOI: 10.18785/grr.0402.05 Follow this and additional works at; http://aquila.usm.edu/gcr Part of the Marine Biology Commons Recommended Citation Phillips^ P. J. 1973. Protozoan Symbionts from the Anemone Bunodosoma cavernata from Galveston Island; Texas. Gulf Research Reports 4 (2): 186-190. Retrieved from http:// aquila.usm.edu/ gcr/vol4/iss2/ 5 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. PROTOZOAN SYMBIONTS FROM THE ANEMONE Bunodosoma cavernata FROM GALVESTON ISLAND, TEXAS by PHILIP J. PHILLIPS! Department of Biology Texas A&M University College Station, Texas 77801 Hargitt (1904), in his classic paper on the medusae of the Woods Hole region, noted the susceptibility of medusoid forms to various pro- tozoan parasites. He stated that “when working upon the regeneration of medusae I found several species of Protozoa very closely associated with them and, under the limitations of the aquaria, often exceedingly troublesome, seriously interfering with the progress of the experi- ments. This suggested the possibility of a parasitic relation.” Unfor- tunately Hargitt made no mention of the taxa of Protozoa involved. Studies regarding protozoan infestations of Cnidaria have been few. Other than the passing references to protozoan infestations by Har- gitt (1904) and Kudo (1966) there have been no intensive studies of protozoan parasites or symbionts of marine Cnidaria. The anemone Bunodosoma cavernata is a common inhabitant of the Texas coast and cursory observation revealed a large complement of protozoan associates. These included Saprodinium, Euplotes^ Pa- raeuplotes, Vorticella, Cohnilembun, Aiiophrys, Uronema and Vahl- kampfia. Approximately forty anemones were collected from jetty rocks on Galveston Island on 3 November 1970. Note was made within one day of protozoans found on the external surfaces, in the coelenteron and in surrounding sea water. The anemones were kept until 7 December 1970 under varying conditions and bi-weekly note was made of the associated protozoan fauna. One group of six anemones was placed in a 29-gallon aquarium in which the salinity was maintained at 28 g/kg and the temperature ranged from 15 to 18 °C. These anemones were considered “normal.” No attempt was made to construct a facsimile of the Galveston littoral environment from which the anemones were removed. Other groups of three to eight anemones were kept in 1.5-gallon aquaria (salinity 28 g/kg, temp. 18 °G) and in 12-inch finger bowls wherein salinity and temperature were varied. For one group (GP 1) the salinity was al- lowed to rise by evaporation from 20 g/kg to 40 g/kg over a 12-day period. In a second group (GP 2) the sea water was flushed out and ! Present address: North Atlantic Division, II.S. Army Corps of Engineers, 99 Church Street, New York, N. Y. 10007. 186 Protozoan Symbionts of Anemone 187 replaced with clean sea water of the same salinity every 3 days for a 3-week interval. In the third group (GP 3), the salinity was constantly maintained at 24 g/kg for 4 weeks. All three groups were maintained at 18 ”C. In a fourth group temperature was maintained at 24° C. Sa- linity was routinely monitored on a daily basis with a Goldberg re- fractometer. Microscopy examinations were made every 2 to 4 days using wet mounts of anemone mucus, coelenteron contents, tissue squashes and the external medium. Protozoans found as epizoites or symbionts on the anemones as well as site of infection and relative abundance are listed in Table 1. Table 1. Protozoa Found in Association with Anemones after Two Weeks in the Laboratory Protozoan Site Wounds of Infection Coelen- Mucus teron Coat Size Range (microns) Vahlkampfia XXX XX XX 7-30 Saprodinium XX X X 21-25 EupZotes XXX XXX XXX 70-140 Paraouplotes X X X 50-80 Varameeiwn X 100-150 VoYtioella XX XX XX C.150 Cohni lembus X X XX 30-80 Anophrys XX XX XX 60-90 Vronema X X X 20-40 X indicates rarity; XX indicates frequent occurrence; XXX indicates dominant organisms numerically. These data were taken 2 weeks after the anemones were removed from the natural environment and introduced into the laboratory. Initially when first placed in aquaria all Protozoa found in association with the anemones occurred in very low numbers when compared to the situa- tion after 2 weeks. Protozoa in habitat sea water included very sparse populations of Eiiplotes, Pammecmm and Vorticella. The normal ane- mones maintained a numerically smaller symbiont population than those in the finger bowls. Those in the 1.5-gallon aquaria had a “nor- mal” symbiont complement also. In GP 2, where the water was flushed out and replaced periodically, the anemones retained the normal fauna 188 Phillips and did not support large populations of any species of protozoan. In the group maintained at 24°C the anemones became moribund after two weeks and developed extremely large populations of ciliates, es- pecia.\\y Eitplotes, Arwphrys and Saprodmiimi, as well as large num- bers of the naegleriid amoeba Vahlkampfia. In addition to being found in necrotic and normal anemone tissues this amoeba was also found in large numbers in the fouled medium. Both cysts and trophozoites of Vahlkampfia were extremely numerous at sites of necrosis. In group 3 (GP 3), wherein conditions were constant, the anemones survived up to 4 weeks at which time they became moribund and died. The same situation prevailed here as with respect to GP 2. In GP 1 increased salinity had no apparent effect on protozoan numbers although indi- vidual EuiAotes showed considerable size increase (up to 250 p in length). Size ranges of all associative Protozoa encountered are given in Table 1. All anemones within the finger bowls died within 5 weeks. In all cases the anemones were infiltrated by large numbers of bacteria and large populations of Enplotes and Vahlkampfia. Approximately half the Euplotes in association with necrotic anemone tissue, were in the process of conjugation. Three morphological types of Vahlkampfia were observed. Most amoebae were 7-10 p in length and formed only one broad, fan-shaped pseudopod which, moreover, consisted mainly of a large hyaline area. P^ormation of a new pseudopod was preceded by flooding the old pseu- dopod with granular endoplasm and the formation of a new^ hyaline bulge. Rate of pseudopod formation was hastened by the heat of a substage illuminator. The second observed trophozoite stage was con- siderably larger (25-30 p) and had numerous psuedopodia. Their morphology approached that of the former type after being on a mi- croscope slide (above an illuminator) for several minutes. These larger amoebae underw'ent division and “reverted’’ to the smaller, mo- nopseudopodial type. On two occasions a nucleus (2-3 p in diameter) with a large endosome was observed in the large amoebic variant. Large numbers of Vahlkampfia cysts were found in decaying and moribund anemones as well as in putrefying media. After being on a microscope slide for about 5 minute.s, above a substage illuminator, these cysts released one amoeba each. Cysts were rounded, consider- ably flattened, had wrinkled surfaces and Nvere between 5-9 p in diam- eter. Vahlkampfia invades anemone tissue and will emerge from tissue fragments when heated by substage illumination for a few minutes. The amoebae can be demonstrated in both normal and necrotic tissue, the greatest concentration of amoebae occurring in necrotic tissues with a large bacterial population. The amoebae were observed to ingest bacteria, indicating that greater degree of association with necrotic tissue may be due to greater availability of food organismvS in necrotic as opposed to healthy tissue. Other Protozoa, especially Eu- Protozoan Symbionts of Anemone 189 plates, showed greater abundance where there w^as high bacterial density. Vahlkampfia and at least a few of the ciliates (Euplotes, Paraeii- plotes, Saprodinium and Vorticella) are not obligate symbionts. Nae- gleriid amoebae are known as facultative parasites from a wide vari- ety of organisms including rodents (Wilson et al. 1967), molluscs (Hogue 1921 and Richards 1970), insects (Page 1970) and man (Cerva and Novak 1968 and Duma et al. 1969). Apparently amoebae of this group can be demonstrated to be symbiotic with most any metazoan. Although Naegleria gruberi is the only naegleriid amoeba definitely known to be a causative agent in human amoebic meningoen- cephalitis (Duma et al, 1969 and Cerva 1970), the remaining naegle- riids cannot be discounted as potential pathogens. Wilson et al. (1967) demonstrated that Vahlkampfia as well as many other genera of amoe- bae can cause systemic amoebiasis in rodents following hypodermic inoculation. Available data indicate that Vahlkampfia, and at least some of the ciliates in this study feed on broken tissue and associated bacterial flora. The external mucus coat or pellicle of the anemone on normal anemones has a relatively large bacterial population and this may account for the establishment of larger populations of Protozoa in this region when compared to other anatomical regions of the anemone. The hymenostomid genera Cohnilemhiis and Anophrys are inter- esting in that they are known to-be intestinal symbionts of echinoids (Kudo 1966). These two genera and Uronema predominantly occur in the coelenteron and mucus coat. Paraeuplotes, found here on anem- ones, is also a known epizoite of corals (Kudo 1966). In the normal littoral habitat the anemones are periodically ex- posed to the air during tidal cycles and there is considerable scouring of the anemone surface as well as drastic short term salinity and tem- perature changes in the immediate environment of the anemone. These environmental factors probably serve to control bacterial and protozoan populations epizootic on the anemones. Conversely the mu- cus sheath of the anemone, which serves for attachment of protective shell fragments and other debris to the anemone surfaces, may pre- vent desiccation at low tide and may act as a suitable substrate for the establishment of epizoites. When exposed the anemones close the oral apparatus, trapping sea water and any associated microfauna possibly allowing survival of some associative forms. It remains to be deter- mined whether or not any of these symbionts or epizoites are depen- dent on the anemone for life cycle completion. Euplotes is by far the most abundant of all the ciliates encoun- tered and conjugation is a commonplace phenomenon especially when putrefaction is well advanced. It is possible that death of the anemone enables or induces conjugation in Euplotes, 190 Phillips Anemones of the littoral zone support an extensive protozoan fauna. Although none of the Protozoa involved are definitely known to be obligate symbionts there is a distinct association. Most of the forms are facultative symbionts, especially VahlJcampfia and Euploteis. Vahlkampfia invasion of anemone tissues can to some extent be asso- ciated with necrotic tissue changes and increased bacterial popula- tions, Most probably bacteria initiate the necrosis and the large bac- terial populations allow for increase in the amoeba population. No causal relationship has been established. ACKNOWLEDGEMENTS I would like to thank Doctors K. Horvath and S. H. Hopkins (De- partment of Biology, Texas A&M University) for critical review of the manuscript. LITERATURE CITED Cerva, L. 1970. Comparative morpholog-y of three pathogenic ^strains of Nacgleria gruberi. Folia Parasit. 17:127-33. Cerva, L. and Novak. 1968. Amoebic meningoencephalitis: sixteen fatalities. Science 160:92. Duma, R. J., H. W. Ferrell, E. C, Nelson and M. M. Jones, 1969. Primary amoebic meningoGneephalitis. New England Journal of Medicine 281:1316-2.3. Hargitt. Ch. W. 1904. The medusae of the Woods Hole region. Bulletin of the Bureau of Fisheries 24:27-79, Hogue, M. J. 1921. Studies on the life history of Vahlkampfia patuxeni n.sp, para- sitic in the oyster with experiments regarding its pathenogenicity. American Journal of Hygiene 1 : 321-45. Kudo, R. K. 1966. Protozoology (Fifth Edition). C. C. Thomas, Illinois. Page, F. C, 1970. Ta.xonomy and morphology of free living amoebae causing meningoencephalitis in man and other animals. Journal of Parasitology. 56:- 257-58. Richard.5, C. S. 1970. Amoebae associated with freshwater molluscs. Journal of Parasitology. 56:282. Wilson, D. E., E, C- Bovee, G. J. Bovee and S. K. Teldord. 1967. Induction of amebiasis in tissues of white mice and rats by subcutaneous inoculation of small free living, inquilinic and parasitic amebas with associated coliform bacteria. Experimental Parasitology. 21:277-86. Gulf Research Reports Volume 4 Issue 2 January 1973 Some Analyses of Twentieth Century Landing Statistics of Marine Shrimp of the South Atlantic and Gulf States of the United States Gordon Gunter Gulf Coast Research Laboratory Katherine McGraw Gulf Coast Research Laboratory DOI: 10.18785/grr.0402.06 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Gunter^ G. and K. McGraw. 1973. Some Analyses of Twentieth Century Landing Statistics of Marine Shrimp of the South Atlantic and Gulf States of the United States. Gulf Research Reports 4 (2): 191-204. Retrieved from http;// aquila.usm.edu/gcr/vol4/iss2/6 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. SOME ANALYSES OF TWENTIETH CENTURY LANDING STATISTICS OF MARINE SHRIMP OF THE SOUTH ATLANTIC AND GULF STATES OF THE UNITED STATES by GORDON GUNTER and KATHERINE McGRAW Gulf Coast Research Laboratory Ocean Springs, Mississippi ABSTRACT There is a strong correlation between the total catch of white and brown shrimp with dockside prices on the United States Gulf Coast since 1902, but there is no significant correlation between South At- lantic production and prices, probably because the South Atlantic shrimp stocks have been over-fished since the 1920s. There is no nega- tive or positive correlation between the catch statistics of brown and white shrimp of the United States, and these species seem to be weak- ly competitive, if at all There is a significant correlation between the annual production of South Atlantic and Gulf white shrimp, but there is none between South Atlantic and Gulf brown shrimp, possibly be- cause the brown shrimp live generally in deeper water and are not so much influenced by short term variations in climatic conditions as the white shrimp are in shallow water. In furtherance of this idea, there is some indication that the brown shrimp production is less variable than the white shrimp production. INTRODUCTION Three species of shrimp of the Family Penaeidae (Genus Pe- naeiis) are present in considerable numbers and in overlapping dis- tributions in the bays and oceanic shallow waters from Cape Hatteras, North Carolina, south to Texas and beyond. These are the white shrimp Penaeus fluviatilis, the brown shrimp P. aztextis, and the pink shrimp P. duorarivm. Another shallow water penaeid Penaeus brasiliensis exists only in such small numbers off Miami, Florida that it was overlooked by bio- logists of the area until discovered there by Eldred (1960), A fifth species of the Penaeidae, XiphopeneMs kroyeri, is almost entirely shal- low oceanic in divStribution with a few entering the bays in cool weath- er (Gunter 1950) . It is not found along the South Atlantic part of the United States in commercial concentrations, but has been fished in the Gulf since boats and seines large enough to fish the shallow offshore waters have been available. 191 192 Gunter and McGraw The white shrimp grows to large size in shallow waters of the bays. The other two species of commercial shrimp, P. aztecus and P. dnorarum, do not grow so large in the bays and shallows and do not school as strongly as the white shrimp and the seabob. They also go into deeper waters when they move into the open ocean. SOME HISTORICAL ANTECEDENTS Indians caught shrimp with the use of dipnets, seines and leafy weirs such as are still employed in the Rio Soto la Marina, Mexico. Shrimp from the North Carolina waters were caught and transported to the Philadelphia market w'hen Thomas Say (1817) first described the North American white shrimp. Catch statistics on the commercial fisheries were collected only after the organization of the United States Fish Commission by S. F. Baird and others in 1871. We may assume with complete assurance, however, that shrimp production grew with the increase in population up until recent years. Even in the early part of this century the catch- ing of shrimp was by means of dipnets, seines, and castnets. For this reason only the white shrimp P. ftiiviatilis and the seabob Xiphope- neiis kroyeri were taken, because they were schooling shrimp. Even so the seabob has been taken in small numbers amounting to about 1.2% of the Gulf catch, (cf. Gunter 1962) partly because of its small size and its open ocean distribution. This shrimp is much more im- portant, relatively, in South American waters (cf. Lindner 1957). The otter trawl came into use along with motor vessels on the South Atlantic Coast during the period of World War I and spread quickly to the Gulf Coast. This permitted the fishing of deep waters and larger shrimp, which move out as they grow older. Thus, produc- tion gradually rose with the increase of demand and the more efficient otter trawl put the large seine crews out of business in Louisiana in the early 1930s. From 1902 the shrimp production in this country increased into the early 1950s. In the 1940s an extreme drought caused a great short- age of white shrimp, especially in Texas waters, and there fishermen turned to the previously unfished brown shrimp which were caught predominantly at night. Most states had laws against shrimping at night for the protection of the white shrimp, the idea being that they should not be harassed all hours of the twenty-four. The large brown shrimp generally bury in the bottom during the day. Recognition of these facts led to exploitation of the brown shrimp and after the early '50s it has yielded more than the white shrimp. This development be- gan in Texas waters in 1947 and spread quickly to other areas on the Gulf and South Atlantic Coast. Even so, the separation of the brown and white shrimp was not begun in the federal fisheries statistics until Shrimp Landing Statistics 193 1957. Therefore, we may say that the shrimp production figures used here were comprised almost entirely of white shrimp from 1903 to 1948, with about being seabobs. From 1948 to 1957 there was a period of production when the brown shrimp and white shrimp were not separated. After 1957 these shrimp have been separated in the catch statistics of the South Atlantic and Gulf Coasts. At that time the seabobs were also separated in the statistics. From 1951 to 1956 inclusive, the heads-off weight of white and brown shrimp produced ranged between 126 and 146 million pounds and in the 1967-71 period it ranged from 125 to 137 million pounds. These are the only years, except for 1963, that the United States shrimp production has ever ranged above 100,000,000 pounds of head- less shrimp. The 1951-56 high production was due to the exploitation of the previously unfished population of brown shrimp plus the white shrimp. The more recent high production seems to be due to an in- crease in the white shrimp population, cau.sed po.s.sibly by a recent hyperfertilization of the bays. DISTRIBUTIONS AND CATCH RECORDS There are many interesting things about the distribution of the shallow water penaeids along the coasts of the South Atlantic and Gulf states and Mexico, but here we are concerned only with the brown shrimp P. aztecus and the white shrimp P. fltiviatilis^ because these two have been the chief commercial producers and they both grow up in estuarine areas. Furthermore the United States population of brown and white shrimp are quite discrete and disconnected from other populations, and we have United States production of these two species unmixed with foreign populations. The white shrimp population of the United States is divided into two distinct parts. The South Atlantic component runs along the coast from North Carolina with the greatest abundance in Georgia and gives out at about the St. Lucie inlet in south Florida (Gunter and Hall 1963). The second population extends from the west Florida panhandle to Aransas Bay, Texas. The brown shrimp has roughly the same distribution but it is le.ss numerous on the Atlantic and extends farther south seasonally in the Mexican waters. Its abundance is greater in the salt waters of Texas than that of the white shrimp, which is most abundant in Lou- isiana because of the lower salinities in that region. In Texas waters brown shrimp are not raised in appreciable numbers farther south than the Aransas-Corpus Christi Bay system, which is connected to the Gulf by Aransas Pass. During the fall both species leave the bays and go to outside waters. Gunter (1962) showed by following the seasonal catch statistics of four areas on the coast that the white and 194 Gunter and McGraw brown shrimp go south on the Texas coast in the early fall and winter. Some go into Mexico and return in diminished numbers in the spring to Texas waters. Catches made off northern Mexico are returned to United States ports. This movement apparently begins off Galveston Bay and covers a distance of some 400 miles and it is virtually a parallel case to the seasonal north to south white shrimp migration and return from Georgia to the region of Cape Canaveral discovered by Weymouth, Lindner and Anderson (1933) (Lindner and Anderson 1956). Pink shrimp exist in fair concentrations off North Carolina and in heavy concentrations off the Tortugas. There are also large concen- trations in the Bay of Campeche, Mexico, which were formerly fished by Florida, Texas, Cuban, and Mexican fishermen, and adequate statis- tics are not available. Former United States catch statistics of this species were confused by Florida and Texas boats bringing in Cam- peche shrimp. Furthermore Gulf and Atlantic catches were confused by shrimpers carrying some shrimp from Tortugas to Atlantic ports. For these reasons we have avoided use of pink shrimp statistics. As grooved shrimp they were mixed with the browns to a small extent in the late 1950s but not enough to vitiate the brown shrimp statistics. SUMMARY OF THE PROBLEM The brown and white shrimp both grow up in the bays of the northern Gulf Coast and the South Atlantic states. They have a differ- ential distribution with relation to salinity and season (Weymouth, Lindner and Anderson 1933, Gunter 1950, 1961, Gunter, Christmas and Killebrew 1964) . The white shrimp come in and move out later in the year. Furthermore the white shrimp grow to larger size in the estuaries and, therefore, are more heavily fished before they move out- side. As a matter of fact the whole shrimp industry grew up in the shallows and gained technical experience on the white shrimp before moving to the open sea. Because of the overlapping life history of these two species of commercial shrimp, both in time and place, the question has arisen concerning their competition. Therefore, some who have been con- cerned with shrimp biology have discussed the.se matters for years, mostly with the suspicion that there was some kind of competition that opposed one shrimp population to the other. These ideas were the genesis of the analyses offered here. All shrimp statistics used here were taken from the annual Fish- ery Statistics of the United States and its predecessors, of which the latest issue is Lyles (1969) , and preliminary pamphlets. Shrimp Landing Statistics 195 PRICES AND PRODUCTION One would think that prices increased with expansion of produc- tion, the demand for shrimp, etc., and such is the case where total United States production and price are concerned. The coefficient of correlation, r, for the figures shown in Table 1 is 0.691 with 39 obser- Table 1. The Total Catch of White and Brown Shrimp of the Gulf and South Atlantic Coasts of the United States in Thousands of Pounds and the Dockside Value in Thousands of Dollars Catch in Catch in Year Pounds Value Year Pounds Value 1902 10,506 1908 11,855 1918 40,632 1923 45,987 1927 64,200 1928 74,986 1929 70,487 1930 57,219 1931 62,628 1932 57,313 1934 77,479 1936 76,520 1937 90,866 1938 96,150 1939 96,150 1940 97,754 1945 122,743 1950 122,048 1951 143,780 1952 145,414 286 - 1953 408 1954 1,746 1955 2,593 1956 3,518 1957 4,550 1958 4,435 1959 2,996 1960 2,731 1961 2,036 1962 3,067 1963 3,778 1964 5,009 1965 4,848 1966 4,848 1967 5,895 1968 21,289 1969 43,144 1970 51,518 1971 54,755 145,414 76,267 172,596 60,535 156,454 61,404 142,297 70,305 90,364 72,438 89,903 71,829 108,548 56,875 112,088 66,143 64,234 50,589 77,788 71,832 112,535 68,785 95,813 69,328 111,643 81,067 107,041 93,784 137,837 99,584 124,480 109,833 126,331 117,317 139,437 119,569 148,125 143,362 vations and 37 degrees of freedom. This means that prices and pro- duction have grown together, and the correlation is significant within the 1 % level. A further breakdown shows that the correlation, r, between price and production on the Gulf Coast amounts to 0.737 which is even more significant (Table 2). The Gulf correlation is higher than that of price 196 Gunter and McGraw Table 2. The Catch of United States Gulf Coast Brown and White Shrimp in Thousands of Pounds and Thousands of Dollars Year Pounds Value Year Pounds Value 1902 8,031 1908 8,156 1918 30,466 1923 30,595 1927 44,725 1928 53,357 1929 50,468 1930 40,203 1931 46,075 1932 42,427 1934 60,621 1936 54,723 1937 73,050 1938 73,108 1939 78,173 1940 83,012 1945 94,444 1950 98,359 1951 125,747 1952 128,745 199 1953 270 1954 1,276 1955 1,771 1956 2,344 1957 3,092 1958 2,986 1959 2,017 1960 1,817 1961 1,400 1962 2,278 1963 2,756 1964 4,181 1965 3,725 1966 3,991 1967 5,141 1968 17,305 1969 33,112 1970 44,136 48,170 1971 145,781 66,336 153,995 53,652 137,923 54,465 125,727 62,499 74,760 63,288 76,992 63,871 94,362 50,348 94,276 57,631 53,574 43,650 64,582 60,557 103,067 63,539 86,139 62,695 96,010 70,907 93,886 82,971 125,862 90,574 109,799 95,837 110,723 101,131 126,897 108,183 129,850 123,770 and production of total shrimp, of the South Atlantic and Gulf com- bined. In contrast, the correlation between price and total catch on the South Atlantic Coast, Table 3, is 0.067, which is not significant at all. This somewhat anomalous conclusion becomes clear if the shrimp of the South Atlantic Coast were over-fished rather early in the develop- ment of this fishery and have been over-fished for years. This explana- tion was advanced by Mr. Milton J. Lindner, whose experience with the South Atlantic shrimp fishery began in 1930. Examination of Table 3 shows that high production in white shrimp on the Atlantic Coast was attained in the 1920s. Apparently these shrimp were fished to the very limit of their yield and have been for a great number of years. This seems to be the only reasonable explanation of the fact that price level and shrimp production have not increased together on Shrimp Landing Statistics 197 Table 3. The Catch and Values of White and Brown Shrimp in Thousands of Pounds and Thousands of Dollars for the South Atlantic Year Pounds Value Year Pounds Value 1902 2,475 87 1953 21,385 9,931 1908 3,699 138 1954 18,601 6,883 1918 10,166 470 1955 18,531 6,939 1923 15,392 822 1956 16,570 7,806 1927 19,475 1,174 1957 15,604 9,150 1928 21,629 1,458 1958 12,911 7,958 1929 20,019 1,449 1959 14,186 6,527 1930 17,016 979 I960 17,812 8,512 1931 16,553 914 1961 10,660 6,939 1932 14,586 636 1962 13,206 11,275 1934 16,858 789 1963 9,468 5,246 1936 21,797 1,022 1964 9,674 6,633 1937 17,816 828 1965 15,633 10,160 1938 17,899 821 1966 13,155 10,813 1939 17,977 857 1967 11,975 9,010 1940 14,742 754 1968 14,681 13,996 1945 28,299 3,984 1969 15,608 16,186 1950 23,689 10,032 1970 12,541 11,386 1951 1952 18,033 16,669 7,382 6,585 1971 18,275 19,592 the South Atlantic Coast, but have increased together on the Gulf Coast. It may be further assumed that if the Gulf fishing continues at a high level with a continued price rise, that the production of Gulf shrimp will reach a limit, if it has not already done so, and that in future times