ECTOPARASITES OF STRIPED MULLET (Mugil cephalus L. ) IN BRACKISH AND FRESHWATER HABITATS By MARK R. COLLINS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1984 ACKNOWLEDGEMENTS Appreciation is expressed to Dr. Ju-shey Ho, University of California at Long Beach; Dr. Roger Cressey, U.S. National Museum; and Dr. Robin Overstreet, Gulf Coast Research Laboratory, for pro- viding identification of parasites. The University of Florida Marine Laboratory provided support in the forms of boats, nets, and travel to scientific meetings. Dr. Claus Buergelt, Veterinary Medicine, University of Florida, arranged for and interpreted histological preparations. Dr. Frank Mature, Jr., and Dr. Frank Nordlie, University of Florida Department of Zoology; Dr. William Seaman, Jr., Florida Sea Grant, University of Florida; and Dr. Ellis Greiner, Veterinary Medicine, University of Florida, pro- vided advice and logistic support. The author is especially grateful to Dr. Carmine Lanciani, University of Florida Department of Zoology, for advice and for comments on the manuscript. TABLE OF CONTENTS Page ACKNOWLEDGEMENTS li ABSTRACT v INTRODUCTION 1 METHODS AND MATERIALS 3 Collection and Maintenance of Fish Samples 3 Laboratory Analysis. . 7 Statistical Tests 10 Identification of Parasites 10 Study Sites 11 Experimental Equipment 12 RESULTS 14 Non-Experimental Mullet 14 Phylum Arthropoda 14 Ergasilus versicolor 14 Ergasilus lizae. - 20 Bomolochus nitidus 27 Naobranchia lizae 27 Rarer species 30 Argulus spp 30 Phylum Platyhelminthes 31 Phylum Annelida 31 Myzobdella lugubris 31 Malmiana philotherma (?) 31 Phylum Nematoda 34 Experimental Mullet 34 SK Mullet Transported and Acclimated Only 38 SK Mullet Caged at SK 38 SK Mullet in Indoor Tank of CR Water 40 SK Mullet in CR Cages 40 CR Mullet Sacrificed and Held in Aerated Freshwater . . 47 CR Mullet Caged at CR 47 CR Mullet Caged at SK 50 CR Mullet in Indoor Tank of Brackish (30 ppt.) Water. . 50 iii Page DISCUSSION 53 Experimental Equipment and Techniques 53 The Parasites 55 Ergasilus spp 55 E^. versicolor 56 E^. lizae 60 Bomolochus nitidus . ... 62 Naobranchia lizae 63 Argulus spp 64 Rarer Species 64 Metamicrocotyla macracantha 65 Myzobdella lugubris 65 Malmiana philotherma (?) 66 Contracaecum sp. (?) 66 Implications and Conclusions 68 SUMMARY 71 LITERATURE CITED 73 BIOGRAPHICAL SKETCH 76 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ECTOPARASITES OF STRIPED MULLET (Mugil cephalus L. ) IN BRACKISH AND FRESHWATER HABITATS by Mark R. Collins December, 1984 Chairman: Carmine A. Lanciani Major Department: Zoology Mugil cephalus (striped mullet) from freshwater (Crystal River) and brackish (Seahorse Key) sites in Florida were examined for gill and body surface parasites. The parasite fauna of the freshwater mullet consisted almost entirely of the copepod Ergasilus versicolor and the leech Myzobdella lugubris. About 32% of E^. versicolor were located at the distal tips of the gill filaments, and both prevalence and intensity were negatively correlated with fish length. The remainder were located proximally between hemibranchs, apparently the first time use of this attachment site has been reported, and intensity was positively correlated with fish length. Intensity of M. lugubris was negatively correlated with fish length. The fauna of brackish water mullet was more diverse. The copepods E. versicolor , E^. lizae , Bomolochus nitidus, and Naobranchia lizae ; adult and immature Argulus spp.; and the monogenetic trematode Met amicrocotyla macracantha were commonly encountered. The leech Malmiana philotherma (?) was found irregularly in the nostrils of mullet, and several additional species of copepods were rarely col- lected. Ergasilus versicolor were found at the diatal tips of gill V filaments, not between hemibranchs. Intensity of E^. versicolor was positively correlated and intensity of E. lizae was negatively cor- related with host length. Subsamples of mullet from both study sites were examined for nematodes (Contracaecum sp. ?). Prevalence and intensity were sig- nificantly greater at the brackish site, although definitive hosts (piscivorous birds) were common at both locations. Intensity was positively correlated with host length at the brackish but not the freshwater location. The kidney was the preferred site of infection at both locations. Experimental translocations of mullet demonstrated that E^. versicolor does not move from filament tips to the proximal area between hemibranchs or vice versa as the salinity changes. At each study site, E^. versicolor apparently attaches at the location typical of that site /salinity and remains there through salinity changes. Each parasite species responded differently to salinity changes en- countered when held at the alternate study site. Differences between study sites in ectoparasite diversity, attachment sites of