MFR PAPER 1144 R. R. Colwell is with the Depart­ ment of Microbiology, University of Maryland, College Park, MD 20742. T. C. Wicks is with the De­ partment of Biology, Georgetown University, Washington, DC 20016. A Comparative Study of the H. S. Tubiash is with the Biological Laboratory, National Marine Bacterial Flora of the Hemolymph Fisheries Service, NOAA, Oxford, of Callinectes sapidus MD 21654.

The isolates from the holding tanks were obtained from animals maintained R. R. COLWELL, T. C. WICKS, and H. S. TUBIASH in the laboratory for more than I week. Crabs obtained from the fish market were held in a tank for I to 2 days before ABSTRACT-The bacterial flora of blue crabs (Callinectes sapidus) from sampling. Crabs from the Rhode River Chesapeake Bay was examined. Hemolymph of normal, healthy blue crabs was 5 in Chesapeake Bay were sampled im­ found non-sterile, with total bacterial counts ranging from <100 to >3.0 x 10 mediately after capture. Crabs from 5 on a seawater-based medium and 0 to >10 on Standard Methods Agar. Marumsco Bar were also sampled in the 3 Counts of 0-10 were observed on TCBS agar. Vibrio spp., including Vibrio field without delay after capture. The parahaemolyticus, were the major taxonomic groups found in the crab largest sample ofanimals examined was hemolymph. A comparison of crab data with that of Chesapeake Bay oysters a catch made offFranklin City, Va. The revealed a qualitative difference in that the crab hemolymph flora was almost hemolymph was sampled 12-16 h after entirely Vibrio spp., whereas the oyster flora included species of a variety of capture. other genera. Total viable counts of aerobic, heterotrophic bacteria in blue crab INTRODUCTION by various investigators (Society for hemolymph were carried out as follows. General Microbiology, 1962; Beers et After removing the fluid with a sterile The normal bacterial flora of blue a!., 1962; Silvestri eta!., 1962;Sokaland 25-gauge needle and syringe wetted crabs (Cal/inectes sapidus) from Sneath, 1963). Colwell and Liston with a sterile solution of 5 percent sodi­ Chesapeake Bay has not been exten­ (1961), in one of the earliest numerical um polyanethol sulfonate (G RO­ sively studied. Interest in the micro­ taxonomy studies, examined marine BAX)I,Z, an anticoagulant reportedly biology of shellfish, traditionally, has bacteria. Subsequently, several other effective for isolating bacteria from focused on human pathogens found in investigators examined the taxonomy of blood (Morello and Ellner, 1969), I ml commercially important species, as in marine bacteria (Floodgate and Hayes, of the hemolymph was withdrawn from the case of the blue crab. Many of the 1963; Hansen et a!., 1965; Liston et aI., each crab, either through the posterior papers published on the microflora of 1963; Pfister and Burkholder, 1965; hinge ofthe carapace or from the base of the blue crab have concerned pathogens Qualding and Colwell, 1964). Results of the coxa ofthe flippers. A dilution series rather than the normal microbial flora these studies proved helpful in identify­ was made, using a sterile diluent com­ present in the healthy animal (Couch, ing and classifying the bacterial flora posed offour salts and distilled water in 1967; Fishbein et a!., 1970; Krantz et from blue crab hemolymph. the concentration: NaCI, 24.0 g; KCI, a!., 1969; Rosen, 1967; Sawyer, 1969; 0.78 g; MgClz, 5.3 g; MgS04 • 7HzO, 7.0 Williams-Walls, 1968). In addition, in­ g; and distilled water, I liter. One-tenth formation has been gathered concerning MATERIALS AND METHODS ml of each dilution was inoculated onto the bacteriology of processing, han­ agar plates, in triplicate, and spread dling, packaging, or storage of crab­ Healthy adult blue crabs (Callinectes evenly with a sterile, bentglass rod. The meat (Loaharanu and Lopez, 1970; sapidus) weighing 150-230 g were ob­ three media used were: MSYE (Pro­ Ulmer, 1961; Ward and Tatro, 1970). tained from seafood markets or col­ teose peptone, I g; yeast