price and shrimp production on the Gulf Coast will no longer show a correlation. PRODUCTION FIGURES BY AREAS AND SPECIES Because of previous correlations noted between the production of white shrimp and rainfall in the State of Texas (Gunter and Ed- wards 1969) and the apparent preference of brown shrimp for higher salinities, we determined the correlations between the catch of whites 198 Gunter and McGraw and browns in the State, r equaled — 0.2151, but with only 14 degrees of freedom it was not significant. Similarly there was no significant correlation between the catch of browns and whites on the South Atlantic Coast, the Gulf Coast, or the total of both areas. This means apparently that the production of these two shrimp are not closely related to one another and that they have different ecological niches and are weakly competitive, if at all. On the other hand, there is a correlation between the total annual production of shrimp of the South Atlantic; wdth the total annual pro- duction in the Gulf, in which r equals 0,3261 with 37 degrees of free- dom (Table 4). This is .significant at the level of 5.0%. This would mean that when conditions are generally good for shrimp production on the Gulf, they are also good on the Atlantic. Most likely these con- Table 4. Comparison of South Atlantic and Gulf Catches of White and Brown Shrimp in Thousands of Pounds Year Atlantic Gulf Year Atlantic Gulf 1902 2,475 8,031 1953 21,355 145,781 1908 3,699 8,156 1954 18,601 153,995 1918 10,166 30,466 1955 18,531 137,923 1923 15,392 30,595 1956 16,570 125,727 1927 19,475 44,725 1957 15,604 74,760 1928 21,629 53,357 1958 12,911 76,992 1929 20,019 50,468 1959 14,186 94,362 1930 17,016 40,203 1960 17,812 94,276 1931 16,553 46,075 1961 10,660 53,574 1932 14,586 42,727 1962 13,206 64,582 1934 16,858 60,621 1963 9,468 103,067 1936 21,797 54,723 1964 9,674 86,139 1937 17,816 73,050 1965 15,633 96,010 1938 17,899 73,108 1966 13,155 93,886 1939 17,977 78,173 1967 11,975 125,862 1940 14,742 83,012 1968 14,681 109,799 1945 28,299 94,444 1969 15,608 110,723 1950 23,689 98,359 1970 12,541 126,897 1951 18,033 125,747 1971 18,275 129,850 1952 16,669 128,745 Shrimp Landing Statistics 199 ditions are of a broad climatic nature, involving such things as cool and warm years, high rainfall and droughts, and even hard cold waves. It would be quite difficult to get some of these factors into figures or numbers, especially comparable figures for statistical calcu- lations, even if the climatic events were recorded years ago as many w'ere not. Therefore, we will pass this question by. Similarly there is a very strong correlation between the white shrimp production of the Atlantic Coast and Gulf Coast (Table 5). Table 5. Catch Figures for South Atlantic and Gulf White Shrimp in Thou- sands of Pounds Year Atlantic Gulf Year Atlantic Gulf 1902 2,475 1908 3,699 1918 10,166 1923 15,392 1927 19,475 1928 21,629 1929 20,019 1930 17,016 1931 16,553 1932 14,586 1934 16,858 1936 21,797 1937 17,816 1938 17,899 1939 17,977 1940 14,742 8,031 1945 8,156 1957 30,466 1958 30,595 1959 44,725 1960 53,357 1961 50,468 1962 40,203 1963 46,075 1964 42,727 1965 60,621 1966 54,723 1967 73,050 1968 73,108 1969 78,173 1970 83,012 1971 28,299 94,444 9,554 11,129 7,204 25,740 8,326 24,574 12,200 28,381 9,113 14,421 7,879 23,166 4,719 47,087 5,272 43,978 10,587 33,599 5,948 29,917 7,020 24,960 11,004 30,918 10,294 44,959 8,111 45 ,962 12,077 42,010 The total series stemming from 1902 to 1971 has 29 degrees of the freedom, because the years 1948 to 1957 were excluded when brown shrimp and white shrimp were not properly separated in the fisheries statistics. The correlation r was found to be 0.655 and significant at the 1% level. In contrast, no such correlation can be shown between the brown shrimp catch of the South Atlantic and Gulf (Table 6). We may spec- 200 Gunter and McGraw Table 6. Atlantic and Gulf Brown Shrimp Production in Thousands of Pounds and the Totals Year Atlantic Gulf Atlantic and Gulf browns combined 1957 6,050 63,631 69,681 1958 5,707 51,252 56,959 1959 5,860 69,788 75,648 1960 5,612 65,895 71,507 1961 1,547 39,153 40,700 1962 7,164 41,416 48,580 1963 4,749 55,980 60,729 1964 4,402 42,161 46,563 1965 5,046 62,411 67,457 1966 7,207 63,969 71,176 1967 4,955 100,902 105,857 1968 3,677 78,881 82,558 1969 5,314 65,764 71,078 1970 4,430 80,934 85,364 1971 6,060 87,788 93,848 ulate here that brown shrimp spend a shorter time in the bays, and live in deeper water in the ocean, and for that reason would be less aifected by climatic variations than the white shrimp in shallower water. Thus production would be less subject to parallel variations induced by climatic variables in shallow water, all leading to greater correlations of the white shrimp catch on the two coasts. We pursued this idea a little further and compared the coefficient of variation of the brown and white shrimp catches (Table 7). The coefficient of variation for the brown shrimp was 25.918 and for the white shrimp was 28.569. A comparison of the significance of differ- ences between two variants showed that this was significant at the classical 95% level. This means that the brown shrimp production is probably less variable than the white shrimp production on the United States coast, and possibly a longer series of data will clarify this point. A list of significant correlations determined in this study and a list of correlations which are not statistically significant are given in Tables 8 and 9, respectively. Shrimp Landing Statistics 201 Table 7. Total Brown and White Shrimp Catches of the United States in Thousands of Pounds Year Browns Whites 1957 69,681 20,683 1958 56,959 32,944 1959 75,648 32,900 1960 71,507 40,581 1961 40,700 23,534 1962 48,580 29,208 1963 60,729 51,806 1964 46,563 49,250 1965 67,457 44,186 1966 71,176 35,865 1967 105,857 31,980 1968 82,558 41,922 1969 71,078 55,253 1970 85,364 54,073 1971 93,849 54,087 Table 8. A List of Significant Correlations Determined in this Study Degrees of Freedom r Signifi- cance 1. South Atlantic and Gulf browns and whites vs. values... 37 0.6912 1.0% 2. Gulf browns and whites vs. values... 37 0.7368 0.1% 3. South Atlantic browns and whites vs. Gulf browns and whites. 37 0.3261 5.0% 4. Atlantic whites vs. Gulf whites ... 32 0.6550 1.0% 202 Gunter and McGraw Table 9. A List of Correlations Determined in this Study that are not Statistically Significant Degrees of Freedom r 1. South Atlantic grooved vs . whites. 15 -0.003 2. Gulf grooved vs. white shrimp 16 0.094 3. South Atlantic browns vs . whites . . 15 -0.065 4. Gulf browns vs. whites..- 16 0.051 5. South Atlantic and Gulf browns vs. South Atlantic and Gulf whites 13 0,2790 6. Texas grooved vs. whites 16 -0.148 7. Texas browns vs. whites 16 -0.215 8. South Atlantic browns and whites vs. value,... 39 0.0674 9. South Atlantic browns vs. Gulf browns 15 0.121 SUMMARY There are five species of commercial penaeid shrimp extending from Cape FTatteras, North Carolina to northern Mexico. One is lo- calized in Biscayne Bay, Florida and one is only produced in low per- centage (less than 2 %) of the total catch in the Gulf of Mexico. A third species, the pink shrimp, has had foreign catches so mixed with the domestic production that local figures on the Gulf Coast for past years are not reliable. Fairly adequate production figures for white shrimp are available for the years 1902 to 1947 and 1958 to the pres- ent, From 1948 to 1967 the brown and white shrimp catches were mixed and to some extent with the pinks. After 1958 all species were separated in the catch records. Shrimp Landing Statistics 203 There is a strong positive correlation between total shrimp pro- duction of the United States and value (dockside price) of the shrimp, and an even more significant correlation between Gulf production and value. In contrast the much smaller South Atlantic shrimp catch shows no correlation with prices, probably because the stock has been fished to capacity since the 1920s, when production limits seem to have been obtained. A strong correlation exists between white shrimp production of the South Atlantic and the Gulf, while none was found for the brown shrimp production of the two areas. A possible explanation for this fact is the deeper w’ater distribution of the brown shrimp, which means a more stable environment, less affected by general climatic oscillations which influence white shrimp in shallow waters and cause similar variations in the two populations. There is no significant correlation between the total United States production of white and brown shrimp, either positively or negatively, nor are there any correlations of the South Atlantic and Gulf areas considered separately. This means that the brown and white shrimp are weakly competitive, if at all. We wish to thank Mr. Paul Poole, data processor at the Gulf Coast Research Laboratory, for his assistance • in the statistical analyses. LITERATURE CITED Eldred, Bonnie, i960. A note on the occurrence of the shrimp, Penaeus brasiliensis Latreille, in Biscayne Bay, Florida. Quarterly .Tournal of the Florida Acad- emy of Sciences 23(2) : 164-65. Gunter, Gordon. 1950. Seasonal population changes and distributions as related to salinity of certain invertebrates of the Texas coast, including the commercial shrimp. Pub. Inst. Mar. Sci. 1(2) t7-51. 1961. Some relations of estuarine organisms to salinity. Limnol. Ocean- og. 6(2) : 182-90. 1962. Shrimp landings and production of the State of Texas for the period 1956-1959, with a comparison of other Gulf states. Publ. Inst. Mar. Sci. 8:216-26. , J. Y. Christmas, and R. Killebrew. 1964. Some relations of salinity to population distributions of motile estuarine organisms with special reference to penaeid shrimp. Ecology 45(1) : 181-85. , and .Judith C. Edwards, 1969. The relation of rainfall and freshwater drainage to the production of the penaeid shrimps (Penaeus fiuviatilis Say and Penaeus aztecus Ives) in Texas and Louisiana waters. FAO Fish. Rep. 57(3) :875-92. , and Gordon E. Hall. 1963. Biological investigations of the St. Lucie es- tuary (Florida) in connection with Lake Okeechobee discharges through the St. Lucie canal. Gulf Res. Rep. 1(5) :189-307, Lindner, Milton J. 1957. Survey of shrimp fisheries of Central and South America. U. S. Fish Wildl. Scrv. Sp. Sci. Rep. Fish. 235 :1-166. , and William W. Anderson, 1956. Growth, migrations, spawning, and 204 Gunter and McGraw size distributions of the shrimp Penaeus setiferus. U. S. Fish Wildl. Serv., Fish. Bull. 106:555-646. Lyles, Charles H. 1969. Fishery statistics of the United States: 1967. U. S. Gov- ernment Printing Office. Washing^ton, D. C. Say, Thomas. 1817. An account of the Crustacea of the United States (continued). Journal of the Academy of Natural Sciences 1(6) :235-353. Weymouth, F. W., M. J. Lindner, and W. W. Anderson. 1933. Preliminary report on the life history of the common shrimp, Penaeus setiferus (Linnaeus). U. S. Dept. Commerce, Bur. Fish. Bull. 14:1-26. Gulf Research Reports Volume 4 Issue 2 January 1973 Effect of Holothurin on Sarcoma 1 80 and B- 1 6 Melanoma Tumors in Mice S.D. Cairns Louisiana State University C.A. Olmstead Louisiana State University DOI: 10.18785/grr.0402.07 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Biology Commons, and the Marine Biology Commons Recommended Citation Cairns, S. and C. Olmstead. 1973. Effect of Holothurin on Sarcoma 180 and B-16 Melanoma Tumors in Mice. Gulf Research Reports 4 (2): 205-213. Retrieved from http://aquila.usm.edu/gcr/vol4/iss2/7 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. EFFECT OF HOLOTHURIN ON SARCOMA 180 AND 8-16 MELANOMA TUMORS IN MICE^ by S. D. CAIRNS 2 and C. A. OLMSTED Department of Biological Sciences, Louisiana State University in New Orleans, Louisiana 70122 ABSTRACT Holothurin, a biotoxic principle from the Cuvierian glands of the Bahamian sea-cucumber, Actinopyga agassizi, was studied as an anti- tumor agent capable of retarding tumor grow’th and prolonging the life of tumor-bearing mice. White Swdss mice injected with Sarcoma 180 had a mean survival time of 18,1 days wdth an average weight gain representing tumor and ascites fluid accumulation amounting to 25.1 grams in 15 days. Of five white Swiss mice with Sarcoma 180 re- ceiving 0.15 mg Holothurin every other day, one was alive at 57 days and the average weight gain of the group was 10.4 grams in 15 days. White Swiss mice with Sarcoma 180 which survived the lethal effects of slightly higher doses of Holothurin also had prolonged survival times and negligible tumor growth. C57-B1-6J mice with B-16 mela- noma tumors did not show increased survival time using the same doses of Holothurin that were effective in w'hite Swiss mice. Toxicity tests indicated that the safe upper limit for intraperitoneal Holothurin injection in white Swiss mice was 0.10 mg/day, 0.15 mg given every other day, and up to 0.30 mg given in a single injection. Similar tests with C57 black mice showed survival with as much as 0.60 mg Holo- thurin in a single subcutaneous injection. Holothurin was found to be 250 to 500 times more effective in causing red blood cell hemolysis than saponin and thus appears to have some action on living cells in addition to its surfactant action. INTRODUCTION Cancer chemotherapy began in 1946 when nitrogen mustai’d was used in treating leukemia patients. During the next 15 years only some thirty drugs were used in cancer chemotherapy but screening of many thousands of compounds was carried out each year (Clark 1961), Al- though marine invertebrates provide a particularly rich source of compound.^ with biological activity in mammalian species, less than ^ This work was supported in part by the Cancer Association of Greater New Orleans, Inc. 2 Present address: Department of Biological Oceanography, the Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, Florida 33129 . 205 206 Cairns and Olmsted 1% of all marine invertebrates thought to possess biotoxic substances have been investigated and of those, only a dozen have been thorough- ly investigated pharmacologically (Halstead 1965). Small concentrations of a crude water extract made from the whole body of the Bahamian sea-cucumber, Actinopyga agassizi, was found to be lethal to mice and fish and to have some tumor cell inhibi- tory action in vitro (Nigrelli and Zahl 1952). The active principle of this crude extract was found to be concentrated in the Cuvierian tub- ules located in the sea-cucumber respiratory tree and was named Holo- thurin (Chanley et al. 1955). Holothurin w’as the first known steroid saponin of animal origin (Nigrelli et al. 1959). Chemical analysis of Holothurin indicates that it is highly soluble in water, non-volatile, heat stable, and exhibits surface-active properties (Nigrelli and Ja- kowska 1960). Holothurin appears to consist of a few steroid agly- cones that are bound individually to four monosaccharide molecules. A provisional formula has been proposed (Alender and Russell 1966). See Fig. 1. Figure 1. Holothurin formula. Proposed structure for Holothurin mole- cule, a steroid saponin of animal origin; the monosaccharide groups rep- resented by Si-S. are, respectively, D-glucose, D-xylose, D-quinovose, and 3-o-methyl glucose. Holothurin Effects on Sarcoma 207 Holothurin is highly toxic to many types of organisms in very small concentrations. It retards onion root tip development (Nigrelli and Jakowska 1960) and fruit fly pupation (Goldsmith* Osburg and Nigrelli 1958) at 1000 ppm, alters regeneration of planarians at 100 ppm (Qiiaglio et al. 1957), is lethal to the “pearr’ fish Campus at 1 ppm and affects sea urchin development at 0.01 ppm (Ruggieri and Nigrelli 1960). Holothurin exhibits a hemolytic effect on red blood cells and reduces the size of a subcutaneously injected Sarcoma 180 tumor in white Swiss mice (Nigrelli 1952). Krebs-2 ascites tumors in white Swiss mice are also inhibited by Holothurin (Sullivan, Ladue and Nigrelli 1955). The present study demonstrates inhibition of Sarcoma 180 tu- mors in white Swiss mice but not of B-16 melanoma tumors in C57 Black 6J mice. Toxicity of Holothurin was estimated by measuring its effects on red blood cell hemolysis and by lethal-dose measurements in mice. It appears that the anti-tumor activity of Holothurin is not only dosc-dcpcndcnt but that the effective dose is determined by the sensi- tivity of the species to the lethal effects of Holothurin. MATERIALS AND METHODS Holothurin : The Holothurin referred to in this paper was a “crude extract” obtained from Dr. Ross Nigrelli and was made from the sun-dried, powdered Cuvierian glands of Bahamian sea-cucum- bers, Actinopyoa agnssizi. Red Blood Cell Hemolysis Tests Using Holothurin and Saponin : Fresh red blood cells w^ere obtained from white Swiss mice just prior to testing the hemolytic effects of Holothurin and saponin of plant origin. The blood was obtained by cardiac puncture while the mice were under light ether anesthesia. Approximately 5 ml of blood was diluted in 100 ml of balanced-saline solution buffered at pH 7.4 (Olm- sted 1967) containing calcium o.xalate (0.1% ) to prevent coagulation. A serial dilution of Holothurin wa.s prepared ranging from 2 mg ml to 0.001 mg/ml and a similar series of dilutions of saponin was made for comparison. One ml of diluted blood was added to 5 ml of each of several con- centrations of both Holothurin and saponin. The test tube was centri- fuged 1 minute after mixing the contents and the optical density of the supernatant was measured using a spectrophotometer set at 540 mu. The degree of hemolysi.s was estimated by the concentration of hemoglobin in the supernatant. These values were compared to 100% hemolysis values obtained either by using water only as the diluent or by using a high concentration of Holothurin or saponin. Holothurin Toxicity in Mice: All Holothurin injections w^ere given at dosages determined by previous tests done in this laboratory 208 Cairns and Olmsted on normal, non-tumor-bearing mice. Tests to determine the lethal and sub-lethal intraperitoneal doses of Holothurin in white Swiss mice were carried out using several concentrations given daily, given on alternate days, and for several concentrations given in a single injec- tion. The dosages of Holothurin given to the white Swiss mice bearing Sarcoma 180 tumors were based on these data. The dosages of Holo- thurin given to the C67 black mice bearing B-16 melanoma tumors was based on the effective doses of Holothurin given to tumor-bearing white Swiss mice. In addition, a test of toxicity of a single subcuta- neous injection of Holothurin was carried out using several concentra- tions of Holothurin given to normal, non-tumor-bearing C57-B1-6J mice. Holotkm'in Administrotio^i to Mice Bearing Sarcoma ISO: Sar- coma 180 cells and the white Swiss mice were obtained from the Gulf South Research Institute in New Orleans. The mice were inoculated intraperitoneally with 0.5 ml of Sarcoma 180 ascites fluid containing 11.6 X 10^ cells/ml. Groups of five mice received daily injections of Holothurin at 0.01, 0.05, 0.10, and 0.15 mg each, and three groups re- ceived injections of Holothurin every other day in the amounts of 0.10, 0.15 and 0.20 mg each. Twelve mice received Sarcoma 180 inocu- lations, but no Holothurin. The survival time was recorded for each mouse and the mice were weighed individually before inoculation with tumor cells and at 5-day intervals during the course of the experi- ments, Holothurin Admmistration to Mice Bearing B~16 Melanoma: A single C57-B1-6J mouse with a B-16 melanoma tumor was used as the source for all of the B-16 melanoma used in this study. The tumor- bearing donor mouse was obtained from the Department of Surgery at Tulane Medical School. The C57-B1-6J mice used in the experi- ments here were obtained from the Roscoe B. Jackson Memorial Lab- oratory in Bar Harbor, Maine. The solid tumor was taken from the donor mouse under light ether anesthesia and was minced with scis- sors in a tissue culture nutrient medium containing serum (Olmsted 1967). Several .small pieces of the minced tumor were trochared sub- cutaneously into the left .side of thirty-five recipient male mice. Thirty of these mice were used for the experiments reported here. Daily in- jections of 0.05, 0.10, and 0.15 mg Holothurin were given to each of the five tumor-bearing mice in three groups and five tumor-bearing mice in another group were each given 0.20 mg Holothurin every other day. Five mice without melanoma, and ten mice with melanoma tumors, were also maintained but not given Holothurin. All Holo- thurin injections were given subcutaneously near the region of tumor implantation in 0.10 ml volumes, and were given over a 13-day period. The survival time was recorded for each mouse, and the mice were weighed in groups of five before tumor implantation and at 6-day in- tervals during the course of the experiments. Holothurin Effects on Sarcoma 209 RESULTS Hemolysis Tests: Minimum concentrations of Holothurin and saponin that produced complete hemolysis were found to be 0.001 mg/ml of Holothurin and 0.25-0.50 mg/ml of saponin indicating that Holothurin was 250 to 500 times more active as a hemolytic agent than saponin. Holothurin Toxicity in Mice: Holothurin treatment of tumor- bearing white Swiss mice at the level of 0,15 mg/day was lethal at 3 to 8 days with a mean of 4.2 days. See Table 1. Toxicity tests on other Table 1. Effects of Holothurin on White Swiss Mice with Sarcoma 180 Body Weight Change No. of Mice Sex Dose (mg) Frequency of iniection Survival Time (days) 5th day (8) 10th day (e) 15th day (e) 12 F 0 0 18.1 (15-21) +2.3 +13.0 +25.1 5 F 0.01 daily 19.2 (15-22) +2.0 +12.7 +27.6 5 F 0.05 daily 17.8 (16-23) 0 +13.8 +33.0 5 F 0.10 daily 16.3 (14-19) -1.4 + 8.2 +30.7 5 F 0.10 alternate days 16.8 (15-20) -1.1 +10.1 +29.4 5 F 0.15 alternate days 26.4 (16-58) -1.0 + 2.9 +10.4 5 F 0.15 dally 4.2 (3-8) died ■“ — — 5 M 0.20 alternate days 22.2 (10-47) -1.8 0 0 groups of normal, non-tumor-bearing white Swiss mice indicated that either 0.10 mg/day or 0.15 mg given on alternate days was a safe upper limit for intraperitoneal Holothurin administration. Addition- ally, 0.30 mg Holothurin in a single injection could be tolerated, but more than this amount in a single injection was lethal. Toxicity tests 210 Cairns and Olmsted OB C57-B1-6J mice without B-16 melanoma tumors indicated that these mice could survive single subcutaneous injections of Holothurin as high as 0.60 mg. Effect of Holothurm o?i White Swiss Mice with Sarcoma 180: Twelve tumor-bearing mice that were not treated with Holothurin lived for an average of 18.1 days after tumor cell inoculation. These mice gained an average of 25.1 g by the fifteenth day after tumor cell inoculation. This increase in body weight represents the extent of in- crease in tumor size and ascites fluid accumulation since body weight increase due to growth of the animal during this time would have been negligible. The tumor-bearing animals receiving 0,15 mg Holo- thurin on alternate days had a mean survival time of 26.4 days repre- senting an increase in survival time of 46%, and a mean w'eight gain at 15 days of 10.4 g representing a 60 to 70% decrease in tumor growth and ascites fluid accumulation. One of the mice in this high Holothurin dose group had a weight gain of only 5.1 g and was still alive after 58 days. Three of the mice receiving 0,20 mg Holothurin every other day died in less than 12 days probably as a result of Holo- thurin overdose. Of the other two mice in the 0.20 mg group, one lived for 30 days and the other for more than 47 days and both had negligible weight gain during this period of time indicating complete tumor suppression. The four groups of tumor-bearing mice receiving less than 0.15 mg of Holothurin per injection did not show any sig- nificant increase in survival time or decrease in weight gain. See Table 1, Effects of Holothurin on C57~Bl-6J Mice with B-16 Melanoma: Ten C57-B1-6J mice not injected with Holothurin lived for an average time of 29.4 days after B-16 melanoma inoculation with a range of 18 to 43 days. The animals with malignant melanoma tumors de- creased in total body weight. Normal, non-tumor-bearing C57-B1-6J mice of comparable age to the tumor-bearing mice all gained w'eight on the diet and the conditions' of this experiment. Measurement of body weight w'as not a valid indicator of tumor growth or of Holo- thurin effectiveness, as in the case of mice with Sarcoma 180, because the melanoma tumor grows in a small solid mass and does not induce ascites fluid accumulation. As the Holothurin dosages used here were comparable to the effective doses for w^hite Swiss mice bearing Sar- coma 180 tumors^ and were well below the toxic levels of Holothurin for C57-B1-6J mice, none of the animals in this experiment showed any increase in the survival time. See Table 2. DISCUSSION The crude extract of Holothurin used here was probably more potent than purified preparations would be. Upon purification, the anti-tumor principle is apparently lost or considerably reduced in con- Holothurin Effects on Sarcoma 211 Table 2. Effects of Holothurin on C57-B1-6J Mice with Melanoma Body Weight Change No. of Mice , Sex 1 Dose (mg) Frequency of injection Survival Time : (days) 5 th day (fi) ! 10th day (8) m I 20 th day (8) 10 M 0 0 29.4 (18-43) +0.4 -0.7 -0.8 -0.3 5 M 0.05 daily 25.6 (20-42) -1.1 -3.4 -3.5 -2.5 5 M 0.10 daily 22.7 (19-26) -4.1 -4.8 -5.4 -4.4 5 M 0.15 daily 1 i 24.4 (20-35) -3.1 -3.5 ! -4.3 -4.5 5 M 0.20 alternate days 29.6 (25-33) -2.6 -3.5 -3.6 -3.3 centration (Chanley et al, 1960). One sample of purified Holothurin showed a tenfold-loss in hemolytic activity, and no anti-tumor activ- ity (Nigrelli and Jakowska 1960). Although the crude extract of Holothurin is effective against var- ious tumors (Nigrelli 1952, and Sullivan, Ladue and Nigrelli 1955), the present study demonstrates that Holothurin is apparently not ef- fective at the same dose levels in differing species of mice. Larionov (1967) suggests that a natural biological product should be more prom- ising as a chemotherapeutic agent than a synthetic compound. The same author has stated that the maximum effective acceptable dosage of an anti-tumor compound is close to the dosage which kills the animal, whereas a decrease in dosage, even slight, leads to a sharp drop in effectiveness. The narrow thre.shold is seen here in regard to the ineffectiveness of the lower concentrations of Holothurin on the Sarcoma 180 tumors, and the lethal effect of 0.20 mg given over a period of several alternate days to the tumor-bearing mice. Due to the narrow range of effective concentrations of Holothurin in relation to its lethal dose, it is likely that the ineffectiveness of Holothurin in the C57-B1-6J mice bearing B-16 melanomas was a reflection of the toler- ance to higher doses of Holothurin in these animals. There is no in- dication that Holothurin, or any other chemotherapeutic agent, would have differing effects in male and female tumor-bearing mice although profound metabolic differences between male and female rats and mice are well-known (Olmsted 1969). However, the differences seen 212 Cairns and Olmsted in this study could also be due to differences in the injection site or in the types of tumor. While this consideration presents many variables, the Sarcoma 180 tumor grows best in white Swiss mice and the B-16 melanoma grows best in C57-B1-6J mice. Further, it should be noted that the mean survival time of the untreated mice with the two dif- ferent types of tumor was about the same, i.e., about S to 4 weeks after tumor inoculation. The hypothesis is further supported by the finding that Holothurin is many times more effective in causing he- molysis than saponin when living red blood cells are used in the test system. Other hemolysis experiments similar to those described here have shown that between 0.04 and 0.10 mg of the crude extract of Holo- thurin would cause complete hemolysis, while 0.08 to 0,10 mg of sa- ponin was required for the same effect (Nigrelli and Jakowska 1960) . The apparent 40 to 100-fold discrepancy between the findings of Nig- relli and Jakowska and the findings of the present study could be ac- counted for by the use of different dilutions of red blood cells, or by differences in the age of red blood cells used, or by species effects. Nigrelli and Jakowska's data suggest that Holothurin is only slightly more effective in causing hemolysis than saponin, while in the present study using red blood cells that were freshly obtained from mice just prior to testing it would appear that Holothurin is some 250 to 500 times more effective than saponin in causing hemolysis. Since a num- ber of studies have shown Holothurin to have profound biological ef- fects in concentrations as low as 0.01 ppm it is not surprising that this biotoxic principle is much more effective than saponin on fresh red blood cells. It would therefore appear that the action of Holo- thurin involves more than a surface tension lowering effect on living cells. Because Holothurin is an effective anti-tumor agent at concen- trations only slightly lower than the lethal dose, it might be tenta- tively hypothesized that Holothurin is acting on some process common to all cells, not just tumor cells. However, no studies have been done on the metabolic effects of Holothurin. The studies reported here indi- cate that the mode of action of Holothurin in retarding tumor growth and prolonging life of tumor-bearing mice is probably different from the action of a surfactant such as saponin. It is likely that the energy requirements of the white Swiss mice with Sarcoma 180 differed from those of the C67-B1-6J mice bearing B-16 melanoma tumors. Other studies from this laboratory on the C67-B1-6J mice with B-16 mela- noma tumors have shown that the growth of melanoma tumor occurs at the expense of the anabolic processes of the host mouse. Structural phospholipids synthesized by the tumor were shown to be derived from synthesis in the liver (Terranova, Hardy and Olmsted 1970). Although several experiments have been reported in the literature regarding .saponin, none have shown saponin to have any anti-tumor activity in vivo. It is likly that saponin destroys cells simply by its Holothurin Effects on Sarcoma 213 action to lower surface tension starting with the plasma membrane and continuing its action on membranes of other cellular constituents in proportion to the concentration of saponin used. While the action of Holothurin has some yet unknown eilect other than, or in addition to, its surfactant effect, it is likely that it causes an inhibition of an active, or energy-dependent, physiological process. LITERATURE CITED Alender, C- B. and F. E, Russell. 