E. versicolor, infection values of Contracaecum sp. (?), sex ratios, and length distributions suggest that in Crystal River a group of larger mullet, primarily females, may reside semi- permanently. There seems to be little intermingling of even smaller mullet between brackish and freshwater areas except during the spawning season. The results of this study may be applicable to mariculture and to mullet management policies. INTRODUCTION Mugil cephalus L. (striped mullet) is one of the most widespread and abundant inshore teleosts (Odum, 1970), occurring in the coastal waters of all seas between 42°N and 42°S (Hoese and Moore, 1977) and in salinities from 0 ppt. (parts per thousand) (Collins, 1981) to 75 ppt. (Simmons, 1957). It supports a valuable commercial fishery in the southeastern U.S., especially in Florida where over 30 million pounds were landed in 1981 (Collins, in press). In many parts of the world this species is important in mariculture (Paperna and Overstreet, 1981; Tang, 1970). Striped mullet spawn at sea, and while they are not truly catadromous, adults do regularly ascend freshwater rivers (Shireman, 1975; Collins, 1981). How long they remain in freshwater habitats is not known. Tagging studies in Flo: ida have concluded that only three lossely knit populations of striped mullet exist along the gulf coast. Also, conversations with commercial mullet fishermen indicate that most believe mullet make only brief excursions into freshwater, soon returning to intermingle with the mullet in salt (brackish) water. However, some information suggests that groups of striped mullet inhabiting freshwater areas may be at least partially isolated because they exhibit different size distributions and sex ratios than fish collected in brackish water (Shireman, 1975; Collins, 1981). „1_ -2- While striped mullet have been included as a host in many general surveys of fish parasites in the U.S. (Wilson, 1936; Pearse, 1952; Causey, 1955), detailed knowledge of their parasite fauna is still limited to a few regions despite the world-wide distribution of M. cephalus. Much of the information has been derived from captive fish in culture operations (Paperna and Overstreet, 1981). In the U.S., works concerning the ecology of striped mullet and their para- sites are very limited, the most notable being studies by Rawson (1973) and Skinner (1975), neither of which dealt with freshwater habitats, and a review by Paperna and Overstreet (1981). Most other reports of the ectoparasites of M. cephalus in the U.S. have been only taxonomic records from relatively small samples. The paucity of ecologically oriented studies of striped mullet parasites is unfortxmate. Such studies are imperative in the preven- tion of parasite epizootics in mono- and polycultural operations involving mullet (Rawson, 1973) and may be valuable in forming mullet management policies since parasites may be used as indicators of diet and migratory behavior (Paperna and Overst reet , 1981). In addition, migratory fishes and their parasites provide excellent opportunities for ecologists to study the effects of changing environments on organisms, especially in anadromous species (Noble, 1960). The present study was an investigation into the occurrence and ecology of the ectoparasites of striped mullet and how the parasite fauna differs between freshwater and brackish habitats. Differences in parasite fauna were then analyzed to provide inferences regarding the length of time mullet remain in a freshwater habitat. MATERIALS AND METHODS Collection and Maintenance of Fish Samples Striped mullet were collected at Seahorse Key (SK) and Crystal River (CR) , Florida, at irregular intervals (Tables 1-3) from July to October 1982 and February to November 1983 with castnet (2.54 cm. square mesh) and trammel net (3.18 cm. square mesh gillnet with 15.24 cm. square mesh double walls) . Attempts to catch at least ten mullet per sample were not always successful. In 1982 all mullet were immediately placed in plastic bags and held on ice until necropsy. In 1983 usually at least five mullet per sample were put on ice immediately as "controls" while the rest, designated "experimental fish," were held alive in covered, aerated tanks (61 x 46 x 46 cm.) aboard the boat. In experi- ments designated to monitor the changes in parasite fauna of mullet introduced to a different habitat, some of these experimental mullet were placed in cages at the location where they were caught. Others were transported by road to the other study site, acclimated to the new salinity (0 ppt. at CR, 22-31 ppt. at KS) , and placed in cages. In addition, a few mullet from both sites were sacrificed immediately after transport and acclimation. Salinities in the transport tanks were adjusted by adding water from the study site at which the mullet were to be acclimated. This was