extract, 1 g; This study was conducted to deter­ lected from several areas in Chesapeake agar, 15 g; four salts solution, as above; mine the total viable, aerobic, hetero­ Bay and Chincoteague Bay during the pH adjusted to 7.4-7.6 with 0.1 N trophic, bacterial flora of blue crab summer of 1970. Male and female crabs NaOH), SMA, the Standard Methods hemolymph in freshly caught, possessing hard carapaces ("soft-shell" Agar (Casein digest, 5.0 g; yeast ex­ aquarium-held, and market crabs during crabs undergoing ecdysis were not in­ tract, 2.5 glucose, 1.0 g; agar, 15 g; dis­ the summer months and to identify and cluded) were sampled to determine total tilled water, I liter; pH adjusted to 7.0 classify the bacterial strains isolated, numbers and types of aerobic, hetero­ with NaOH) and TCBS, the Thio­ using the methods of numerical tax­ trophic bacteria comprising their nor­ sulfate-citrate-bile salts-sucrose agar3 onomy. Numerical taxonomy as a mal microflora during the summer method for the analysis of taxonomic months when the water temperature I Roche Diagnostics, HotTman-LaRoche, Nutley, data for bacteria was first proposed by ranged between 23°-32°C. Sources of N.J. Sneath (1957) and has been used, with crabs, bacterial isolates, and dates of 'Reference to trade names does not imply en­ dorsement by lhe National Marine Fisheries Ser­ modifications of the original methods, sampling are given in Table I. vice, NOAA. 29 Table I.-Source and date of Isolation of bacterial methyl carbinol, and urease; production Table 2.-Bacterial counts of the hemolymph of crabs strains Included In the numerics' taxonomy analysis. collected at several locations on the media employed of ammonia from peptone, NOz, and in the study. Counts given are per milliliter hemolymph Strain number Date of isolation Source' of Individual crabs. Figures given for each sample set ; N0 3 hydrolytic activity on starch, are mean values, except where otherwise indicated. 1-10 7/15f70 HT gelatin and casein; methyl red test; and 11-27 7119170 FM No. 28-33 7/21/70 RR utilization of alanine, proline, glutamic Source sampled MYSE SMA TCBS 34-39 8/22170 MB acid or methionine as sole carbon and 40 8/27/70 FC Laboratory 41 8/27170 FC nitrogen sources. holding tank 3 1.8 x lOS 3 X 10' 10 42 8/27170 FC Drug sensitivity of each pure culture Fish market 2 6 x 1()4 4 x 103 1 43 8/27/70 FC Rhode River 6 2.1 x lOS 60 0 44 8/27/70 FC was determined using penicillin, 10 Marumsco Bar 4 1.3 x lOS 2 X 105 100 45 8/27170 FC units; Chloromycetin, 30 J.1g; tetracy­ Franklin 46 8/27170 FC Cily. Va. 33 3.1 x 10' 2 X 10' 33 47 8/27170 FC cline, 30 J.1g; dihydrostreptomycin, 10 Total range 48 -:100--·3.0 x 10' (}_>105 0-10' 48 8/27/70 FC J.1g; and colimycin, 10 )J.g5. Sensitivity 49 8/27170 FC was recorded without quantification of colony type appearing on TCBS agar 'HT ~ Holding tank. Georgetown University: FM ~ Fish market. District of Columbia; MB = Marumsco Bar. Md. the diameter ofthe inhibition zone. Sen­ was yellow, 2 mm in diameter, and simi­ (Chesapeake Bay): RR = Rhode River, Md. (Chesapeake sitivity to the pteridine compound lar to Vibrio alginolyticus. Results of Bay): FC = Franklin City, Va. (Chincoteague Bay). (0/129) was also recorded (Shewan et total viable counts are given in Table 2. (yeast extract, 5.0 g; polypeptone pep­ a!., 1954). No change in counts was observed be­ tone, 10 g; sodium citrate, 10.0 g; All media, with the exception of the tween 48 hand 7 days. sodium thiosulfate, 10.0 g; oxgall, 5.0 g; blood agar, were prepared with the salt The strain clusters obtained from the sodium cholate, 3.0 g; sucrose, 20.0 g; solution, described above, as diluent. In computer analysis of the taxonomic NaCI, 10.0 g; iron citrate, 1.0 g; thymol most cases, dehydrated medium was data are shown in Figure 1. All strains blue, 0.04 g; bromthymol blue, 0.04 g; mixed directly with the salt solution as grouped atS ;;3 57 percent and the major agar, 14.0 g; distilled water, I liter; pH diluent, but some