1966. In Physiology of Echinoderms, R. A. Boolootian (ed.), p. 531. Interscience, New York. Chanley, J. D., S. K. Kohn, K. Nigrclli and H. Sobotka. 1955. Zoologica, 40:99. Chanley, J. D., R. T.edeen, J, Wax, R. Nigrelli and H. Sobotka. 1959. J. Am. Chem. Soc., 81:5180-3. Chanley, J. D., J. Perlstein, R. Nigrelli and H. Sobotka. 1960. Ann. N. Y. Acad. Sci,. 90:902-5. Friess, S. L., F. G. Standaert, B. R. Whitcomb, R. Nigrelli, J. D. Chanley, and II. Sobotka. 1959. J. Pharmacol., 126:323-9. Clarke, R. L. 1961. Cancer Chemotherapy, Chas. C, Thomas, Springfield. Goldsmith, B. D., H. B. Osburg and R. Nigrelli. 1958. Anat. Rec., 130:411-12. Halstead, B. W. 1965. Poisonous and Venomous Marine Animals of the World, U. S. Government Printing Office, Washington, D. C., 1 :994. Larionov, L. F. 1967. Recent Results in Cancer Research, Vol. 8, P, Rentchnick (ed.), Springer, New York. Nigrelli, R. 1952. Zoologica, 37:89-90. Nigrelli, R. and P. A. Zahl. 1952. Proc. Soc. Exp. Biol. Med.. 81:379-80. Nigrelli, R., J. D. Chanley, S. K. Kohn and H. Sobotka. 1955. Zoologica, 40:47-8. Nigrelli, R. and S. Jakowska. 1960. Ann. N. Y. Acad. Sci., 90:884-91. Olmsted, C. A. 1967. J. Cell Biol., 48:283-99 Olmsted, C, A. 1969. Lipids., 4:401-12. Quaglio, N. D„ S. F. Ndan, A. M. Veltri, P. M. Murray, S. Jakowska and R. Nigrelli. 1957. Anat. Rec., 128:604-5. Ruggieri, G. D. and R. Nigrelli, 1960. Zoologica, 46:1-16. Sullivan, T. D., K. T. Ladue and R. Nigrelli. 1955. Zoologica, 40:49—52. Terranova, J., K. Hardy and C. A. Olmsted. 1970. J. Am. Oil Chem. Soc., 47; 78 A. Gulf Research Reports Volume 4 Issue 2 January 1973 A Geochemical Study of a Marsh Environment Thomas R Lytle Gulf Coast Research Laboratory Julia Sever Lytle Louisiana State University Patrick L. Parker University of Texas DOI: 10.18785/grr.0402.08 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Geochemistry Commons, and the Marine Biology Commons Recommended Citation Lytle; T. R, J. S. Lytle and R L. Parker. 1973. A Geochemical Study of a Marsh Environment. Gulf Research Reports 4 (2); 214-232. Retrieved from http:/ / aquila.usm.edu/gcr/vol4/iss2/ 8 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell^usm.edu. A GEOCHEMICAL STUDY OF A MARSH ENVIRONMENT by THOMAS F. LYTLE Gulf Coast Research Laboratory Ocean Springs, Mississippi JULIA SEVER LYTLE Department of Chemistry Louisiana State University Baton Rouge, Louisiana and PATRICK L. PARKER Marine Science Institute The University of Texas Port Aransas, Texas INTRODUCTION The study of the fate and distribution of carbon compounds and their associated trace elements in contemporary environments as well as ancient sediments has added information of importance to geology and the ocean sciences. These recent advances have been spurred on by such interests as the search for deposits of fossil fuels, the control of pollution, the understanding of the origins of life, the analysis of lunar samples, and the hopes for analyses of the surface of Mars. The organic matter found in Holocene sediments is derived from once living organisms. The complex molecules synthesized by orga- nisms and their degradation products are the object of study of organ- ic geochemistry. After the death of an organism^ most of the organic matter in it is utilized by other organisms or oxidized to CO2. A small amount of the organic matter is not destroyed but is trapped in the top sediment. Biological activity gradually ceases, but slow chemical and physical transformations are continuous. Since the carbon-carbon bond is a strong bond, portions of the original molecules deposited in sediments often survive for long periods of time. Detection of many organic molecules in the concentrations found in nature was impos- sible before recent improvements in instrumental methods of separa- tion, characterization and quantitation. With the present tools of mass spectrometry and gas chromatography the field of organic geo- chemistry has undergone a tremendous surge of activity. The trace metal components of sediments provide a rich area for the study of geochemistry with such a diverse array of mechanisms responsible for their inclusion in the sedimentary environment. The biogenic source of trace metals in sediments is usually assumed but 214 Geochemical Marsh Study 215 rarely stimulates a great deal of interest. However, the biogenic com- ponent may play a significant role in trace metal fixation in highly productive areas. Because trace metals in a sediment may undergo radical changes in concentration levels with only slight changes in the overall chemistry of the sediment, they should be very sensitive indi- cators of certain microscale and macroscale chemical and biological factors in sedimentary environments. All organisms rely to a greater or lesser extent upon a constant supply of trace metals for the func- tioning of their metabolic processes. So profound is the need for cer- tain trace metals in all organisms that they are termed essential trace metals whereas others may form an essential ingredient in the nutri- ents of only a few organisms. Others may perform no known func- tion or be toxic to most organisms. Even so, plants and animals have the capacity to concentrate most trace metals from their environ- mental sources. When organisms die, most of the trace metals are re- leased after decay, but a fraction may be retained in the organic debris trapped in the sediments and be preserved as a remnant of the pre-existing life forms. To study the trace metal transfer between sediments and plants, analytical tools of high sensitivity are required to cope with trace concentrations. Fortunately methods and instru- ments — particularly atomic absorption spectrophotometry — having the required sensitivity are available at modest costs to those inter- ested in conducting trace metal analysis and have greatly facilitated this type of study. Harbor Island, located approximately 1 mile north of Port Aran- sas, Texas (Fig. 1), was chosen as an environment to investigate in an organic and trace metal geochemical study. Situated at the head of Aransas Pass Channel, this tidal delta is characterized by a series of salt marshes inundated by tidal creeks and tidal channels. Maximum relief on the island is but 12 feet, but even slight variations in eleva- tion are denoted by the variation of plant species dominating each zone of elevation. The plants are limited to the more hardy herbaceous species which withstand the extended periods of low rainfall and sedi- ment dessication found in the intertidal and supratidal areas. Most of the identified species of plants are also native to the salt marshes of the off-shore islands of Mississippi. The fauna was restricted primar- ily to members of the lower phyla of the animal kingdom. Several factors made this island ideally suited for geochemical investigations. The topography reveals rather simple and small scale physiographic processes. The island is protected both from effects of major oceanic disturbances by the barrier islands and from the effects of man by its relative inaccessibility. The minimal number of external parameters was helpful in facilitating the interpretation of the geo- chemical data collected. The goal of this study was twofold: 1) chemically to charac- terize a specific salt marsh, the sediments and associated biota and 2) 216 Lytle, Lytle and Parker Figure 1, Map of collecting area. Geochemical Marsh Study 217 to establish clear relationships between the chemical substances re- siding in the sediments and similar or identical substances occurring in the biological specimens. In this particular study it was felt that the hydrocarbons would yield the most significant organic geochemical information. These compounds are ubiquitous but minor components of all organisms. Though their function is not entirely understood, it is known that they are concentrated in the w^axy coatings of plants and mo.st likely aid in the protective mechanisms of plants. Among the several classes of bio- chemical materials, the hydrocarbons exhibit probably the greatest resistance to biological and chemical degradation and therefore may be preferentially concentrated and preserved in sedimentary environ- ments. An extreme variet^^ of specific hydrocarbons occurs naturally in plants and hence the possible combinations and distributions of these hydrocarbons are limitless. The stability and unique distribu- tions of hydrocarbons ranks them as a very important tool in the cor- relation of biolipids and geolipids. In choosing the trace metals to be studied, preference was given to those that are known to be essential to all organisms. Those ele- ments fitting this description were copper, zinc, molybdenum, man- ganese and iron. In addition cobalt known to be essential at least to blue-green algae was included. To monitor the biogenic contribution of non-essential trace metals, nickel, cadmium and lead w'ere also analyzed in the samples. With the diverse functions of these various elements and occurrence at easily measurable levels in most organ- isms, they should provide useful information about the plant-sediment interplay in a salt marsh. EXPERIMENTAL CoUevlion and Handling A shallow depression outlined by a band of Batis maritiina (Salt- wort) and Salicomia bigelovii (Glass wort) was chosen as the site for sediment and plant collection for organic analysis. The sediments were black indicating a reducing environment and contained root mot- tles and plant debris. Algal mats were forming at the edges of the de- pression and benthic organisms could be seen as they fed on ditritus present at the surface. Sediment samples chosen for trace metal analy- sis were surface sediments from three intertidal zones of the island. In addition one subtidal, subsurface .sample was taken from the lit- toral zone of the island. This latter sample was taken to define the trace element distribution in an area not influenced by the marsh plants. For the organic analysis, seven ‘marsh plants were collected, washed with distilled water and extracted immediately after collec- 218 Lytle, Lytle and Parker tion. After washing the whole plants with distilled water to remove loose debris and epiphytes, the plants were minced in a Waring blen- der with methanol. The material was filtered and the methanol saved. The minced material was extracted ultrasonically for 15 minutes with 200 ml of chloroform with constant stirring. In addition to these seven plants, seven more species were collected for trace metal analysis (Table 1). Only herbaceous portions of the plants were saved. These Table 1. Marsh Plants Analyzed HYDROCARBONS AND TRACE METALS Limoniwn aavoli-nianim (Walt.) Britton (Sea lavender) Batis mca^ltima L. (Saltwort) Salioomia higelovii Torr. (Glasswort) Lyoiym carolinianum Walt. (Christmas berry) Sesuvium mapitimm. (Walt.) BSP (Sea Purslane) Oenothera drummondii Hooker (Evening primrose) Bovpiohia frutesoens (L.) DC. (Sea ox-eyes) TRACE METALS Spartina altevni flora Loisel (Cord grass) Machaeranthera phyllooephala (DC.) Shinners Spartina patens (Ait.) Muhl, (Salt-grass) Monanthochloe littoralie Engelm. (Key-grass) Hedyotis nigidoans (Lam. ) Fosberg Avioennia germinans (L.) (Black mangrove) Distiohlis spioata (L.) Greene (Marsh spike-grass) portions were quickly rinsed in double distilled water and oven dried at 60°C for 48 hrs. Following ashing in a muffle furnace at 450'’C for 24 hrs., the samples were leached with 3N HCl and the filtrate over glass fiber filters was saved. Sediment samples for all analyses were hand collected, and all necessary precautions were taken to eliminate contaminations and to arrest bacterial action. The sediment samples for hydrocarbon analy- sis were digested with dilute HCl to remove carbonates and w’ashed to remove inorganic salts. Sediments for trace metal analysis were Geochemical Marsh Study 219 oven dried at 60 °C, weighed, ashed at 450°C digested in 3N HCl, fil- tered on glass wool and the filtrate saved. The extracts from plants and sediments were further separated as follows : Hydrocarbon Separation and Characterization Each lipid sample (plant and sediment) was saponified by reflux- ing with 0.5N KOH-MeOH for 1 hour. Nonsaponifiable components were removed by extracting the alkaline solution with benzene. After- wards, the alkaline solution was acidified with dilute HCl to pH 3, and the fatty acids were extracted into benzene. Methyl esters of the fatty acids were prepared using BF.,-MeOH (Metcalfe and Schmitz 1961). Silica gel (Woelm, Grade 200, Act. I) was packed beneath 25 ml alumina (Woelm Neutral, Grade 100, Act. I) in a 43 cm X 2.5 cm (o.d.) column. The nonsaponifiable residue was fractionated on the column into four parts : n-hexane fraction contained aliphatic hydro- carbons; benzene fraction contained aromatic hydrocarbons; chloro- form-methanol (4:1, v/v) fraction contained alcohols; methanol frac- tion contained glycerides and the polar lipids. The aliphatic hydrocarbons were identified and measured by gas chromatography on columns of SE-30, Apiezon L, and FFAP (Va- rian). Standard hydrocarbons were used to calibrate the Perkin- Elmer 880 gas chromatograph equipped with hydrogen flame ioniza- tion detectors. The columns were 8 feet by 1/8 inch o.d. copper tubing. The support was 80/100 mesh Chromosorb G, acid washed, dichloro- di methyl silane treated (Johns Manville). Fatty acid methyl esters were identified using the same in.strument but using columns with FFAP, SE-30, and diethylene glycol succinate (DEGS). Linear-log plots of the retention times yielded straight lines for both hydrocar- bons and acids and were useful for identification when standards for each carbon number were unavailable. Coinjection of standards were used to clarify some identifications. All hydrocarbon samples were run at programmed temperatures from 100° to 260° at 6®C per minute holding at 265°C. All fatty acid methyl esters were run from 150° to 265'' at 6°C per minute holding at 265°C. Mass spectra were obtained with a modified Consolidated Electrodynamics Corp., Model 21-1 03C mass spectrometer. The spectra w’ere run at 70eV. Additional sample treatment helped in some identifications. Urea adduction enriched the branched from the nonbranched components. Sulfuric acid and/or bromine-CCb treatment (Morrison and Boyd 1969) identified unsaturated component peaks. The organic carbon content of the sediment was determined by a combustion technique using a Leco gasometric carbon analyzer. 220 Lytle, Lytle and Parker Trace Metal Separation and Determination Iron was removed from the sediment filtrate by extraction of the iron-chloro complex from an 8N HCl solution of the total trace metals with isopropyl ether. Separation of all the trace elements of interest from matrix materials except Mo and Fe w'as effected by ad- justing: the pH to ca 3, adding sufficient ammonium pyrrolidine dithio- carbamate (APDC) to make the final solution 0.5% in APDC and ex- tracting the APDC-metal complexes with methyl isobutyl ketone (MIBK). Molybdenum w'as extracted from a fresh batch of filtrate by adding KCNS and SnCla to make a final IN HCl solution 2% in both. The molybdenum-CNS complex was extracted with isopropyl ether. All metals but molybdenum were analyzed using the Perkin- Elmer 303 atomic absorption spectrophotometer. The fuel was acety- lene and the oxidant, compressed air. The MIBK solution of cobalt, cadmium, copper, manganese, nickel and zinc and the aqueous solution of iron were aspirated directly using the following wavelength set- tings (in nm) : Cd-229, Co-241, Cu-325, Fe-248, Mn-279, Ni-232, Pb-283 and Zn-214. The color intensity of the Mo-CNS complex was measured at 460 nm on the Beckman Model DU spectrophotometer to determine molybdenum concentrations. Appropriate standards and blanks were prepared for all determinations and any necessary cor- rections were applied. RESULTS Organic The hydrocarbons from the marsh plants and the blue-green algal mats were assumed to be the main source of the hydrocarbons for the ecosystem studied. Microorganisms were present, but their contribu- tion is taken to be small in an organic-rich sink. Table 2 summarizes the analytical results. Normal straight-chain hydrocarbons ranging from C15 to C33 were identified in the sediments. Pristane and phytane were present in low concentrations. Phytane was detected only in trace amounts. The branched hydrocarbons were only a very small percentage of the total hydrocarbon fraction. These branched hydrocarbons were al- most completely olefinic. The sediment hydrocarbon distribution was bimodal with maxi- mums at Cl7 and C29 (Fig. 2) . However, the largest concentration of hydrocarbons was in the C27 to C31 range. This reflects the hydro- carbon pattern found in the two prominent contributors to the organic matter, blue-green algae and higher plant life. Normal alkanes iso- lated from the marsh plants showed an odd-carbon number predomi- Geochemical Marsh Study Table 2. Marsh Plant and Sediment Analytical Analyses 221 Marine Plant % lipid dry wt % HC dry wt % HC lipid wt Ma j or Compo- nent Limon'iion carolinianum 2.02 0.057 2.8 n-C29 Batis mavitima 2.31 0.008 0.37 n-C27 Salicomia higelovii 2.14 0.005 0.26 n-C31 Lyaium aavolinianum 3.41 0.085 2.3 I1-C29 Sesuvium maritirnwn 1.91 0.005 0.32 n-C25 Oenothera dmmiond'i'i 4.72 0.11 2.3 n-C29 Bond chi a frutesoens 4.91 0.043 0.75 n-C29 Sediment 0.15 0.0011 0.75 n-C29 Figure 2. Histogram for normal hydrocarbons of Harbor Island sedi- ments. 222 Lytle, Lytle and Parker nance with C27, C29, and C31 the most prominent alkanes while C17 was the most prominent alkane in blue-green algae. The urea adducted fraction of the hydrocarbons indicated that the cluster of peaks in the CIS range were not isoprenoid alkanes, but were branched-hydrocarbons with their branching points near the end of the chain. The large number of peaks in the C19 to C20 range were completely absent from the urea adducted fraction indicating that they were more highly branched. These peaks disappeared almost completely after treatment with H.SO, indicating that they w^ere ole- fmic. Infrared spectra of urea adducted and urea nonadducted frac- tions confirmed the evidence that the olefins present in the sediment hj'drocarbons are primarily multibranched olefins, in conclusion, the following generalizations can be made about the hydrocarbon results for the sediments analyzed : (1) All samples e.xhibitcd an odd-carbon preference. (2) A bimodal distribution was exhibited in the hydrocarbon patterns of sediments whose organic matter was derived from both blue-green algae and terrestrial plant sources. (3) The amount of extracted lipid material in plants was on the order of 25 times as great as that extracted from the sedi- ments. (4) Pristane and phytane were found in all samples. Normal alkane distributions in the plants were demonstrated be- tween Cl 3 and C35 but for the most part ranged between C23 and C31. Isoprenoid hydrocarbons, farnesane, pristane, and phytane, were identified, but this fraction represented less than 2% of the total frac- tions. Branched-chain hydrocarbons, iso-C27, C29 and C31, anteiso- C26, C28 and C30, were tentatively identified. There were some un- identified peaks in each chromatogram, but those peaks represented less than 10% of the total hydrocarbons. Borrickia fr'utescens con- tained a large number of branched and oleftnic peaks in the C15 to C18 molecular weight range, none of w'hich was identified (approxi- mately 21% of the total hydrocarbons fell in this Cl 5 to C18 range). In all other plants analyzed the lower molecular weight hydrocarbons represented less than 5% of the total weight. There was a definite odd-carbon preference in every plant sample. The largest component in four of the species was C29. The distribu- tions were slightly different for each species. Gas chromatographic analyses, tabulated in terms of individual normal hydrocarbons from C15 to 033, are presented in Fig. 3. Olefins have been reported in the hydrocarbons of many plants (Stransky and Streibl 1969), but there was little evidence of olefins PERCENT Caf/POSITIDN Geochemical Marsh Study 223 HYDROCARBON Figure 3. Histograms of normal hydrocarbons in marsh plants. 224 Lytle, Lytle and Parker being present in greater than trace amounts in any sample except Borrichia frutescens. Microhydrogenation of olefins in this particular sample yielded normal alkanes. However, some peaks in this sample were not changed by hydrogenation; nor were they enriched in the urea iionadducted fractions. Hence, they were tentatively identified as branched hydrocarbons having their branching points near the end of the carbon chain. Gas chromatograms are shown for two plants which contained unidentified peaks (Fig. 4). Tentative identification for the iso- and cwiteiso-alkanes as was given in Fig. 5 was based on their response to urea adduction, their inertness to HoSO^, and the absence of func- tional groups in the IR spectra. Plots of retention time vs. carbon num- ber for the three series, normal alkanes, ABC, and XYZ yields three parallel lines which indicates that the latter series are two distinct, homologous alkane families. Trace Metals In the sediments selected for trace metal analysis it was assumed that primary sources of biogenic material were the marsh plants. By choosing intertidal and supratidal sediments a large contribution from blue-green algae was hopefully excluded. A summary of the trace ele- ment data obtained from three surface and one subsurface sediments is contained in Table 3* Included are the results of organic car- bon determinations. Overall the levels of trace metals are quite low when compared to an “average” sediment from the near-shore area such as provided by Chester (i965). The distribution among the sedi- ments of any one element when compared to another shows similar trends so that it appears that no highly significant variations in trace metal fixations exist among the four sediments. However in inter- sediment comparisons it can be seen that significant differences do exist among the individual sediments, indicating degree rather than type as the more important aspect in investigating the effects respon- sible for trace metal fixation in these sediments. The fourteen marsh plants have a trace metal distribution which is shown in Table 4. In Fig. 5 are displayed the ranges of trace metals in the marsh plants and some results of other workers in summarizing available data on terrestrial and on marine plants. Fitting certain characteristics of both groups, it might be expected that the marsh plants would assume some sort of “middle ground”. These marsh plants appear to be slightly depicted in most elements except for molybdenum with respect to the terrestrial-marine range. With the principal source of heavy metals coming from the soil solution, it is not surprising from viewing the trace metal array in the sediments that these relatively low levels would be reflected in the biota of the regions. BATIS MARITIMA SESUVUM MARITINUM TIME MIN Figure 4. Chromatograms of Batis maritima and Sesuvium maritinum hydro carbon extract: Operating Coyiditions Column coating: FFAP Temperature program: 150°- Column dimensions: 150 ft. 210"^ at 4°C/min X 0.02 in Helium flow: 4 cc/min X-tso-C25 Key A-anteiso-C26 Y-iso-C27 B-anteiso-C2S Z-iso-C29 C-a7itei8o-CS0 226 Lytle, Lytle and Parker M n Fe Co Ni Cu 2n Mo Cd Pb 0.05 01 0.5 1.0 5,0 10.0 500 lOO.O 500.0 1000.0 Concentration (;jg/6 dry wt.) Figure 5. Ranges of trace metals in marsh plants — The range of each metal among the plants is indicated by the solid bar for that metal ex- tending from the minimum value to the maximum value- Also included are bars extending from averages of terrestrial (T) to those of marine (M) plants reported by Bowen (1966). 