accomplished by allowing water from a 6 1. container to trickle into o CO u o 0) CO cd u as a u s 3 O u c 0) c o 6 e o u CO cu CO c •H o ^^ 0) -u J3 CO e 43 H 03 ' — i CO i-( ■H > 0) in I I u o rH o O ■H CO ^ > CD 3 ■-! •H 03 CD 3 ^ •H CO CD 00 U td CO J= o c l-l Xl •H E OJ J= c (1) 01 3 u 0) 43 o <• VO O CJ> u-1 VD 3- -3- O cn CD •H Q o to 00 1 CM 1 CN CN ON CJN 1 c 1 CSI 1 vD o 1 o 1 1 C O CN CM CM 1 r^i 1 m OO 00 00 00 00 I vO O I OO <• cn C^J 1 1 1 o VO CO On O 1 O 1 o O m 1 1 00 00 00 CO -5- CO J 1 CO fi4 5D M rrl Cu J_J 'H CO (-* G QJ H rC Jn4 (Tl lU w rri Tl n\ s H 01 (J r" w fii QJ 4J CO 1 [„j (-" M ■H 1 CO Cj *H n 1 1 [ 1 1^ \J Q /-^ QJ II r-* GJ j_j +j r-< /-\ w 'H •rH CJ Q) td rH t~l CO O 4-J O CO 01 X) (U (U 4-1 CO CO 1-1 oooo<»cocx)aocDcooo -6- Table 3. Number of infested hosts and intensities of leeches and monogenetic trematodes for each Seahorse Key collection. Date // Fish Metamicrocotvla macracantha Malmiana philotherma (? # Infested Intensity // Infested Intensity 82-07-18 8 3 1.0 82-09-03 9 2 1.5 82-09-17 •• 11 6 2.8 82-09-29 10 4 2,3 - 82-10-17 5 3 2.3 82-10-20 9 4 2.3 5 1.0 83-04-30 (a) 10 5 2.0 0 83-05-14 2 2 10.0 0 83-06-10 10 2 1.5 4 3.0 83-07-08 6 1 1.0 1 1.0 83-07-25 10 2 1.0 1 1.0 83-08-09 5 1 1.0 0 83-10-17 5 2 1.0 1 1.0 83-11-02 5 1 1.0 0 (a) 1 fish with 5 M. lugubris . -7- the tank at a rate of 6 1. per 25 min. with a pause of from 10 to 60 min. each time the salinity in the tank changed by 4 ppt. as deter- mined by ref ractometer . Total acclimation times of 4, 8, or 12 hrs. were used to determine if the rate of salinity change affected parasites or hosts. Caged fish were held from one to forty days. Additionally, a similar series of experiments lasting up to 26 days was run using an indoor tank with CR mullet held in filtered sea water (30 ppt.) and SK mullet in filtered CR water (0 ppt.). These experiments eliminated reinf estation as a factor while addressing the effects of salinity changes on parasite behavior. Also, an experiment was carried out to determine whether certain parasites abandon dead hosts, in which 5 dead CR mullet were held in an aerated aquarium for 6 hrs. and the distribu- tion of their ectoparasites was compared to that of control fish that were necropsied immediately after death. Laboratory Analysis In the laboratory standard length (SL) was measured, gills from both sides were removed and preserved in 70% EtOH, and the body sur- face, mouth, and inner opercle surfaces (and nostrils of most fish) were examined under magnification. Subsamples of non-experimental mullet were chosen at random to examine for nematodes in the liver and kidney. The preserved gills were inspected with the aid of a dis- secting microscope. The sites of attached parasites were noted by subdividing the gills (Figure 1). Three dorso-ventral sections (dorsal, medial, ventral) and three filament areas (proximal, medial, distal) were designated for each of three major regions (lateral 00 t CO 0) 0) iH c •H U 1-: to iH C O i rH CO c o O •H 4J C O o 01 •H CO 1 « CO CO c o 0) u tn IS a 0) •H CO D- > (U 1-1 o o u •H 0) CO CO e M CO M " M JZ CO O a CO N •H CO CO 0) E •H U CN ■H <; cu r-l CO n 2 CO 3 0 M 00 -H O -H X n o l-l 3- 3 00 •H -10- surface, medial surface, and between hemibranchs) for a total of 28 sections per arch when the gill rakers are also included. In order to determine the existence of a pathological response to erga- silid parasites, gill filaments were sectioned and stained with the standard hematoxylin and eosin. Histological preparations were made and evaluated by Dr. Claus Buergelt (Univ. of Florida). Statistical Tests All data were recorded on computer for analysis, and all statis- tical tests were non-parametric (Siegel, 1956). The Spearman Rank test was used to test for correlations of parasite intensity (number per parasitized fish) and prevalence (percent of parasitized fish in the sample) with host length and with the salinity at SK on the date of collection. Contingency Chi-square tests were employed to determine the significance of attachment site preferences by the parasites. KoJmogorov-Smirnov and Mann-Whitney tests were used in comparisons of intensity and prevalence between study sites and between experimental fish and controls, respectively. For all statistical tests, if p were greater than .05 the difference is stated to be not significant. Pooling of experimental fish was often necessary due to small sample sizes . Identification of Parasites Identification of many fish parasites, even to genus, can be difficult, especially considering the frequent taxonomic changes that -11- occur. Several keys and descriptions were utilized including Sawyer at al. (1975) for leeches, and Schell (1970) and Hargis (1956) for monogenetic trematodes. Preliminary identifications of ergasilid copepods were based on works by Roberts (1969, 1970) and Johnson