Novel experimental technique to confirm fish hosts for ?large river mussels?. 
 
by 
 
Michael Andrew Hart 
 
 
 
 
A thesis submitted to the Graduate Faculty of 
Auburn University 
in partial fulfillment of the 
requirements for the Degree of 
 Master of Science 
 
Auburn, Alabama 
August 2, 2014 
 
 
 
 
Keywords: Freshwater mussel, host confirmation, migratory fish, gill excision. 
 
 
Copyright 2014 by Michael Andrew Hart 
 
 
Approved by 
 
James Stoeckel, Chair, Associate Professor, School of Fisheries, Aquaculture and Aquatic Sciences 
Stephen Bullard, Associate Professor, School of Fisheries, Aquaculture and Aquatic Sciences 
Matthew Catalano, Assistant Professor, School of Fisheries, Aquaculture and Aquatic Sciences  
ii 
 
 
 
 
 
 
Abstract 
 
 
 Unionid mussels are declining in North America.  The larval stage (glochidium) 
represents a life-history bottleneck because glochidia are parasitic and require an appropriate fish 
host.  However, appropriate fish host species for many mussel species are still unknown, making 
it difficult to implement effective conservation measures.  Large-river migratory fish are 
suspected to be important hosts for mussels, but are difficult to work with and few have been 
positively confirmed as hosts in laboratory trials.  I applied standard fisheries techniques to 
successfully capture, haul, and hold two problematic, large-river fish species; Alosa 
chrysochloris (skipjack herring) and Alosa alabamae (Alabama shad).  I describe a new 
technique (gill excision) which facilitated the use of problematic fish for host assays and allowed 
me to provide the first laboratory conformation of fish host identity for Reginaia ebenus and 
Elliptio crassidens, as well as confirming that glochidia can successfully metamorphose from 
dead fish gills.   
 
 
 
 
 
 
 
iii 
 
 
 
 
Acknowledgments 
 
 Funding for this project was provided by the USDA through a grant provided by Dr. 
Wendell Haag (USDA Forest Service, Center for Bottomlands Hardwoods Research).  I would 
like to first thank my advisor Dr. James A. Stoeckel for giving me the opportunity to be a 
graduate student under his guidance and also for allowing me to participate in a wide array of 
different projects during my time as a graduate student.  Many people provided assistance to me 
while working on my thesis and I would like to thank all of them. I would like to thank my 
committee members, Dr. Stephen Bullard, and Dr. Matt Catalano for their contributions to the 
development, analysis, and writing of this thesis.  I would like to thank Dr. Steve Sammons for 
his advice on where to catch the skipjack herring used in this study.  I would like to thank Brian 
Helms for his assistance with collecting and hauling our study animals.  I would like to thank 
Dave Werneke for his assistance with identifying all of the small stream fish used in this study.  I 
would like to thank the USFWS Warm Springs Fish Hatchery for loaning of the fish hauler and 
also the Georgia Department of Natural Resources for their help in capturing the Alabama shad 
used in this study.  Without the assistance of my fellow lab members in the Crustacean and 
Molluscan Ecology Lab (CAMEL); Tyler Mosley, Catlin Ames, Andrew Gascho-Landis, Ian 
Palmer, Daniel Foree and Brad Staton, this project would not have been possible.  Finally I 
would like to thank my parents; Andy and Jena for their support, love, and patience during my 
time as a graduate student as well as my brother Elliot who was always able to watch my dog 
Bender while I was out in the field.  
iv 
 
 
 
 
 
 
 
Table of Contents 
 
 
Abstract ......................................................................................................................................... ii 
Acknowledgments ....................................................................................................................... iii 
List of Tables ............................................................................................................................... vi 
List of Figures ............................................................................................................................. vii 
Chapter 1: Confirmation of host fish for Reginaia ebenus using newly developed gill excision 
technique   ......................................................................................................................... 1 
 
 Introduction   ..................................................................................................................... 1 
 Methods  ........................................................................................................................... 4 
 Results   ........................................................................................................................... 14 
 Discussion   ..................................................................................................................... 16 
Chapter 2: Use of previously described gill excision technique to confirm host fish species for 
Elliptio crassidens   ......................................................................................................... 21 
 
 Introduction   ................................................................................................................... 21 
 Methods  ......................................................................................................................... 24 
 Results   ........................................................................................................................... 31 
 Discussion   ..................................................................................................................... 33 
Chapter 3: Assay to determine if glochidia can modify their host environment (gill tissue) to 
slow decomposition to allow for metamorphosis   ......................................................... 37 
 
 Introduction   ................................................................................................................... 37 
 Methods  ......................................................................................................................... 38 
v 
 
 Results   ........................................................................................................................... 41 
 Discussion   ..................................................................................................................... 42 
 Literature Cited   ............................................................................................................. 44 
 
 
 
vi 
 
 
 
 
 
 
List of Tables 
 
 
Table 1: Skipjack herring survival from collection site to lab in experiment 1  ......................... 53 
Table 2: Results of gill inspections for encapsulated glochidia of Reginaia ebenus on all                      
 fish taxa tested ................................................................................................................ 54 
 
Table 3: Alabama shad and skipjack herring survival from collection site to lab  
 in experiment 2  .............................................................................................................. 55 
 
Table 4: Mussel length, percent viable glochidia, and viable brood size of mussels              
utilized for glochidial baths on two dates  ...................................................................... 56 
 
Table 5: Number of sloughed glochidia form bluegill infected with elephant ear  .................... 57 
Table 6: Percent viable glochidia, and viable brood size of mussels utilized for glochidial  
bath.  ................................................................................................................................ 58 
 
vii 
 
 
 
 
 
 
List of Figures 
 
 
Figure 1: a) 5700 liter rectangular fiberglass tanks used to hold large river fish.  b) Baffle and 
airlifts used to generate circular water flow  ................................................................... 59 
 
Figure 2: 8500 liter round fiberglass tank and cover used to hold large river fish  .................... 60 
Figure 3: 1900 liter round fish hauler with oxygen tanks used to haul fish from collection site to 
lab  ................................................................................................................................... 61 
 
Figure 4: Picture and schematic diagram showing collection of sloughed glochidia and juveniles 
from AHAB tanks  .......................................................................................................... 62 
 
Figure 5: Schematic diagram showing collection of sloughed glochidia and juveniles from cone 
tanks  ............................................................................................................................... 63 
 
Figure 6: Photo of infected skipjack herring gills under cross polarized lighting ...................... 64 
Figure 7: Number of juveniles produced from skipjack herring gills on each day following 
excision  .......................................................................................................................... 65 
 
Figure 8: Total juvenile production from Alabama shad gills excised on days 10, 13, 14, 17, 20, 
and 23 post infection  ...................................................................................................... 66 
 
Figure 9: Daily juvenile production from Alabama shad gill arches excised on a) Day 10, b) Day 
13, c) Day 14, d) Day 17, e) Day 20, and f) Day 23 after infection  .............................. 67 
 
Figure 10: Total juvenile production from skipjack herring gills excised on day 16 post excision 
......................................................................................................................................... 68 
 
Figure 11: Daily juvenile production from skipjack herring gill arches excised on day 16 after 
infection .......................................................................................................................... 69 
 
Figure 12: Images showing representation of a ?gill unit? analyzed in this study ..................... 70 
Figure 13: Images showing state of gill filaments on gill units across time treatments ............. 71 
Figure 14: Graph showing results of lamellae counts for gill units across time treatments ....... 72 
Ch. 1) Confirmation of Reginaia ebenus as a host specialist using a pairing of traditional and 
novel techniques. 
Introduction: 
Freshwater Unionid mussels (mussels) have a complex reproductive strategy in which 
parasitic glochidia require a vertebrate host (usually fish) to metamorphose into a juvenile 
mussel.  Because of this host requirement, the larval stage represents a bottleneck that limits 
many mussel populations (Strayer 2008).  The regional distribution of mussel species is linked to 
the distribution of their host fish (Watters 1992, Schwalb et al 2011, Schwalb et al 2013, Haag 
and Warren 1998). Within a region, extinction and recolonization rates of local mussel 
populations are strongly impacted by larval life-history traits (Vaughn 2012).  Within a site, 
coexistence or competitive exclusion of mussel species with shared host taxa may be driven by 
differing efficiencies in host infection and relative abundance of shared host taxa (Rashleigh and 
DeAngelis 2007).  Host specificity can affect how a mussel population responds to a decrease in 
abundance, or complete loss, of specific fish-host species.  Some mussel species are generalists, 
able to utilize a wide range of fish taxa as hosts, whereas others are specialists, able to utilize 
only a small number of closely related fish taxa.  Therefore after the loss of a specific host 
species, generalist mussels can continue to utilize remaining fishes as hosts, whereas a specialist 
that loses its host species will not be able to reproduce, persisting until the population succumbs 
to old age (Douda et al. 2012). 
 A major contributing factor in the decline of freshwater unionids is the construction of 
dams and other impoundments on natural river systems (Vaughn and Taylor 1999; Watters 
1996).  Impounded reaches in a river system can lead to lower fish species richness when 
compared to free flowing sections (Santucci et al 2005) due to disruption of migratory patterns.  
2 
 
 It has long been recognized that dams block migration of anadromous fish (Coker 1930), 
and populations of many migratory fish are not nearly as abundant and as widespread as they 
once were, particularly in heavily-dammed river systems (Ely et al 2008).  This may not create a 
bottleneck for generalist mussels that can continue to utilize the remaining fish species.  
However, specialist mussels lose the ability to recruit if their required host is eliminated from the 
local fish community.  Many mussel species that are suspected to be specialists on large-river 
migratory host fish have suffered declines above dams and some have even been extirpated 
(Kelner and Sietman 2000, Haag and Warren 2010, Hinck et al 2012).   
Reginaia ebenus (ebony shell) is a large-river mussel typically found at depths between 4 
to 7 meters in sandy gravel substrate and areas of swift current protected from excessive silt 
deposition (Williams et al 2008).  It is a short-term brooder (tachytictic), typically gravid 
between April and September (Surber 1912; Howard 1914a; Coker et al. 1921), and suspected to 
use Alosa chrysochloris (skipjack herring) as its host fish based on observations of naturally 
infected fish (Watters 1994; Surber 1913; Howard 1914b). This host relationship, however, has 
never been confirmed in a controlled laboratory setting. 
 Skipjack herring is a euryhaline species typically found in deep, clear waters with swift 
current (Boschung et al. 2004).  It is native to the Gulf of Mexico and associated river drainages, 
and migrates upriver in spring in order to spawn between March and April (Etnier and Starnes 
1993).  This species has historically been found in the Mississippi River drainage as far north as 
Minnesota.  However, dam construction in the Mississippi River has severely impeded its 
migration routes, reducing populations in the upper Mississippi River (Kelner and Seitman 2000; 
Neebling and Quist 2008).  Skipjack herring are now considered critically imperiled in 
Wisconsin and of special concern in Minnesota, where they were considered extirpated for many 
3 
 
years (Natureserve 2013).  In Alabama, skipjack herring have been collected in nearly every 
river system and are considered secure (Mettee et. al. 1996, Natureserve 2013) 
 The distribution of ebony shell mirrors that of skipjack herring.  Previously abundant in 
the upper Mississippi River system, populations that are now cut off from their presumed 
migratory host species are in danger of extirpation (Kelner & Sietman 2000; Coker 1930; 
Natureserve 2013).  This pattern is not limited to the northern states.  In the Mobile River basin 
of Alabama, the historic distribution of ebony shell included the main stem of the Coosa River 
up to Three Island Shoals. Today, however, ebony shell can no longer be found in the Coosa 
River above Jordan Dam (Gangloff 2003); presumably due to the inability of skipjack herring to 
migrate upstream past hydroelectric dams.  However, ebony shell is still common in the Coastal 
plain reaches of large Mobile Basin Rivers (downstream of hydroelectric dams) where it is 
typically the most common mussel species (Lydeard et al 1999, Williams et al. 2008) and is 
considered stable in Alabama (Williams et al. 1993).  
Although ebony shell is assumed to be a specialist, and declines of the species are 
generally assumed to be associated with the interference of skipjack herring migratory patterns 
(Payne and Miller 2000), the relationship between ebony shell and skipjack herring has never 
been confirmed with laboratory trials.  Nor has any other fish host for ebony shell been 
confirmed with laboratory trials (Howard 1914c).  Skipjack herrings are very difficult to 
transport and hold in captivity, and, to our knowledge, have never been successfully utilized in 
laboratory host trials for any mussel species.  For example, in a previous study looking at 
potential host fish for Elliptio crassidens, Hauswald (1997) was unable to keep skipjack herrings 
alive in transit, much less utilize this species in laboratory trials.  In this study, I hypothesize that 
ebony shell is a specialist and utilizes skipjack herring as its primary host. 
4 
 