1 1 1 r- f 1 M “1 —i T T M T M M T T M M T M T T . M J 1 1_ 1 1 — 1 1 L_ DISCUSSION Hydrocarbons The odd-carbon predominance of the hydrocarbons in the Harbor Island sediments reflected the hydrocarbon composition of the marine plants, in particular the distribution in the C25-C31 range. The large concentration of CT7 reflected the blue-green algal contribution to the sediment hydrocarbons (Winter.s, Parker and Van Baalen 1969, Gelpi et al. 1970). Thus, the correlation between biological and geological lipids was found to exist, the sediments retaining the biological infor- mation needed in order to recognize what type of life dominated their surroundings. This correlation between biological and geological hy- drocarbons can be extended to include the isoprenoid hydrocarbons. Isoprenoids are abundant in the marine environment (Blumer 1965, Clark and Blumer 1967, Blumer, Mullin and Thomas 1964, Blumer and Thomas 1965). Pristane in marine algae was reported by Clark (1966). Pristane and phytane were found in photosynthetic and non- photosynthetic bacteria by Han et al. (1968). Isoprenoid hydrocar- Distribution of Trace Metals in Harbor Island Sediments Geochemical Marsh Study u u o £ Ti u N P u o u 'T3 0 ) CO VO CM vO CO CM ■ •H }-l • • • • 1 0) • p c c c > « 1 H p fo • P o p ■u P ■H 00 TO B P 00 o VO -a* • *H P es CvJ t— 1 1— 1 CM 00 X •H •H P > rH W P CO cn X) o CNJ VO o\ 4-> CU fo • • • • • 0) ■M ,Q X m <1- CM VO e cn u •H p 00 r-H •H X 00 00 p ov O O 0) p • • • • • CO $-1 iH p o VO lO m CO cn P CO 1— 1 P Xi cn P P 0) 0) 00 cn p tH P o O OV o CM cn tH P U • • • • • cu cn ■U P m CO CM CM CO u > p > g < cu 0) cn p p VO 'd- r— I m O o X ov (N CO CM I— 1 m CM • cn C P •H O m VO r— 1 •H cn •H P 1 1— 1 pH 00 > U u P P cn o X P o o X) X p ‘H p P 4-1 cn P X • P CTJ X p "V o U (U p 00 ■u •H 1 •U p p CO U ' c p p cs CO tt 0) p m p > •H a o > 00 > p *H p p 0) 1 Q 1 o CJ 227 detected 228 Lytle, Lytle and Parker CO 00 OS CO CM CO cr> CO CO 00 VD O CM 00 CO o i-H CM 1— 1 i-H CO ^ CO 1— 1 i-H ^ CO <3" >— 1 •. CM os CO r-. CTs m 00 o os "sT ♦ m • • VO VO 00 00 O OS CO o m CM CM O "d- CO 1-H CO CM 1-H m 1-H CM CO O 00 CO <1- 00 00 o CM m CO 1-H ■vT <3- I-H CO OS CM CM in I-H I-H m 00 CM 'd- CO OS m 00 r-H m LD m VO VO m 00 00 00 CM 00 in 00 CM CM 1-H o 1-H CM 1—1 1-H o o o O 1— 1 1-H o 1— 1 CO vD os I-H os in CO CO sr vO CM os CM VO t— 1 VO VO CO CM 1-H O O o I-H o O 'd- o o o o O O o o O d o o o O 00 O o CO I-H r-. O OS OS o m O o 00 O CM CM -d- OS -d- O OS •d- OS in 1-H CM OS OS 00 I-H CM I-H I-H CM m CM VO r- I-H r-. •d- in 1-H d- 1-H OS 00 O CM CO 1-H CM -d- CM CM CM 1-H cT CM 1-H 1-H CM The standard deviations for metal analysis by plant are added and then averaged to arrive at the Average Standard Deviation. Geochemical Marsh Study 229 bons are found in petroleum and crude oils in very high concentra- tions (Bendoraitis, Brown and Hepner 1962, Han and Calvin 1969). The presence of only trace amounts of pristane and phytane in Har- bor Island sediments indicates lack of pollution from these sources. This simple model should be useful. It could be of economic signifi- cance by its use in understanding the origin of fossil fuels. It could be of social significance by its use in understanding the fate of organic matter. With man’s environment being polluted at a rapid pace, it is essential that we understand the fate of organic matter whether it be in the geosphere, atmosphere or hydrosphere if we are to gain control over our environment. T race Metals In an attempt to emphasize the fine differences in trace metal levels in the Harbor Island Sediments a plot of trace metal vs organic carbon was made and shown in Fig. 6. This selection of X-Y variables was made assuming the percentage organic carbon is a measure of bio- genic input to the sediments. The trends and apparent deviations from these trends were observed rather than absolute levels to establish the correlation of organic matter and trace metal levels. Since a slightly different selection of marsh plants prevail in each sediment area, a semi-quantitative estimate was made of the actual biomass of plant material in a 25-foot radius of the three surface sedi- ments and the ratios of the various plants in those areas revealed the following plant contributors in decreasing order of importance: Sediment No. 1 — Salicornia higel-ovii, Batis maritima, Svartina alterniflora, Lycmm carolinianurri (Christmas berry), Li- monium caroliniamim, Borrichia friitescens (Sea ox-eyes) and Oenothera drummondii. Sediment No. 2--Bpartina alterniffora, Avicennia germinans (Black mangrove), Borrichia f^nttescens, Ma^haeranthera phyllocephala, Monanthochloe littoralis (Key-grass), Hedy- otis nig't'icans, Distichlis spicata (Marsh spike-grass) and Spartma patens (Salt-grass) . Sediment No. 3 — Salicornia bigelovii and Batis maritima. Defining the relationship of individual plants or groups of plants and the sediments for specific heavy metals was, in this limited study, impossible. However, by comparing the plant data to the data in Fig. 6 and observing the sediment-plant association groups some rather in- teresting information may be gained. With respect to all elements but molybdenum (not shown in Fig. 6) a relative slump is seen in Sediment No. 4, the subsurface sample with little contribution from marsh plants. Still other effects such as ABUNDANCE 0 230 Lytle, Lytle and Parker Figure 6. Metals organic carbon in Harbor Island sediments — Num- bers 1-4 on the organic carbon axis correspond to the Harbor Island soil sediment numbers from Table 3. Both ordinate and abscissa values are on a linear but arbitrary scale. Only the ordinate uses the origin as a value of zero. Organic carbon values: 1 — 0.66%, 2 — ^0.31%, 3 — ^0.22%, 4—0.41%. Geochemical Marsh Study 231 migration of trace elements to the surface and concentration there is a known phenomenon and might account for the surface enrichment. It might be expected that this effect would be noticeable even in the absence of the marsh plants. To a degree the marsh plants still could contribute a significant portion of the trace metals. For any release by dissolution by bacterial action of trace metals beneath the surface of an intertidal or supratidal zone would certainly be impeded by the con- tinual percolation downward of rainwater and receding tidal waters. Most elements show a positive correlation with organic matter. This previously has been deduced by several workers (Jenne 1968). However, as Jenne pointed out, this contribution by organic matter acts really as a secondary effect. In sediments not having an overwhelming load of organic matter the organic matter with its trace metal load can establish the proper conditions for inclusion of trace metals in ferromanganese oxides. These oxides present in most oxidizing sedi- ments are known to act as very effective trace metal accumulators. The marsh plants may not complete the fixation of trace metals in sediments but at least may start the process by acting as a source and sediment conditioner. Iron appears quite enriched in Sediments No. 1 and 3 relative to Sediments No. 2 or 4. It seems that any explanation based on a physio- graphical or geological basis would also alter the distribution of other elements more so than observed, It could very well be that the marsh plants are more active in the transport of this element than the others. Neither Sediment No. 2 nor 4 received any sizeable contribu- tion from Salieornia higelovU or Batis maritinum or others having the highest concentrations of iron. Baas Becking and Moore (1959) concluded that the majority of iron in marine sediments exists as com- plexed iron so that it is possible that these two plants supply substan- tial amounts of complexed iron in their plant litter. Though molybdenum is extremely depleted in the surface sedi- ments (Table 3), this metal maintains a normal level in the subsur- face sediment. In the marsh plants the levels of molybdenum are the only ranges above and including those predicted from averages of ter- restrial and marine plants (Fig. 5). It is known that in surface, oxi- dizing sediments molybdenum may be lost through its conversion to the soluble molybdate ion whereas in subsurface reducing sediments molybdenum as well as copper, iron, zinc et al. may be retained as their insoluble sulfides. The marsh plants then must speed the deple- tion of molybdenum from surface samples which after decay of the plants is either leached out of the sediments or transported to reduc- ing layers deeper in the sediments. Lead is greatly enriched in Sediment No. 2. Some of the plants in this area contain the highest levels of lead though their relative en- richment over those species in other areas could not fully account for 232 Lytle, Lytle and Parker this high level. The possibility of some lead or other metal contamina- tion cannot be entirely dismissed in view^ of the debris washed in by tidal action. The marsh plants as a whole have exhibited an ability about as strong as terrestrial and marine plants in accumulating a variety of trace metals from their environment. In some cases the marsh plants may act as sources for trace metal enrichment in the sediments and in other instances, notably molybdenum, as an active depleting agent. The coastal marsh environment widespread along the Gulf and Atlantic coastlines will undoubtedly continue to be a primary target of heavy metal pollution. The varying degrees of enrichment of the various ele- ments both essential and nonessential in marsh plants should make them of value in establishing base-line evaluations of heavy metal in- ventories in a coastal area. LITERATURE CITED Baas Becking, L. G. M. and D. Moore. 1959. The relation between iron and or- ganic matter in sediments. J, Sediment. Petrol. 29:454. Bendoraitis, J. G., B. L. Brown and L. S. Hepner. 1962, Isoprenoid hydrocarbons in petroleum. Anal. Chem. 34:4. Blumer, M. 1965. Organic pigments; their long-term fate. Science 149:722. Blumer, M., M. M. Mullin and D. W. Thomas, l.%4. Pri.stane in the marine en- vironment, Helgolander wis.s. Meeresunters 10:187. Blumer, M. and D, W. Thomas- 1965. Phytadienes in zooplankton. Science 147:- 1148. Bowen, H. J. M. 1966. In Trace Elements in Biochemistry, Academic Press, New York. Chester, R. 1965. Elemental geochemistry of marine sediments. In Chemical Oceanography Vol. 2, (editors, J. P. Riley and G. Skirrow), Academic Press. Clark, R, C., Jr. 1966, Master's thesis, Woods Hole Oceanography Institution, Woods Hole, Mass. Clark, R. C. and M. Blumer. 1967. Distribution of paraffins in marine organisms and sediment. Limnology and Oceanography 12:79. Gelpi, E., H. Schneider, J. Mann and J. Oro. 1970. Hydrocarbons of geochemical significance in microscopic algae. Phytochemistry 9:603. Han, J. and M. Calvin. 1969. Occurrence of Ci> 2 -C 2 fl isoprenoids in Bell Creek crude oil, Geochim. Cosmochim. Acta 33:733. Han, J., E. D. McCarthy, M. Calvin and M. H. Benn. 1968. Hydrocarbon con- stituents of the blue-green algae Nostoc muscorum, Anacystis nidulans, Phor- midiivm Luriduvi and Cklorogloea fitsekii. J. Chem. Soc. (Sect. C), 2785. Jenne, E, A. 1968. Controls on Mn, Fe, Co, Ni, Cu and Zn concentrations in soils and water: the significant role of hydrous Mn and Fe oxides. In Trace Inor- ganics in Water, American Chemical Society, Washington. Metcalfe, L. D. and A. A. Schmitz. 1961. The rapid preparation of fatty acid esters for gas chromatographic analysis. Anal. Chem. 33:363. Morrison, T. R. and R. N. Boyd. 1969. In Organic Chemistry, 2nd Ed, Allyn and Bacon, Inc., Boston. Stransky, K. and M. .Streibl- 1969. On natural waxes, Xll. Composition of hydro- carbons in morphologically different plants. Collection Czschoslov. Chem. Commun. 34:103. Winters, K., P. L. Parker and C. Van Baalen. 1969. Hydrocarbons of blue-green algae: geochemieal significance. Science 163:467. Gulf Research Reports Volume 4 Issue 2 January 1973 An Examination of Legislation for the Protection of the Wetlands of the Atlantic and Gulf Coast States Anthony J. Haueisen University of Mississippi DOI: 10.18785/grr.0402.09 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Environmental Law Commons, and the Marine Biology Commons Recommended Citation Haueisen, A. J. 1973. An Examination of Legislation for the Protection of the Wetlands of the Atlantic and Gulf Coast States. Gulf Research Reports 4 (2): 233-263. Retrieved from http:// aquila.usm.edu/gcr/vol4/iss2/9 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. AN EXAMINATION OF LEGISLATION FOR THE PROTECTION OF THE WETLANDS OF THE ATLANTIC AND GULF COAST STATES by ANTHONY J. HAUEISEN, J.D. School of Law The University of Mississippi Oxford, Mississippi INTRODUCTION The most useful aquatic areas in the world are in serious danger of destruction. The estuaries, where fresh water, land and sea meet in a dynamic and highly productive zone, are today gravely threatened through unwise and unplanned excessive use of their valuable but finite capacities. Estuaries include the coastal zone which is affected by both the run-off of fresh water from the land and the salt w^ater from the sea. This zone includes tidal rivers, marshes, bays and river mouths. The value of these estuarine regions has been well established by biologists. However, this value is fully appreciated by only a hand- ful of people. The intense uses to w^hich the coastal zone is being placed are so expan.sive, so competitive and potentially so destructive. Much shipping for industrial and military purposes begins and ends in es- tuaries. The waste products of the industries which crowd the coastal zone and of one-third of the population of this country are daily being poured into these waters. Estuaries are directly linked to suitable con- ditions needed for the development of three-fourths of the fish and shellfish taken for food production and recreational fishing. In addi- tion, the use of these so important coastal regions is ever increasing by our growing population for aesthetic and recreational purposes. Each of the above uses is an important human use. However, each of these uses is potentially destructive. 'Even aesthetic uses, which have heretofore been above reproach, can irretrievably destroy this fragile ecosystem when vast areas are developed for housing projects by dredging and filling in the land. In fact, this action may destroy the very rea.son why people wish to move to the shore. All of these uses, and others, have developed without sufficient comprehension of their effects and interactions and totally without planning for an optimal balance for present and future human uses. Present and po- tential human uses involve vast and complicated economic problems, political and geographic difficulties, and grave social and legal com- plications. It is with the latter, the legal problems of wetlands pre- servation and utilization, that this paper is concerned. It is proper that I should here acknowledge the work of Mr. Lionel Eleuterius of the Gulf Coast Research Laboratory who first 233 234 Haueisen visualized the need for a collection and examination of the laws of the various coastal states in regard to the preservation of their wetlands and who personally assembled the raw materials out of which this paper is formed. Furthermore, I should acknowledge Dr. Thomas Lytle, also of the Gulf Coast Research Laboratory, who recommended this project to me and guided my research. This paper will be divided into several distinct sections. It is initially important that we define just what we are examining; there- fore, the first part of this paper will describe what wetlands are and attempt to explain something of the ecology of this region. The next part of the paper will examine the laws of the coastal states to deter- mine what, if anything, they have done to preserve and utilize their wetlands. The last part of the paper will examine what states may do to protect their wetlands, contain a legislative model for a wetlands act, and discuss possible implementation of this proposed legislation. WHAT ARE WETLANDS? Wetlands is a general term used to describe the water-land inter- face, whether ocean and shore or river and bank. However, in this paper we will be more concerned with the ocean-shore interface be- cause of the extreme richness of this particular area and its vulner- ability. Wetlands may occur as a dividing fringe along shorelines in- terposed between permanent dry land and open surface water expanses of rivers, estuaries, etc. They may also exist in another configuration ; as rather extensive tracts continuing for hundreds of miles and acres in area. There are numerous local names by which wetlands are known. Some of these are salt marsh, tidal marsh, marshland, tideland, sub- merged land, swamp, slough, bog, mud flats, wet meadow, or flood plain. People tend sometimes to use these terms interchangeably, although ecologists differentiate between many of these terms. The following clas.sification system is derived from an extensive analysis of wetlands in Maryland.^ The classification system briefly describe.^ each type of wetland and outlines very generally the more important physical and ecological characteristics distinguishing each. In examining wetlands, we are concerned with two major distinct groupings: 1.) those wet- lands which occur in inland, freshwater areas, and 2.) those which occur in coastal areas and are our especial area of interest. I. Inland, freshivater wetlands. In this category of freshwater, non-tidal wetlands there are seven types of wetlands included. They are : 1 Maryland State Planning Department, Draft Report-W etlands In Maryland, January, 1970, p. V3. Wetlands Legislation 235 1. Seasonally flooded basins and flats 2. Inland fresh meadows 3. Inland shallow fresh marsh 4. Inland open fresh water — ponds, lakes, etc. 5. Shrub swamp — found alonj? sluggish streams. 6. Wooded .swamp — often occurs on poorly drained uplands. 7. Bogs — usually waterlogged and having a spongy covering of mosses. II. Coastal Wetlands This category can be further subdivided into two distinct areas. These are fresh and saline marshes, which are distinguished by differ- ences in shoreline elevation and the consequent varied influence of the tide. This tidal activity partially accounts for the high productivity of the coastal wetland as will be described later. There are seven types of coastal wetland, three of which are in freshwater areas and four of which are in saline areas.” They are: A. Fresh Areas 1. Coastal shallow fresh marsh — average mean high tide may cover this area with up to 6 inches of water. 2. Coastal deep fresh marsh — average mean high tide may cover this area with from 6 inches to 3 feet of water.^ 3. Coastal open fresh marsh — these are usually more or less enclosed tidal ponds, and vegetation is usually scarce or lacking because of the turbidity of the water due to the tidal cycles and currents which keep sediment and detri- tus in suspension. B. Saline Areas 1. Coastal salt meadow — this soil is always waterlogged; but because of its elevation, it is rarely covered by tidal waters. 2. Irregularly flooded salt marsh — the soil in this area is covered by a few inches of water at irregular intervals. 3. Regularly flooded salt marshes — ^the soil is covered at mean high tide by one-half foot or more of water. This is the area with which we are specifically interested. This area is important as nesting areas for gulls and rails, as feeding areas for herons, as habitat for mussels, snails, and crabs and is used by fish and crustaceans. 4. Sounds and bays — this type consists of submerged land under the open waters of sounds or bays. Vegetation is 2 Ibid., p. V-7. 3 The difference in depth of saltwater inundation determines the types of vege- tation and animal life found in these areas. 236 Haueisen usually scarce. However, this area is important because it is the habitat of fishes, oysters, mussels, shrimp, clams, crabs, and many other invertebrates upon which these species feed. This necessarily brief description of the terms used in defining wet- lands will serve to illustrate just how minute and complex terminolo- gies can become when dealing with any involved and intricate natural system. BASIC WETLANDS ECOLOGY The term ecology has very recently come into vogue. However, the language and techniques of ecology and ecological research are largely unknown to the public. Ecology may be defined as the study of the interrelationships of organisms to one another and to their en- vironment. It is possible to divide a study of an ecological system into two distinct parts: 1.) the structure of the ecosystem itself, including quantity and distribution of plants and animals and the physical char- acteristics such as temperature, light, pH, salinity, dissolved oxygen, etc.; and 2.) the function of the ecosystem, including the rate and amount of biomass production'* and the cycling of nutrients within the biotic community. Without a knowledge of these ecological princi- ples and other information, it would be impossible for well-intentioned individuals to formulate sound natural resources management policies. Therefore, I will briefly describe some general principles of ecology and some concepts concerning the marsh ecosystem. Since tidal marshes are ecological formations resulting from the invasion of shallow water by land vegetation, it would perhaps be best to examine those areas best suited to this invasion. The most obvious place for land vegetation to invade the shallow sea is along the edges of the vast coastal plain which makes up a part of the continental shelf. This coastal plain extends along most of the Eastern and Gulf coasts of the United States but is largely lacking along the Pacific Coast. In early geological history, the present continental shelf was, for the most part, also the coastal plain. Today, the continental land mass is less expansive and much of the former coastal zone is sub- merged. Some submergence is still occurring, at a rate of about one foot per century; but the coastal area is just about holding its own due to the seaward transport and deposition by rivers of sediments from inland. However, the levee-building activity of the Army Corps of Engineers is diverting this sediment from the coastal zone and ac- tually forcing it out to the edge of the continental shelf where it is lost for human use for all practical purposes. In some coastal areas, 4 Biomuss may be defined as the total quantity at a given time of living orga- nisms of one or more species per unit of space (species biomass), or of all the species in a community (community biomass). 238 Haueisen marshes. Among marine species only a relatively few can adjust to the rapid salinity changes which occur with each tide. Those few species that can endure the tidal marsh conditions, however, are rela- tively free of the kinds of competitors and enemies that harass related species in nearby waters. As in other ecosystems, the same general relationships and com- ponents exist in a tidal marsh. Plants utilize sunlight energy in their growd:h and reproduction and are the primary producer component of the ecosystem. They are a source of food for grazing animals. These ani- mals in turn*furnish food for carnivores. Bacteria and fungi comprise the component which reduces or decomposes the dead organisms to inorganic levels, returning such nutrients as phosphates and nitrates to the marsh system where they are re-used by the plants in the man- ufacture of more plant tissue. In conjunction with the re-use of the nutrients and the one-way flow and ultimate release of the energy first stored by the plant life, it is necessary to add to this simple structure other natural trends and cycles which affect the marsh ecosystem. The tides redistribute nutrients and sediments throughout the tidal marsh. They also affect the overall primary productivity by decreasing or in- creasing exposure of the microscopic algae and marsh plants or the quantity of phytoplankton that is favorably exposed to sunlight as the volume of water over the marsh changes. It has been claimed that “tidal marshes and estuaries . . . are among the most fertile areas in the world in terms of energy, calories, proteins, carbohydrates and vitamins.’’ ^ Ecological studies have shown that “gross income from the best market farms . . . runs as high as $2,000 per acre per year. On a comparative basis, the major marshlands are producing $4,000 worth of nutrients per acre per year.” ^ Fertility of estuaries also results from year-round primary production. Wetlands ecosystems tend to maintain a constant rate of production during seasonal environments. For example, marsh grass produces at least tw'o crops per year as compared with wheat which grows only a few months with zero growth for many months. The an- nual production of a marsh, or an estuary as a whole, may be double or triple that of ordinary agricultural land simply because it produces two or three times as long each year. Now that we have very briefly examined some basics of marsh ecology and its importance, it is time to turn to the next major area of this paper, an examination of the various coastal states with re- gard to those laws which have been enacted to protect their wetlands. ® Robert L. Dow, Maine*s Coastal Marshlands: Their Values, Present and Fu- ture (Augusta: Maine Department of Sea and Shore Fisheries, 1962), p. 3. ® Robert L. Dow, Economic Yields of So^ne Maine Coastal Wetlands (Augusta: Maine Department of Sea and Shore Fisheries, 1966) , p. 1. Wetlands Legislation 239 LEGAL APPROACHES TO THE PROTECTION OF WETLANDS In this chapter, I shall examine statutes from the states on the Atlantic Coast and on the Gulf Coast of the United States and shall emphasize those which act to protect wetlands. First, I will examine the Federal laws relative to this ai'ea; and then, I will examine the laws of the states in geographical order from Maine to Texas. A. FEDERAL LEGISLATION The English Common Law has come into all of the United States, with the exception of Louisiana, and has evolved through case inter- pretation by the Supreme Courts of these various cases. The essential common law principle is that title to soil under navigable waters is in the sovereign except as far as private rights have been acquired in an expressed grant from the original sovereign. During the time of the original thirteen colonies, the tidelands and the submerged lands were under the complete control of the Crown and they were held by the Crown in trust for the people as a whole. The riparian owner has own- ership ordinarily only to the mean high water mark. After the Revolutionary War, the American states succeeded the Crown as Sovereign. The States continued to hold the tidelands and submerged lands in a sovereign capacity in trust for the people sub- ject to the public purposes of navigation, commerce, fishing, boating, recreation and enjoyment free from obstruction and interference. State ow'nership, and in some cases private ownership, of tide- lands and submerged lands are subject to the paramount right of con- trol by the Federal Government under the U.S. Constitution for com- merce and navigation. As stated, all real property w^as originally in the Sovereign. It w'as then granted in one w^ay or another to private individuals. There are three types of grants from the Sovereign: 1.) the first, was by the Lords Proprietors j 2.) by the Crown itself up until the time of the Revolutionary War; and 3.) from the Revolu- tionary War to the present in the form of State Grants, The States of the Atlantic and the Gulf Coast have absolute title to submerged lands. The state has prima facie title (i.e. it goes into the court with a presumption of title) to the tidelands, i.e., the area betw'een the mean high water mark and the mean low water mark. A claimant to private owmership of tideland.s must come into court with a chain of title, tracing hi.s title back to the original grant from the Sovereign. This is sometimes difficult to do due to the destruction of the intervening records. A claimant mu.st produce an original grant which Is then presented to the court as a question of law as a con- struction of that grant. The Federal Government has jurisdiction over these tidelands, submerged lands and navigable waters under the Navigation and 240 Haueisen Commerce Clauses of the Federal Constitution. There is joint or con- current jurisdiction between the Federal Government and the Gov- ernment of the particular State over the navigable waters of the United States which are also navigable waters of that particular State. The navigable waters of a State include all of the navigable waters of that State, whereas, the Federal Government only has juris- diction over a certain percentage of tho.se waters which have been classified, either by the Federal Congress or by some rule and regula- tion, as being Federal navigable waters. These w'aters are usually waters that are in continuous connection between different states or between the state and the open ocean. The other State waters are under the complete jurisdiction of that particular State. The law, as far as the Federal Government is concerned, was relatively stable up until 1947 w'hen after a period of time the U.S. Department of Interior through the U.S. Attorney Gen- eral’s Office began litigation first against the State of California and later involving decisions in 1950 against the States of Texas and Lou- isiana attempting to take over actual Federal ownership of all tide- lands and submerged lands. The decisions of the U.S. Supreme Court in effect completely reversed all prior law and stated that the Federal Government was the absolute owner of these areas. The Congress passed the Submerged Lands Act of 1953 w'hereby the Federal Gov- ernment, in effect, abandoned the title to the area lying below the mean high water line and left the actual ownership in the hands of the State as determined by the State law. At the present time, the only interest the Federal Government has in these areas is this easement or servitude on the part of the Federal Government for the control of commerce and navigation which goes up to the mean high water mark. Congress also has an interest, of course, in any of these areas which affect National Defense, international affairs, flood control and elec- tric power production. Some of these Federal Agencies with various powers of regula- tion are the Federal Power Commission, U.S. Navy, U.S. Army Corps of Engineers (which has essential control over interstate navigation and the intracoa.stal w^aterway), and the U.S. Coast Guard wdth the enforcement of revenue laws and boating laws. In 1953, the Federal Government also passed the Outer Conti- nental Shelf Lands Act which stated that the Federal Government had absolute jurisdiction betw^een the outer limits of the state boundary, which is usually taken to be three miles, but in the case of Texas and Florida extends out to three marine leagues (which is ten and a half miles) outside of the state's mainland area. The Federal Government is supreme between the state boundary and the extent of the Conti- nental Shelf, extending out in some areas 120 to 150 miles. The State jurisdiction over the tidelands, submerged lands, and navigable waters are co-extensive with the Federal Government in- Wetlands Legislation 241 volving both the state navigable waters and the Federal navigable waters. The State has exclusive jurisdiction over the state navigable waters. Within each State there are a number of agencies and depart- ments that have overlapping jurisdiction, regulation and control of the tidelands and submerged lands. The focus of the Federal Government with regard to the protec- tion of the wetlands has been directed in two major areas: pollution control and preservation of migratory bird nesting and feeding areas. Both of these approaches may be helpful in protecting marshlands. First, let us examine the role of the Federal Government in pollution control. The Federal Government entered into the area of water pollution control in 1899 with the enactment of the Rivers and Harbors Act.' This statute made it a criminal offense to deposit any refuse matter of any kind or description into any navigable water of the United States. In 1966. the U.S. Supreme Court held that “refuse" was not limited to materials of no value but included any substance, whether or not usable by industrial standards, which has a deleterious effect on navi- gable waters,® The Oil Pollution Act of 1924 was an enactment to deal with the problem of oil discharges from vessels into coastal waters damaging aquatic life, harbors and dock and recreational facilities.