and Rogers (1972, 1973). All final identifications of copepods were made by Dr. Ju-shey Ho (Univ. of California at Long Beach) and Dr. Robert F. Cressey (U.S. National Museum). Dr. Robin M. Overstreet (Gulf Coast Research Laboratory) provided identifications of leeches and monogenetic trematodes. Crustacean parasites were preserved and shipped in glycerin-ethanol , and leeches and monogeneans were fixed in hot aceto-f ormol-ethanol and shipped in 70% ethanol. Study Sites Collections were made at two locations. Crystal River and Seahorse Key. Crystal River flows generally westward into the Gulf of Mexico, has a total length of approximately 9 km. , and is primarily spring-fed. All sampling was done in the vicinity of the springs where the salinity was 0 ppt. and the water temperature was 23 ± 2°C at 2 m. The area is tidally influenced, and the substrate is sandy mud with some rock outcroppings. The dominant submergent vegetation is Hydrilla sp., which often fills the entire water column, but substantial patches of Vallisneria sp. also occur. Seahorse Key is a small island north of Crystal River about 4 km. offshore. Salinities commonly range from 12-32 ppt. , but remained at 22-31 ppt. during the study. Extensive beds of -12- Thalassia testudinum surround the island, and two small mangrove- lined tidal creeks flow into the cove on the north face of the island where collections were made. Oyster bars are also common on the north face of the island. The substrate varies from sand in the areas of T^. testudinum to fine mud in most of the cove. Water temperature varied drastically depending on location and time of day, but the maximum measured during collecting operations was 30.0°C. The estuarine nature of this section of Florida's gulf coast is main- tained by freshwater input from several rivers, including the Homossassa, Crystal, Withlacoochee, Waccassassa, and Suwanee Rivers in the 60 km. section around the study sites. Thus, while collec- tions were not made in the immediate vicinity of the mouth of Crystal River, the habitat at Seahorse Key is not dissimilar. Experimental Equipment Mullet were transported by road between sites and acclimated in 61 X 46 X 46 cm. Nalgene tanks containing approximately 50 1. of water and 3-4 fish. Each tank was aerated by a recirculating system consisting of a 12v 1890 l./hr. submersible bilge pump that removed water from the bottom of the tank and sprayed it onto the surface through a perforated plastic tube. Experimental cages were constructed from 2.5 cm. poultry mesh (chicken wire) with a frame of cypress strips. The dimensions were 183 X 91 X 61 cm., and the cages were weighted to keep them on the bottom. (One cage was constructed of 1.9 cm. mesh nylon netting -13- stretched over a frame of 1.9 cm. O.D. PVC pipe, but this cage was discarded after a preliminary trial due to excessive algal growth on the netting.) A maximum of five fish at a time was held in each cage. The 305 x 610 x 91 cm. indoor concrete tank had a four-corner airlift-powered subsand filter system. Filtration rate was increased by the addition of a 1134 l./hr. pump, wh/ch removed water from beneath the sand and sprayed it onto the water surface, thus also increasing aeration. In addition, two floss and charcoal filters (46 x 46 x 46 cm.), which were powered by another 1134 l./hr. pump, filtered water that was pumped directly from the water column. Before a new group of mullet was introduced, the tank was drained, the sides were scrubbed, and all surfaces including the sand were flamed with a propane torch to kill any parasites remaining from the previous group. A maximum of ten mullet was kept in the tank at one time. RESULTS Non-Experimental Mullet Of 198 non-experimental mullet necropsied, 105 were from SK and 93 from CR. No individual was completely free of ectoparasites. Table 4 contains the prevalence and intensities of commonly en- countered parasites from the two study sites and Tables 5 and 6 the prevalence and intensity in host length classes. Phylum Arthropoda Ergasilus versicolor. This parasite was the most nvimerous at both study sites and was always found on the gill filaments. Preva- lence was 100% at both sites, and intensity varied from 2-217 with a mean of 35.0 at SK and from 3-139 with a mean of AO.l at CR. At both sites the anterior arches were preferred and intensity significantly decreased from anterior (arch 1) to posterior (arch 4) (Chi-square: p < .001) (Table 7). Differences in the characteristics of infesta- tion between study sites were noted. At SK all E. versicolor were found near the tips (distal portion) of the gill filaments (except on one fish) , but at CR only about 32% were on the distal portion of the filaments. The majority were located between hemibranchs on the proximal portion of the filaments and wert not visible without I . 