media required spe­ phenons detected are shown in Figure 8.6). The inoculated media were incu­ cial preparation. Media for the casein I, A through 1. Strains clustered in Phe­ bated at 25°C for 48 h, after which hydrolysis, methyl red, Voges­ non A were not found to be closely re­ counts were made and the media again Proskauer, and carbohydrate utilization lated, indicated by the low similarity incubated for an additional 5 days for a tests were prepared double strength in values (62-64 percent). Phenon B con­ repeat count. distilled water, and a double strength, sisted of strains more closely related to Forty-nine cultures were isolated, sterile salt solution was added (I: I each other (5 = 72-74 percent). Phenon purified, and subjected to the testing volJvol) after sterilization. C comprised the largest group, consist­ procedure used for numerical tax­ A total of 126 features used in the ing of strains 41 through II (Figure I) onomy. Colonies were selected ran­ computer analysis were scored, with clustering atS ;;3 75 percent. Three sub­ domly and were streaked three times to the code: 0 (negative), I (positive), or 3 groups (I, II, III) were observed in ensure purity of the cultures. After (not tested or not applicable). The Phenon C (Figure 1). purification, all cultures were tested for coded data were entered for analysis in Phenon D, strain 9-24 (Figure I) ability to grow on a medium (MSYE) an IBM 360/40 system equipped with formed at 5 values ;;3 71-75 percent, in­ containing the major ionic constituents discs and magnetic tape drives. dicating relatively high relationship of seawater. All cultures isolated from Georgetown University Taxonomy among the strains. Phenon E formed at the SMA and TCBS media also grew on Programs (GTP-I, 2, 4, and 5) were 72 percent 5 and was considered dis­ MSYE and were, therefore, maintained used in the analysis (Colwell, 1964; tinct from Phenon D. Phenons F and G on MSYE. Colwell and Liston, 1961; Moffett and formed at relatively low 5 values, 65-67 Taxonomic tests were carried out on Colwell, (968). The programs have percent 5 and 68 percent 5, respec­ each pure culture for morphology and been documented for the IBM Compu­ tively. Phenon H (intra 5 values of motility, using wet-mount preparations ter Users Library. 62-63 percent) and Phenon I completed of 24-h broth cultures under phase con­ the groupings obtained in the analysis. trast; gram stain; growth characteristics RESULTS on MSYE agar after 48 h incubation at A wide range in bacterial counts for DISCUSSION 25°C; growth at 4, 15,25,37, and 45°C; the blue crab hemolymph was ob­ growth at NaCI concentrations of 0.0, Counts on MSYE were comparable served. On the MSYE medium, counts among sampling locations. Greater 3.0,7.0, and 10.0 percent; growth at pH ranged from < 100/ml (five animals) to variations were noted for counts on 4.0,5.0,6.0,7.0,8.0, and 9.0; hemolysis >3.0 x 105/ml(fiveanimals). Counts on SMA, with the lowest counts on SMA of human blood; utilization of citrate; SMA were lower, O/ml (II animals) to aerobic and anaerobic utilization ofglu­ noted for Rhode River animals. Crabs > 105 mJ (two animals), indicating that cose and sucrose, employing MOF from Marumsco Bar gave highest many of the bacteria found in the counts on SMA. The salinity of the medium 4 (Leifson, 1963); utilization of hemolymph require the major salts galactose, mannitol, ribose, lactose, water at Rhode River was lower than at found in seawater for growth. On mannose; production of catalase, ox­ Marumsco Bar and both the Rhode TCBS, the counts were from 0 (14 ani­ idase (Kovacs, 1956), indole, acetyl River and Marumsco Bar crabs were mals) to 103 (3 animals). The dominant sampled immediately after capture. 3Baltimore Biological Laboratories, Baltimore, Hence, no correlation was noted be­ Md. 'Baltimore Biological Laboratory, Bioquest Divi­ 'Difco Laboratories, Detroit, Mich. sion of Becton-Dickinson. Inc., Sensi-Disks. tween time of sampling after capture 30 100

