140 



Fishery Bulletin 98(1) 



tion known as bitter crab disease in snow and Tanner 

 crabs (Meyers et al., 1987). Low prevalences (1-4%) 

 of another species, H. australis, have been reported 

 in sand (Portunus pelagicus) and mud (Scylla ser- 

 rata) crabs from Australia (Shields, 1992; Hudson 

 and Shields, 1994). 



Infections of Hematodinium spp. or Hematodin- 

 jum-like species have been reported from a variety of 

 different hosts (see Shields, 1994, for review). There 

 are, however, only two described species oi Hemato- 

 dinium: H. perezi Chatton and Poisson, 1931, and H. 

 australis Hudson and Shields, 1994. By convention 

 (Newman and Johnson, 1975; MacLean and Rud- 

 dell, 1978) and from its distinct morphological fea- 

 tures, we concur that Hematodinium perezi is the 

 infectious species in the American blue crab. 



Blue crabs sustain one of the largest fisheries in 

 Chesapeake Bay. Current management plans and 

 state regulations are based on population assess- 

 ments that include numbers of juvenile and adult 

 crabs found during winter, spring, and summer sur- 

 veys (Lipcius and Van Engel, 1990; Abbe and Stagg, 

 1996; Rugolo et al., 1998). Although these projections 

 include estimates of natural mortalities, they do 

 not account for the potential epizootics and mortali- 

 ties caused by Hematodinium perezi. In this study, 

 we examined host mortality in controlled laboratory 

 experiments and documented changes in the hemo- 

 grams (total cell counts, and differential counts) of 

 inoculated crabs versus uninfected crabs. We also 

 examined proliferative growth of//, perezi at approx- 

 imately weekly intervals and made observations on 

 the biology and life history of the parasite. 



Materials and methods 



Blue crabs were collected from Chesapeake Bay and 

 several of its subestuaries during the annual VIMS 

 Winter Dredge Survey (part of the Chesapeake Bay 

 Stock Assessment Program) with a 1.83-m-wide Vir- 

 ginia crab dredge fitted with 0.5-inch ( 1.25-cm ) Vexar 

 mesh dragged on the bottom for one minute at three 

 knots. Crabs were also taken with commercial crab 

 pots from two reference locations on the Delmarva 

 Peninsula, Red Bank and Hungars Creeks, Virginia. 

 Uninfected crabs were housed together for three to 

 seven days prior to treatment to ensure acclimation 

 and absence of overt bacterial or protozoal diseases 

 (as assayed below). During the experiments, crabs 

 were fed fish and squid semiweekly and held individ- 

 ually in aquaria (5 gal., 19 liter) at 20° to 21°C, and 24 

 ppt salinity. Although H. perezi infects both sexes, only 

 mature, nonovigerous female crabs (healthy, orange 

 maturing gonads, little to no shel] damage, 120-160 



mm carapace width including epibranchial spines) 

 were used in the experiments. Females were used 

 to limit the number of treatment effects (e.g. poten- 

 tial differences between sexes) and to improve sample 

 sizes given the laborious nature of the experiments. 



Hematodinium perezi was maintained in the lab- 

 oratory by serial passage of infected hemolymph. 

 Hemolymph from naturally infected crabs was in- 

 jected directly into uninfected crabs. Naive (unex- 

 posed) crabs and crabs used for inoculation experi- 

 ments were obtained from low-salinity non-enzootic 

 locations. Infected and inoculated crabs were housed 

 separately and used as hemolymph donors to inject 

 naive hosts (10^-10^ parasites per host). Injections 

 were given in the arthrodial membrane of the fifth 

 leg at the juncture of the basis with the cara- 

 pace. We have maintained //. perezi for over seven 

 months using this method with no apparent loss from 

 pathogenicity. 



Two mortality experiments and one early life his- 

 tory experiment were undertaken. The mortality-I 

 experiment used raw, infected hemolymph as the inoc- 

 ulant. Although appropriate for maintaining infec- 

 tions in the laboratory, raw hemolymph cannot be 

 adjusted to manipulate parasite densities without 

 the use of physiological buffers, nor can it be guar- 

 anteed as sterile without appropriate assessment 

 (see Welsh and Sizemore, 1985). Preliminary experi- 

 ments with sterile sea water, physiological buffers, 

 and infected hemolymph indicated that buffer- washed 

 parasites remained infectious, and could, therefore, 

 be adjusted to consistent densities appropriate to 

 controlled experiments. The mortality-II experiment 

 used buffer-washed parasites adjusted to a density 

 similar to that used in the mortality-I experiment. 

 Mortality-II experiment closely resembled mortality-I 

 experiment except for 1 ) handling (buffer washes with 

 centrifugation) and 2) the use of plasmodial versus 

 uninucleate stages of the parasite. Uninfected crabs 

 served as controls in both experiments. Controls were 

 used to assess handling effects and to establish base- 

 line densities of hemocytes. The early infection exper- 

 iment was designed to examine the effects of early 

 infections on the hematology of the host and the 

 early life history of the parasite. Experimental den- 

 sities in the early infection experiment were four 

 times higher than those in the previous experiments 

 (4.1 X 10'' vs. approx. 1.0 x 10'^ parasites/crab, respec- 

 tively) and were arbitrarily higher to insure obser- 

 vation of parasites prior to their proliferation. 



In the mortality-I and mortality-II experiments 

 different proportions of trophonts and plasmodia 

 were used (for definitions see below). The mortal- 

 ity-I experiment consisted of a control group of unin- 

 fected crabs (n=22) injected individually with 100 pL 