4-1 CO O PM t/i tn C O -1 s to • > • V. t4 4-1 O ~i CJ 3 • u •a o 4J U CO C cn n! c S 4J i — in ON o CN O o o CN T— t 0^ O -3- cs m o o —1 in o o V V a. to c tn en tn c c c o CN 00 CM 00 m m -H —I CM u o i-H tn o 3 0) u T3 nj ■H iH M tn •H i-l tn tn 3 •H 3 3 ^ "o tn i-H tJ c •H ~l O 03 )-i tn :n r-l M t3 o ,n P4 00 to B o u o W W cq 2 tn -3 ■H 0) 3 3 -a < CO o r-l •H "d. O O V V a. Ou +* +" tn to tn 01 tn to 1 1 1 1 1 C C c C C c 1 1 1 1 cr> CM r— ( r— 1 Csl o CM c •H t3 O O lU CO E 3 o 0) nj -3 O s o 01 u 2 ^ -16- C c O -u 3 > • p tH Cm u u O -H U 3 •o o ■u u W (-1 c to w c 0) 0) S -u c 03 c CM o o o V a V Q. CM 00 00 V u 0 o O ■-I ^ — ^ r-l o o 0] o ^ — N o o u o •H •H C •H 03 03 (U 03 ,D B u •H OJ CD OJ E > > 1— ( > 03 •H 0) ^ M 3 3 c r-1 ' — V ^ r-l r-l •H 1—1 •H to •H 01 CO IB M 4J 03 ut O 03 ■i-J M O 00 •H OJ H T3 ! Cd N >^ — ^ ^ ' in O V a 03 03 1 c c c ^ — X — > , — ^ — < CSl CM o O CD ■ — - r~ CO -H CM CM p >, 01 (U ?> c r-l -H a\ ON S3- GO o> vO ON VD ^3 CL 0) 03 CO > CO OJ T3 o CO C E O 0) o Z ^ -17- o CO 05 n! u • ►J 4J CO c • g a u •H 1 c iH (U C rH O o a a >M o II >, 4-1 •H CO C CS u OJ PM C/) 60 c o tsi o en I O 00 O 00 cs CM o vo CM O tN o CO O -3- O tM 00 in vo rvi ov 00 — I rn -a- 00 O — < tN n O 0^ <— < U-1 CO ^ o o o -a- O o <■ o On Cv CV vO 00 vO 00 m 1 O VD CN 00 1— ^ 00 —I in — ' o m <• CN in o — o <• O O 00 O vD 00 O — ' in O O O O r-- O O o -a- vo o -3- O (M o — ' o —I CN] o o O <^ vo 0-1 o 00 cs in CO CM —I O m o o O — ' Cfl a) ■s. •rt O cc •H rH (J cn cn Cfl o f— ' "cj tn CU •H CJ •H •H rH u •j: ■H K t — t -5 — ! 3 K cn to •H (0 3 -a a. ■i u N ■U 00 < 1 •H ■H > z 1 — 1 < D 4-1 « E O CM ~3- O O O —I CM CM rH rH CM CM -3- CM O CM CM PL, i-i CO rH 4-1 O to 'J x. 4-1 V. c o tfl tn u •H to Uh n u HJ tJ to p z -3- CM CM CO O CM -18- m 0) *j •H n cd u « a tj (U P-i •1-4 t-i 3 00 o O CD M W O c a) I— I CO > to -H 0 c c m CO 'U U C ^ M •H e (U as u c tJ m dj ^ i-i CO ^ c CD (U C J-J •H C CO (0 (U •rH tJ Q C T— I CO > I-I Cl. o cn 2 -H o 0^ cn m -a- m o CN — < o u-l 00 o o VO CM 32l STANDARD LENGTH (mm) CO CO •H o u c ca to ^ to e •H (U •a ^ E o M-l u o iH O a •H CO M 0) > wl •o V-i to ■a c CO to • o u ,c o )-l • i-1 CO 0) > U U 0) to rH 3 to l-l > •H 05 to O 4J c >> -. to u u u to o 0 4-1 1 00 I c O OJ CO u 9 •H (14 -24- (0 > Cd 4-1 4-1 01 0) iH •H i-t O H U 0) 14-1 > ■H PS O r-( rH O C 4J •H 0) CO >, 1-1 (U u > o CO <4-4 4-1 o c 4-1 rH )-l 0) iH CO u •H T) ao an 4-1 14-1 4-1 c o CO 0) CJ CO 4-1 u a CO 0) •H O Ph 4J 0) 60 F>4 -26- -27- appearance to E^. versicolor , this species exhibited a very different microhabitat preference. The preferred sites for E^. lizae were on the gill rakers or on the proximal parts of the gill filaments on the lateral and medial surfaces; i.e., never between hemibranchs. Of the total number of E. lizae collected, 43.5% were on the lateral surface and 56.5% on the medial, which represents a significant difference from a random distribution (Chi-square: p < .01). Also, the dorsal and ventral thirds of the arches were preferred to the middle third on both lateral and medial surfaces (Chi-square: p < .001 in both cases). Arches 2 and 3 were preferred over 1 and 4, and arch 1 was utilized least (Table 7). Intensity of infestation by E. lizae was negatively correlated with host length (Spreaman rank: p < .05) (Figure 5), but was not significantly correlated with salinity, and prevalence was not significantly correlated with salinity or host length (Tables 4-5). Bomolochus nitidus. This copepod was collected only from SK mullet, primarily from the inner surfaces of the opercles but occa- sionally from the gill rakers or the proximal portions of the filaments. Prevalence was 45.7%, maximum intensity was 14, and the mean intensity of infested fish was 2.6. There was no significant correlation of intensity or prevalence with salinity or host length. Naobranchia lizae. Occurring only at SK, the prevalence of N. lizae was 29.5%, maximum intensity was 12, and mean intensity was 2.8 per infested fish. This species occurred on the lateral and medial surfaces of all arches, never between hemibranchs. There was no significant correlation between intensity or prevalence and salinity or host length. There was, however, a definite avoidance of arch 4, with a total of 31 removed from arch 1, 28 from arch 2, 26 from arch Figure 5. Mean intensity of E. lizae on Seahorse Key mullet vs. host standard length. Bars indicate standard error. V <240 241-260 261-280 281-300 301-320 >32l STANDARD LENGTH (mm) -30- 3, and only 4 from arch 4 (significant, Chi-square: p < .001) (Table 7). Rarer species. Several other copepods occurred rarely enough to make prevalence and intensity values inappropriate. Also, the iden- tifications of all but the first species in this group are tentative, and all were collected from SK mullet only. Lernaenicus longiventris — 12 specimens from fins, especially