My objectives were as follows:  
1) Evaluate the efficacy of standard fisheries techniques to capture and haul skipjack herring 
from river to laboratory with minimal mortality. 
2) Evaluate techniques to infect and hold skipjack herring during host trials.  
3) Evaluate the host specificity of ebony shell by testing the suitability of 14 fishes, across 7 
families, as hosts. 
Methods: 
Collection and holding of study animals 
 Collection of an array of potential host fishes and gravid unionids presented logistical 
challenges related to time of availability, and acclimation needs.  Stream fishes were most easily 
collected in summer and early fall to avoid high water events associated with spring and late fall.  
Skipjack herring were most easily collected in early spring, when they concentrated below dams 
during upstream spawning migrations.  Ebony shell do not become gravid until late spring/early 
summer and could only be collected during low river stage.  Short term brooders such as ebony 
shell typically abort their broods when disturbed or handled.  Thus the full complement of fish 
species to be tested had to be acclimated to laboratory conditions and ready to use as soon as 
ebony shell were collected.  This entailed a lengthy period of collection and holding in the lab for 
most species. 
 Specimens of 14 fishes of 7 families were collected for use in laboratory host trials.  
pretty shiners (Lythurus bellus), fathead minnows (Pimephales promelas), blackspotted 
topminnows (Fundulus olivaceous), orangefin shiners (Notropis ammophilus), largescale 
stonerollers (Campostoma oligolepis), striped shiners (Luxilus chrysocephalus), and darters 
(Etheostoma sp. and Percina sp.) were collected via seining and backpack shocking from 
5 
 
Hillabee and Choctafaula creeks which are tributaries of the Tallapoosa river. Ebony shell is not 
known to be present in these streams (Williams et al 2008, Gangloff 2003).  All fish were 
collected in fall, 2010, returned to the lab in aerated coolers and held in a recirculating aquaria 
system with dechlorinated tap water at approximately 18 ?C for ? 7 months.   
 Bluegill (Lepomis macrochirus), and largemouth bass (Micropterus salmoides)  were 
collected via electroshocking from a water supply reservoir at the E. W. Shell Fisheries Research 
Center (Auburn University) that has few to no mussels present (personal observation). Channel 
catfish (Ictalurus punctatus, ?Marion strain?) were obtained from a separate experiment on the 
North Auburn Fisheries Station.  Bass, bluegill, and catfish were held in insulated rectangular 
fiberglass tanks (Length 59? x Width 21.5? x Height 14?) with a baffle dividing the length in 
order to create circular water flow. The tanks were filled with dechlorinated tap water and 
maintained at approximately 21?C.  Fish were fed frozen brine shrimp, bloodworms, flake food, 
commercial fish pellets and fathead minnows as appropriate.  Water quality (Ammonia, Nitrite, 
and Nitrate) was measured at least once per week using Tetra? EasyStrips.  These fish were held 
for ? 7 months prior to being used in experiments.  
 The large river fish species tested in this study; freshwater drum (Applodinotus 
grunniens), gizzard shad (Dorosoma cepedianum), and skipjack herring (Alosa chrysochloris), 
are seldom utilized in host studies due to challenges associated with collecting and holding them 
in captivity.  In this study, all three large-river species were held in large, outdoor, fiberglass 
tanks prior to host trials.  Two of the tanks were rectangular (3.5m L x 1.8m W x 0.9m H,) with a 
volume of 5,700 liters). The rectangular tanks were equipped with a central, longitudinal baffle 
and airlifts were used to generate a circular water flow (Fig. 1a, b). The third tank was circular 
(diameter = 3 meters, depth = 1.2 meters) with a volume of 8,500 liters (Fig. 2).  Internal bio 
6 
 
filters were installed in both tanks to maintain water quality.  The circular tank was part of a 
larger recirculating system that had a total volume of 23,000 liters.  Water was pumped into the 
circular tank from an external biofilter with a volume of 2,300 liters and provided circular flow 
around the central baffle. The rectangular tanks were located under an awning whereas the 
circular tank was located a short distance away in direct sun.    
  Freshwater drum were collected via hook and line from Lake Guntersville in 
Guntersville, AL during December of 2010 and hauled back to the lab in a 3,600 liter rectangular 
fish hauler filled with water from the catch site. Upon return to the South Auburn Fisheries 
Research Station they were held in one of the 5,700 liter rectangular tanks at a stocking density 
of 2.64 fish/m3. The tank was filled with dechlorinated tap water and water quality was 
monitored weekly for ammonia, nitrite, and nitrate using Tetra? EasyStrips.  Drum were held in 
these tanks for ? 6 months prior to infection and maintained on a diet of Corbicula and crayfish.   
 Gizzard shad were collected via electroshocking from the Chattahoochee River below 
Eagle Phenix Dam in Columbus, GA.  Gizzard shad (5) were returned to the lab on March 15 
2011 in a 3,600 liter rectangular fish hauler filled with river water at the collection site and salted 
with Kent Marine Sure-Haul and sodium chloride according to product directions.  Gizzard shad 
were subsequently held in the round tank salted to 3-5 ? with Morton Pool Salt for ~12 weeks 
prior to infection.  They were not directly fed during this time although fathead minnows were 
added to the tank as food for co-occurring skipjack herring (see below), and they may have 
preyed on minnows during this time.  Also, detritus from occasional algal blooms had settled on 
the bottom and served as a food source for gizzard shad.   
  To my knowledge, skipjack herring have never been successfully utilized in previous 
host trials due to the inherent difficulty of handling and holding this species. Thus, methods for 
7 
 
this species will be described in more detail compared to the other two large river fish species 
utilized in this study. Skipjack herring were collected from the Chattahoochee River just below 
Eagle Phenix Dam in Columbus GA and identified according to Boschung et al 2004).  In our 
initial collecting and hauling attempt we coordinated with the Georgia Department of Natural 
Resources to obtain fish that were collected as part of an annual fish survey.  During this initial 
collection, I noted that electroshocking was not very effective for skipjack herring capture as the 
fish were nearly always deep and immediately sank rather than floating to the surface. After 
several hours of shocking I had only collected 6 fish.  I also noted that the 3,600 liter rectangular 
hauling tank was not suitable for skipjack herring. Upon arrival at the lab (~ 1 hr. hauling time), 
two of the six skipjack herring were dead, and the surviving fish had lesions near their mouths 
and along the sides of their bodies, presumably from beating themselves against the tank corners 
and walls.  
For subsequent attempts (March 24th and 29th), I therefore switched to collecting skipjack 
herring with hook and line, using small white jigs.  To reduce physical trauma to the fish during 
transport caused by swimming into corners and sloshing of water, I switched to using a 1900 
liter, round, fish hauler loaned from the Warm Springs Fish Hatchery (Fig. 3).  This tank was 
filled with water close to the top and sealed with a water-tight lid.  This kept sloshing to a 
minimum which has been shown to be important in reducing stress in hauled fish (Winkler 
1987).  Skipjack herring were reeled to the boat as quickly as possible and immediately placed in 
a 570 liter oval stock tank with no corners or protrusions. The stock tank was situated on the 
deck of the boat and was filled with water from the collection site.  Within ~30 minutes of 
capture, skipjack herring were transferred to the round hauling tank filled with river water at the 
boat ramp.  In order to reduce the amount of osmoregulatory stress experienced by the fish 
8 
 
captured for this trial I salted the water using Kent sure-Haul, a commercial product containing a 
mixture of hauling salts, combined with sodium chloride (Morton Pool salt) to a final salinity of 
~5?.  The use of salts in hauling water has been shown to reduce the amount of osmoregulatory 
stress experienced by hauled fish as well as to help them recover more quickly (Murphy and 
Willis 1996; Collins and Hulsey 1963; Mazik et al. 1991; Swanson et al. 1996; Wydoski and 
Wedemeyer 1976, Weirich and Tomasso 1991).  Oxygen was gently bubbled into the hauling 
tank during collection and subsequent hauling.  The time of travel from leaving the boat ramp to 
arriving at the station was approximately 1 hour.  Once the skipjack herring arrived at the South 
Auburn Fisheries Station they were distributed among the large circular tank and the two 
rectangular tanks.  Netting material was placed over both rectangular tanks to keep fish from 
jumping.  The circular tank had a cover made from off-white shade cloth (SYNTHESIS 
COMMERCIAL 95?) to prevent fish from jumping and to shade the tank and prevent it from 
overheating.  All tanks were filled with dechlorinated tap water and sodium chloride was added 
to maintain 3-5 ? salinity (Morton Pool salt) in order to lessen the stress from transferring fish 
from hauling tank to holding tank water.  Water quality was monitored biweekly in these systems 
and partial water changes were implemented whenever ammonia rose above 0.5 mg/L or nitrites 
rose above 1mg/L.  Skipjack herring were fed fathead minnows biweekly.  
Gravid ebony shell were collected from the Alabama River in Wilcox County on June 15 
and 24 of 2011. Mussels were collected using a hookah breathing apparatus (Keene engineering) 
and tactile searching of the substrate at approximately 1.5 to 2.5 meters deep.  Mussels were 
checked for gravidity on the bank of the river by gently opening the shell with reverse pliers and 
looking for swollen creamy white or pink to red gills (Williams et al 2008), and gravid 
individuals were transported to the laboratory (approximately 2.5 hours travel time) in a cooler 
9 
 
with a small amount of ice.  On June 6th, 18 of 31 collected mussels were transported wrapped in 
moist paper towels, while the remaining 13 mussels were transported in individual gallon 
Ziploc? bags of river water.  Upon arrival at the laboratory, the bags were immediately 
submerged in a recirculating aquarium system maintained at 18?C and an airstone was placed 
into each individual bag. Mussels wrapped in paper towels were kept moist and cool.  All 
mussels were held overnight prior to use in host assays.  Mussels collected on June 23 were all 
transported in Ziploc bags in coolers with a small amount of ice, and held overnight in the same 
manner as for the previous collection.    
Host Trials 
 Approximately one week before trials, fish were transferred to their respective trial tanks 
to acclimate.  Trial tanks were filled with dechlorinated tap water and salinity was gradually 
decreased to < 0.5 ? during the acclimation period.  Trials f or all fish species other than 
freshwater drum, gizzard shad, and skipjack herring were conducted in Aquatic Habitat (AHAB) 
research tanks of two sizes (1.5 L, and 3 L) according to individual fish size.  AHAB tanks were 
held within two, independent, recirculating systems holding 36 tanks each.  Within each 
recirculating system, water flowed out of each tank through a 105?m filter cup to retain sloughed 
glochidia and juveniles (Fig. 4), and into a sump/biofilter.  From there water was subjected to 
coarse mechanical filtration and ultraviolet light sterilization. After sterilization, water passed 
through a secondary 100?m filter to catch any stray glochidia or juveniles that had bypassed the 
previous filters.  Water then flowed into a manifold system that distributed water into each tank 
to start the cycle over again.  Water temperature was maintained at 18-21o C by means of an in-
line chiller.  Water quality was monitored from a sample collected from the sumps weekly using 
10 
 