® The pro- hibitions in this Act proved to be quite difficult to police as a practical matter. The Water Pollution Control Act of 1948 was also the really first modern identifiable Federal program for water pollution control.^® It stated that the pollution of interstate waters which endangered the health or welfare of persons of another state is to be considered a public nuisance subject to abatement and that the States were recog- nized to have primary responsibility and the right to control water pollution while the Federal Government had jurisdiction over the na- tion’s interstate waterw'ays and their tributaries.*^ The Surgeon Gen- eral was directed to coordinate and encourage cooperation among all levels of government involved in pollution control and to engage in joint activities with State and interstate agencies.*® The Act also en- couraged the adoption of uniform State law’s and the creation of inter- state pollution control compacts.*^ The Surgeon General was autho- rized to prepare and adopt comprehensive programs to eliminate 7 Rivers and Harbors Act of 1899, 38 U.S.C, §§401-413 (1899). s United StaUR v. Standard Oil Co., 384 U.S. 223 (1966) . » Oil Pollution Act, 33 U.S.C. §§431-437 (1924). 10 Water Pollution Control Act, 62 Stat. 1155, 33 U.S.C. §§466 (1948). ** Ibid. i2/6wi., at §466(b) (1952). i3/6id., at §466(c) (1964). 242 Haueisen pollution in waterways and support and aid technical research to de- vise and perfect methods of treatment of industrial wastes.^^ In 1956, amendments to the original 1948 Act were passed which provided for a much more intensive and well-organized Federal pollu- tion abatement program than did the earlier law.'*^ They emphasized the basic policy that water pollution problems were best solved at the local level. Grants to States and to interstate agenciexS were authorized for administration of water pollution control programs including com- prehensive river basin programs involving control, research, and en- forcement. It provided for technical assistance, the encouragement of interstate compacts and uniform State laws, the appointment of a Fed- eral Water Quality Advisory Board, and a cooperative program for the control of pollution from Federal installations. In 1961, the Federal Water Pollution Control Act Amendments were enacted. These amendments extended Federal pollution abate- ment authority to all interstate or navigable waters, increased con- struction grants and authorized research facilities in various parts of the country and the conducting of water quality studies in the Great Lakes. It should also be noted that the amendments transferred the administrative responsibilities for the program from the Surgeon General to the Secretary of Health, Education and Welfare.^"^ In 1965, the Federal Water Quality Act was signed into law.^^ Briefly, this Act provided for the adoption and enforcement of water quality standards for interstate waters and set up the Federal Water Pollution Control Administration in the Department of Health, Edu- cation and Welfare. It increased construction grants and authorized research and development grants for preventing discharge of untreat- ed wastes from storm sewers or combination storm-sanitary sewers. In 1966, the Federal Clean Water Restoration Act became law.^® This Act provided incentives to the States to adopt water quality standards for pollution control and provided for Federal reimburse- ment for qualified construction projects commenced at a time when Federal grant funds were not sufficient to pay the full Federal share. The purpose of this provision was to encourage a State prepared to move ahead with sewage treatment plant projects to start building without waiting for a Federal grant. Under this program, if the proj- ect is Federally approved, the local governments may advance the Federal share of the project themselves and be reimbursed with funds as they become available. It authorized the use of Federal enforce- Ibid. Water Pollution Control Act Amendments of 1956, 70 Stat. 499, 33 U.S.C. §466-466(k) (1964). 16 Ibid. 17 Ibid. 18 Water Quality Act of 1965, 79 Stat. 903 (1965). i*> Clean Water Eestoration Act of 1966, 80 Stat. 1246 (1966). Wetlands Legislation 243 merit equipment with relation to international boundary waters and transferred responsibility for administration of the Oil Pollution Act to the Secretary of Interior and changed that Act to include inland waters. In 1970, the Water Quality Improvement Act was enacted.^o This Act forbade oil discharges into navigable waters, adjoining shorelines and the waters of the contiguous zone and requires a National Con- tingency Plan for removal of any spills. This Act provides that the owner of a polluting facility can be fined up to eight million dollars for the clean-up costs or more if there is willful negligence or willful misconduct. The second focus of Federal programs is concerned with migra- tory wildfowl. The Migratory Bird Treaty Act of 1918 assigned to the Federal Government primary jurisdiction over the protection of mig- ratory birds, including wildfowl. This was followed in 1929 by the Migratory Bird Conservation Act which provided the authority and funds for the establishment of Migratory Bird Refuges. The first wa- terfowl hunting stamp (at a cost of $1) was required by the Migra- tory Bird Hunting Stamp Act of 1934. This Act helped finance the refuge program. The Federal Aid in Wildlife Restoration Act (Pittman-Robert- son) of 1937 made it possible for many states to initiate programs of wetlands acquisition and development. In 1949, an amendment to the Hunting Stamp Act raised the price of the “duck stamp'’ to $2; an- other in 1958 further increased it to ^3. The latter amendment also specified that duck stamp funds be used exclusively for the purchase of waterfowl production areas and suitable areas for migratory bird refuges; also that as much as 40% of a refuge may be opened to hunt- ing of migratory birds. In spite of these efforts, loss of w^etlands continues at an alarm- ing rate. In recognition of the growing problem, Congress in 1961 passed a bill which has become known as the Accelerated Wetlands Acquisition Act (Public Law 87-383, approved Oct. 4, 1961). This Act makes possible the stepped-up purchase of essential wetlands now, while they still exist. In effect, Congress could loan the conservation movement $105 million which, plus the revenues from duck stamp sales, could be used for this purpose. This money is to be used to pro- mote the conservation of migratory waterfowl and to offset or prevent the serious loss of important wetlands and waterfowl habitat essential to the preservation of such waterfowl. At the end of 7 years, 75 % of each year’s duck stamp revenues must be used for repayment of the loan. The task of this legislation is obvious — to save at least an essen- tial minimum of the nation’s wetlands. “t* Water Quality Improvement Act of 1970, 84 Stat. 91 (1970). 244 Haueisen B, STATE LEGISLATION 1. Maine Maine. Revised Statutes Annotated, Title 12, ch. 421, § §4701—4709 ( Additional Supp. 1967 ) No person, agency or municipality may remove, fill, dredge or drain sanitary sewage into, or otherwise alter any swamp, marsh, bog, beach, flat or other wetland bordering coastal waters, or fill, dredge or drain sanitary sewage into such waters within such area, without fil- ing written notice to do so, with plans, to the m\micipal officers and the Wetlands Control Board. A public hearing is held after receipt of such notice. After the hearing, a permit is issued if the Wetlands Con- trol Board approves. Approval may be conditioned upon the applicant amending his plans. Approval may be withheld by the municipality or the Board if either body finds the plans damaging to fish life. An ap- peal is provided for. Anyone violating any provision of this statute is subject to a maximum fine of $100. A continuing violation of this statute may be enjoined. Maine Revised Statutes Annotated, Title 30, ch. 229, ^4001 (Addi- tional Supp. 1967 ). Wetlands may be ‘*taken” by a municipality with the consent of the owner and payment of compensation. Maine Revised Statutes Annotated, Title 30, ch. 229, §300J (Addi- tional supp. 1967 ). Any municipality may receive wetlands as devises or gifts. 2. New Hampshire Wetlands regulation had a modest beginning in New Hampshire in 1955 when the legislature made it illegal to create land by filling in great ponds — lakes or ponds over 10 acres in size — ^without permission of the Governor and Council. Before that time, apparently a person could convert the state’s water into his own land by the simple ex- pedient of filling it in. This statute dealt with public inland waters, not wetlands as such and was the extent of any regulatory effort until 1965. Then, in an act regulating sewage disposal systems on islands, it was provided that no one could fill in a marsh bordering on or adjacent to a great pond of the State for building purposes without approval of the sewage dis- 21 New Hampshire Revised Statutes Annotated 482:41-a to 41-d. Wetlands Legislation 245 posal system in accordance with local zoning ordinances of the mu- nicipality, or in absence of same, the Water Supply and Pollution Con- trol Commission. The emphasis of this statute, though, was on the prevention of pollution and not the preservation of wetlands* In 1967, New Hampshire got its first full-fiedged wetland regula- tions — three dredge and fill laws. The first prohibited any person, firm or corporation from excavating or dredging any bank, flat, marsh, swamp or lake bed that lies below the natural mean high water mark of any fresh public waters of the state without petitioning the Water Resources Board. The second replaced the 1955 law regulating plac- ing fill in fresh public waters with a more up-to-date version. The third prohibited persons from excavating, removing filling or dredg- ing any bank, flat, marsh or swamp in and adjacent to tidal waters without approval of the New Hampshire Port Authority.®''^ In 1971, Bill No. 228 was introduced into the New Hampshire House of Representatives and dealt with excavating, filling, mining and construction in the inland waters of the state and established an inland wetlands authority to which anyone seeking to alter any of the existing interior wetlands of the state would have to make petition for a permit to do so.^® 3. Massachusetts Massachusetts General Laws ch. 131, §40 (1967) entitled: Protec- tion of Flood Plains. No person shall remove, fill or dredge any bank, flat, marsh, mea- dow, or swamp bordering on any inland waters without filing notice of hia intention to do so, with plans of the proposed activity, wuth the local authority and with State departments of public works and nat- ural resources, A public hearing is provided for. The local authority may recommend protective measures in the public interest, which are submitted to the Commissioner of Natural Resources. If the area where the w’ork is to be done is essential to proper flood control, the Commissioner may impose conditions necessary to protect the public interest, which must be complied with. A continuing violation of this section may be enjoined. Massachusetts General Laws ch. 130^ ^105 (Additional supp. 1966) Because of the urgent necessity of protecting coastal w’etlands, the Commissioner of Natural Resources was given the power to adopt, 22 New Hampshire Revised Statutes Annotated 149 :4^^. 23 New Hampshire Revised Statutes Annotated 488-A (1967). 24 New Hampshire Revised Statutes Annotated 482:41-e to 41-i. 25 New Hampshire Revised Statutes Annotated 483-A, §1-5. 2« New Hampshire Revised Statutes Annotated 483-B, §1-26. 246 Haueisen amend, modify or repeal orders regulating, restricting or prohibiting dredging, filling, removing or otherwise altering, or polluting coastal wetlands.Violations of the Commissioner’s order is punishable by a fine of between $10 and $60 and/or maximum imprisonment of one month. However, if the Commissioner’s order so restricts the use of the property as to deprive the owner of its practical use, a court may de- cree that the order does not apply to that owner’s land. In such a case, the Department of Natural Resources may take the land for the State by eminent domain. Massachusetts General Laws ch. 40 ^8(c) (Additional supp. 1968) This section empowers a city or town to acquire by gift, pur- chase, grant, bequest, devise, lease, or otherwise the fee or any lesser interest in wetlands and open spaces. It also empowers a city or town to take land by eminent domain for conservation purposes. Massachusetts General Laws ch. 132 A, §11 (Additional supp. 1968) The State may reimburse a city or town up to 60% of the cost of acquiring land for conservation purposes pursuant to §8(c) of ch. 40. Massachusetts General Laws ch. 130, §2 7 A (Additional supp, 1966) No person shall remove, fill or dredge any bank, flat, marsh, meadow or swamp bordering on coastal waters without filing written notice of his intention to do so with the local licensing authority, the State Department of Public Works, and the Director of Marine Fish- eries. Restrictions may be placed on such work and such work must be done subject to these restrictions. Violations of this section are punishable by a maximum fine of $100, imprisonment for not more than 6 months, or both. A continuing violation of this section may be enjoined. A case concerning this section held that it does not authorize an absolute prohibition against the filling in of a privately owned marsh- land if the result would be that the owner would be so deprived of the practical uses of his land so as to amount to a taking of his land with- out compensation. Whether there had been such a deprivation of the practical uses of the marshland as to be equivalent to a taking without compensation depended upon the uses to which the marshland could be put without violating the statutory prohibition against the marsh. Since the evi- dence at trial on this issue was lacking, the injunctive decree of the 27 Commisaioner of Natural Resources v. S. Volpe & Co., 206 N.E. 2d 666 (Mass. 1966). Wetlands Legislation 247 trial court was reversed and the case was remanded for additional findings on the above issue. 4. Rhode Island Rhode Island General Laws Annotated ^^2-1-13 to 2-1-17 (Addi- tional supp. 1967 ) These acts established a public policy of preserving the coastal wetlands of the State. The Department of Natural Resources may, after public hearing, designate coastal wetlands or parts thereof, the ecology of which shall not be disturbed. This designation will be re- corded in the registry of deeds in each city or town where the land is located. The right of appeal is allowed for 2 years after recordation. Provision is made for award of damages. Rhode Island General Laws Annotated ^^11 -46. 1-1 (Additional supp, 1967) Anyone who dumps or deposits mud, dirt or rubbish upon, or who excavates and disturbs the ecology of intertidal saltmarshes, without first obtaining a permit therefor issued by the Department of Natural Resources shall be fined for each offense $100: $50 to the State and $50 to the complainant. The Director of Natural Resources shall re- fuse to issue such a permit if in his judgment the dumping or deposit- ing of mud, dirt or rubbish or excavation would disturb the ecology of intertidal saltmarshes. 5. Connecticut Conn. Gen. Stat. Ann. Title 25, ch. 473, §26-17a. (1967) The State Board of Fisheries and Game is empowered to do the following: (1) Acquire wetlands, or any easements, interests or rights therein, by purchase, exchange, condemnation, gift, devise, lease or otherwise. (2) Enter into agreements with owners of wetlands to conserve wetlands. (3) Enter into leases with an option to purchase wetlands, pro- vided : a. approval of the Commissioners of Agriculture and Natu- ral Resources is obtained, and b. the lease does not exceed 10 years. 248 Haijeisen (4) Take wetlands by eminent domain. (5) Secure title to wetlands by paying to a municipality the amount of the municipality's tax liens on such wetlands, where the municipality's property tax on such wetlands is unpaid for 6 years. Conn. Gen, Stat. Ann. Title 25, ch. 473, ^25-10 to §25-77 (1967) These statutes provide for the dredging of sand and gravel from lands under tidal and coastal waters. This is regulated by the Water Resources Commission, supplemented by a member designated by the Shellfish Commission. Public hearings must be held. Shore erosion, navigation, and living resources must be considered. Local zoning for marshland protection has been unsuccessfully attempted in Connecticut. 6. New York Long Island Wetlands AcC^ This Act permits the State to enter into cooperative agreement with towns, villages, or counties for the purpose of preserving and maintaining wetlands on Long Lsland which have been dedicated for conservation purposes. The State is empowered to provide one-half of the cost of maintaining such areas. New York Conservation Law, §429a-g (1967 ) These sections require the issuance of a permit by the Water Re- sources Commission before anyone may alter the waters of the State. The Commission may issue the permit subject to conditions upon which the work must be done. The following activities require a per- mit: altering the channel of a stream, removing materials from a stream, excavating or filling in navigable waters, erecting an im- poundment structure, dock or wharf in or across a natural stream or watercourse. A violation of this statute constitutes a misdemeanor, punishable by a maximum fine of $500, maximum imprisonment of 1 year, or both. New York Public Lands Law, art. 2, §3.5 (Additional supp. 1967) The Commissioner of General Services may license and regulate the business of taking sand, gravel or other materials in or upon lands under water and may prescribe the terms and conditions under which Dooley v. Town Zoning Commission, 157 Conn. 304, 197 A.2d 770 (1964). 29 New York Conservation Law, §394 (1967). Wetlands Legislation 249 the same may be taken. After adoption of regulations by the Com- missioner, it shall be unlawful to take or remove from lands of the State under water any sand, gravel, or other material, without a license. !\etv York Public Lands Law^ art. 6, §§75-7^ (Additional supp. 1967) Empowers the Commissioner of General Services to grant lands under water to a county, city, town or village for conservation and other purposes* 7. New Jersey The W^etlands Act of 1970^^ This Act recognizes the ecological importance of the coastal wet- lands, and states that it is necessary to prevent the deterioration and destruction of these lands in order to preserve the ecological balance of the coastal area. The Act directs the Commissioner of the Department of Environ- mental Protection to make an inventory and maps of the wetlands. The Act authorizes the Commissioner to make regulations restricting or prohibiting dredging, filling, removing, or otherwise altering or polluting the coastal wetlands. The Act requires the Commissioner to hold a public hearing before adopting any regulations concerning the wetlands. The Act prohibits any regulated activity from being carried on without a permit from the Department of Environmental Protection. The Commissioner must consider the ecological effect of the work to be performed before issuing a permit. The Act provides that any per- son having a recorded interest in land affected by the Commissioner's regulations may file a complaint in the Superior Court to determine if the regulations deprive him of the practical use of his land. The Act provides a fine of $1,000 to be levied against violators of the regula- tions promulgated by the Commissioner, and also makes violators liable for the cost of restoration of the affected coastal wetland to its condition prior to the violation in so far as is possible. iS.J. ST AT. AI\I\. 12:5-3 to 12:5-3 (1914) The Board of Commerce and Navigation must pass on all plans for development of waterfront which involves the construction or al- teration of a dock, wharf, pier, bulkhead, bridge, pipeline, or any 1 could find no other citation. In the official legislative copy, all that is said is that it may be cited as the Wetlands Act of 1970. 250 Haueisen other similar or dissimilar waterfront development. Public hearings may be held. No provision is made for eminent domain. No reference is made to protection of natural resources. NJ. ST AT. ANN. 13:8A-I to 13:8A-18 (1961 ) This is New Jersey’s Green Acres Land Acquisition Act of 1961. The Act provides for purchase of lands for public recreation and con- servation of natural resources. A sum of $60 million was made avail- able by a Green Acres Bond referendum. The acquisition program is under the direction of the Commissioner of Conservation and Econom- ic Development. Of the total amount available, $20 million was for the purpose of supporting local acquisition. In addition to fee simple ac- quisitions, acquisition of conservation easements is permitted. 8. Delaware Delaware Coastal Zone AcC^ This Act states that the coastal areas of Delaware are the most critical areas for the future of the State in terms of the quality of life in the State. It, therefore, declared that it is the public policy of the State of Delaware to control the location^ extent and type of industrial development in Delaw^are's coastal area. In so doing, it is thought that the State can better protect the natural environment of its bay and coastal areas and safeguard their use primarily for recreation and tourism. Specifically, this chapter seeks to prohibit entirely the con- struction of new' heavy industry in its coastal areas, which industry is determined to be incompatible with the protection of that natural en- vironment in those areas. While it is the declared public policy of the State to encourage the introduction of new industry into Delaware, the protection of the environment natural beauty and recreation po- tential of the State is also of great concern. In order to strike the cor- rect balance between these two policies, careful planning based on a thorough understanding of Delaware's potential and her needs is re- quired. Therefore, control of industrial development other than that of heavy industry in the Coastal Zone of Delaware through a permit system at the State level is called for. It is further determined that offshore bulk product transfer facilities represent a significant danger of pollution of the Coastal Zone and generate pressure for the con- struction of industrial plants in the Coastal Zone, which construction is declared to be against public policy. This Act, as seen from the above, is quite a radical departure from the historic and traditional policy of every state to encourage any industry it can to build and operate from that state. Since the Act •■^1 DEL. CODE ch. 70, §7001-7014. Wetlands Legislation 251 is so new, it cannot as yet be determined what economic effect this Act will have on the State of Delaware; but it appears to be a serious and determined effort to protect the coastal zone. 9. Maryland ANN, CODE of MD. §§715-731 (1967 Replacement Vol.) This Act declares that in many areas of the State much of the wetlands have been lost or despoiled by unregulated dredging, dump- ing, filling, etc. and that the remaining wetlands of the State are in jeopardy of being lost. It declares that it is the public policy of the State to preserve the wetlands and to prevent their despoliation and destruction, §719 defines **state wetlands” and ‘'private wetlands”. §721 declares that it is unlawful for anyone to dredge or fill in State wetlands without a license to do so by the Board of Public Works. Any- one violating the provisions of this section is deemed guilty of a mis- demeanor and may be fined not less than $500 and not moro than $1,000. Each violation shall be a separate and distinct offense, and in the case of a continuing violation, each day’s continuance thereof will be deemed to be a separate and distinct offense. Any land created in violation of this section will be the property of the State. §723 states that the Secretary of Natural Resources may from time to time promulgate rules and regulations governing dredging, filling, removing or otherwise altering or polluting private wetlands. Provisions are made for a permit to carry out any of the above activi- ties on private wetlands. Appeal may be taken to the Board of Review of the Department of Natural Re.sources and from there to a circuit court in the county in which the land is located. If the court rules that the restrictions under the Secretary’s rules and regulations constitute a taking of land without compensation, the Secretary of Natural Re- sources may proceed to condemn the land or interests therein and take it by eminent domain. 10. Virginia Virginia has no specific statutes in the Code relative to coastal wetlands protection. There are some statutes and a constitutional ar- ticle that do provide some measure of protection to some of the wet- lands. Two statutes and the constitutional provision listed below pertaining to ownership of lands are particularly pertinent. The Con- stitution of Virginia states that: The natural oyster beds, rocks and shoals, in the waters of this State shall not be leased, rented or sold but shall be held in trust for the benefit of the people of this State, subject to 252 riAUEISEN such regulations and restrictions as the General Assembly may prescribe, but the General Assembly may, from time to time, define and determine such natural beds, rocks or shoals by surveys or otherwise.