caudal; L. polyceraeus — 1 specimen from anal fin; Caligus sp. — 4 specimens from fins and body surface; Neobrachiella longimana — 8 specimens from mouth and base of pectorals . Argulus spp. While occurring with low prevalence and intensity, argulids were collected regularly from SK mullet, but not at CR. Thirty-three adults were collected from the mouth, fins, and body surface of 22 fish (prevalence = 20.9%) with a mean intensity of 1.5 and a maximum of 4 per fish. Immature argulids, which were trans- luscent and lacking in pigment except the eye spots, were found only on the gill filaments. Forty-one specimens were removed from 24 mullet for a prevalence of 22.9%. The maximum intensity was 3 per fish, and the mean intensity was 1.8. The occurrence of neither adult nor immature argulids was correlated with salinity or host length (Tables 4-5) , and there was no apparent arch preference by the immature specimens (Table 7). -31- Phylum Platyhelminthes All monogenetic trematodes examined were identified as Metamicrocotyla macracantha, but because the gill arches were pre- served before removal of parasites, many individuals were broken and not definitely identifiable. Monogeneans were found only on the gill filaments and, with the exception of one specimen, only on SK mullet. Eighty-eight were removed from 38 SK mullet for a prevalence of 36.2%, maximum intensity of 18, and mean intensity of 2.3 per infested fish. Neither prevalence nor intensity was correlated with host length or salinity. There was a definite preference for the anterior arches, especially arch 1. A total of 65 were removed from arch 1, 13 from arch 2, 9 from arch 3, and 1 from arch 4 (significant, Chi-square: p < .001) (Table 5). Phylum Annelida Myzobdella lugubris. While occurring on only one SK mullet, this parasite was quite common at CR with 253 removed from 65 fish. Prevalence was 69.9%, maximum intensity was 24, and the mean intensity of infested fish was 3.9. Intensity but not prevalence was negatively correlated with host length (Spearman rank: p < .05) (Figure 6). M. lugubris occurred on all external surfaces, in the mouth, and rarely on the gill arches, but the most frequently infested sites were the fins (Tables 4-6). Malmiana philotherma (?). In October 1982 it was notieced that a leech was partially protruding from a nostril of a SK mullet. Both u 3 T3 B U u -o 0) C > n) •H 4-t CO i-H 0) ffl 4-1 4J CO CO o t^-H o c •H c O CO CO cfl •H pq 3 • 00 J 3 4J iH 60 C • 0) o u ■M C •H to CO 4-> C to CU 4-> U C CO • •"^ c o to • (U to u s > 32l STANDARD LENGTH (mm) -37- Table 8. Total nematodes (Contracaecum mult ipapillat urn ?) per infection site and host sex for both study locations. Location Host Sex No. Hosts Liver Kidney Seahorse Key Female 23 113 169 Male 27 89 99 Crystal River Female 43 39 82 Male 23 17 41 -38- the rest either escaped, died, or were removed from cages by vandals. Of the mullet caught at SK, 22 were caged at CR, 11 were kept in CR water in an indoor tank, 14 were caged at SK, and 7 were transported and acclimated to 0 ppt. but then sacrificed. Of the mullet collected at CR, 8 were caged at SK, 5 were caged at CR, 9 were held in brackish (30 ppt.) water in an indoor tank, and 5 were sacrificed and then held in an aerated aquarium for 6 hours. SK Mullet Transported and Acclimated Only No effects on any parasites due to the transport and acclimation procedure were apparent (Table 9). Ergasilus lizae, E^. versicolor, _L. longiventris , B^. nitidus, M. philotherma, M. macracantha, and imma- ture argulids were all present on one or more of these seven mullet and were all in good condition and at sites on the body listed pre- viously for SK mullet. While intensities of parasites were lower on experimental fish than on controls, none of the differences were statistically significant. SK Mullet Caged at SK Despite being caged at a location where the water depth was less than 6 inches on low spring tides, the water temperature highly variable, and D.O. levels often low, these 14 fish remained the healthiest-looking of all caged mullet. All parasites remained in the expected locations and there was no significant difference in intensity of any parasite between pooled experimental fish and -39- Table 9. Intensity of and number of fish infested by commonly encountered parasites on Seahorse Key mullet sacrificed after acclimation and transport (Experimental) and necropsied immediately (Control). Control ExperimenCal No. Fish: • 22 7 Ergasilus versicolor No. Infested: Intensity : Ergasilus lizae No. Infested: Intensity : Bomolochus nltidus No. Infested Intensity: Naobranchia lizae No. Infested: Intensity : Argulid (imraat . ) No. Infested: Intensity : 22 56.9 7 23.0 21 8.1 6 6.7 7 2.0 4 1.8 6 1.7 11 2.3 2 1.0 Metamicrocotyla macracantha No. Infested: Intensity : 6 4.2 3 2.3 -AO- controls. On only one host