Tetra? EasyStrips for ammonia, nitrites, and nitrates with water changes being performed as 
needed.   
Freshwater drum trials were conducted in a cone tank system (Fig. 5) that consisted of 4 
cone tanks that were 55 liters in volume with all four outflows going into a 100 liter 
sump/biofilter.  The gizzard shad trials were conducted in a similar system consisting of 4 cone 
tanks that were 150 liters in volume and emptied into a common 200 liter sump. Water was 
pumped from the common sumps through a manifold back into the cone tanks.  The inflow line 
for each tank was angled against the side of the cone tank to create a circular flow.  The cone 
tanks were fitted in a double stand pipe configuration which allowed the water exiting the cone 
tanks to come from the bottom.  This water then flowed out of each individual cone tank though 
a drain pipe and into a 100?m nylon filter sock attached to the outflow of each drain pipe to 
catch any sloughed glochidia or juveniles.  Cone-tanks were held at ambient room temperature 
(19-21o C), and water quality was monitored in each cone tank system bi-weekly using Tetra? 
EasyStrips for ammonia, nitrites, and nitrates. 
 Prior to using the cone tank systems for the gizzard shad and drum we tested their 
suitability for skipjack herring.  All skipjack herring held individually in 55 or 150 liter cone 
tanks died within 24 hrs.  I therefore decided to hold infected skipjack herring in the previously 
described rectangular outdoor tanks.  Partial water changes with dechlorinated tap water were 
used to reduce salinity in these tanks during the acclimation period.  The water temperature range 
for the skipjack herring tanks during the trial was 24-26?C and was monitored throughout the 
trial using a HOBO? Pendant temperature data logger.  Water quality was monitored in each 
tank weekly for ammonia, nitrites, and nitrates using Tetra? EasyStrips, and water exchanges 
were performed as needed.   
11 
 
 Although all of the mussels returned to the lab had swollen gills, indicating fertilization, 
the stage of embryonic development was unknown.  Therefore, on the day following collection, I 
extracted a small amount of fluid from a gill of each mussel using an 18 gauge needle and 
examined the contents under 10X magnification.  Embryonic stage was placed into one of four 
categories: 1) fertilized eggs (observable cell division, but not developed into glochidia), 2) early 
stage glochidia (embryos shaped like glochidia but not fully formed), 3) free glochidia (fully 
formed, not bound in membranes), or 4) bound glochidia (fully formed, enclosed in a 
membrane).  Only broods dominated by free glochidia, were utilized for host trials.  Glochidia 
for host trials were collected by flushing excised gills into a beaker, using a syringe with a 20 
gauge needle and artificial freshwater (AFW).  Glochidia viability was evaluated by means of a 
salt test (Fritts et al. 2014).  Glochidia in the beaker were diluted to 2000 ml with AFW.  The 
solution was mixed with a plunger to suspend glochidia, while avoiding the creation of a vortex, 
and 100 ?L subsamples were withdrawn and placed on a counting tray.  Subsamples were 
examined under a dissecting microscope at 10X and the number of open and closed glochidia 
was counted for each subsample.  A few drops of 35? artificial saltwater (Instant Ocean? Sea 
Salt) were added to each subsample, and the number of open and closed glochidia in each 
subsample was recorded for a second time.  Glochidial viability was recorded as the proportion 
of glochidia snapping shut after addition of salt water:  
Proportion of viable glochidia) = [(number open initially) ? (number open after NaCl)] ? (total 
number of glochidia).   
The total number of viable glochidia in   a given brood or batch (multiple broods combined) was 
calculated as:  
 Vn = G/s * V * P 
12 
 
Where Vn = total number of viable glochidia in a brood or batch, G = number of glochidia (open 
plus closed) in each subsample, s = combined volume of subsamples (ml), V = sample volume 
(ml) and P = proportion of viable glochidia.  
  Skipjack herring were infected on June 16, 2011 and all other fish were infected on June 
24, 2011. All fish species were inoculated by immersion for 15 minutes in an aerated glochidial 
bath of dechlorinated tap water containing 2,000 glochidia / L.  Skipjack herring were infected in 
a bath made using combined broods of 4 mussels, while all other fishes were infected using baths 
made from the combined broods of 2 mussels.  Multiple fish were held within a single glochidial 
bath, and any given glochidial bath was only used once.  Bath volume varied according to fish 
size and sensitivity to handling.  Fishes to be tested in AHAB tanks were inoculated in 2 L 
beakers or 19 L buckets.  Drum and gizzard shad were inoculated in large, 140 L coolers with the 
lids closed during inoculation.  Skipjack herring were inoculated in large, round tanks covered 
with shade cloth material and containing 570 L of water from the tanks in which they were 
housed.  After the standard 15 minute inoculation period, fish were returned to the same host trial 
tanks to which they had been acclimated (see previous section). 
 To determine whether glochidia were successfully encapsulated on host fish, gills of all 
fish species except for skipjack herring were analyzed for presence or absence of encapsulated 
glochidia 4-6 days post infection (Zale and Neves 1982).  One fish from each of the other species 
was anesthetized using a bath made with AQUI-S 10 at the recommended rates specified by the 
manufacturer (5-10ml/100L).  Both sets of gills were then, inspected under a dissecting 
microscope at 10-15X by gently prying back the operculum.  Individuals of a given species were 
inspected sequentially until one fish was observed with encapsulated glochidia.  Once a fish had 
been inspected it was placed into a bath of fresh water and allowed to recuperate from the 
13 
 
anesthesia and handling event before being placed back into their respective tanks.  Because I 
saw no evidence of encapsulated glochidia on any fish of any species other than Skipjack herring 
(Table 2), I did not monitor for sloughed glochidia or juvenile production after Day 6.  
 Because of their sensitivity to confinement and handling, skipjack herring were housed in 
large communal tanks and were not examined for glochidial encapsulation within the first week 
of infection.  To check for juvenile production, I siphoned a 3.2 m2 portion of the bottom of each 
communal tank on day 10 post infection.  Siphoned water was passed through a 100?m filter 
sock to collect sloughed glochidia and/or juveniles and filtrate preserved in 75% ethanol.  
Subsamples were examined under a dissecting microscope equipped with cross-polarized 
lighting (Johnson 1995).  Even with the use of cross-polarized lighting, it quickly became 
apparent that detection of juvenile mussels would be very difficult due to the sediment, detritus, 
and fish waste accumulating on the tank bottoms.   
Therefore, on day 11 a single skipjack herring (TL 37.5cm and WW 305.6g) was 
removed from one of the tanks to inspect the gills for encapsulated glochidia and determine if 
subsequent siphoning was warranted.  The fish was anesthetized in a 140 L cooler dosed with 
AQUI-S 10 (5-10ml/100L) to minimize stress during handling.  Gills were inspected in the same 
fashion as for the other species and the fish was placed in an aerated water bath free of AQUI-S 
10 to recuperate from the anesthesia.  However, the fish never recovered and died within 5 
minutes of being placed in the recuperation bath.  Because I had observed encapsulated 
glochidia, I immediately excised its gill arches and preserved half of them in 70% ethyl alcohol 
for documentation.  The remaining gill arches were placed in a 1.5 L AHAB tank within the 
same system used for the non-large river fish species, with a 105 ?m filter cup situated under the 
outflow of the tank.  The filter cup was examined every 1-2 days for sloughed glochidia and/or 
14 
 
juveniles for 8 days post excision.  Juveniles were differentiated from sloughed glochidia by 
valve movement and the presence and movement of a foot.  The remainder of the skipjack 
herring from both tanks were inspected for encapsulated glochidia at 19 days post infection.   
Because I had observed encapsulated glochidia on Day 11, I siphoned a separate 3.2 m2 
portion of the tanks on Day 12 post infection and then siphoned the entire tank bottoms on Day 
19.  The Day 12 sample contained a substantial amount of detritus, and was preserved in ethanol 
for later analysis.  The Day 19 sample was relatively clean since the entire tank bottom had 
already been siphoned once, and I examined this sample immediately for active juveniles rather 
than preserving it.  
Results:  
The combination of hook and line sampling, use of a circular hauling tank, use of hauling 
salts, and provision of bubbled oxygen, proved very effective for the successful capture and 
transport of skipjack herring.  Using these techniques, survival from capture to arrival at the lab 
ranged from 96% (23/24 fish) to 86% (25/29 fish) (Table 1).  Using the initial technique of 
electroshocking and hauling in a rectangular hauler, I captured only 6 skipjack herring with 4 
surviving the transport to the lab.  It was noted that the fish hauled in the round hauling tank 
upon arrival at the lab looked better and had fewer lesions on their bodies than fish hauled in the 
rectangular tank.  
Water quality monitoring showed that our large, outdoor recirculating system had high 
nitrite levels (>10ppm) for roughly 13 days after fish were introduced into this system.  Through 
the use of frequent water changes using dechlorinated tap water (approximately 10% of total 
system volume 2-3X per day) throughout this initial period nitrite levels were brought down to 
around ~3ppm and continued to decrease towards zero as the biofilter became established.  To 
15 
 
minimize stress and effects of high nitrites I maintained a salinity level of around 4-5? using 
sodium chloride (Boyd 1990)  
Feeding activity of the skipjack herring varied between the two different tank types used.  
In the rectangular tanks located under the awning actively feeding skipjack herring were seldom 
witnessed, while in the large round tank located in the direct sun, skipjack herring were 
frequently observed swimming up and grabbing a minnow before returning to the depths of the 
tank while a portion of the cover was lifted during feeding events.  
Results from host trials 
Five of the mussels collected on June 16th and held in bags of creek water released a 
portion of their gill contents while being held overnight.  Gill extractions the following day on all 
mussels revealed 12 with fertilized eggs that had not yet developed into glochidia, 8 with early 
stage glochidia, and 11 mussels with free, mature glochidia.  Broods were randomly assigned to 
one of three batches consisting of 4, 4, and 3 combined broods respectively.  The first batch 
exhibited a salt test viability of 85% whereas the subsequent batches exhibited declining 
viabilities 52% and 41%.  Only the first batch was used to infect skipjack herring.  Because the 
skipjack herring required a large volume glochidial bath, there were not enough glochidia left 
over to infect additional fish species. 
Ten of the 20 mussels collected on June 23rd released at least some of their gill contents 
while being held in bags overnight.  Three mussels broods were dominated by fertilized eggs, 6 
broods were dominated by glochidia in membranes, 2 broods were dominated by free, mature 
glochidia, and the remaining broods were released as conglutinates.  The two broods dominated 
by free, mature glochidia were combined to create glochidial baths to infect all fish species other 
than skipjack herring.  Glochidial viability of baths made from these two broods was 67%.    
16 
 