^^ Some wetlands and shallows are included within the area thus protected. The Code states that: All of the beds of the bays, rivers, creeks and the shores of the sea within the jurisdiction of this Commonwealth, and not conveyed by special grant or compact according to law, shall continue and remain the property of the Commonwealth of Virginia, and may be used as a common by all the people of the State for the purpose of fishing and fowling, and of tak- ing and catching oysters and other shellfish . . The Code further states that; All unappropriated marsh or meadow lands lying on the Eastern Shore of Virginia, which have remained ungranted, and which have been used as a common by the people of this State, shall continue as such common, and remain ungranted, and no land warrant shall be located upon the same . . However, an accurate designation of those lands used as a com- mon has been lost over the years and only now are state historians trying to relocate these long obscured lands from old deeds and grants. This work could take several years. Title 62.1, §62.1-3, at first glance, seems to offer protection to coastal wetlands. Closer reading, however, shows that to a large extent the Marine Kesources Commission does not have discretionary power relative to filling or over construction of private docks and landings for non-commercial use. The Commission does not have any authority over dredging by a riparian owner. Altogether, Virginia's statutory provisions are highly inadequate for protecting the coastal wetlands. 11. North Carolina The North Carolina Department of Conservation and Develop- ment shall pass on all excavations and filling proposals. If any state agency raises an objection to action of the Department, a meeting of a Review Board composed of the Directors of other state agencies may be held. The Review Board may affirm, modify, or overrule the action of the Department of Conservation. Provisions for appeal to the 32 VA. CONST, art. 175. 33 CODE of VA. tit. 62, §62-1. 34 CODE OF VA. tit. 41, §41-81. Wetlands Legislation 253 courts are provided. No provision is made for taking of any land by eminent domain.^® A limited acquisition program is in effect funded by part of the State Motor Vehicle Tax Fund.^** ]2. South Carolina The State of South Carolina in seeking to protect its coastal wet- lands has taken the approach of defining its jurisdiction and own- ership of the tidelands, submerged lands and waters located in the coastal region of the State. The State of South Carolina has declared that it has absolute title to Submerged Lands (the area below the mean low water mark) in the navigable waters of the State. The State has prima facie title to Tidelands (marshes) (the area between the mean high water mark and the mean low water mark), in and ad- jacent to the navigable waters of the State. The State holds the Tide- lands, Submerged Lands and Navigable Waters in trust for and sub- ject to public purposes and rights of navigation, commerce, fishing, bathing, recreation or enjojmient, and other public and useful pur- poses, or such other rights as are incident to public waters at common law, free from obstructions and interference by private persons. On the basis of this general interpretation of the State’s definition of tidelands and tideland ownership, there has arisen a legal conflict concerning which lands are actually owned by the State and held in public trust for the people, and which lands are actually owned and/or operated by private individuals. With the exception of isolated cases brought to settle legal disputes to title of specific acreages of marsh- lands, there has, up to the present, been little or no effort on the part of the State to inventory on behalf of the people of South Carolina the extent of the State’s claim to ownership of lands held in trust. The continued lack of an applied formula concerning legal interpretation of ownership of these tidelands has led to an ever increasing number of conflicts between the State and private individuals. In a pilot proj- ect in one of South Carolina’s counties, an evaluation of ownership claims has pointed out that approximately 90 % of tidelands, marshes, and coastal w'aterw'ays are now claimed by private individuals. These areas so claimed have in many cases undergone extensive improve- ments relating to water control and management to provide recre- ational use and development, channel construction, and dredge and fill operations for private real estate development. The South Carolina Water Resources Commission on behalf of the State has collected and evaluated information relating to the tidelands ownership question. This and additional information is available to the State for considera- tion in resolving the tideland’s ownership question. 35 1969 Adv, Legislative Service No, 7, §113-229 (effective January 1, 1970). N.C. gen. STAT. ch 105, §446.2 (1967 amended 1969). 254 Haueisen §1-357 of the Code of Laws of South Carolina, 1962 embodies the principles expressed above, It states, in part, that the State Budget and Control Board is charged with the complete control, regulation and leasing of all State lands and all public trust properties. For the pur- pose of leasing oyster and clam rights, it shall use as its agent and advisor the Wildlife Resources Commission and funds derived from the Commission shall be used by the Marine Resources Division of the Wildlife Resources Department for research and management of its marine resources. No person or agency, public or private, will con- struct or in any manner place upon or within tidelands or submerged lands any pier, w'harf, or other structure of any nature or excavate, dig or in any manner create a dock, ditch, canal, or w’atercourse of any nature within or upon such tidelands and submerged lands or place any material upon or change in any manner the natural condi- tions of such lands without first obtaining a permit from the Board. Public notice shall be made by the Board of all permits issued. 13. Georgia An attempt to establish a Coastal Wetland Protective Board failed to pass the 1969 Georgia General Assembly. Otherwise only the usual fish and game laws and water pollution laws are germane to wetland protection. 14. Florida Chapter 67-393 of the General Acts of 1967 This Act amended the Florida Bulkhead Act of 1957 (Florida Statutes, §253.12 et seq.). §253.12. Vests title in tidal lands in the trustees of the Internal Improvement Trust Fund. The trustees may sell such lands, provided that they determine that the sale would not adversely affect public interests, including the preservation of fish and other natural re- sources. Notice of the sale must be published in the county newspaper. If objections are filed to the sale, a public hearing must be held. If it appears as a result of the hearing that the public interest w’ould be ad- versely affected by the sale, then the trustees must withdraw the land from sale. A biological and an ecological survey of the land to be sold must be made in determining whether the sale of the land would ad- versely affect the public interest. §253.122. The Board of County Commissioners of each county or governing body of any municipality, after obtaining a biological and an ecological survey, may locate and fix bulkhead lines. Any extension of land outward into the waters of the county is deemed an interfer- ence with navigation and the conservation of natural resources. Wetlands Legislation 255 §253.123 The removal of sand, rock or earth from the navigable waters of the State and the submerged bottoms thereof lying channel- ward of bulkhead lines is not permitted. Certain exceptions are pro- vided for. §253.124. Anyone desiring to add to existing land bordering on the navigable waters of the State must apply for a permit to do so. A permit will be issued only if a biological and an ecological survey re- veal that the public interest will not be adversely affected. The permit may be revoked for non-compliance with its terms. Anyone who vio- lates this section is guilty of a misdemeanor and upon conviction shall be fined a maximum of $500, imprisoned for not more than 6 months, or both. The trustees have the authority to require the person to re- move the fill. Zahel V. Pmellas County Water and Nav. Control AuthoritVy 171 So.2d 376 (1965), on remand 179 So.2d 370, dealt with §§253.122 and 253.124. In this case the local authority lost in its attempt to prevent owners of bottom land from filling approximately 11.5 acres of land to be used as a trailer park. The court held that denial of permission to fill the land amounted to a taking of property without just com- pensation, because it was not established that granting the permit would materially and adversely affect the public interest. Florida Statutes Annotated ch. .175 ( Additional supp. 1966) This chapter, entitled “Outdoor Recreation and Conservation,” empowers the trustees of the Internal Improvement Fund to acquire wetlands and floodlands by purchase, lease-purchase agreements, or otherwise, with funds from the Land Acquisition Trust Fund. 15. Alabama The State of Alabama has taken the position, as has South Caro- lina and other States, that the state does own absolutely the submerged lands or the area lying below the mean low water mark, which would include the navigable waters. They have also taken the position that the State owns the tidelands, the area between the mean high water mark and the mean low water mark. They are in the process of litiga- tion, as in all other states. The claimants are now going into court and are attempting to prove by grants, in their chain of title, that their private ownership extends to the mean low water mark, because this tidelands area is now valuable. The State has entered into litigation involving cases w'here the tidelands have been filled up by natural accretion. Other cases involve the action of the water in the reliction, or the washing away, of for- merly high and dry lands, or formerly tidelands, that are now com- pletely submerged. Ordinarily where there is accretion or reliction, 1 256 Haueisen the boundary line of the fast land owner changes with this change in the mean high water line. ALA, CODE tit. 8, §§ 232-252 (1932) The Director of Conservation is vested with authority to develop State-owned swamplands. These laws are designed to encourage ex- ploitation. ALA. CODE tit. 38, 19-122. (1932) These statutes set forth the right of riparian owners. These au- thorize and encourage riparian owners to develop lands abutting on tidelands owned by the State by filling and improving these tideldnds. The Department of Conservation is authorized to acquire lands in connection with fish and game programs. 16. Mississippi MISS. CODE A!\!\. ^^549.7-01 and 7605-09 (1942) These sections give Port Commissioners or County Port Authori- ties, respectively, full jurisdiction and control over lands below mean high tide, including filling and dredging operations. The title to oil and gas remains in the State. These statutes are designed to encourage development of the submerged lands. At the regular session of the 1972 Mississippi Legislature, a bill was proposed in the House entitled the ^'Mississippi Coastal Wetlands Protection Act.” This Act recognizes that the coastal wetlands are of great ecological importance and states that it is necessary to prevent the deterioration and destruction of these lands in order to preserve the ecological balance in the coastal area. It declared that the remain- ing coastal wetlands of the State are in jeopardy of being lost or de- spoiled by unwise and unplanned activities; that such loss or despolia- tion will adversely affect, if not entirely eliminate, the value of such wetlands as sources of nutrients to finfish, Crustacea and shellfish of significant economic value; that such loss or despoliation will destroy the ecological system of such wetlands as habitats for plants and ani- mals of significant economic value and will eliminate or substantially reduce marine commerce, recreation and aesthetic enjoyment; and that such loss or despoliation will, in most cases, disturb the natural ability of tidal wetlands to reduce flood damage and adversely affect the public health and welfare; that such loss or despoliation will sub- stantially reduce the capacity of such wetlands to absorb silt and will thus result in the increased silting of channels and harbor areas to the detriment of free navigation. Therefore, it was declared to be the public policy of the State of Mississippi, taking into account varying Wetlands Legislation 257 scientific, ecological, economic, developmental, recreational and aesthet- ic values, to preserve the natural state of coastal wetlands and to pre- vent the despoliation and destruction of these wetlands. The Act prohibits any regulated activity from being carried on without a permit. It is the duty of any person proposing to perform any regulated activity to ascertain whether such work affects wet- lands. Any person who violates the provisions of this Act will be civilly liable to the State for the restoration of the affected wetlands to their condition prior to such violation, insofar as restoration is possible. In addition to civil liability under this Act, a violation of this Act is a mis- demeanor and will be punished by a fine of not less than $500 and not more than $1,000 or by imprisonment of not more than 30 days, or both. The Mississippi State Legislature enacted an amended “Missis- sippi Coastal Wetlands Protection Act” during its February 1973 session. 17. Louisiana In addition to general water pollution control legislation and leg- islation for control of fishing, legislation relative to mineral leasing (oil wells) is the only pertinent legislation in Louisiana. 18. Texas REV. CIV. ST AT. TEXAS arts. 4051 through 4056a These statutes give the Texas Parks and Wildlife Commi.s.sion management control over marl, sand, gravel and shell deposits in the navigable streams, bays, bayous, and the Gulf of Mexico within the jurisdiction of the State. Prior to issuing dredging permits, the Com- mission must consider possible damage to oysters, oyster beds and fish. CONCLUSIONS Several states have statutory provisions relating to wetland, marsh and submerged lands and flood plains. A general categorization of the legal approaches various states have taken based upon the above examination of these statutes indicates the following: (1) legislation which enables a state to acquire wetlands or any easement, interest or rights therein by the following means: eminent domain, purchase, exchange, gift, devise, lease, lease with an option to purchase, payment of unpaid tax liens on the land. (2) legislation which prohibits certain activities in wetlands areas. 258 Haueisen (a) many statutes provide that a project which involves fill- ing, dredging, obstructing or altering the course of wa- ters in wetlands areas may not commence without ob- taining a permit therefor; any conditions placed upon the work in the permit must be complied with. Many of these statutes provide for fines and imprisonment, and violations are subject to injunction or abatement. (b) a few' statutes prohibit uses of wetlands inconsistent with conservation by zoning wetlands for conservation purposes. (c) one statute prohibits the use of earth-moving equip- ment in wetland areas, unless such equipment is regis- tered with the Department of Water Resources. (3) the Long Island (New York) Wetlands Act is unique. It pro- vides that the State may enter into cooperative agreements with counties to maintain wetlands and may furnish one-half the cost of maintenance. (4) legislation not directly related to wetlands, but affecting flood plains, has been enacted by some states. Such legisla- tion requires that a county zone its flood plains to prevent encroachment and consequent damages. I would call special attention to legislation of the following States : 1. Connecticut — wetlands acquisition 2. Florida — sale of tidal lands 3. Maine — regulation of dredging and filling 4. Massachusetts — regulation of dredging and filling 5. New Hampshire — regulation of dredging and filling 6. New York — cooperative management agreements 7. Rhode Island — wetlands zoning and regulation of dredging and filling Recent court decisions in Massachusetts suggest that a legal basis exists for State regulation of marshland use in instances where private ownership and use rights exist w’hich may be in conflict with the pub- lic purpose of protection of marine resources. Commissioner of Natu- ral Resources v. 5. Volpe and Co., 206 N. E. 2d 666 (Mass. 1965) Reg- ulation, through the police powers of the State, is the most direct legal approach to control of use of wetland, marsh and submerged land. This is short of outright acquisition or control of development rights. Therefore, the significance of the Massachusetts court decision cannot be ignored in analyzing other states^ laws and policies which clearly support the doctrine that protection of marine re.sources is a public purpose. Wetlands Legislation 259 A LEGISLATIVE PROPOSAL FOR PROTECTION OF THE WETLANDS An act to provide for the orderly preservation and development of the coastal wetlands; to provide procedures for obtaining permits to alter wetlands; to provide penalties for violation of this act; and for related purposes. Be it enacted by the legislature of the State of : SECTION 1. This act is to be known as the “Coastal Wetlands Protection Act'" and may be so cited. SECTION 2. It is declared that much of the coastal wetlands of the State of have been lost or despoiled by unregu- lated dredging, dumping, filling and the like activities, and that the remaining coastal wetlands of this State are in jeopardy of being lost or despoiled by these and other activities; that such loss or despolia- tion will adversely affect, if not entirely eliminate, the value of such wetlands as sources of nutrients to finfish, Crustacea and shellfish of significant economic value; that such loss or despoliation will destroy the ecological system of such wetlands as habitats for plants and ani- mals of significant economic value and will eliminate or substantially reduce marine commerce, recreation and aesthetic enjoyment; and that such loss or despoliation will, in most cases, disturb the natural ability of tidal wetlands to reduce flooding and adversely affect the public health and welfare; that such loss or despoliation will substan- tially reduce the capacity of such wetlands to absorb silt and will thus result in the increase silting of channels and harbor areas to the det- riment of free navigation. Therefore, it is declared to be the policy of this State, taking into account varying scientific, ecological, economic, developmental, recreational and aesthetic values, to preserve the nat- ural state of coastal wetlands and to prevent the despoliation and de- struction thereof. SECTION 3. For purposes of this act: (a) “Coastal wetlands,’* “tidal wetlands,” or “wetlands” shall mean those areas w'hich border on or lie beneath tidal waters, such as but not limited to banks, bogs, salt marsh, swamps, meadows, flats or other lowlands subject to tidal action, including those areas now or formerly connected to tidal waters, and the surface of which is at or below an elevation of one (1) foot above local extreme high w^ater. (b) “Regulated activity” means any of the following: draining, dredging, excavation or removal of soil, mud, sand, gravel, aggregate of any kind, or rubbish from any w^ctland or the dumping, filling or depositing thereon of any soil, stones, sand, gravel, mud, aggregate of any kind, rubbish or similar material either directly or otherwise, and the erection of structures, driving of pilings, or placing of obstruc- 260 Haueisen tions, whether or not changing the tidal ebb and flow. Notwithstana- ing the foregoing, “regulated activity” shall not include the construc- tion or maintenance of aids to navigation which are authorized by governmental authority; the accomplishment of emergency decrees of any duly appointed health officer of a municipality acting to protect the public health; conservation of soil, vegetation, water, fish, shellfish and wildlife performed by duly authorized governmental agencies; or trapping, hunting, fishing and shellfishing where otherwise legally permitted. (c) “Dredging" means the removal or displacement by any means of soil, sand, gravel, shell or other material ; whether of intrinsic value or not, from wetlands. (d) “Filling” means either the displacement of waters by the deposition into wetlands of soil, sand, gravel, shell or other material ; or the artificial alteration of water levels by physical structures, drain- age ditches or otherwise. (e) “Person” means any natural person, partnership, joint stock company, unincorporated association or society, or the State and any agency thereof, or municipal or political subdivisions or other corpora- tion of any character whatsoever, (f) “Commission” shall mean the Natural Resources and Conser- vation Commission, the director of said commission or his duly au- thorized representative. SECTION 4. No regulated activity shall be conducted upon any wetland without a permit. Any person proposing to conduct or cause to be conducted a regulated activity upon any wetland shall file an ap- plication for a permit with the commission, in such form and with such information as the commission may prescribe. Such application shall include a detailed description of the proposed work and a map showing the area of wetland directly affected, with the location of the proposed work thereon, together with the names of the owners of record of adjacent land and known claimants of water rights in or ad- jacent to the wetland of whom the applicant has notice. The commis- sion shall cause a copy of such application to be mailed to the chief administrative officer in the town or towns where the proposed work, or any part thereof, is located, and to the Director of the State Game and Fish Commission, the county attorney of the county or counties in which any part of such proposed work may occur or which may be affected by such work, the district attorney of any such county or coun- ties, the boards of supervisors of any such county or counties, and the Marine Resources and Fisheries Conservation Commi.ssion, No sooner than thirty (30) days and not later than sixty (60) days from the receipt of such application, the commission shall hold a public hearing on such application. The following shall be notified of the hearing by mail not less than fifteen (16) days prior to the date set for the hear- Wetlands Legislation 261 ing: all of those persons and agencies who are entitled to receive a copy of such application in accordance with the terms hereof and all owners of record of adjacent land and known claimants to water rights in or adjacent to the wetlands of whom the applicant has notice. The commission shall cause notice of such hearing to be published at least once not more than thirty (30) days and not fewer than ten (10) days before the date set for the hearing in the newspaper having a general circulation in each county where the proposed work, or any part thereof, is located. All applications and maps and documents re- lating thereto shall be open for public inspection at the office of the commission. At such hearing any person or persons may appear and be heard. It shall be the duty of any person proposing to perform any reg- ulated activity, including the performance of any contract with any state agency for dredging, sale or removal of shells, gravel, sand or other such materials, to ascertain whether such work affects wetlands. SECTION 5. In granting, denying or limiting any permit the commission shall consider the effect of the proposed work with refer- ence to the public health and welfare, marine fisheries, shell-fisheries, wildlife, the protection of life and property from flood, hurricane and other natural disasters, and the public policy set forth in Section 1 of thisAct. The commission shall require a bond in an amount and with surety and satisfactory conditions securing to the state compliance with the conditions and limitations set forth in the permit. The com- mission may suspend or revoke a permit if the commission finds that the applicant has not complied with any of the conditions or limita- tions set forth in the permit or has exceeded the scope of the work as set forth in the application. The commission may suspend a permit if the applicant fails to comply with the terms and conditions set forth in the application. The commission shall state, upon his record, his findings and reasons for all actions taken pursuant to this section. The commission shall cause notice of any order in issuance, denial, revoca- tion or suspension of a permit to be published in a daily newspaper having a circulation in the county or counties wherein the wetland lies. SECTION 6. (a) An appeal may be taken by the applicant or any person or corporation, municipal corporation or interested com- munity group other than the applicant who has been aggrieved by such order from the denial, suspension or revocation of a permit or the is- suance of a permit or conditional permit within thirty (30) days after publication of such issuance, denial, suspension or revocation of any such permit to the court of any county having jurisdiction over the property which may be affected by any such proposed work authorized by such permit. If the court finds that the action appealed from is an unreason- able exercise of the police power, it may set aside the order. If the 262 Haueisen court so finds that the action appealed from constitutes the equivalent of a taking without compensation, and the land so regulated otherwise meets the interests and objectives of Section 1 of thisAct, it may, at the election of the commission, (1) set aside the order or (2) proceed to award damages as provided by Section 9 of thisAct. (b) Such appeal shall be brought by a complaint in writing, stat- ing fully the reasons therefor, with a proper citation, signed by a com- petent authority, and shall be served at least twelve (12) days before the return date upon the commission and upon all parties having an interest adverse to the appellant. Such appeals shall be brought to the next return day of the court after the filing of such appeal. The com- mission shall forthwith, after service of notice of any appeal, prepare and file in said court a copy of such portions of the record of the case from which such appeal has been taken as may appear to the commis- sion to be pertinent to such appeal, with such additions as may be claimed by any party of interest to be essential thereto, certified by the commission. The court, upon such appeal in making its determina- tions as provided in subsection (a) of this section, shall review, upon the record so certified, the proceedings of the commission and examine the question of the legality of the action of the commission and the propriety of said action. If, upon hearing such appeal, it appears to the court that any testimony has been improperly excluded by the com- mission or that the facts disclosed by the record are insufficient for the equitable disposition of the appeal, it shall refer the case back to the commission to take such evidence as it may direct and report the same to the court, with the commission’s finding of fact and conclusions of law. Such appeal shall have precedence in the order of trial. SECTION 7. In determining the propriety of issuing permits for any regulated activity under thisAct, the commission and courts are to interpret broadly the provisions of thisAct in favor of the pre- servation of wetlands as opposed to any alteration of the character of such wetlands, and to favor the best public interest as opposed to pri- vate or corporate pecuniary interest. SECTION 8. The Attorney General, district attorney or county attorney having jurisdiction may institute civil and/or criminal ac- tions or proceedings against any person believed to be in violation of thisAct. Such action shall be brought in the court of any county in which the alleged violation occurs or in which property affected by such alleged violation is located in the manner of other proceedings. SECTION 9. Any person who violates the provisions of thisAct shall be civilly liable to the State for the restoration of the affected wetland to its condition prior to such violation, insofar as restoration is possible. The appropriate court shall specify a reasonable time for the completion of restoration. In addition to civil liability under thisAct, a violation of thisAct Wetlands Legislation 263 is a misdemeanor and shall be punished by a fine of not less than Five Hundred Dollars ($500.00) and not more than One Thousand Dollars ($1,000.00) or by imprisonment of not more than thirty (30) days, or both. In the case of continuing violations, each day shall constitute a separate charge ; however, separate violations under this Act need not be severed for trial when an identity of parties and location exists. SECTION 10. If any clause, sentence, paragraph or part of this Act shall for any reason be adjudged by any court of competent juris- diction to be invalid, such judgment shall not affect, impair or invali- date the remainder of thisAct, but shall be confined in its operation to the clause, sentence, paragraph or part thereof directly involved in the controversy in which judgment shall have been rendered. SECTION 11. This Act shall take effect and be in force from and after its passage. Gulf Research Reports Volume 4 Issue 2 January 1973 The Seasonal Occurrence and Abundance of Chaetognatha in Mississippi Sound Mohammed Saeed Mulkana Gulf Coast Research Laboratory Thomas D. Mcllwain Gulf Coast Research Laboratory DOI; 10.18785/grr.0402.10 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Recommended Citation Mulkana, M. S. and T. D. Mcllwain. 