were any parasites present in much greater numbers than on control fish caught at the same time and sacrificed immediately; 291 JE. versicolor were removed from the distal parts of the filaments of a mullet held for one month, while the maximum number on any control fish caught during that period was 58 (Table 10) . SK Mullet in Indoor Tank of CR Water Eleven SK mullet were held indoors, one for 7 days, nine for lA days, and one for 26 days (Table 11). All were in good condition. There was no change in the location of JE. versicolor and no statis- tically significant difference in the nxmbers of E^. versicolor or JE. lizae on the pooled experimental mullet compared to 10 control fish collected on the same dates. While monogeneans were absent from the "26 day" mullet, they were present and apparently in good condi- tion on fish held up to lA days. On the "7 day" fish all JE. lizae and N. lizae were in good condition, but on the other experimental fish most E^. lizae had lost their egg sacks, and all five N. lizae had empty egg sacks. No B^. nitidus were present on experimental mullet, but six of the controls were infested with a total of 21 specimens, a significant difference (Chi-square: p < .01) (Table 11). SK Mullet in CR Cages Of the 22 SK mullet caged at CR, various numbers were held for 3 days, 7 days, 12 days, lA days, and 30 days (Table 12). Several of the fish were emaciated and sluggish and many had reddish lesions -41- Table 10. Intensity of and number of fish infested by parasites commonly encountered on Seahorse Key mullet caged at Seahorse Key. Days in Cage : No. Fish: 16 lA 17 30 40 Ergasilus versicolor No. Infested: 16 5 5 2 11 Intensity: 33.4 34.4 30.0 50.0 291.0 16.0 Ergasilus lizae No. Infested: 16 5 5 2 1 1 Intensity: 8.9 18.6 5.8 11.5 12.0 21.0 Bomolochus nitidus No. Infested: 7 2 3 2 0 0 Intensity: 2,4 3.0 1.0 5.0 Naobranchia lizae No. Infested: 3 0 1 0 0 0 Intensity: 1.3 1,0 Argulid (adult) No. Infested 3 0 2 2 1 Intensity: 1.3 1.0 4.5 4.0 Argulid (immat.) No. Infested: 4 0 0 0 0 Intensity: 1.0 Metamicrocotyla macracantha No. Infested: 6 2 3 Q i Intensity: 1.8 3.5 2.7 4.0 1 1.0 -42- Table 11. Intensity of and number of fish infested by parasites commonly encountered on Seahorse Key mullet held in a tank of filtered Crystal River water. Days in Tank: 0 7 14 26 No. Fish: 10 1 q 1 1 Ergasilus versicolor (a) No. Infested: Intensity: 10 39.8 1 30.0 9 29.4 1 23.0 Ergasilus lizae No. Infested: Intensity : 10 4,8 1 10.0 7 7.0 1 10.0 Bomolochus nitidus No. Infested: Intensity: 6 3.5 0 0 0 Naobranchia lizae No. Infested: Intensity : 2 1.5 1 1.0 3 1.7 0 Argulid (immat.) No. Infested: Intensity : 2 1.5 0 0 0 Metamicrocotyla macracantha No. Infested: Intensity : 3 1.0 1 5.0 2 1.0 0 (a) all E. versicolor were at distal tips of filaments. -43- Table 12. Intensity of and number of fish infested by parasites commonly encountered on Seahorse Key mullet caged at Crystal River. Days in Cage: 0 3 7 12 lA 30 No. Fish: 27 5 7 4 3 3 Ergasilus versicolor Between hemibr . No. Infested: 0 12 2 3 3 Intensity: 22.0 22.0 1.5 10.3 5.3 Distal tips No. Infested: 27 5 7 4 3 3 Intensity: 46.0 45.6 128.0 135.5 36.0 20.3 Ergasilus lizae No. Infested: 24 5 6 1 3 2 Intensity: 7.6 11.2 5.5 6.0 2.3 4.5 Bomolochus nitidus No. Infested: 12 0 0 0 0 0 Intensity: 2.3 Naobranchia lizae No. Infested: 9 2 2 0 0 0 Intensity: 3.1 2.0 1.0 Argulid (adult) No. Infested: 2 0 0 10 0 Intensity: 1.0 6.0 Argulid (immat.) No. Infested: 12 2 0 0 0 0 Intensity: 2.3 1.0 Metamicrocotyla macracantha No. Infested: ^ 2 2 0 0 0 Intensity: 3.7 2.0 2.5 Myzobdella lugubris No. Infested: 0 0 0 2 2 3 Intensity: 5 5 ^ 5 53 -44- where scales were lost. No B^. nitidus were present on any of the mullet, including those held for only three days, whereas 48.5% of 33 SK control fish for the period were infested. Similarly, 42.4% of the controls had immature argulids on their gill filaments while none were found on experimental fish held for more than 3 days. Naobranchia lizae and monogeneans were present on fish held up to seven days and appeared to be in good condition, except that the one IJ. lizae held for seven days had empty egg sacks. Monogeneans and adult argulids occurred irregularly on controls during this period, so no conclusion as to their tolerance to freshwater can be made, t^zobdella lugubris appeared only on fish held 12 days or longer in freshwater with six of 10 fish held for this time period infested with an average of 6 leeches each. Leeches did not occur on the SK controls. Ergasilus lizae were present on all but four of the experimental fish. While there was no statistically significant difference in intensity on pooled experimental fish compared to controls, many lost their egg sacks. Combining data from cage and indoor tank experiments, the percent of E. lizae with eggs declined from 95.6% for controls to 0% for all fish held six days or longer (Figure 8). There was a nega- tive correlation between percent of E. lizae with egg sacks and number of days in freshwater, significant at p < .01 (Spearman rank). All experimental mullet were infested with E. versicolor on the distal parts of the gill filaments. Intensity was unexpectedly high compared to controls on only two fish, one held 7 days and one 12 days, from which 441 and 426 were removed. There was no statistically significant difference in intensity of E^. versicolor on filament tips l-l 0) CO to & to •a CO 0) M-i o u c •H -i XI tH C (U ,C to 0) > M 0^ 0) 4-1 (U iH 4-1 iH •H & i R-E. King, and B.L. Jacobs. 