Gill inspections (Table 2) performed at 4-6 days post infection on all species other than 
skipjack herring revealed no individuals with encapsulated glochidia.  In contrast, the gill 
inspection performed on the single skipjack herring on Day 11 post infection revealed numerous 
encapsulated glochidia (Fig. 5).  The excised gills from this fish produced 1,160 active juveniles 
over the course of 8 days (Fig. 6).  Gills of the remaining 7 skipjack herring in the communal 
holding tank that were examined on Day 19  revealed no encapsulated glochidia and thus were 
never excised  and held for juvenile production (Table 2).   
Siphoning of the tanks prior to Day 11 revealed no sloughed glochidia or juveniles, but 
the large amount of detritus on the tank bottoms prevented an accurate analysis of the samples.  
Samples collected via siphoning after Day 11 were cleaner but still difficult to analyze.  However 
I did observe one juvenile with its foot extended from samples taken on day 19 post infection 
indicating that juveniles were successfully produced by live fish in the communal tanks.  
Discussion: 
 As far as I am aware (Science Citation Index; keywords: Fusconaia ebena, Reginaia 
ebenus, host study, life history, skipjack herring, Alosa chrysochloris), this study represents the 
first successful collection, holding, and use of skipjack herring for a mussel host study.  Previous 
researchers either didn?t try to use skipjack herring due to difficulty in holding this species 
(Surber 1913) or were unsuccessful in their attempts.  For example, although not explicitly 
described in Hauswald (1997), the author hauled skipjack herring in rectangular coolers filled 
with river water (no hauling salts) with aeration.  All skipjack herring died before reaching the 
lab (Hauswald personal communication).   
It is very important to minimize stress during the process of capturing, hauling, and 
holding of fish for research.  Stress takes many different forms and can have many different 
17 
 
effects.  Some fish tend to swim into tank corners and cause physical harm to themselves as they 
beat against tank walls. Stress associated with temperature shock can suppress the immune 
system (Noga et al 2000).  Stress associated with handling or capture, can cause adrenaline to be 
released into the bloodstream, ultimately leading to loss of ions in the blood and problems with 
maintaining osmoregulatory homeostasis. (Ports et al. 2006).  Stress can also increase oxygen 
consumption rates and deplete energy reserves (Barton and Schreck 1987; Harmon 2009).  Scope 
for Activity (SFA) is the energetic difference between standard and active metabolism (Fry 
1971; Fry 1947) and represents the energy available for growth, reproduction, swimming, and 
dealing with stress.  Stress depletes the SFA, leading to a negative feedback between stress and 
energetic health (Barton and Iwama 1991).   
The high survival of skipjack herring through standard hauling and holding procedures 
demonstrated that they are not as sensitive to handling as commonly assumed.  In this study, we 
collected fish by hook and line rather than using standard electroshocking techniques because it 
was a much more efficient way to collect large numbers of skipjack herring (see Methods).  I 
was concerned that handling associated with landing of fish would be more stressful than 
electroshocking.  However, all fish collected in this manner survived collection, holding in the 
boat, and transport to the on-shore hauling tank.  Subsequent high survivorship showed that 
skipjack herring can be successfully collected and held for host assays as long as the following 
standard fisheries techniques are carefully utilized and followed: 1) use of round hauling tanks, 
2) use of professional hauling salts, 3) use of bubbled oxygen during hauling, and 4) use of large 
tanks with circular flow for holding.   
One thing I overlooked in this study was the amount of time necessary to establish 
sufficient bio-filtration in our large outdoor recirculating system.  I didn?t set up our biofilter far 
18 
 
enough ahead of time, due mainly to freezing temperatures prior to collection of skipjack herring 
in March.  I also salted our large outdoor recirculating system down to match the hauling tank 
waters? salinity and this caused what little biofilter I had established to crash before it eventually 
acclimatized to higher salinities.  This led to harmful spikes in ammonia and nitrites.  In 
subsequent trials I established the bio-filtration well ahead of time by seeding it with an 
ammonia solution and adding salt well ahead of the fish collection date so that a salt-tolerant 
biofilter was already well established by the time I added fish to the system. 
Although I was able to successfully hold skipjack herring in large communal tanks, the 
tanks were not optimal for host studies for several reasons.  Air blowers or large pumps are 
required to maintain constant water flow, generating high utility bills.  The tanks themselves are 
expensive and require a large amount of space, limiting the number available for host studies.  
Feeding and holding a large number of fish within a tank generates a lot of waste, which makes 
siphoning juveniles from the bottoms of the tanks problematic.  Also, the housing of multiple 
infected fish in a communal tank prevents the collection of transformation data from individual 
fish.    
 Gill excision represents a valuable, new technique to confirm problematic, large-river 
fishes as hosts for mussels.  I was able to produce 1,160 active ebony shell juveniles from half of 
the excised gills of a skipjack herring, demonstrating that it is a viable technique.  The main 
benefits of this technique are that it allows problematic large river fishes to be successfully 
utilized in host trials using a minimum number of large tanks, it allows for the collection of 
transformation data from individual fish, and it allows for the collection of data from fish dying 
during traditional host-trials that would otherwise simply be discarded.  Future studies 
demonstrating the success of this technique across a range of fish taxa, and comparing ?live gill? 
19 
 
to excised gill production are warranted.  Also, research is needed to determine the length of time 
glochidia need to develop on gills before they are competent to metamorphose from excised 
gills.  
This study represents the first laboratory determination of a host species for ebony shell.  
It appears to be a specialist on skipjack herring based on the evidence from this and previous 
studies.  Surber (1913) attempted artificial infections of ebony shell with a range of fishes 
obtained in the vicinity of Fairport Iowa (species list not provided in paper), but with no success.  
He did not attempt artificial infections with skipjack herring likely due to difficulty of handling.  
However he was able to obtain specimens of skipjack herring that were naturally infected with 
glochidia that he identified as ebony shell based on morphology.  Howard (1914c) nominally 
identified the glochidia of ebony shell encapsulated on wild-caught largemouth bass 
(Micropterus salmoides), black crappie (Pomoxis nigromaculatus), and white crappie (Pomoxis 
annularis).  However, subsequent attempts to confirm these hosts through laboratory methods 
ultimately failed.  Similar to Howard (1914c), I did not observe successful transformation of 
ebony shell on largemouth bass.  These results illustrate the danger in identifying hosts based 
solely on observations of wild-caught animals.  Glochidia may be misidentified based on 
morphological characteristics only, and observations of encapsulation do not necessarily indicate 
successful metamorphosis.   
Specialization of ebony shell on a large-river migratory fish species may explain the 
recent decline and current distribution patterns of this species.  In Alabama ebony shell have 
been extirpated in the Coosa River upstream of Jordan Dam likely due to the inability of skipjack 
herring to migrate upstream past the dams (Gangloff 2003).  The inability of skipjack herring to 
migrate upstream past dams (Coker 1930) could also explain the existence of relic populations 
20 
 
such as those observed upstream of Lock and Dam 19 on the upper Mississippi River (Kelner 
and Seitman 2000).  Maintenance and restoration of ebony shell populations will require 
restoration of longitudinal connectivity within river systems currently fragmented by dams.  
Although some small dams may be removed completely, other methods to increase fish passage 
may be more practical in many situations.  For example Ely et al. (2012) showed that ?attraction 
flow? created by pumps in lock chambers facilitated the upstream migration of Alabama shad 
(Alosa alabamae) through lock and dam systems in the Apalachicola River.  This, and similar 
strategies, may allow for passage of migratory fish species through some existing dams without 
requiring dam removal or installation of expensive fish passage structures.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21 
 
Ch. 2) Use of gill excision to identify hosts for Elliptio crassidens:  metamorphosis success 
and time of excision. 
Introduction: 
Dams have been implicated in the decline of many freshwater mussel species native to 
North America, particularly those which rely on migratory fish species as hosts for their parasitic 
larvae (Watters 1996; Vaughn and Taylor 1999).  Dams facilitate mussel declines by 1) 
interfering with dispersal potential of mussel larvae when host fish are present but constrained in 
their historic migration patterns (Watters 1996), and 2) by reducing or eliminating recruitment 
altogether when preferred host fish are prevented from migrating past dependent mussel beds 
(Douda et al. 2012; Hinck et al. 2012; Larinier 2000).  In some cases loss of fish hosts results in 
the creation of relic populations:  populations of large, old individuals with little to no 
recruitment that persist until they die out of old age.  
Elliptio crassidens (elephantear) (Lamarck, 1819) is a freshwater mussel species found in 
creeks and rivers with a moderate current (Heard 1979).  In larger impounded rivers it is 
typically found in the tail waters of dams as well as stable areas composed of firm sand and mud 
where there is little silt deposition (Williams et al. 2008).  It is a short-term brooder (tachytictic), 
reported to be gravid from April to June in the Mississippi River (Coker et al., 1921) and from 
April through August in the Apalachicola Basin (Brimbox and Williams, 2000; O?Brien et al., 
2003).  Similar to Reginaia ebenus (ebony shell; Chapter 1) recruitment of elephantear 
populations is thought to be dependent on availability of large-river, migratory fish hosts.  Alosa 
chrysochloris (skipjack herring) is a suspected host based on historic observations of natural 
infection (Howard 1914a).  It is typically anadromous but can complete its life cycle in fresh 
water if blocked from the sea (Etnier and Starnes 1993), spawning from early March through late 
22 
 
April on sand and gravel bars within large rivers.  Alosa alabamae (Alabama shad), a congener 
of skipjack herring, has also been suggested as a host species (Hauswald 1997).  Alabama shad is 
an anadromous fish species that migrates upriver from late February through April to spawn in 
habitats with coarse sand /gravel substrates and moderate current.   
Historically, elephantear was widely distributed in North America, reaching from 
Pennsylvania west to Wisconsin and from Quebec Canada all the way south to the panhandle of 
Florida (Parmalee and Bogan 1998).  In the early 1900?s it was so abundant in certain reaches of 
the Cumberland River that it was considered a nuisance species due to its shell being unsuitable 
for use in the button industry.  Culled shells from the button industry were so abundant, they 
were considered for use as flux in metal foundries along the river (Wilson and Clark 1914).  
Since that time, elephantear has declined in the northern portion of its range (Natureserve 2014).  
The northern decline of elephantear is thought to be due in large part to the inability of its 
host fish to migrate northward through numerous dams and impoundments.  In Minnesota, 
skipjack herring populations declined with the initiation of dam construction along the 
Mississippi River in 1913.  The species was considered extirpated from Minnesota after 1928, 
but has been collected in low numbers since 1986 (www.dnr.mn.us; Bell Museum of Natural 
History, fish collection database).  Skipjack herring is currently endangered in Wisconsin 
(www.dnr.wi.gov) and considered ?unrankable? in Missouri (http://mdc4.mdc.mo.gov/ 
applications/mofwis/Mofwis_Summary.aspx?id=0100086) due to a lack of information and/or 
substantially conflicting information about status or trends.  Alabama shad were previously 
numerous enough to support commercial fisheries as far north as Iowa, but are now considered 
globally imperiled (Natureserve, 2014).  Alabama shad are currently depleted in most of the 
Mississippi River basin (Lee et al. 1980, Page and Burr 1991) and have been extirpated from 
23 
 