95

90

85 r f- a::: 80

~ 0

70

65

60 A

55 36 48 51 17 37 47 34 42 30 16 29 4 32 43 5 10 14 9 3 25 26 27 22 13 20 45 12 8 18 41 49 50 36 28 46 2 6 31 33 44 7 I II 39 40 24 23 15 21 19 35

Flgure 1.-Sorted outputlrom the taxonomic analy.l. by computer. Groups were formed by highest linkage sorting. and the salinity and the number of bac­ tant to dihydrostreptomycin; being duced acid in sucrose, galactose, man­ teria in crab hemolymph detected on oxidase positive, indole negative, and nose, mannitol, and ribose and were lac­ SMA, a medium employed for detecting citrate positive; utilizing alanine and tose negative. All strains hydrolyzed

nonmarine bacteria. Also, no correla­ proline as sole carbon and nitrogen starch, produced NH3 from peptone, tion between sex and total viable count sources; growing in 7-10 percent NaCl; reduced nitrates, produced catalase, on SMA was detected. However, it was and were classified as Bacillus sp. hydrolyzed casein, utilized citrate as a clear that the proportion of nonmarine Phenon B included five strains, one of sole carbon source, were oxidase posi­ bacteria in crab hemolymph was higher which was the reference strain of Vibrio tive and methyl red positive. None of for fish market and Marumsco Bar parahaemolyticus (strain Sak-3). the strains of Phenon C produced crabs. The holding tank and Franklin Phenon C was the largest group, with urease or acetylmethylcarbinol, but City samples exhibited approximately 27 strains comprising the cluster. All of half produced indole. All were sensitive the same proportion of nonmarine bac­ the strains of Phenon C were round to Chloromycetin; 64 percent were sen­ teria (bacteria capable of growth on ended, motile, gram-negative rods, sitive to dihydrostreptomycin; 36 per­ SMA). 1.3-3.0 Mm long x 0.5-1.3 Mm wide and cent to colimycin, and 16 percent to The genera of bacteria found in crab occurring singly, not as pairs or chains. penicillin. Phenon C was classified as hemolymph were obtained from the These strains grew at pH 4.0-9.0 and at Vibrio spp. taxonomic data analyses, and each of temperatures of 15°-45°C, with half of Phenon 0 strains were straight, the eight phenons were classified. the isolates capable ofgrowth at4°C and gram-negative rods, 1.3-3.0 Mm long, Strains 38 and 12 (Phenon A) were in 3-10 percent NaCI. However, 92 per­ occurring singly and in short chains gram-positive, motile rods, 1.3-3.0 Mm cent of the isolates were capable of (less than five cells). Colonies 2-5 mm in long; capable of utilizing glucose and growth without added NaCI. The diameter and a translucent yellow or sucrose aerobically and anaerobically; strains formed convex, entire, translu­ white were formed on agar. Glucose growing at temperatures of 4°_37°C and cent colonies yellow to white in color on and sucrose were utilized aerobically pH 7.0-9.0; sensitive to penicillin, MYSE agar. All fermented glucose and anaerobically, without gas. Acid Chloromycetin, and colimycin; resis- anaerogenically and, in general, pro- was produced from mannose; nitrate 31 was reduced and NH3 produced from of the animals proved to be interesting. members of which did not predominate peptone. All of the strains of Phenon 0 The holding tank isolates were all crab hemolymph. The results suggest grew in 3-10 percent NaCI and most Vibrio spp., including phenons B, C that the Phenon 0 Vibrio is more