1973. The Seasonal Occurrence and Abundance of Chaetognatha in Mississippi Sound. Gulf Research Reports 4 (2): 264-271. Retrieved from http:// aquila.usm.edu/gcr/vol4/iss2/ 10 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. THE SEASONAL OCCURRENCE AND ABUNDANCE OF CHAETOGNATHA [N MISSISSIPPI SOUND by MOHAMMED SAEED MULKANA^ and THOMAS D. McILWAIN Gulf Coast Research Laboratory Ocean Springs, Mississippi INTRODUCTION The Chaetognatha as “biological indicators" have attracted atten- tion from various fields of oceanography and fisheries biology because various species of Chaetognatha are associated with different water masses, and their distribution and numerical abundance are used to infer biological productivity and movements of water with their in- herent temperature, salinity, sediment load pH, and other variables (Legare and Zoppi 1961, Redfield 1950). The Chaetognatha popula- tions, therefore, tend to fluctuate following the seasonal changes and the influx of water masses from different sources. The present work is a study of seasonal changes in the numerical abundance of the Chaetognatha species during a yearly cycle in Mis- sissippi Sound. A comparison is made with the seasonal abundance of calanoid copepods, an important zooplankton component, for a possi- ble correlation. MATERIAL AND METHODS Samples were collected at one station near the middle of Missis- sippi Sound at approximately 30° IT N and 80° 45' W. The area re- ceives open Gulf waters through the Dog Keys Pass and brackish water from the Bay of Biloxi. Plankton samples were collected at monthly intervals with Clarke-Bumpus plankton samplers, equipped with silk bolting nets :^2 and #10. The nets were towed 15 rain, for each haul, and the amount of water filtered was recorded by means of calibrated flow meters attached to the samplers. Both types of nets were used at the surface and at a depth of 10 ft. which is near the bottom. Occasionally the efficiency of the plankton nets decreased due to large amounts of suspended clay particles, phytoplankton blooms, ctenophores and med- usoid coelenterates. Data on temperature, salinity and transparency were recorded at each sampling. Surface temperatures were taken with a precision mercury thermometer (Fig. IL Water transparency was recorded by 1 Present address, Jackson County Campus, Mississippi Gulf Coast Junior College, Gautier, Mississippi 39553. 264 Chaetognatha in Mississippi Sound 265 30 Figure 1. Seasonal variations in temperature, salinity and numerical abundance of chaetognaths in Mississippi Sound from January to Decem- ber 1965. 266 Mulkana and McIlwain means of a Secchi disc. Salinity was determined from hydrometer readings. The identification of Chaetognalha is difficult when body sizes overlap, and morphological characters vary with age and size. The important characters used in classification were size, extent of collar- ette, number of hooks, number of anterior and posterior teeth, size and shape of the seminal vesicles, presence or absence of anterior fin, T.C. ratio (the ratio of posterior fin along the trunk to the posterior fins along the caudal segment multiplied by 100) the completeness of the fins, the ventral ganglion and presence or absence of the gut diver- ticulum (Cf. Grant 1963, Legare and Zoppi 1961, Pierce 1947, and Thomson 1947). RESULTS AND DISCUSSIONS Eight species of Chaetognatha belonging to two genera, Sagitta and Krohnitta, were identified in the samples (Table 1). Pierce (1947) reported the occurrence of Sagitta tenuis, S. his- pida, S, enfUita, S, hdenae and Krohnitta pacifica off the west coast of Florida. All these species were present in our samples. Sagitta tennis and S. hispida were most abundant and formed the major part of the chaetognath populations (Tables 1 and 2). Although both of these species are considered warm water neritic forms, they also occur in the Atlantic, but their northern limits have not been determined. S. tenuis is found as far north as Delaware Bay and S. hispida has been reported to occur as far north as Cape Hatteras in the Atlantic (Deevey 1960) . Sagitta enflata and S. bipimctata were comparatively less abundant, Sagitta enflata is an epiplanktonic, warm-water form, but occurs north of Cape Hatteras, where it seems to be a stray form from the Gulf stream. Sagitta enflata is abundant in Florida waters (Grant 1963) . Sagitta bipimctata is an oceanic form and its presence indicates mixed oceanic and neritic waters. Grant (personal communi- cation) has suggested that the Gulf of Mexico chaetognath fauna in- cludes moat, if not all, the forms found in the central and equatorial Atlantic. Sagitta helenae has restricted distribution and is limited to the waters off the Atlantic coast of the United States and the Gulf of Mexico. Sagitta hexaptera, S. serratodentata and Krohnitta paciflea were found in fewer number and probably occur as stray forms. Surface and bottom samples were analyzed separately to see if there was any variation in these shallow depths. A significant numeri- cal abundance was observed only during the summer period (Table 2) . In June and August a greater number of organisms occurred in the bottom samples, while in July a comparatively higher number was found in the surface samples. In June and Augu.st transparency was low (McIlwain 1968), apparently because of pronounced mixing of S C 03 3 ec B o ««-i A X 03 C 5£ -a 5 C 3 “•a tw fil ® ■« « .2 § J is i-s > »ri (£) ss^ I •a w 03 Q 4) U cs c3 XJ e s 4» W S O) 3 u u o Krohnitta paoifica II 1 1 1 1 v£> 1 1 1 1 Sagitta serrato- dentata II 1 1 1 1 in 1 1 I 1 +i \ B Ss « oa 01 to 'Ki II 1 I v£> 1 1 1 1 1 1 y 1 +!> § « Co rQ 4^ I i 119 i 208 t Sagitta he lenae 1 —1 1 cvj I 1 CTi ^ . 4_l . . . C^>-lM{>>c:»HW3eU4J>U tocucvjp^coppDcutjoaj »nP*^S L C N C I H ifTsm) Figure 8. Monthly tail length-frequency curves of males and females of brown shrimp on which all the sampling period is represented. Mean length values of successive generations are identified by letters. The corresponding total length scale (Lj) is also shown below. 290 Chavez Table 1. Monthly age classes determined from the analysis of frequencies made with probability paper. They are represented as abdominal length, in ram. Each generation is identified by letters. (These values are also shown in Figure 12) Generation Mean size values of monthly age classes I II III IV V VI A 103.5 108.0 B 88.5 94.5 94.5 99.0 C 82.0 87.0 91.5 96.0 105.0 D 93.0 E 81.0 91.5 94.5 F 81.0 88.0 96.0 G 79.0 87.0 96.0 H 79.0 87.0 94.5 105.0 I 79.0 87.0 91.5 96.0 100.0 J 78.0 K 77.0 88.0 L 89.5 95.5 106.5 M 78.0 88.5 97.5 102.0 N 87.0 94.5 103.5 0 78.0 87.0 93.0 96.0 103.5 P 85.5 Q 87.0 96.0 102.0 R 87.0 93.0 S 85.0 Average 79.2 87.3 94.1 98.4 102.8 108.0 the regressions previously determined. But in the case of males and with the considerations formerly made, it seemed pertinent to take the results of males and females computed together instead of those of males only in order to reduce the risk of working with wrong data, mainly when there is the antecedent of no marked differences between the sexes (Chin 1960). Therefore, the results of growth in weight are shown in the following expressions : Males: w = 46 [1 — e-o-2567(t+o.238a)]3 Females: w =:^113 [1 — e—^. 162 ( 1 + 0 . 750 )] s The expressions seem to be similar to those of the length growth. Growth of Penaev.s aztecus aztecus 291 Figure 9. Ford-Walford graph of abdominal length in mm at age t against length at age t+1 for males and females of brown shrimp. Nevertheless, the differences existing between Loo of each sex are equivalent to quite distinct weights of specimens of the same age but different sex, variations that increase with age. This can be seen in Fig, 11 where the growth in weight is represented by sex. The sizes and weights of each monthly age class calculated by •means of the von Bertalanffy growth model are shown in Table 2, whose variations explain well the differences in growth found by Williams (1955), Loesch (1965), and St. Amant et al, and Joyce (both after Cook and Lindner 1970) upon juvenile brown shrimp populations, somewhat larger growth rates than those found during the present analysis for similar size ranges. Finally, bearing in mind that only global data are frequently available, for example, the Gulf Coast Shrimp Data, in which it is not 292 Chavez Figure 10. Von Bertalanffy growth curves, in length, found for each sex of brown shrimp. Observed values of age classes are shown by small circles. possible to separate the differences due to sex, it was decided to cal- culate growth parameters of the population as a whole, which was done in the following way : with the values of Leo and W co determined for each sex, new average values were obtained and were considered as Loc and W oc of the population, and the age classes were those de- termined by making an average of the resulting pairs of values form- erly analyzed after the growth rate was considered for each sex sep- arately. Therefore, the latest obtained results are as follows : Loo = 207 mm Woo = 70 g k = 0.1904 to = —0.872 The graphic representation of individual growth rate of the Pe- naeiis aztecus aztecus population studied is shown in Fig. 12, and in Table 3 the length and weight values for each month of age are also indicated. 294 Chavez Table 2. Growth rate in length and weight of Penaens aztecus aztecus Ives calculated for each sex and month during the first fourteen months of age MALES Age (Months) Length (mm) Increase (mm) Weight (g) Increase 1 48.5 37.9 0.9 0.9 2 75.7 27.2 3.7 2.8 3 100.6 24.9 8.3 4.6 4 118.3 17.7 13.4 5,1 5 131.8 13.5 18.6 5.2 6 142.0 10.2 23.4 4.8 7 150.4 8.4 27.7 4.3 8 156.6 6.2 31.3 3.6 9 161.4 4.8 34.3 3.0 10 165.3 36.7 2.4 11 168.2 38.7 2.0 12 170.4 40.3 1.6 13 172.1 1.7 41.5 1.2 14 173.5 1.4 42.5 1.0 FEMALES 1 58.3 30.9 1.7 1.5 2 85.0 26.7 5.3 3.6 3 107.6 22.6 10.7 5.4 4 127,0 19.4 17.5 6.8 5 143.3 16.3 25.2 7.7 6 157.1 13.8 33.3 8.1 7 169.0 1 11.9 41.4 8.1 8 179.0 10.0 49.2 7.8 9 188,0 9.0 56.6 7.4 10 195.0 7.0 63.4 6.8 11 200.9 5.9 69.7 6.3 12 205,0 4.1 75.3 5.6 13 210.6 5.6 80.3 5.0 14 214.4 3.8 84.7 4.4 Growth of Penaeus azteciis aztecun 295 AGE, MONTHS Figure 12. V^on Bertalanffy growth curves found for all brown shrimp population sampled (both sexes combined). The curves show the longi- tudinal and ponderal average growth of individuals. DISCUSSION OF RESULTS AND CONCLUSIONS As a corollary of biometric study of the analyzed population, it is concluded that the total length-abdominal length ratio shows a mostly straight relationship in both sexes, and that the observed dif- ferences in each sex are very slight. For this reason, it is supposed that the calculated population formula is representative of such a transformation, a characteristic that on the other hand means that the shrimp growth is isometric respectively to these two factors. The calculated, relationship was made on a basis of the need of knowing the total length of specimens formerly sampled, and it was necessary to make a reference calculated in terms of total lengths of the shrimp. Concerning the length-weight relationship, it is pertinent to point out that the exponent of length usually is a value which fluctu- ates about three, and in this case described an isometric growth^ char- acterized from Ricker (1958), because the specific gravity and the body form remain constant, regardless of the size of the organism. Former experiences with shrimp (Hall 1962, Butler 1970, Chin 1960, 296 Chavez Table 3. Monthly growth rate in length and weight of brown shrimp calcu- lated for all population samples (both sexes combined) Age (Months) Length (mm) Increase (ram) Weight (g) Increase (g) 1 61.8 i 30.1 1.9 1.7 2 87.0 25.2 ; 5.3 3.4 3 105.5 18.5 9.9 4.6 4 121.1 15.6 15.5 5.6 5 139.6 18.5 21.3 5.8 6 151,1 11.5 27.2 5.9 7 160.8 9.7 32.8 5.6 8 168.5 7.7 37.9 5.1 9 175.2 6.7 42.6 4.7 10 181.0 5.8 46.7 4.1 11 185.4 4.4 50.3 3.6 12 189.1 3.7 53.4 3.1 13 192.2 3.1 56.1 2.7 14 195.0 2.8 58.3 2.2 Kutkuhn 1966, George 1970a and 1970b, Nikolic' and Garcia 1970, Cruz-Morejon and Cadima 1970, Angelescu and Boschi 1959, and Chavez and Rodriguez-de la Cruz (1971), suggest that at least in this group, the growth shows a remarkable tendency to isometry, where the calculated exponents fluctuate around three, ranging from 2.65 to 3.25. These variations are probably due to insufficient data, small number of samples, to variations in condition coefficient of shrimp, or to the influence of other physiologically responsible factors. Gulland (1969) points out that there is a wide number of growth equations, but none is entirely satisfying in all possible situations. Dickie (from Tesch 1968), asserts that growth curves are valuable for descriptive purposes, but the biological interpretation of models and their parameters still present great diflficulties. The von Bcrta- lanffy growd^h model in the author's opinion, has been shown to be quite satisfactory; besides, it should be kept in mind that this model is the best known and the most widely used in production studies on species of economic value ; it satisfies reasonably the two most impor- tant criteria : it fits most of the observed data on growth, and can be readily incorporated into stock assessment models. The growth analysis developed in the present paper seems to offer Growth of Penaens aztecus azteciis 297 a good outlook for further application to the penaeid shrimp group. This group presents serious difficulties for the determination of age and growth of the species belonging to it, because of its variability due to environmental changes. There is also the impossibility of re- ferring the age classes to growing marks to check the analytical infer- ences made from the information acquired with the samples. With the comparative study of size-frequency curves, it is possi- ble to figure out that in analyzed samples there is a small percentage of shrimp whose age ranges are from 2 to 13 months old in males, approximately, with lengths ranging from 93 to 172 mm in each case; in females, all seems to show that age classes ranging from 2 to 15 months old are present (the number of extreme size classes is also small), whose lengths lie between 91 and 216 mm. It is supposed that the presence of larger shrimp is a random occurrence rather than the result of an ecological factor. These large sizes were found from June through August in both sexes, from January until March in males, and from November to March in females. The smaller shrimp, approx- imately 2 months old, were observed quite well chronologically located. In spite of the fact that Gulf of Mexico shrimps viiTually repro- duce the year-round, the presence of the smallest shrimp is neces- sarily interpreted as a maximum of reproduction existing at least 2 months before their recruitment. If recruitment of such tiny shrimp occurs mainly from June through August, it is supposed-that they be- long to a generation born from March through May (it is necessary to add 15 days to the apparent age, because of the larval stage dura- tion). In part, this confirms the opinion of Kutkuhn (1962) about the increased spawning activity during March- April and September- October. The results obtained by the use of probability paper to determine the size classes suggests more objectivity than the use of the size- frequency curves only (Chavez and Rodriguez-de la Cruz 1971), es- pecially when the number of samples is small. For instance, either one of these methods may be profitable, but the accuracy of the results is determined by the experience of the analyst. It is necessary to keep in mind that the present study was under- taken with the idea of determining the average individual growth rate, regardless of the seasonal valuations, due to the fluctuating changes of ecological factors determining the habitat of P. azteciis azteciis populations. The study was made in that way because on in- corporating growth parameters into stock assessment models, extreme variations capable of modifying the results are discarded. SUMMARY A study of brown shrimp {Penaeus aztecus aztecus Ives, 1891) was undertaken. The data were obtained from samplings made twice 298 Chavez or thrice a week in Tampico (Mexico) canneries, during the period from June 1967 to March 1969. The data consist of abdominal length records of 20,003 specimens of both sexes, 6,879 males and 13,124 females. An additional sampling of 500 specimens was made in July 1971 in which total length, tail length and weight were recorded. These data were analyzed as logarilhmical regressions establishing the corresponding relationships, of which the formulas are as follows : Total length (Y) — abdominal length (X) ratio: Males: Y = 2.05 r = 0.999 Females: Y = 1.531 r = 0.999 Both sexes: Y = 1.621 ^ ^ o.998 Weight (W) — length (1) ratio: Males: W = 0.000214 r = 0.968 Females: W = 0.000010 r = 0.997 Both sexes: W 0.000023 j, ^ o.986 Tail length data were grouped and represented as monthly length-frequency curves, to which mean values of age classes found in each month were incorporated. These values were obtained after analyzing each monthly group of data on probability paper, Once the mean values of age classes were obtained, a Ford-Walford plot was made and a von Bertalanffy growth curve was fitted. The constants are a.s follows: Males : Loo = 178.1 mm Woo ^ 46 g k = 0.2567 to ^ —0.2388 Females : Loc = 236 mm Woo = 113 g k = 0.162 to ^ —0.759 Both sexes; Loo = 207 mm Woo = 70 g k = 0.1904 to = —0.872 LITERATURE CITED Angelescu, V. and E. E. Boschi. 1959. Estudio biologico pesquero del langoatino del Mar del Plata en conexion con la operacion nivel medio. Servicio de Hi- drografla Naval, Argentina, 1017:1-135. Beverton, R. and S. Holt. 1957. On the dynamics of exploited fish populations. Min. Agr. and Fish., Fish. Invest. Lond., Series 2, 10. Butler, T. H. 1970. Synopsis of biological data on the prawn Pandalus platyceros Brandt, 1851. FAO Fish. Rep. (57)4:1280-315. Cassie, R, M. 19.54. Some uses of probability paper in the analysis of size fre- quency distributions. Australian J. Marine and Freshwater Res., 5:513-22. Chavez, E. A. and M. C. Rodriguez-de la Cruz. 1971. Estudio sobre el crecimiento Growth of Penaeus azteciis azteciis 299 del camaron cafe (Penaeus callfondensis Holmes) del Golfo de California. Rev. Soc. Mex. Hist. Nat. 32:111-27. Chin, E. 1.960. The bait shrimp fishery of Galveston Bay, Texas. Trans. Amer. Fish. Soc., 89(2) :135-4l. Cook, H. L. and M. J. Lindner. 1970. Synopsis of biological data on the brown shrimp Penaeus aztecus aztecus lyes, 1891. FAQ Fish, Rep. 57(4): 1471-97. Cruz-Morejon and E. Cadima. 1970. Relaciones entre largos y pesos de camarones capturados en la plataforma cubana. FAO Fish. Rep. (57)2:539-48. Cushing, D. H. 1968. Fisheries biology. The Uniyersity of Wisconsin Press. George, M. J. 1970a. Synopsis of biological data on the penaeid prawn MeAuve- naeus monoceros (Fabricius 1798). FAO Fish. Rep. (57)4:1539—57. 1970b. Synopsis of biological data on the penaeid prawn Metapenaeus brevicornis (II. Milne Edwards, 1837). FAO Fish. Rep. (57)4:1559-73. Gulland, J. A. 1969. Manual of methods for fish stock assessment. Part I. Fish population analysis. FAO Man. Fish, Sci. Hall, D, N, F. 1962. Obseryations on the taxonomy and biology of some Indo- west Pacific Penaeidae (Crustacea, Decapoda). Fish. Pubis. Colon. Off., 17:- 1-229. Klima, E. F. 1964. Mark recapture experiments with brown and white shrimp in the northern Gulf of Mexico. Proc. Gulf. Carib. Fish. Inst, 16:52-64, Kutkuhn, J, H. 1962. Gulf of Mexico commercial shrimp populations, trends and characteristics, 1956-59. U. S. Fish Wildl. Sery., Fishery Bull. 62(212) :343- 402. 1966. Dynamics of a penaeid .shrimp population and management im- plications. U, S. Fish. Wildl. Sery., Fishery Bull. 65(2) : 313-38. Loesch, H. 1965. Distribution and growth of penaeid shrimp in Mobile Bay, Ala- bama, Publ. Inst. Mar. Sci. (Texas), 10:41-58. Nikolic’, H. and R. Garcia. 1970. Diez viajes de exploraci6n con el barco Camaron I en la plataforma sudoriental de Cuba. FAO Fish. Rep. (57)2:571-87. Osborn, K. W., B. W. Maghan, and S, B. Drummond. 1969. Gulf of Mexico shrimp atlas. U. S. Dept. Int. Bur Com. Fish. Circular 312:1-20. Ricker, W. E, 1958. Handbook of computations for biological statistics of fish populations. Bur. Fish. Res. Bd. Canada (119) : 1-300. Tesch, F. W. 1968. Age and growth. In Methods for assessment of fish production in fresh waters. W. E. Ricker (Ed.). I. B. P. Handbook No. 3:93-123. Black- well. Williams, A. B. 1955. A contribution to the life histories of commercial shrimps (Penaeidae) in North Carolina. Bull. Mar. Sci, Gulf, Caribb. 5:116-46. Gulf Research Reports Volume 4 Issue 2 January 1973 Nutritional Components of the Standing Plankton Crop in Mississippi Sound Mohammed Saeed Mulkana Mississippi Gulf Coast Junior College Walter Abbott Gulf Coast Research Laboratory DOI: 10.18785/grr.0402.13 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Mulkana, M. S. and W. Abbott. 1973. Nutritional Components of the Standing Plankton Crop in Mississippi Sound. Gulf Research Reports 4 (2): 300-317. Retrieved from http:// aquila.usm.edu/gcr/vol4/iss2/ 13 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized administrator of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. NUTRITIONAL COMPONENTS OF THE STANDING PLANKTON CROP IN MISSISSIPPI SOUND^ by MOHAMMED SAEED MULKANA^ and WALTER ABBOTT Gulf Coast Research Laboratory Ocean Springs, Mississippi 39564 ABSTRACT A study on seasonal changes in the nutritional components of standing plankton biomass was made from 20 April 1965 to 6 Septem- ber 1966. Plankton were separated into net plankton and nannoplank- ton fractions. Nannoplankton'* standing biomass exceeded by 72 times that of net plankton. Although protein and carbohydrate levels were higher in net plankton on a unit weight basis, total nutrients available from nan- noplankton were substantially higher. Some seasonal trends were ap- parent as changes in the standing biomass in net plankton. Nanno- plankton exhibited no seasonal trends. Lipid and caloric values of net plankton showed seasonal trends similar to those shown by dry weight. INTRODUCTION Early studies on marine plankton were directed primarily toward species composition, geographical distribution and seasonal succes- sions. The importance of plankton to fisheries led to studies of plank- ton biomass, with considerable recent emphasis on plankton produc- tion as indicated by primary photosynthetic productivity. There are relatively few data, however, on specific nutritional components — carbohydrate, protein, and lipid — available from plankton and on quantitative seasonal changes in these components. Furthermore, al- though nannoplankton comprises the major fraction of the total plank- ton and exceeds the net plankton manyfold (Banse 1964, Pomeroy and Johannes 1966), those nutritional component data available relate only to net plankton. 1 This work was supported in part by National Science Foundation Grant No. BG-3452, and is extracted from a Ph.D. Thesis submitted to Department of Zool- ogy, Mississippi State University. 2 Jackson County Campus, Miss. Gulf Coast Jr. College, Gautier, Mississippi 39563. 3 It should be noted that the authors did not Separate the nannoplankton from what has been called the nannodetritus and both components are included in the word nannoplankton, as used in this paper. Ed. 300 Mississippi Plankton Crop 301 Mississippi Sound is regarded as a highly productive body of water (Gunter 1962, 1963, Christmas, Gunter, and Musgrave 1966) because of the influence of nutrient rich waters from Mississippi River (Riley 1937). This report is concerned with measurements of net plankton and nannoplankton dry weight, protein, and carbohy- drate, and with lipid and caloric content of net plankton during a yearly cycle in Mississippi Sound. MATERIALS AND METHODS From April 20, 1965 to September 6, 1966, plankton was col- lected as near monthly as logistics and weather permitted from 6 sampling stations along two transects across Mississippi Sound — from Marsh Point to Horn Island (Fig. 1) and from Ship Island to Biloxi Beach (Fig. 1). The order of running the transects was sys- tematically reversed from cruise to cruise to avoid sampling bias, but on each cruise one transect was w'orked by day and the other by night. Composite net plankton samples were taken by pumping water from surface, mid-depth, and near bottom into a No. 20 plankton net Figure 1. Map of Mississippi Sound showing sampling stations and depths. GCRL represents the site of the Gulf Coast Research Laboratory. 