Iowa (Harlan and Speaker 1969).  Missouri still hosts some of the last spawning populations in 
the Mississippi River (Pflieger 1997) but the species is not abundant and considered to be 
imperiled in the state (Missouri Natural Heritage Program).  Similar to its suspected hosts, 
elephantear is now endangered in Minnesota, Missouri, and Wisconsin, and is threatened in 
Illinois (Cummings 2011, USFWS 2006).  In Alabama, elephantear and skipjack herring are still 
abundant in the Tennessee River and parts of some other drainages (Williams et al. 2008; 
Natureserve 2014), but Alabama shad are rare (Mirarchi et al 2004).   
Although circumstantial evidence suggests that the decline of northern elephantear 
populations is due to dams and impediment of suspected host species migration, elephantear has 
never been successfully produced in host trials from any fish species (e.g. Hauswald).  This is 
likely due to elephantear being a host specialist: they do not transform on easily cultured fish 
species and their suspected hosts have never been successfully utilized in host trials due to 
difficulties in handling and holding.  In the previous chapter, I demonstrated that skipjack herring 
could be successfully collected and held if established fish hauling protocols were scrupulously 
utilized and followed.  I also demonstrated the utility of a promising new technique ? gill 
excision ? for conducting host trials with problematic fish species.     
 The success of this technique raised further questions regarding its viability for other fish 
and mussel species, as well as the optimal time during which gills should be excised in order to 
ensure transformation of encapsulated glochidia.  There is presumably a ?window of 
opportunity? in which the glochidia are competent enough to metamorphose but have not all 
dropped off of the gill tissue.  Prior to this time, glochidia may not be sufficiently developed to 
metamorphose from dead gills.  In this chapter I describe my attempts to refine the gill excision 
24 
 
technique using elephantear and two of its suspected host fish species:  skipjack herring and 
Alabama shad.   
Objectives were as follows: 
1) Determine if the standard fisheries techniques successfully utilized for hauling, holding 
and infecting skipjack herring will also work for Alabama shad. 
2) Determine whether skipjack herring and Alabama shad are viable hosts for elephantear. 
3) Determine the period of time within which gills can be excised from a host and produce 
juvenile mussels. 
4) Determine the length of time glochidia will metamorphose from gills after excision. 
Methods: 
Collection and holding of study animals  
 We utilized the same large, outdoor holding tanks described in chapter one for this study.  
In order to establish a functioning biofilter that was already adjusted to enhanced salinity levels 
prior to stocking fish in the tanks, we filled all tanks with dechlorinated tap water 3 weeks prior 
to the first fish collection and brought them up to a salinity of 3-5 ppt with Morton Pool salt. 
Commercially available ammonia (RED MAX brand) was added daily to each tank to maintain a 
concentration of 3 ppm until a nitrite pulse was observed, indicating the establishment of a 
functioning biofilter microbial community.  Ammonia additions were then halted.   
 Three species of fish representing 2 families (one suspected host family and one 
suspected non-host family) were collected for use in this host trial.  Bluegill (Lepomis 
macrochirus) were collected at the Auburn University South Auburn Fisheries Research Station 
(SAFRS) via trapping in a reservoir that did not contain mussels.  Bluegill were held in a 
fiberglass tank (Length 1.5m x Width .55m x Height .35m) with a volume of 290 liters and a 52 
25 
 
liter biofilter at ambient room temperature ( ~20?C) for approximately 1 month prior to infection.  
Bluegill were fed frozen bloodworms and flake food three times per week and water quality was 
checked weekly with aquarium water test strips.  Water exchanges were performed if nitrates 
exceeded 40 mg/L. 
 Alabama shad were collected below the Jim Woodruff Lock and Dam (JWLD) during 
their annual spring spawning run on March 22, 2012 via electrofishing by the Georgia 
Department of Natural Resources.  The JWLD impounds the confluence of the Chattahoochee, 
and Flint Rivers. Fish were transported to SAFRS in a 1900 liter, round, fish hauler (Warm 
Springs Fish Hatchery, Warm Springs, GA) filled with water from the collection site that was 
treated with Kent Sure-haul and Morton Pool salt to reach a final salinity of ~5?.  Oxygen was 
gently bubbled into the tank through an ultra-fine pore ceramic plate air diffuser that sat nearly 
flush with the bottom of the tank to minimize contact with the fish being hauled.  Before shad 
were placed into the round fish hauler, they were weighed and measured by the Georgia DNR as 
part of a separate study.  Water quality parameters (salinity, dissolved oxygen, and temperature) 
in the hauling water were measured just prior to, during, and just after transport to SAFRS.  
Length of the trip was 4 hours and 40 minutes.  Upon arrival (3:00pm CST) at SAFRS the 
number of live and dead shad was recorded, and live shad placed into a large 8500 liter round 
(3m diameter, 1.2m height) fiberglass holding tank incorporated within a larger recirculating 
system (total volume of ~23000 liters).  The holding tank was equipped with a central biofilter 
constructed from two 120 liter trashcans filled with BIO-FILL? (Aquatic eco-systems) 
connected at the bottoms and weighed down with cinder blocks.  The internal biofilter had a 2? 
PVC pipe fitted inside the two trashcans to serve as an airlift inside to provide water flow across 
the bio filter media.  The internal biofilter was utilized for backup bio filtration/aeration in the 
26 
 
case of pump failure for the main system.  The internal bio filter also acted as a central baffle that 
caused the inflow from the larger recirculating system to create a circular current within the 
holding tank.  The holding tank was covered with off-white shade cloth (SYNTHESIS 
COMMERCIAL 95?) stretched across a PVC frame to prevent fish from jumping out and to 
keep the system relatively cool in full sun.  All of the other tanks within the larger recirculating 
system were also covered with corrugated plastic roof panels to keep the temperature down and 
reduce algal blooms.  The bio filtration for the overall recirculating system was provided by two 
external sumps that were filled with BIO-FILL? (Aquatic eco-systems) bio media and had a 
volume of 2,300 liters combined (.65m wide x .47m deep x 3.65m long each).  Water quality was 
assessed weekly and partial water changes were implemented whenever nitrates rose above 
40mg/L.  Alabama shad were fed a biweekly ration of mosquito fish (Gambusia affinis) collected 
from an earthen pond at the SAFRS.  
 Skipjack herring were collected in late March, 2012 during the annual spring spawning 
run in the tail waters of the Eagle Phenix Dam in Columbus Georgia.  Forty four skipjack herring 
were captured by hook and line using light colored, plastic grubs and jig heads on March 20.  
Skipjack herring were quickly reeled in and immediately placed in an oval (~570 liter) stock tank 
filled with water from the catch site.  To avoid overcrowding in the stock tank, skipjack herring 
were taken back to the boat ramp and placed in a round hauler in batches of 10 or less.  The 
round fish hauler (~1,900 liters) was filled with water from the catch site and salted down with 
Kent Sure-Haul and Morton pool salt (uniodized) to a salinity of approximately 5?.   Oxygen 
was bubbled into the hauler using an ultra-fine ceramic plate air diffuser that was flush with the 
bottom of the tank to reduce contact and injury of the fish being transported.  Once all 44 fish 
had been transported to the hauler they were driven back to the SAFRS (~1 hr. transport time).  
27 
 
Upon arrival at the SAFRS, survival and mortality was recorded and live fish were split between 
two, outdoor, rectangular tanks (3.5m L x 1.8m W x 0.9m H) situated under an awning. Tanks 
had a volume of 5,700 liters each and were covered with 1? mesh fish netting to prevent skipjack 
herring from jumping from the tanks.  Both tanks had been drained and thoroughly cleaned of all 
sediment using a wet-dry vacuum prior to filling.  Each tank was fitted with an internal central 
longitudinal baffle made from lengths of 10? diameter PVC pipes and corrugated plastic roof 
panels.  Airlifts made from lengths of 4? diameter PVC pipe fitted with a 90? fitting at the top 
horizontal to the surface of the tank produced a circular current around the central baffle.  A 920 
liter, external bio filter was connected to each rectangular tank via a 1.25? diameter PVC airlift 
with a distribution pipe leading down into the bottom of the bio filter which caused the water to 
pass upward through the bio media.  The water then flowed back into the main rectangular tank 
though a 2? diameter PVC overflow.  The bio reactors were filled with KALDNES? filter 
media, and larger bio barrels to provide plentiful surface area for the nitrifying bacteria.  A large 
airstone was affixed to a 2? diameter PVC standpipe running up the center of these bio reactors 
creating an upwelling current which agitated and aerated the bio media.  Water quality 
parameters were measured weekly and partial water changes (approximately 30-50%) were 
implemented whenever nitrates rose above 40mg/L.  Skipjack herring were fed bi-weekly ration 
of mosquito fish collected from an earthen pond at the SAFRS.       
Elephantear were collected on April 3rd of 2012 from Chewacla creek in Macon County 
by tactile searching and brought to the bank.  Mussels were gently pried open with reverse pliers 
and identified as gravid if the outer gills appeared creamy white and slightly inflated compared 
to non-marsupial gills (Britton and Fuller, 1980; Williams et al., 2008).  Nine gravid mussels 
were identified and placed into individual 1 gallon Ziploc? bags filled with creek water 
28 
 
(temperature 21.1?C pH: 6.8). Bags were held in a cooler and transported (~20 minutes) back to 
the SAFRS.  Upon arrival at the laboratory the bags were placed into a recirculating aquarium 
system maintained at 18?C and an airstone was placed into each bag to provide aeration.  Three 
of the mussels began releasing glochidia within several hours of arriving at the lab and were 
utilized for host trials that same day.  The remaining mussels were held (in bags) overnight in the 
system.  Only two still retained glochidia the following day and were suitable for use in host 
trials. 
Host trials 
One week prior to initiation of the host trials, fish were transferred among tanks such that 
rectangular tank #1 held eight skipjack herring and eight Alabama shad while rectangular tank #2 
held seven skipjack herring and seven Alabama shad.  These fish were used in the host trials.  
The remaining fish were transferred to the large circular tank.  To avoid negative effects of 
salinity on glochidia, the salinity of the two rectangular tanks was gradually lowered via water 
changes over a one week period until salinities were approximately <1 ppt.   
 To initiate host trials, glochidia were extracted from gravid mussels by puncturing the 
marsupia with a 22 gauge needle and then flushing out the glochidia (Dodd et al. 2005) into 
separate beakers brought up to a final volume of  2,000 mL with AFW.  Glochidial viability of 
each brood was evaluated by means of a salt test (Fritts et al. 2014).  Each 2,000 mL batch of 
glochidia was mixed using a plunger type stirrer to keep the glochidia uniformly suspended in 
the solution while avoiding a vortex.  Ten 200 ?L subsamples were withdrawn and placed into a 
10-well glass plate.  Each subsample was examined under a dissecting microscope, and the 
number of open and closed glochidia recorded.  Several drops of 35? synthetic salt water were 
then added to each subsample and the number of open and closed glochidia was quantified again. 
29 
 