com­ strains utilized citrate and were indole (subgroups I, II and III), and O. The mon in oysters than in crabs, with the and methyl red negative. Phenon 0 fish market isolates were Vibrio spp., phenon C Vibrio being more common in strains were catalase positive and hy­ including the B, C, and 0 phenons. crabs. Further support for this observa­ drolyzed casein. This phenon was also However, representatives of the other tion can be drawn from the observations identified as comprising members ofthe genera were also isolated from the fish of Lovelace et al. (1968) who did a bac­ genus Vibrio. market samples. Rhode River isolates teriological analysis ofoysters collected Phenon E consisted of two cultures, were all identified as Vibrio spp. (Phe­ from the Marumsco Bar and Eastern strains 23 and 26, clustering at 72 per­ non C, subgroups I and II). The Bay areas ofChesapeake Bay and found cent 5. These strains were gram­ Marumsco Bar isolates included phe­ a predominance of Vibrio, Pseudomon­ positive, motile rods, occurring singly, nons A, C, 0, and H (Vibrio, Bacillus, as, Achromobacter spp., in that order, 1.3-3.0 11m long x 0.7-1.5 11m wide. and unidentified strains). The Franklin with a smaller percentage of Coryne­ Both strains grew at 15°_45°C and fer­ City isolates were predominantly of bacterium, Cytophaga-Flavobac­ mented glucose and sucrose and pro­ Phenon C (subgroups I and II) with terium, Micrococcus-Bacillus, and duced acid from mannose. They did not some Band 0 phenons also observed to enteric species. Pseudomonas, A chro­ hydrolyze casein and were oxidase and comprise the flora. Thus, the common mobacter and Corynebacterium spp. catalase negative. They were sensitive feature of the crab hemolymph was the were not isolated from crab hemolymph to penicillin, chloromycetin, tetracy­ preponderance of Vibrio spp., particu­ in this study and only a few Flavobac­ cline, and colimycin, but not to dihy­ larly Phenon C vibrios which were pres­ terium, Bacillus, and other unidenti­ drostreptomycin. The strains did not ent in crab hemolymph from all loca­ fied species were noted. However, the produce indole, were methyl red nega­ tions. A significant difference observed number ofstrains included in this study tive and citrate positive, did not pro­ between the Marumsco Bar and fish was relatively small. Nevertheless, the duce urease, and grew in 7-10 percent market crabs and crabs from other loca­ indication is that the composition ofthe NaC!. Phenon E was identified as a tions in Chesapeake Bay was the pres­ microbial flora ofthe blue crab does not Bacillus sp. ence of Bacillus, Acinetobacter, reflect geographical differences so The two strains, 15 and 27 (Phenon F) Flavobacterium, and other unidentified strongly nor is so varied as that of the were identified as A cinetobacter sp. bacteria in the hemolymph. oyster (Lovelace et aI., 1968). based on their inability to utilize the The analysis was carried further in In summary, the total number of via­ carbohydrates tested, their oxidase that the data for the 49 strains of this ble, aerobic, heterotrophic bacteria negativity, and catalase positivity. study were combined with data ob­ found in hemolymph of healthy blue The distinguishing features of Phe­ tained from 181 other isolates from crabs is large and variable. Analysis of non G (strains 21 and 22) which clus­ Chesapeake Bay blue