1963. Ecology of the gill parasites of Gillichthys mirabilis Cooper. Ecology 44:295-305. Odum, W.E. 1970. Utilization of the direct grazing and plant detritus food chains by the striped mullet Mugil cephalus. pp. 222-240 in: Marine Food Chains. J.H. Steele (ed.). Oliver and Boyd, London. Papema, I., and R.M. Overstreet. 1981. Parasites and diseases of mullets (Mugilidae). pp. 411-493 in: Aquaculture of Grey Mullets. O.H. Oren (ed.). International Biological Program 26, Cambridge Univ. Press, Cambridge. Pearse, A.S. 1952. Parasitic Crustacea from the Texas coast. Publ. Inst. Mar. Sci. Univ. Tex. 2:6-42. Polyanski, Y.I. 1958. Ecology of parasites of marine fishes. pp. 48-842 in: Parasitology of Fishes. V.A. Dogiel, K. Petrushev- ski, and Y.I. Polyanski (eds.). Leningrad Univ. Press, Leningrad. Raibaut, A., O.K. Ben Hassine, and G. Prunus. 1975. Study of the Infestation of Mugil (Mugil) cephalus Linne ' , 1758 (Pisces, Teleostei, Mugilidae) by the copepod Ergasilus nanus van Beneden, 1870 in Lake Ischkeul (Tunisia). Soc. Zool. France 100:427-436. -75- Rawson, M.V. 1973. The development and seasonal abundance of the parasites of striped mullet, Mugil cephalus L. , and mummichogs, Fundulus heteroclltus (L.). Dissertation, Univ. of Georgia. 100 pp. Rivas, L.R. 1980. Sjmopsis of knowledge on the taxonomy, biology, distribution, and fishery of the Gulf of Mexico mullets (Pisces: Mugilidae) . pp. 34-53 in: Proceedings of a workshop for poten- tial fishery resources of the northern Gulf of Mexico. M. Flandorfer and L. Skupien (eds.). Miss. -Ala. Sea Grant Consort. Publ. MASGP-80-012. Roberts, L.S. 1969. Ergasilus arthrosis n. sp. (Copopoda: Cyclopoida) and the taxonomic status of Ergasilus versicolor Wilson, 1911, Ergasilus elegans Wilson, 1916, and Ergasilus celestis Mueller, 1936, from North American fishes. J. Fish. Res. Bd. Canada 26:997-1011. • 1970. Ergasilus (Copepoda: Cyclopoida): Revision and key to species in North America. Trans. Amer. Micros. Sco. 89:134-161. Sawyer, R.T. , A.R. Lawler, and R.M. Overstreet. 1975. Marine leeches of the eastern United States and the Gulf of Mexico with a key to the species. J. Nat. Hist. 9:633-667. Schell, S.C. 1970. How to Know the Trematodes. Wm. C. Brown Co., Dubuque, lA. Shireman, J.V. 1975. Gonodal development of striped mullet (Mugil cephalus) in fresh water. Prog. Fish-Cult. 37:205-208. Siegel, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill Book Co., New York. 312 pp. Simmons, E.G. 1957. Ecological survey of the upper Laguna Madre of Texas. Publ. Inst. Mar. Sci. Univ. Tex. 4:156-200. Skinner, R. 1975. Parasites of the striped mullet, Mugil cephalus, from Biscayne Bay, Florida, with descriptions of a new genus and three new species of trematodes. Bull. Mar. Sci. 25:318-345. Smith, R.F. 1949. Notes on Ergasilus parasites from the New Brunswick, New Jersey area, with a check list of all species and hosts east of the Mississippi River. Zoologica 34:127-132. Tang, Y.A. 1970. Evaluation of balance between fishes and available fish foods in multispecies fish culture ponds in Taiwan. Tran. Amer. Fish. Soc. 99:708-718. Wilson, C.B. 1936. Parasitic copcpods from the Dry Tortugas. Pap Tort. Lab. 29:327-347. BIOGRAPHICAL SKETCH Mark R. Collins was born January 4, 1956, in Melbourne, Florida, and graduated from Westside High School, Anderson, South Carolina. He received a Bachelor of Science degree in marine science from the University of South Carolina in 1978 and a Master of Science degree from the University of Florida Department of Zoology in 1980. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Carmine A. Lanciani, Chairman Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Frank J. Matur^, Jr. Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Frank G. Nordlie Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. j This dissertation was submitted to the Graduate Faculty of the Department of Zoology in the College of Liberal Arts and Sciences and to the Graduate School, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1984 William Seaman, Jr. Associate Professor of Forest Resources and Conservation Ellis C. Greiner Associate Professor of Veterinary Medicine Dean for Graduate Studies and Research