The proportion of viable glochidia in each brood was calculated as:  [(number open initially) ? 
(number open after NaCl)] ? (total number of glochidia) 
The total number of viable glochidia in each brood was calculated as:  
 Vn = [(G/S) *V]* P 
Where Vn = total number of viable glochidia in a brood, G = number of glochidia (open plus 
closed) in all subsamples combined, S= combined volume of all subsamples (ml), V = total 
volume of sample (ml), and P = proportion of viable glochidia. 
Infection 
 Fish were infected on two consecutive days.  On April 3, 2012 a 560 L (280 L holding 
tank water plus 280 L dechlorinated tap water) glochidial bath was made to a concentration of 
4,000 viable glochidia per liter using the combined glochidia from 3 elephantear broods.  Seven 
skipjack herring, and seven Alabama shad were held simultaneously in the bath for 15 minutes. 
The bath was covered with 1? fish netting and a layer of shade cloth to minimize disturbance of 
the fish.  Two large air stones provided aeration and helped keep glochidia in suspension.  
Following the 15 minute holding period, fish were transferred back to the original holding tank 
and the glochidial bath discarded 
 On April 4, 2012 a second glochidial bath was made as previously described, and eight 
skipjack herring, eight Alabama shad, and five bluegill infected simultaneously.  Following the 
15 minute holding period, skipjack herring and shad were transferred back to their original 
holding tank.  Bluegill were transferred to individual 1.5 L AHAB tanks in a recirculating 
system.  AHAB tanks were filled with dechlorinated tap water and maintained at 18-21?C.  
Water flowing out of each tank passed through 105 ?m filter cups to catch any sloughed 
glochidia or juvenile mussels, then drained into a common biofilter.  From the biofilter, water 
30 
 
was pumped through a coarse mechanical filter fitted with an ultraviolet (UV) sterilization, 
through a chiller, through a fine (100 ?m) filter, and then through a manifold system for 
redistribution through the AHAB tanks.   
To determine the window of opportunity within which glochidia could successfully 
metamorphose from excised gill tissues of skipjack herring and Alabama shad, we excised the 
gills of one infected fish/species/tank every 2-3 days from day 5 through day 36 or until we ran 
out of live fish, whichever came first.  To excise gills we sacrificed the fish via quick severing of 
the spinal cord with scissors.  The first two gill arches on each side of the fish were then removed 
with a pair of scissors and forceps and placed immediately into a 1.5 L AHAB tank that was 
labeled according to fish species and date of excision.  
Gills were not excised from bluegill.  Filter cups for all bluegill in the AHAB system 
were checked for sloughed glochidia and juvenile mussels 2 days post infection and every day 
thereafter up until day 16.  On day 16 all bluegill were removed from their tanks and their gills 
inspected.  Encapsulated glochidia were not observed on any bluegill at this time (see results), 
but they were returned to their tanks and filter cups were checked again on day 23 as a 
precaution.   
Filter cups associated with the AHAB tanks containing excised gills were checked 1 days 
post excision and every other day thereafter for sloughed glochidia and juvenile mussels.  All 
filter cups were monitored until no glochidia or juveniles were found during 5 sequential 
inspections.  Filter cups were analyzed by first rinsing the contents of the filter cups into a 
counting tray and inspecting the counting tray under a dissecting microscope (10X 
magnification).  Juveniles were identified as such by the presence of a foot and were recorded as 
31 
 
active juveniles (locomotion observed) or questionable juveniles (closed and with tissue, but no 
observed movement or activity).  Dead glochidia or juveniles were recorded as empty shells. 
To determine whether juveniles metamorphosed off of live fish in the holding tanks prior 
to gill excision, we siphoned a 1 m2 area of the bottom of tank # 2 twenty five days after 
infection.  Siphoned water was first filtered through a 100 ?m filter bag to retain glochidia and 
juveniles (130-150 ?m length x 141-160 ?m height).  Material collected on the 100 ?m filter was 
then passed through a 200 ?m filter screen to remove large particles.  The filtrate was then 
brought to a volume of 125 mL and mixed with a plunger.  Juveniles and glochidia were 
enumerated in 15 consecutive 5 ml subsamples, using a dissecting microscope equipped with 
cross-polarized lighting to make the shells stand out from sediment and debris.  Juvenile density 
of the holding tank (#/m2) was calculated by the following formula: 
D = J * (V/v) / SA  
Where D = juvenile density; J = total # juveniles counted; V = total volume of sample; v = 
volume of subsamples; and SA = surface area of holding tank sampled. 
Results:  
Hauling success  
 Use of a round, circular hauler and hauling salts proved successful for Alabama shad as 
well as for skipjack herring.  Survival from capture to arrival at the SAFRS was 96% (29/30 fish) 
for Alabama shad after a 5 hour haul and 97% (43/44 fish) after a 1 hour haul for skipjack 
herring (Table 3).   
32 
 
Glochidial viability of gravid elephantear 
 The salt test assay indicated glochidial viabilities of 91, 96, and 96% for the three mussel 
broods used on April 3rd, and 94, and 97% for the two mussel broods used on April 4th.  Brood 
size ranged from 386,000 to 1,506,000 viable glochidia per mussel (Table 4). 
Transformation Success 
 Bluegill produced an average of 83.8 ? 49.2 sloughed glochidia per fish, but no viable 
juveniles.  Over 92% of all glochidia had sloughed off of the bluegill by the second day post 
infection (Table 5).  On day 16, direct observation revealed no encapsulated glochidia within the 
gill filaments of any bluegill, nor had any sloughed glochidia or juveniles accumulated in filter 
cups during the subsequent week (June 23 observation; Table 5). 
 Gill arches of Alabama shad excised on days 5, 6, 7 and 8 after infection produced no 
juveniles.  Gills excised on days 10, 13, and 14 produced 2, 4, and 9 active juveniles 
respectively.  Gills excised on days 17, 20, and 23 produced a total of 326, 328, and 293 active 
juveniles respectively.  Gills excised on day 28 produced only 5 empty shells.  Gills excised on 
days 30 and 36 did not produce any live, dead, or questionable juveniles (Fig. 1). Similarly, 
transformation efficiency of excised gills (active juveniles / total gill production after excision) 
was zero for gills excised prior to day 10, less than 1% from gills excised on days 10-14, 50-60% 
for gills excised on days 17-23, and NA for gills excised on or after day 28 (Fig. 1). 
 Empty shells (dead) were typically produced by gills over a period of 2-3 weeks after 
excision, but the majority of active juvenile production occurred within the first 10 days of 
excision, with peak juvenile production observed within the first 4 days of excision (Fig. 2).  
Relatively large numbers of ?questionable? juveniles were produced by gills excised on days 13 
33 
 
and 14,  while juveniles produced by gills excised on days 17, 20, and 23 were nearly all active 
(Fig. 2). 
 Gill arches of skipjack herring that were excised on days 5 - 13 produced no active 
juveniles.  Gills excised on day 16 produced a total of 85 active juveniles with a transformation 
efficiency of 28% (Fig. 3).  Questionable juvenile production dominated the first 3 days after 
excision.  Active juvenile production occurred over a ten day period with peak production 
occurring on days 6-8 post excision.  Empty shell production was observed for 19 days post 
excision (Fig. 3).  Due to mortality of skipjack herring during the holding period prior to the 
period of observed juvenile production, no infected skipjack herring remained in the holding 
tanks after day 16. 
 Siphoning the bottom of holding tank # 2 on day 25 following infection yielded an 
estimated juvenile density of 187 / m2, or 1,134 juveniles per the entire 6.3 m2 of tank bottom.  
Because the tank sides sloped at the bottom and juveniles likely crawled partly up the sides of 
the tank, this should be viewed as a conservative estimate of juvenile production.  Regardless of 
the total number of juveniles produced, observation of active juveniles on the holding tank 
bottom demonstrates that juveniles were produced by live fish, most likely by both species.  
Because all deaths of infected fish occurred prior to the period of active juvenile production, and 
dead fish were quickly removed and discarded, juveniles were most likely produced from live 
fish rather than from the gills of fish that died in the tanks.  
Discussion:  
After capture by electrofishing, Alabama shad were weighed in a plastic battery box 
before being transferred into the round hauler.  Despite this initial rough handling, the 
survivorship of the Alabama shad was high.  This showed that, similar to skipjack herring, 
34 
 
Alabama shad are less delicate than commonly assumed.  Alabama shad can successfully be 
captured, hauled, and held for use in host fish assays as long as the following standard fisheries 
techniques are followed: 1) use of round hauling tanks, 2) use of professional hauling salts, 3) 
use of bubbled oxygen during hauling, and 4) use of large tanks with circular flow for holding.   
 The collection of gravid females from our local population of elephantear coincided with 
the reported periods of gravidity in the Mississippi river between April and June (Coker et al. 
1921) as well as in the Apalachicola river basin between April and August (Brim Box and 
Williams 2000; O?Brien et al., 2003).  Although some of the mussels held in Ziploc bags 
released glochidia into their bags within several hours of collection, it did not seem to affect the 
quality of their glochidia as viability of all broods was > 90%.   
 To our knowledge this is the first laboratory determination of host fish identity for 
elephantear. Skipjack herring have long been suspected as a host (Howard 1914a, Fuller 1974), 
but never confirmed due to the inherent difficulty in working with this fish species.  In a previous 
study attempting to identify fish hosts for elephantear, 7 different families (Catostomidae, 
Centrarchidae, Clupeidae, Cyprinidae, Ictaluridae, Percidae, and Sciaenidae) were tested but 
none produced juveniles (Hauswald 1997).  Similar to that study, we observed no production of 
elephantear from bluegill.  The Clupeidae species tested by Hauswald (1997) was Dorosoma 
cepedianum, whereas we tested skipjack herring and Alabama shad.  Our results combined with 
that of Hauswald (1997) suggests that elephantear is a host specialist on large river migratory 
fish of Clupeidae, particularly those in Alosa.   
Determination of migratory Alosa as hosts for elephantear has important implications for 
conservation.  While elephantear currently has lowest concern status on the IUCN red list 
(Natureserve 2014) and is considered stable across its entire range (Williams et al. 1993), 
35 
 