302 Mulkana and Abbott (76 ja mesh size) submerged in a water-filled container. The water was pumped at the rate of 190-240 liters/min. for 30 min. with a pre- calibrated pump. Leong (1967) has discussed plankton sampling by pumping in oceanographic .situations. Our conditions were less severe because we used 5 cm pump instead of 1,27 pump and pumped water from a maximum depth of 8 meters rather than from 120 meters. As a result there should be less churning effect and less fragmentation of the planktonic organisms. A 19-liter aliquot of net filtrate was taken at each station as a nannoplankton sample. Both this aliquot and the net plankton sample were kepi chilled by ice until return to the labor- atory. The nannoplankton samples were flocculated with potassium alum (KAl(SOi).,) to precipitate all suspended matter, plankton centri- fuges having proved unsatisfactory for this purpose. The floes \vere dissolved with just sufficient 3N hydrochloric acid and centrifuged at 1500 HPM for 15 minutes. Both net and nannoplankton concentrates were freed from water and salt by filtration on 0.8 ju, Millipore type A A filters with a rapid wash with isotonic ammonium formate (Par- sons, Stephens and Strickland 1961). Humphrey and Wootton (1966) have shown by pigment studies with Gijmnodinium, Nnnnochloris, and sea water samples that delicate, small phytoplankton are retained as well by this Millipore filter as by those of lesser pore size. Samples were dried to constant weight at 105° C (Curl 1962) against a Milli- pore filter blank, scraped from the filter, and homogenized by grind- ing. For measurement of protein content, portions of plankton homo- genates, and albumin, used as standard, were hydrolyzed in a sealed vial with 6N HCl for 12 hours at 120“ C (Welcher 1963). Hydroly- sates were filtered, dried on a steam bath, and diluted to a standard volume. Amino acid content of plankton fractions and albumin hydro- lysates was determined colorimetrically a.s leucine equivalents using a modification of the ninhydrin technique of Landua and Awapara (1949), The albumin equivalent of leucine, determined experimental- ly, was 6.97 /xmol leucine/mg albumin. Plankton carbohydrates were determined by hydrolyzing por- tions of homogenates, and oyster glycogen, used as standard, with 3N HCl for 1^2 hours at 95° C. Hydrolysates were dried and glucose content was determined colorimetrically by the method of Folin and Wu (Department of the Army 1951 ) . Nannoplankton samples were too .small for routine lipid analysis. Lipid from net plankton wa.s extracted with chloroform-methanol- water (2:2:1) in a screw-capped Waring blender jar (Bligh and Dyer 1959). The homogenates were filtered and allowed to .stand in separa- tory funnels with Teflon stopcocks until phases separated. The chloro- form layer was removed, washed with chloroform-methanol-water Mississippi Plankton Crop 303 (3:48:47) (Folch, Lees and Stanley 1957), and dried with infrared lamp. Total lipid in the aliquots was determined spectrophotometrically by the method of Snyder and Stephens (1959) using triolein as stan- dard. Caloric measurements on net plankton were made according to Parr Manual No. 122 (1951) in a Parr Peroxide Bomb Calorimeter, Model 1401. Dried net plankton samples were ground to pass a 60 mesh screen. Particle size is important, because combustion reactions occur in a few seconds and large particles, if present, may not burn completely. Benzoic acid was used as combustion aid. Ash content of plankton samples w’as determined by hot nitric acid digestion of the dried homogenate followed by ignition to con- stant weight in a muffle furnace at 550*^ C. Salinity and temperature were determined m situ at each station by means of a Beckman Induction Salinometer. Figure 2 shows mean surface water temperatures during the course of this study. All data for biomass, nutrients, and caloric values have been com- puted for a water column of 1 sq m area. 4 —I 1 1— T 1 ( 1 r 1 r — ^ — I 1 1 1 I 1 1 1 r AMJ JASONDJ FMAMJJASO 1965 ' --1966 Figure 2. Mean surface temperature at six sampling stations from 20 April 1965 to 6 September 1966. 304 Mulkana and Abbott RESULTS Dry Weight: The net plankton dry weight values of 637 mg/sq m and 452 mg/sq m (Fig. 3, Table 1) observed in April and May 1965, o X E NANNOPLANKTON NET PLANKTON AMJ J ASONDJ FMAMJ JASO -1965 1966 Figure 3. Seasonal variation in mean values of dry biomass of net plank- ton and nannoplankton. respectively, derived from a plankton increase because of vernal warming in Mississippi Sound (Thomas and Simmons 1960), The peak of 401 mg/sq m in August 1965, appeared to be due to increased herbivore populations. The magnitude of spring and summer peaks in 1966 was considerably higher than peaks in 1965. Deevey (1960) indi- cated that the zooplankton seasonal cycle in nearshore waters may be extremely variable from one year to the next. On several occasions in summer 1966 concentrations of a blue-green alga, Trichodesmium sp., were observed in surface waters. The algal abundance would contrib- ute markedly to the upswing in standing biomass. Beginning in late summer 1965, a gradual decrease in standing dry weight ensued, with a minimum of 43 mg/sq m in January 1966. A similar trend was noted again in 1966, when average values showed a sharp decline during September, decreasing from 1560 mg/sq m to 681 mg/sq m (Table 1). Table 1 Summary of Mean Values for Dry Weight, Nutrients, Ash Content and Caloric Values in the Standing Biomass of Net Plankton. Means have been Computed from Results at Six Stations Date Dry Weight mg/sq m Protein mg/sq m albumin Carbohydrate mg/sq m glycogen Lipid mg/sq m triolein Ash mg/sq m Total nutrient and ash as per- cent dry weight kcal/ sq m 1965 20 April 637 158 29 15 217 65 1.7 4 May 228 62 9 8 113 84 0.5 11 May 452 100 7 15 230 78 0.9 25 June 284 96 9 11 147 92 0.6 31 July 282 63 6 8 171 88 1.1 14 August 401 104 7 11 208 82 0.8 26 August 337 66 6 16 130 65 1.1 7 October 285 73 4 4 166 87 0.6 4 November 225 47 4 5 129 82 0.5 20 December 194 17 10 1 no 71 1966 23 January 43 15 1 1 22 90 8 April 508 92 24 9 175 59 1.3 28 May 898 172 15 11 463 74 2.2 17 June 444 76 7 16 219 72 1.0 1 July 739 217 37 12 389 88 1.9 15 July 581 197 34 16 276 90 1.1 19 July 740 271 20 19 402 96 1.8 13 August 566 185 32 14 292 92 1.2 26 August 1560 470 32 48 778 85 3.3 6 September 681 235 16 34 310 87 1.7 Grand Mean 504 136 15 14 247 81 1.3 Mississippi Plankton Crop 305 306 Mulkana and Abbott The nannoplankton dry weight exceeded the net plankton dry weight manyfold, and showed a mean ratio to net plankton of 140:1, with the maximum in winter and the minimum in summer (Tables 1, 2 and 3). Yentsch and Ryther (1959), on the basis of chlorophyll a values, and Johannes (1964), on the basis of dry biomass, pointed out that nannoplankton comprised the major fraction of the total plank- ton. The nannoplankton crop showed no definite seasonal trend, bio- mass peaks at various stations appearing in spring, fall, midsummer, and midwinter (Fig. 3). Mean dry weight values ranged from 7.50 g/sq m to 21.8 g/sq m in spring and from 7.9 g/sq m to 25.4 g/sq m Table 2. Summary of Mean Values for Dry Weight and Nutrients in Nannoplankton. Means have been Computed from Results at Six Stations Date Dry Weight g/sq m Protein g/sq ID albumin Carbohydrates g/sq m glycogen 1965 20 April 21.8 1.4 0.5 4 May 7.5 0.8 0.1 25 June 10.9 1.3 0.5 31 July 25.4 1.-8 0.6 14 August 8.0 0.9 0.3 26 August 7.9 0.5 0.3 7 October 37.0 1.6 1.1 4 November 46.7 1.3 1.4 20 December 67.6 1.1 0.5 1966 23 January 42.8 1.0 2.3 28 May 95.2 2.7 0.5 17 June 123.5 6.3 6.0 1 July 22.8 2.3 1.4 15 July 28.5 1.7 1.0 29 July 61,2 3.5 1.3 13 August 17.2 3.2 0.7 26 August 12.4 2.5 0.4 6 September 14.8 2.9 0.6 Grand Mean 36.2 2.0 1.9 Mississippi Plankton Crop 307 Table 3. Seasonal Variation in NannopIankton-to-Net Plankton Ratios for Dry Weight, Proteins, and Carbohydrates Date Dry Weight nanno/net Proteins nanno/net Carbohydrate nanno/net 1965 20 April 34 9 17 4 May 33 13 17 25 June 38 13 62 31 July 90 29 100 14 August 20 9 48 26 August 24 7 57 7 October 130 22 278 4 November 208 27 310 20 December 348 64 55 1966 23 January 996 65 2086 28 May 106 16 34 17 June 278 82 903 1 July 31 11 39 15 July 49 9 87 29 July 83 13 65 13 August 30 17 22 26 August 8 5 14 6 September 22 12 37 in summer 1965 (Table 3). An average nannoplankton rise was no- ticed starting in fall and reaching a maximum, 67.6 g/sq m, in Decem- ber 1965. The nannoplankton-to-net plankton dry weight ratio was 996:1 in January 1966 and 8:1 in late August 1966, the highest and lowest for the entire study period (Table 3). Such an extreme differ- ence in ratio probably resulted from low production of net plankton and little or no spawning and breeding activity of herbivores at a time when production at the nannoplankton level was high. Roth frac- tions maintained higher average standing dry weights in summer than in winter, with a differential, larger increase in net plankton. Protein : Protein comprised the largest component of net plank- ton nutrients. The dry weigh t-to-protein ratio was 3.7:1, equivalent to 27% protein. Parsons, Stephens and Strickland (1961), Raymont 308 Mulkana and Abbott (1963), and Blazka (1966) have shown considerably higher relative protein levels in net plankton, apparently because their samples were not contaminated by the large amounts of detritus, clay, and silt usu- ally present in collections from Mississippi Sound. The high ash con- tent in our net plankton samples should account for the low relative protein level. Mean net plankton protein for all samples was 136 mg/sq m. The level varied from 62 mg/sq m to 158 mg/sq m between April and August 1965 (Fig. 4). With the seasonal temperature decline, protein 1965 —1966 Figure 4. Seasonal variation in mean values of protein of net plankton and nannoplankton. values decreased to 15 mg/sq m in January 1966. As temperatures rose again with the onset of spring, net plankton protein values again increased, reaching 470 mg/sq m in August 1966. The summer maxi- mum for 1966 was considerably higher than the summer values for 1966 and followed the same trend as net plankton dry weight. The dry weight-to~protein ratio in nannoplankton was 18:1, equivalent to 5.5% protein. Similar low levels of protein in smaller plankton organisms have been reported by Blazka (1966), Raymont (1963), and Parsons, Stephens and Strickland (1961). This supports the concept that nannoplankton are mainly composed of photosyn- Mississippi Plankton Crop 309 thesising elements and, hence, are mainly carbohydrate, not protein (Fig. 5) . Nannoplankton protein ranged from 0.5 g/sq m to 1.8 g/sq m during the April-December 1965 period, with maximum and mini- mum occurring in the summer (Table 2). Data for 1966 indicated greater variability in available protein. Values fluctuated between 1.0 g/sq m and 6.3 g/sq m, with marked variations during the summer. Although the protein level in nannoplankton was extremely low, the nannoplankton-to-net plankton mean protein ratio on a water column basis was 24:1 (Table 3). ^ LI PIDS ililii C AP BO H Y DRAT ES NANNOPLANKTON 20 10 0 10 20 Figure 5. Average nutrient composition and dry weight of Mississippi Sound plankton. 310 Mulkana and Abbott Carbohydrate : Carbohydrate composed 3% of the dry net plank- ton sample, Riving a carbohydrate-to-protein ratio of 1 :9. Various biochemical studies on plankton agree that carbohydrate level in net plankton is relatively low and protein makes up most of the nutrient component (Raymont and Conover 1961, Parsons, Stephens and Strickland 1961, Blazka 1966). Carbohydrate values for April 1965 and April 1966 were 29 mg/sq m and 24 mg/sq m, respectively (Table 1). From late spring to winter 1965, values remained at or below 10 mg/sq m with a minimum of 1 mg/sq m in January 1966 (Fig. 6). I nannoplankton 6i - 5 - 4 ■ 3 ■ 2 ■ 1 ■ r A 1965 ^ 1966 Figure 6. Seasonal variation in mean values of carbohydrate of net plankton and nannoplankton. Nannoplankton mean carbohydrate content was 1.9 g/sq m, of the dry weight, with a mean ratio of 1 :1 :18 for carbohydrate : pro- tein :dry weight (Table 2). Carbohydrate and protein formed only about 10% of the dry weight, the low nutrient content in nannoplank- ton apparently resulting from large amounts of detritus and clay ad- mixed with nannoplankton samples. Ash weight determinations on 15% of the samples gave values as high as 85% ash. Although nanno- plankton showed a lower percentage nutrient content, on a unit area basis average carbohydrate level was a total of 235 times that of net plankton. Protein and carbohydrate maintained almost the same levels in the nannoplankton and showed similar seasonal fluctuations. Mississippi Plankton Crop 311 Livid: The average net plankton lipid was of the dry weight. The maximum, 16 mg/sq m, and minimum, 1 mg/sq m, values were observed in August and December, respectively, in 1965 (Fig. 7). In 1966, minimum, 1 mg/sq m, and maximum, 48 mg/sq m, oc- curred respectively, in January and August. From August 1965 on- ward, lipid values declined sharply to a minimum in winter and grad- ually increased with the onset of vernal conditions in 1966. The lipid content relative to the dry weight, during August lipid peaks, was 4.7% in 1965 and only 3% in 1966 when a considerably higher stand- ing crop was noted (Table 1). Wimpenny (1938) reported similar results from the North Sea, and indicated that lipid content was lower when higher standing biomass was present. Fisher (1962) pointed out that size, spawning, maturity, and season affect lipid content of plank- ton, but he attributed the difference in lipid content to a differential in Species composition from one year to the next. Lipid content in spring samples, which consisted predominantly of phytoplankton, was lower than in summer samples which included considerable zooplank- ton (Table 1). Blazka (1966) indicated that lipid content per unit bio- mass increases from algae to zooplankton in fresh water habitats. Possibly this is true in marine environments as well. Nannoplankton samples were too small for routine determination of lipid content, but based upon a single cruise, when 18 gallons of net filtrate from each of the six stations were processed, lipid content was found to be 1.2% of nannoplankton dry weight. 1965 — 1966 Figure 7. Seasonal variation in mean value of lipids of net plankton. 312 Mulkana and Abbott Figure 4 summarizes dry weight and nutrient composition data for net plankton and nannoplankton. Caloric values and ash content : With one exception, net plankton caloric values^ averaged for six stations on each date, fell within the range 3. 6-5. 4 kcal/ash-free g (Fig. 8). This is a remarkable consis- (M 1965 1966 Figure 8. Seasonal variation in mean caloric values and ash weights of net plankton. tency^ considering that net plankton dry weights in the same sample series varied from 225 mg/sq m to 1560 mg/sq m. The mean value for the entire series w^as 4.85 kcal/ash-free g. Richman (1958) and Golley (1961) reported similar values for Daphnia, but caloric values ob- tained for Copepoda and Cladocera by Sitaramiah (1967) were as high as 7.02 kcal/ash-free g. Paine (1964) indicated that higher values are usually found for eggs and resting stages, although well- fed zooplankton also show a high caloric content periodically. The ex- Mississippi Plankton Crop 313 treme value found in the mixed net plankton samples in the present analyses was 9.9 kcal/ash-free g. This extremely high result derives from an anomalous datum from one station on one date, and may re- flect a technical error. The plankton samples yielded high ash contents throughout the period of study. Paine (1964) suggested that in organisms high in ash content, such as sponges and opisthobranchs, at least some low caloric estimates may result from difficulties in determining accurate- ly the amount of noncombustible material present. Corrections for en- dothermic combustion reactions due to high ash content may be made as suggested by Paine (1966), but plankton samples may at times con- tain large amounts of low energy-yielding organic detritus, thus caus- ing low estimates. The w^et-ashing procedure employed in this investi- gation destroys carbonates and leads to low ash estimates, thus giving rise to slightly low results w'hen caloric values per unit weight are computed on an ash-free basis. Caloric content on a unit area basis followed a general trend simi- lar to that shown by other parameters. Thus, the highest caloric value for 1965, 1.7 kcal/sq m on April 20, corresponded to the highest value for dry w^eight in 1965. In 1966, highest values for all parameters ex- cept carbohydrate corresponded to highest standing dry weight. Al- though caloric data for winter and early .spring are not available, a trend towards decline in values w^as apparent after the peak in sum- mer 1965. This may be associated with seasonal decrease in tempera- ture. Vernal rise in caloric content w'as again noted in 1966, along with the rise in other parameters. The caloric value obtained in Au- gust 1966, 3.3 kcal/sq m, was the highest for the year. Other para- meters in 1966 showed similar trends. DISCUSSION Nannoplankton, by definition, is that portion of the total plank- ton not obtained in tow-net collections. Since it is difficult to collect in quantity and even more difficult to study by classical, systematics- oriented methods, nannoplankton, historically, has been ignored to a large extent in favor of net plankton. Nevertheless, various workers, approaching the assay of the total plankton from several standpoints, have all agreed that, at least in temperate and tropical w^aters, nannoplankton dominates the primary producer trophic level. Thus^ Pomeroy and Johannes (1966) found that 94-99% of total plankton respiration resulted from organisms too small to be retained by a fine net; Banse (1964) estimated that nannoplankton made up more than half of the total plankton; and Yentsch and Ryther (1959) reported that, in Vineyard Sound, net plankton represented, in various samples, 2-47% of total plankton 314 Mulkana and Abbott biomass. The latter authors also tabulated other earlier results, in- cluding- a report by Riley from Tortugas that net plankton amounted to only one percent of total plankton. The study described here relates to observations on the total plankton crop of Mississippi Sound. Turbulent conditions are the rule in this body as in other Gulf Coast estuaries and lagoons, because of shallow waters, extensive wind stirring, river discharge from the mainland, and tidal mixing from the Gulf of Mexico through various passes between barrier islands. As a result, clay particles and detritus stay in su.spension and become admixed with plankton samples, there- by giving rise to high ash values. Further, limited data Indicate that, probably because of fine clays, ash is differentially increased in the nannoplankton fraction. Despite this, nannoplankton, on a unit-area, total-water-column basis shows a total organic content greater by orders of magnitude than net plankton organic content. Thus, for an 18-month period in 1965 and 1966, mean nutritional composition of net plankton was 136 mg/sq m protein, as albumin, 15 mg/sq m carbohydrate, as glycogen, and 14 mg/sq m lipid, as triolein. During the same period, nanno- plankton samples from the same stations and times showed means of 2.0 g/sq m protein and 1.9 g/sq m carbohydrate. These values were associated with mean dry weights of 604 mg/sq m for net plankton and 36.2 g/sq m for nannoplankton, including ash contents of 49% for net plankton and, based on very limited samples, 65% for nanno- plankton. The net plankton nutritional values amount, on a dry weight basis, to 27% protein, 3% carbohydrate, and 3% lipid. These levels are considerably lower than published re.sults based on clay-free sam- ples or individual species (Parsons, Stephens and Strickland 1961, Linford 1965, Raymont and Linford 1966, Blazka 1966). For nanno- plankton, comparable data are not at hand. However, it has Vjeen pointed out by Gunter (1938, 1941, 1967) that most abundant species of fishes on the northern Gulf Coast feed at the ba.se of the food chain. These are the menhaden, Brevoortia patroniis and B, gzmteri, the an- chovy, Anchoa mitckilli^ and the mullet, Mugil cephalus. Quite prob- ably a considerable portion of their food is nannoplankton but there are no good data, although Peck (1894) said that menhaden filtered out dinoflagellates and minute plankton. There seems to be little doubt that oysters and other pelecypods feed to a considerable extent on nannoplankton (Nelson 1925, 1947) and filter feeding smaller crusta- ceans may also be in part based on nannoplankton. No geographically uniform seasonal trend was found for nanno- plankton standing crop, although net plankton did exhibit such a trend, a result in accord with the findings of Yentsch and Ryther (1959). Unfortunately, no method was available for estimating the Mississippi Plankton Crop 315 relative contribution of salt marsh detritus to the two plankton frac- tions. This detritus might either emphasize or mask a trend, since Odum and de la Cruz (1967) suggested that outpourings from Georgia salt marshes provided the major organic load in Georgia estuaries, and that the majority of this w'as “nanno detritus.'" Study on the year- ly cycle of nannoplankton carbohydrate in Louisiana estuaries by Mul- kana (1969), however, shows a normal seasonal trend associated with rise and decline of temperature. Net plankton caloric values, on an ash-free weight basis, were consistent with various reported values in the literature. On the aver- age, the caloric value of 1 g, ash-free, of net plankton was equivalent to 1.3 g glucose. On an area basi.s, caloric values were relatively steady. There was one extreme value of 3.3 kcal/sq m in August 1966, but the remainder fell in the range 0. 5-2.2 kcal/sq m. Nannoplankton samples were insufficient to permit caloric deter- minations, but should show larger values than net plankton per unit area consistent with the much larger standing crop of nannoplankton. SUMMARY 1. On a unit area basis, mean nannoplankton dry weight was 72 times that of net plankton from April 1965 to September 1966 in Missis.sippi Sound, 2. The mean net plankton protein, carbohydrate and lipid were found to be 136 mg/sq m, 15 mg/sq m and 14 mg/sq m, respectively. The mean nannoplankton protein and carbohydrate were, respectively, 2.0 g/sq m and 1.9 g/sq m. 3. Mean levels of net plankton protein, carbohydrate and lipid were 27, 3, and 2.8 7 / > respectively, of net plankton dry weight. In nan- noplankton, protein and carbohydrate were 5.5 and 5%, respec- tively, relative to dry weight. 4. Although nutrient levels were higher in net plankton on unit weight basis, total nutrients available from nannoplankton were substantially higher. The mean nannoplankton-to-net plankton protein and carbohydrate ratios per unit water column were 24:1 and 235:1, respectively. 5. Some seasonal trends were apparent as changes in the standing biomass of net plankton, associated with temperature rhythm. Nannoplankton exhibited no definite seasonal trends. Present re- sults suggest that factors other than temperature, such as grazing activity of herbivores and detritus influx, strongly influence nan- noplankton dry weight. 6. Caloric values of net plankton showed seasonal trends similar to 316 Mulkana and Abbott those shown by dry weight. Average net plankton caloric value during the study period was 1.3 kcal/sq m. LITERATURE CITED B'Mise, K. 1964. On the vertical distribution of zooplankton in the sea. hi: Prog- ress in Oceanography. Ed. by M, Sears. Perganion Press, New York, pp. 53- 12.5, Blazka, P. 1966. The ratio of crude protein, glycogen and fat in the individual steps of the production chain, hi: Hydrobiological Studies. Ed. by J. Hrbacek. Academia Pub. House, Czechoslovac Acad. Sci., pp. 395-408. Bligh, E. G., and W. J, Dyer. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, pp, 911-17, Christmas, J. Y,, G. Gunter and P. Musgrave. 1966. Studies on annual abundance of postlarval penaeid shrimp in the estuarine waters of Mississippi, as re- lated to subsequent commercial catches. Gulf Res. Rep. 2:177-212. Curl, H. Jr, 1962. Standing crops of carbon, nitrogen, and phosphorus and trans- fer between trophic levels in continental shelf waters south of New York. Rapp. P.— v. Reun Cons. perm. int. Explor, Mer 153, pp. 183-89. Department of the Army. 1951. Methods for Medical Laboratory Technicians. Manual No, TM8-227. Superintendent of Documents, U. S. Printing Office, Washington 25, D. C. Deevey, G. B. 1960. The zooplankton of the surface waters of the Delaware Bay region. Bull. Bingham occanogr. Coll. 17, pp. 6-63. Fisher, L. R, 1962, The total lipid material in some species of marine zooplankton. Rapp. P.-v. Reun Cons, perm, int. Explor. Mer 153, pp. 129-35. Folch, J., M. Lees and G. H. Sloane Stanley. 1957. A simple method for the isola- tion and purification of total lipids from animal tissues. J, Biol. Chem. 226, pp. 497-509. Golley, F. B. 1961, Energy values of ecological materials. Ecology 42, pp. 581-83. Gunter, G. 1938. The relative numbers of species of marine fish on the Louisiana coast. The American Naturalist, 72:77-83. . 1941. The relative number of shallow fishes of the northern Gulf of Mexico, with some records of rare fishes from the Texas coast. The American Midland Naturalist, 26(1) : 194-200. 1962. Shrimp landings and production of the State of Texas for the period 1956-1959, with a comparison with other Gulf .states. Publ. Inst. Ma- rine Sci., Univ. Texas. 8:216-26. 1963. The fertile fisheries crescent. J, Miss. Acad. Sci. 9:286-290. 1967. Some relationships of estuaries to the fisheries of the Gulf of Mexico. Part IX Fisheries, pp. 621-638, Estuaries, George H. Lauff, Ed., Publication No. 83, Amer. Assoc. Advan. Sci., Washington, D. C. 1967. Humphrey, G. F„ and M. Wootton. 1966. Comparison of the techniques used in the determination of phytoplankton pigments. Monographs on Oceanographic Methodology, UNESCO 1, pp. 37-63. Johannes, R. E. 1964. Phosphorus excretion and body size in marine animals: microzooplankton and nutrient regeneration. Science, N. Y. 146, 928-24. Landua, A, J., and J, Awapara. 1949. Use of modified ninhydrin reagent in quan- titative determination of amino acids by paper chromatography. Science, N. Y. 109, 385. Leong, R. 1967. Evaluation of a pump and reeled hose system for studying the vertical distribution of small plankon. U. S. Fish Wildl. Serv. Spec. Rept. Fisheries No, 545. Linford, E. 1965. Biochemical studies on marine zooplankton. II. Variations in the lipid content of some Mysidaeea. J. Cons. perm. int. Explor. Mer 30,: 16-27. Mississippi Plankton Crop 317 Mulkana, M. S. 1969. An annual nannoplankton carbohydrate cycle in Bartaria Bay, Louisiana, and its adjacent areas, J, La. Acad, Sci. 32:62-8. Nelson, T. C. 1925. The nannoplankton as source of the oyster’s food. Report; Agricultural College Experimental Station, Brunswick, New Jersey, 243^ 1947. Some contribution from the land in determining conditions of the life in the sea. Ecol. Monogr., 17:337-46. Odum, E. P., and A. de la Cruz. 1967. Particulate organic detritus in a Georgia salt marsh-estuarine ecosystem, AAAS Publ. 83, 383-8. Paine, R. T. 1964. Ash and caloric determinations of sponge and opisthobranch tissues. Ecology 45, 384-7. 1966, Endothermy in bomb calorimetry. Limnol. Oceanogr, 11, 126-9. Parsons, T. R., K, Stephens and J. D, Strickland. 1961. On the chemical composi- tion of eleven species of marine phytoplanktera. J, Fish. Res. Bd. Can. 18, 1002-15. Peck, J. I. 1894. On the food of the menhaden. Bull, U. S. Fish Commn. (1893) XIII, 113-126, PI. 1-8. Pomeroy, L. R., and R. E. Johannes. 1966. Total plankton respiration. Deep Sea Re.s. 13, 971-3. Raymont, J. E. G. 1963. Plankton and Productivity in the Oceans. The Macmillan Company, New York. , and R. J. Conover. 1961. Further investigation on the carbohydrate con- tent of marine zooplankton. Limnol, Oceanogr. 6, 154 64. , and E. Linford. 1966. A note on the biochemical composition of some Mediterranean zooplankton. Int, Kevue gcs. Hydrobiol. 51, 485-8, Richman, S. 1958. The transformation of energy by Daphnia pulex. Ecol. Monogr. 28, 273-91. Riley, G. A. 1937. The signiticance of the Mississippi River drainage for biologi- cal conditions in the northern Gulf of Mexico^ J. Marine Re.s. 1 :60-74. Sitaramiah, P. 1967. Water, nitrogen and calorific values of freshwater orga- nisms. J. Cons, perm. int. Explor. Mer 31, 27-30. Synder, F., and N, Stephens. 1959, A simplified spectrophotometric determination of ester groups in lipids. Biochem. biophys, Acta 34, 244-45. Thomas, W. H,, and E. G. Simmons, i960. Phytoplankton production in the Missis- sippi Delta, hi: Recent Sediments, Northwestern Gulf of Mexico. Symposium, Amer, Petrol. Inat, Project 51, 1951-8. Welcher, F. J. 1963. Standard methods of chemical analysis. Part A. 2, 920-2. Wimpenny, R. S. 1938, A routine method for the estimation of fat in plankton and its application to herring tissues. J. Cons. perm. int. Explor. Mer 13, 338-48. Yentsch, C. S., and J. H. Ryther. 1969. Relative significance of the net phyto- plankton and nannoplankton in the waters of Vineyard Sound. J. Cons. perm, int. Explor. Mer 24, 232-8.