crabs (unpub­ taxonomic data by computer revealed tered at 68 percent S were variable gram lished data). The data for the 230 strains the predominant genus present in the stain, rod shape and yellow colonies. were subjected to computer analysis hemolymph to be Vibrio, with members The strains fermented glucose and su­ and the results showed 26 clusters at of Bacillus, Acinetobacter and crose, but did not reduce nitrates )lor 5;361 percent with the largest cluster Flavobacterium present in much small­ hydrolyze casein. They were oxidase containing 68 strains. The 49 strains er numbers. Clearly the hemolymph of positive, sensitive to dihydrostrep­ from the study reported here showed 22 most healthy blue crabs is not sterile. tomycin, indole and methyl red nega­ strains clustered in the large, 68 strain tive, citrate positive, urease negative phenon, identified as the genus Vibrio. ACKNOWLEDGMENTS and grew in the presence of 0-10 per­ Thus, the results of the extended tax­ cent NaCI. Phenon G was identified as onomic analysis substantiates the ob­ This study was supported by Sea a Flavobacterium sp. servation that the predominant bacteria Grant Project #04-3-158-7, National Phenons H and I were not identified. in the hemolymph of blue crabs are Oceanic and Atmospheric Administra­ They were motile, gram-negative rods Vibrio spp. tion, and Grant #GB-35261 X from the which produced yellow to white col­ An additional comparison was made National Science Foundation. The au­ onies on MSYE agar at 15°_37°C and at of the data for the 49 crab isolates with thors wish to acknowledge the helpful pH 7.0-9.0. Strains in Phenons H and I data for 161 strains isolated from the advice and kind cooperation of Aaron fermented glucose and sucrose. The Eastern oyster, Crassostrea virginica, Rosenfield, National Marine Fisheries other characteristics tested varied which is found in Chesapeake Bay (un­ Service, NOAA, Oxford, Md. among the strains of the two groups. published data). The entire set of data In summary, the isolates from the for the 210 strains was analyzed by LITERATURE CITED hemolymph ofthe blue crab belonged to computer and it was found that the crab Beers, R. J., J. Fisher, S. MeGraw, and W. R. the genera Vibrio, A cinetobacter, isolates, for the most part, did not clus­ Lockhart. 1%2. A comparison of methods for computer taxonomy. J. Gen. MicrobioI. Flavobacterium, and Bacillus. The ter with and were distinct from the oys­ 28:641-652. largest number of isolates, 38 of the 49 ter strain clusters. Seven of the crab Colwell, R. R. 1964. A study of features used in the diagnosis of Pseudomonas aeruginosa. J. isolates examined, were identified as strains did cluster with oyster strains at Gen. Microbio!. 37: 181-194. Vibrio spp. 5 ;361 percent. These seven strains Colwell, R. R., and J. Liston. 1961. Taxonomic relationships among the pseudomonads. J. Bac­ A taxonomic distribution by source were largely of one phenon, Phenon 0, terio!. 82: 1-14. 32 Couch, J. A. 1967. A new species of Loaharanu, P., and A. Lopez. 1970. Bacte­ Shewan, J., W. Hodgkiss, and J. Liston. 1954. A Lagenophrys (Ciliata: Peritrichida Lageno­ riological and shelf-life characteristics ofcanned, method for the rapid differentiation of certain phryidae) from a marine crab, Cal/inectes pasteurized crab cake mix. Appl. Microbiol. non-pathogenic. asporogenous baciUi. Nature sapidus. Trans. Am. Microsc. Soc. 86:204-211. 