northern populations in Wisconsin and Minnesota have virtually disappeared (Natureserve 2014; 
Hauswald 1997; Mathiak 1979; Sietman 2003).  Additionally, many extant populations in the 
southern part of its range appear to be relic populations with little to no recruitment (Miller and 
Payne 2000, Gangloff 2003).   Persistence of old individuals coupled with a lack of recruitment 
can be explained by blockage of upstream migration of large-river migratory fish hosts by dams 
(Kelner and Sietman 2000; Watters 1996; Fuller 1974).  My study suggests that installation of 
fish passage mechanisms for skipjack herring and Alabama shad have the potential to restore 
elephantear populations upstream of dams.  
The gill excision technique described in this study is a valuable tool for conducting host 
assays with problematic fish species that require large tanks for holding.  While large tanks can 
be successfully siphoned for juveniles if kept very clean, there are disadvantages to this method. 
Most importantly, large tanks take up a lot of space and many research facilities are space-
limited.  This requires that either multiple individuals of the same species be held in the same 
tank, or, as was the case in this study, multiple species may be held in the same tank.  Siphoning 
of tank bottoms precludes the collection of transformation data from individual fish.  Gill 
excision allows for collection of data from individual fish even if those fish were housed together 
in multi-species tanks.   
A critical aspect of the gill excision technique, however, is that it is only successful 
within a limited time period.  This window of opportunity starts when encapsulated glochidia 
have developed for a sufficient amount of time that they can metamorphose from excised gills, 
and ends when they have all metamorphosed from the gills of live, infected fish.  In this study, 
the window of opportunity for gill excision lasted for nearly two weeks (days 10 ? 25 post 
infection) for Alabama shad, with greatest transformation success occurring from gills excised 
36 
 
on days 17-23, from which 50-60% of encapsulated glochidia successfully metamorphosed into 
active juveniles.  Due to mortality of skipjack herring in our trial we were unable to determine 
the full window of opportunity but did show that the optimal period of excision began at 16 days 
post infection.   
Surprisingly, gills did not rapidly decompose to the point where all encapsulated 
glochidia metamorphosed or sloughed off within a few days.  Although the majority of active 
juveniles were produced within the first four days, successful metamorphosis occurred for as 
long as 1-2 weeks after excision and dead shells sloughed off of gills for 3 weeks or more, post 
excision.  We hypothesized that encapsulated glochidia were somehow slowing the 
decomposition rate of gill tissue, giving them time to complete metamorphosis.  This hypothesis 
is tested in Chapter 3. 
 
37 
 
Ch. 3) Effect of glochidial infection on decomposition rate of excised fish gills   
Introduction: 
 Parasites can modify their environment (host) in various ways in order to increase the 
likelihood of their survival and reproduction.  Some parasites are able to modify the behavior of 
their hosts in ways that make them more susceptible to predation by their final host (Combes 
2001; Moore 2002; Lafferty 1999).  Other parasites are able to modify their environment down 
to the cellular level in order to make their ultimate survival and reproduction more likely. For 
example, certain intracellular parasites such as Cryptosporidium parvum, Leishmania spp., 
Trypanosoma cruzi, Thelleria spp., Toxoplasma gondii, and Plasmodium spp. are able to inhibit 
apoptosis of cells in which they inhabit (Carmen et al. 2007). 
 During experiments described in the previous two chapters, an interesting observation 
was made regarding the gill tissue that had been excised and held within the AHAB tanks.  Gill 
tissue seemed to degrade very slowly in the AHAB system and was still recognizable as gill 
arches with filaments for > 7days after excision.  This raised several questions concerning 
potential effects of encapsulated glochidia on parasitized gill tissue.  Perhaps the glochidia were 
somehow altering their environment in a way that prevented or slowed down the gill tissue 
breakdown which would allow them to persist in the tissue until metamorphosis.  This would be 
advantageous to glochidia whose host died before the glochidia were able to metamorphose from 
live gills.   
 In this chapter I tested whether or not encapsulated glochidia slow the decomposition rate 
of the excised gill tissue.  I hypothesized that gill tissue infected with glochidia would 
decompose more slowly than uninfected gills.   
 
38 
 
Methods: 
Collection and holding of study animals. 
 I utilized Lampsilis straminea (southern fatmucket) and one of their hosts, Micropterus 
salmoides (largemouth bass), in this study because both species are abundant, easily collected, 
and easily held in a laboratory environment.  Largemouth bass were collected from a water 
supply reservoir on the North Auburn Fisheries Research Station (NAFRS) that contains few to 
no Unionids.  This increased the chance that our wild-caught host fish were na?ve and had not 
acquired immunity to mussel glochidia (Rogers and Dimock 2003).  Small fish (approx. 5-6?) 
were used to facilitate capture, hauling and holding.  Fish were captured by use of a large seine 
net and transported back to the South Auburn Fisheries Research Station (SAFRS) in large 140 L 
coolers. 
 Bass were held individually in 3 L tanks within a recirculating AHAB tank system for 
approximately 1 month prior to infection.  Water flowing out of each AHAB tank passed through 
a filter cup (105?m mesh) to retain any sloughed glochidia or juveniles.  Water then drained into 
a common sump containing bio barrels that served as a biofilter.  From there, water was pumped 
through a coarse mechanical filter that also contains a UV sterilization bulb and then through a 
fine mechanical filter (100?m) to catch any stray glochidia or juveniles that had bypassed the 
initial filter cup.  Filtered water was then distributed to AHAB tanks via a manifold.  
Pond temperature during the time of fish collection was 11?C.  Water in the AHAB 
system was originally brought down to this temperature to avoid temperature shock to the fish, 
and subsequently raised 1 degree per day until the system reached 21?C  and then held constant 
for the remainder of the trial.  Bass were fed mosquito fish (Gambusia sp.) on a bi-weekly basis.  
39 
 
Water quality was monitored on a weekly basis using aquarium water test strips and water 
exchanges were performed when nitrate levels exceeded 40 mg/L.   
 Gravid Lampsilis straminea were collected in April, 2013 from Line Creek (Union 
Springs, Bullock County, Alabama) by tactile searching of the substrate.  Mussels were 
embedded in a sandy silt substrate approximately 1-2 meters deep in a fairly swift current.  
Southern fatmuckets are sexually dimorphic, allowing for easy identification of females.  
Mussels were identified as gravid by the presence of swollen marsupial gills, creamy white to tan 
in color (Williams et al. 2008).  Three gravid mussels were collected and placed in 1 quart Ziploc 
bags filled with water from the collection site.  The bags were held in a cooler and transported 
(~1 hour) back to the Auburn University South Auburn Fisheries Research Station (SAFRS).  
Upon arrival the bags were placed in a recirculating aquarium system maintained at 18?C with an 
airstone placed into each bag to provide aeration and water circulation. These mussels were held 
overnight and used the following day for the trial.    
Host trials 
 Glochidia were extracted from the three gravid mussels using a 22 gauge needle attached 
to a syringe filled with Artificial Fresh Water (AFW).  Water was injected into marsupia to flush 
glochidia into individual 1,000 mL beakers (one beaker per mussel) and the resultant glochidial 
solutions were diluted up to 1,000 mL.   
 Glochidial viability of each brood was determined via a standard salt test (Fritts et al. 
2014). Each 1,000 mL beaker was mixed using a plunger style stir rod to prevent a vortex and to 
ensure even distribution of glochidia throughout the beaker.  Ten 200 ?L subsamples were 
withdrawn from a given beaker and placed into a 10-well glass plate.  The number of open and 
closed glochidia were counted and recorded in each subsample using a dissecting microscope.  
40 
 
 Several drops of 35? synthetic salt water were then added to each subsample and the 
number of open and closed glochidia was counted again.  The proportion of viable glochidia in 
each brood was calculated as: [(number open initially) ? (number open after NaCl)] ? (total 
number of glochidia). 
The total number of viable glochidia in each brood was calculated as:  
Vn = [(G/S) *V]* P 
Where Vn = total number of viable glochidia in a brood, G = number of glochidia (open plus 
closed) in all subsamples combined, S= combined volume of all subsamples (ml), V = total 
number of mL in solution, and P = proportion of viable glochidia. 
Infection 
  All three beakers containing glochidia were combined to make a master stock solution.  
One liter of the master stock solution was diluted to 19 L to create a glochidial bath at a 
concentration of 4,000 glochidia/L.  Six largemouth bass were placed in this stock solution for 
15 minutes before being transferred back to their respective AHAB tanks.  Another group of 6 
bass was placed in a bucket containing 19 L of AFW with no glochidia in order to expose them 
to similar handling stress while remaining free of glochidia infection.  After 15 minutes, 
uninfected largemouth bass were also returned to their respective tanks and all tanks labeled as to 
whether they contained infected or uninfected fish.  Fish were then held in AHAB tanks for 7 
days to allow sufficient time for encapsulation, but not enough time for metamorphosis.  
Previous assays showed L. straminea glochidia metamorphosed from live largemouth bass 14-16 
days after infection at ~21 C (M. Hart, unpublished data).   
41 
 
Gill excision 
 At day 8 post infection, all fish were removed from their AHAB tanks and quickly 
euthanized by severing of the spinal cord with surgical scissors.  Once fish were dead the three 
outermost gill arches on each side of each fish were excised using forceps and cut in half to 
produce 12 ?gill units? (Figure 1).  The gill units were then returned to their respective AHAB 
tanks and fish were discarded.  Three gill units were taken out of each AHAB tank on days; 0, 3, 
6, and 12 days post excision and preserved in a 10% neutral buffered formalin solution (NBF).   
To quantify decomposition, each preserved gill unit was examined under 40X magnification,   
and the number of identifiable lamellae on three individual gill filaments was recorded.  (Figure 
1).  The results of this analysis were analyzed using a Two-way ANOVA (Minitab 16).  
Results: 
Glochidial viability of L. straminea 
 Glochidial viabilities of the three broods used to make up the master stock solution were 
94, 88, and 67%.  Brood sizes ranged from 32500, to 111500 viable glochidia per gravid mussel 
(Table 6). 
Juvenile production 
 No juveniles were produced from infected gills excised on day 8 and held in the AHAB 
chambers for up to twelve additional days. 
Gill decomposition 
 Gill units from all time treatments appeared to the naked eye to be in good condition 
(identifiable gill filaments).  However upon examination with a microscope (40X) it became 
apparent that the tissue had decayed substantially.  After day 6 post-excision, identifiable 
lamellae were no longer present (Figure 2).  
42 
 
Results of gill unit lamellae counts 
 There was no significant difference in lamellae counts between infected and non-infected 
gills through time (Fig. 3; p = 0.472)  
Discussion: 
  My hypothesis that encapsulated glochidia slowed the decomposition rate of gill 
filaments was rejected.  Within 6 days gill lamellae had decomposed to the point where they 
could no longer be identified, regardless of treatment.  Gill units at every time treatment however 
still appeared to be in ?good shape? with the naked eye and also appeared to be decomposing 
slowly as observed in Chapters 1 and 2.  
 The slow breakdown of gill tissue observed in chapters I and II may have been due to 
constant flow of clean water over the gills.  During the gill decomposition experiment, one of the 
manifold valves became clogged and the associated AHAB tank lost flow for ~ 3 days.  The gill 
tissue in this tank quickly became covered in bacteria and fungi, and degraded to the point where 
it became unidentifiable as gill arches with filaments.    
 This chapter and the chapters preceding it have shown that glochidia can successfully 
metamorphose off of dead gill tissue from their host fish in the lab.  In nature, metamorphosis of 
glochidia from an infected fish that died is possible, but would likely require specific conditions.  
First, the infected fish would have to die at a time that coincides with the window of opportunity 
as described in chapter 2.  The timing and length of this window likely varies with species and 
temperature, but I show that it can last for a week or more.  Second, based on the anecdotal 
observation of this chapter, the fish may have to die and be retained in a riffle or other location 
that provides a constant flow.  Additionally, the fish may have to die with its gills flared so as to 
expose them to water flow.  A congruence of all of these circumstances may be unlikely, but I 
43 
 
have occasionally observed recently dead fish, with gills flared, in flowing sections of local 
streams.  It is unknown whether these conditions occur often enough to provide substantial 
recruitment from dead gills in natural systems.  However, as discussed in previous chapters, even 
if not likely to occur in nature, the ability of glochidia to metamorphose from dead gills allows 
for a valuable research tool in mussel host assays.  Fish that happen to die in routine host trials 
may still be able to provide valuable data.  
44 
 