19:734-741. (Lond.) 173:208-209. Fishbein, M., I. J. Mehlman, and J. Pitcher. Lovelace, T. E., H. TUbiash, and R. R. Colwell. Silvestri, L, M. Turri, L R. Hill, and E. Gilardi. 1970. Isolation of Vibrio porahaemolyticus 1968. Quantitative and qualitative commensal 1962. A quantitative approach to the systemat­ from the processed meat ofChesapeake Bay blue bacterial "ora of Crassostrea virginica in ics ofactinomycetes based on overall similarity. crabs. Appl. Microbiol. 20: 176-178. Chesapeake Bay. Proc. Natl. Shellfish. Assoc. In Microbial Classification. Twelfth Symp. Soc. Floodgate, G. D., and P. R. Hayes. 1%3. The 58:82-87. Gen. Microbiol. 12:333-360. Adansonian taxonomy of some yellow pig­ Moffett, M. L, and R. R. Colwell. 1%8. Sneath, P. H. A. 1957. The application of com­ mented marine bacteria. J. Gen. Microbiol. Adansonian analysis of the Rhizobia­ puters to taxonomy. J. Gen. Microbiol. 30:237-244. ceae. J. Gen. Microbiol. 51:245-266. 17:201-226. Hansen, A. J., O. B. Weeks, and R. R. Colwell. Morello, J. A., and P. D. Ellner. 1969. New Society for General Microbiology. 1962. 1965. Taxonomy of Pseudomonas piscicida medium for blood cultures. Appl. Microbiol. Microbial classification. Twelfth Symp. Soc. (Bein) Buck, Meyers, Liefson. r Bacteriol. 17:68-70. Gen. Microbiol., Univ. Press, Cambridge, 483 89:752-761. Pfister, R. M., and P. R. Burkholder. p. Kovacs, N. 1956. Identification of Pseudomonas 1%5. Numerical taxonomy of some bacteria Sokal, R. R., and P. H. A. Sneath. 1%3. pyocyanea by the oxidase reaction. Nature isolated from antarctic and tropical seawaters. J. Principles of numerical taxonomy. W. H. (Lond.) 178:703. Bacteriol. 90:863-872. Freeman and Co., San Francisco, 359 p. Krantz, G. E., R. R. Colwell, and E. Lovelace. Quadling, c., and R. R. Colwell. 1%4. Theuseof Ulmer, D. H. B.,Jr. 1961. Preparationofchilled 1969. Vibrio parahaemolyticus from the blue numerical methods in characterizing unknown meat from Atlantic blue crab. U.S. Fish. Wild!. crab Callinectes sapidus in Chesapeake Bay. isolates. Dev. Ind. Microbiol. 5: 151-161. Serv., Fish. Ind. Res. 2:21-45. Science (Wash., D.C.) 164: 1286-1287. Rosen, B. 1967. Shell disease of the blue crab, Ward, J. F., and M. C. Tatro. 1970. Bacte­ Leifson, E. 1963. Determination ofcarbohydrate Cal/inectes sapidus. J. Invertebr. Pathol. riological examination of whole crab samples. metabolism of marine bacteria. J. Bacteriol. 9:348-353. Nat. Resour. Inst., Univ. Md., Contrib. 423, p. 85: 1183-1184. Sawyer, T. K. 1%9. Preliminary study on the 195-196. . Liston, J., W. Wiebe, and R. R. Colwell. epizootiology and host-parasite relationship of Williams-Walls, N. J. 1968. Clostridium 1963. Quantitative approach to the study of Paramoeba sp. in the blue crab, Callinectes botulinum Type F: Isolation from crabs. Science bacterial species. J. Bacteriol. 85: 1061-1070. sapidus. Proc. Natl. Shellfish. Assoc. 59:6Q..64. (Wash., D.C.) 162:375-376.

MFR Paper 1144. From Marine Fisherie,> Review, Vol. 37, Nos. 5~, May.,}une 1975. Copies of this paper, in limited numbers, are available from 083, Technical Information Division, Environmental Science Infor­ mation Center, NOAA, Washington, DC 20235. Copies of Marine Fisheries Review are available from the Superintendent of Documents, U.S. Gov­ ernment Printing Office, WaShington, DC 20402 for $1.10 each.

33