Literature Cited 
Barton, B. A., and G. K. Iwama. 1991. Physiological changes in fish from stress in aquaculture 
with emphasis on the response and effects of corticosteroids. Annual Review of Fish 
Diseases, pp 3-26. 
Barton, B. A., and C. B. Schreck 1987. Metabolic Cost of Acute Physical Stress in Juvenile 
Steelhead. Transactions of the American Fisheries Society. 116:2 pp257-263. 
Brim Box, J., and J. D. Williams. 2000. Unionid mollusks of the Apalachicola Basin in Alabama, 
Florida, and Georgia. Alabama Museum of Natural History Bulletin 21:1-143.  
Britton, J. C., & S. L. H. Fuller. 1980. The freshwater bivalve Mollusca (Unionidae, Sphaeriidae, 
Corbiculidae) of the Savannah River Plant, South Carolina. Savannah River Plant 
Publication SRO-NERP-3. Department of Energy, Aiken, South Carolina, 37pp. + 5 
maps.  
Boyd, C. E. 1990. Water Quality in Ponds for Aquaculture. Birmingham Publishing Co.  
Boschung J.R., H.T., R. L. Mayden, E.O. Wilson, and J.R. Tomelleri. 2004. Fishes of Alabama. 
Smithsonian Books.  
Cummings, K. 2011. Elliptio crassidens. In: IUCN 2013. IUCN Red List of Threatened Species. 
Version 2013.2. <www.iucnredlist.org>. Downloaded on 29 May 2014. 
Carmen, J. C., and A. P. Sinal. 2007. Suicide prevention: disruption of apoptotic pathways by 
 protozoan parasites. Molecular Microbiology 64(4), 904-916.  
Collins, J. L. and H. H. Hulsey. 1963. Hauling Mortality of Threadfin Shad Reduced with M.S. 
222 and Salt. The Progressive Fish-Culturist, 25:2, pp 105-106. 
Combes, C. 2001. Parasitism: the ecology and evolution of intimate interactions. University of 
 Chicago Press, London, UK. 
45 
 
Coker, R. E. 1930. Keokuk Dam and the fisheries of the upper Mississippi River. Bull. Bur. Fish. 
45:87-139. 
Coker, R. E., A. F. Shira, H. W. Clark, and A. D. Howard. 1921. Natural history and propagation 
of fresh-water mussels. Bulletin of the U.S. Bureau of Fisheries 37(1919-1920):77-181, 
17 plates. 
Dodd, B. J., M. C. Barnhart, C. L. Rogers-Lowery, T. B. Fobian, and R. V. Dimock Jr. 2005. 
?Cross-resistance of largemouth bass to glochidia of unionid mussels. Journal of 
Parasitology 91(5): 1064-1072. 
Douda, K.P. Hork? and M. B?l?. 2012. Host limitation of the thick-shelled river mussel: 
identifying the threats to declining affiliate species. Animal Conservation 15:536-544. 
Ely P.C., S.P. Young, and J.J. Isely. 2008. Population Size and Relative Abundance of Adult 
Alabama Shad Reaching Jim Woodruff Lock and Dam, Apalachicola River, Florida. 
Journal of Fisheries Management 28:827-831.  
Etnier, D. A. and W. C. Starnes. 1993. The Fishes of Tennessee. The University of Tennessee 
Press, Knoxville, Tennessee, USA. 681pp.  
Fritts, A. K., M. C. Barnhart, M. Bradley, N. Liu, W. G. Cope, E. Hammer, and R. B. Bringolf. 
2014. Assessment of toxicity test endpoints for freshwater mussel larvae (Glochidia) 
Environmental Toxicology and Chemistry, Vol. 33, No. 1, pp. 199-207. 
Fry, F. E. J. 1947. Effects of the environment on animal activity. University of Toronto 
Biological Series 55:1-62. 
Fry, F. E. J. 1971. The effect of environmental factors on the physiology of fish. Pages 1-99 in 
W. S. Hoar and D. J. Randall, editors. Fish physiology, volume 6. Academic Press, New 
York. 
46 
 
 Fuller, S. L. H. 1974. Clams and mussels (Mollusca: Bivalvia). Pollution ecology of freshwater 
invertebrates. C. Hart and S. L. H. Fuller, editors. Academic Press, New York, 215-273. 
Gangloff, M. M.  2003.   The status, physical habitat associations, and parasites of freshwater 
mussels in the upper Alabama River Drainage, Alabama.  Ph.D. Dissertation, Auburn 
University. 
Haag, W.R. and M.L. Warren, Jr. 1998. Role of ecological factors and reproductive strategies in 
structuring freshwater mussel communities. Canadian Journal of Fisheries and Aquatic 
Sciences. 55:297-306. 
Haag W.R. and M.L. Warren Jr. 2010. Diversity, abundance, and size structure of bivalve 
assemblages in the Sipsey River, Alabama. Aquatic Conservation of Marine and 
Freshwater Ecosystems 20: 655-667. 
Harmon, T. S. 2009. Methods for reducing stressors and maintaining water quality associated 
with live fish transport in tanks: a review of the basics. Reviews in Aquaculture 1, 58-66. 
Hauswald, C.L. 1997. Life History and Conservation of Elliptio crassidens from the Blue River, 
Indiana. M.S. Dissertation, University of Louisville. 
Heard, W.H. 1979. Identification manual of the fresh water clams of Florida. State of Florida, 
Department of Environmental Regulation, Technical Series, 4(2): 1-82. 
Hinck J.E., S.E. McMurray, A.D. Roberts, M.C. Barnhart, C.G. Ingersoll, N. Wang, T. 
Augspurger. 2012. Spatial and Temporal Trends of Freshwater Mussel Assemblages in 
the Meramec River Basin, Missouri, USA. Journal of Fish and Wildlife Management. 
Vol. 3(2):319-331 
 
47 
 
Howard, A. D. 1914a. "Experiments in Propagation of Freshwater Mussels of the Quadrula 
Group." Report of the U.S. Commissioner of Fisheries for 1913 Appendix 4 1-52. 
Howard, A. D. 1914b. A second case of metamorphosis without parasitism in the Unionidae. 
 Science 40(1027):353-355. 
Howard, A. D. 1914c. Experiments in propagation of fresh-water mussels of the Quadrula 
group. Appendix IV to the Report of the U.S. Commissioner of Fisheries for 1913:1-52, 6 
plates [Issued separately as U.S. Bureau of Fisheries Document 801.] 
Johnson, L. E. 1995. Enhanced early detection and enumeration of zebra mussel (Dreissena spp.)  
Kelner D.E. & Bernard E. Sietman 2000. ?Relic Populations of the Ebony 
 Shell, Fusconaia ebena (Bivalvia: Unionidae), in the Upper Mississippi River Drainage.? 
Journal of Freshwater Ecology, 15:3, 371-377. 
Lafferty, K. D. 1999. The evolution of trophic transmission. Parasitology Today 15:111-115. 
Larinier, M. 2000. Dams and Fish Migration. World Commission on Dams: Environmental 
Issues, Dams and Fish Migration, Final Draft. Contributing paper.  
Lee, D. S., C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E. McAllister, and J. R. Stauffer, Jr. 
1980. Atlas of North American freshwater fishes. North Carolina State Museum of 
Natural History, Raleigh, North Carolina. I-x + 854 pp. 
Lydeard, C., J. T. Garner, P. Hartfield and J. D. Williams. 1999. Freshwater mussels in the Gulf 
region: Alabama. Gulf of Mexico Science 1999(2): 125-134. 
Mathiak, H.A. 1979. A river survey of the unionid mussels of Wisconsin 1973?1977. 
 Sand Shell Press. Horicon, Wisconsin. 75 pp. 
48 
 
Mazik, P. M., B. A. Simco, and N. C. Parker. 1991. Influence of Water Hardness and Salts on 
Survival and Physiological Characteristics of Striped Bass during and after Transport. 
Transactions of the American Fisheries Society, 120:1, pp 121-126. 
Mettee, M. F., P. E. O?Neil, and J. M. Pierson.  1996.  Fishes of Alabama and the Mobile Basin. 
Oxmoor House, Birmingham, Alabama. 820 pp. 
Miller, A. C. and B. S. Payne. 2000. An Analysis of Freshwater Mussels (Unionidae) in the 
Upper Ohio River near Huntington, West Virginia, 1998 Studies. U.S. Army Corp of 
Engineers Report. 45pp. 
Mirarchi, R. E., J. T. Garner, M. F. Mettee, and P. E. O?Neil. 2004. Alabama wildlife. Volume 2. 
Imperiled aquatic mollusks and fishes. University of Alabama Press, Tuscaloosa, 
Alabama. Xii + 255 pp.  
Missouri Natural Heritage Program. 2004. Missouri Species and Communities of Conservation 
Concern Checklist. Missouri Department of Conservation. Jefferson City, Missouri. 47 
Pp. 
Moore, J. 2002. Parasites and the behavior of animals. Oxford University Press, New York, New 
 York, USA. 
NatureServe. 2013. NatureServe Explorer: An online encyclopedia of life [web application]. 
Version 7.1. NatureServe, Arlington, Virginia. Available 
http://www.natureserve.org/explorer. (Accessed: October 9, 2013) 
NatureServe. 2014. NatureServe Explorer: An online encyclopedia of life [web application]. 
Version 7.1. NatureServe, Arlington, Virginia. Available http://explorer.natureserve.org. 
(Accessed: February 26, 2014). 
49 
 
Neebling, T. E., and M. C. Quist. 2008. Observations on the Distribution and Status of Western 
Sand Darter, Spotted Gar, and Skipjack herring in Iowa Rivers. 115(1-4):24-27 
Noga, E. J. 2000. Fish Disease: Diagnosis and Treatment. Iowa State University Press, Ames. 
O?Brien, C. A., J. D. Williams, and M. A. Hoggarth. 2003. Morphological variation in glochidia 
shells of six species of Elliptio from Gulf of Mexico and Atlantic Coast drainages in the 
southeastern United States. Proceedings of the biological society of Washington 
116(3):719-731.  
Page, L. M., and B. M. Burr. 1991. Peterson field guide to freshwater fishes of North America 
north of Mexico. Second edition. Houghton Mifflin Harcourt, Boston. Xix + 663 pp. 
Parmalee, P.W. and A.E. Bogan. 1998. The Freshwater Mussels of Tennessee. University of 
Tennessee Press: Knoxville, Tennessee. 328 pp. 
Pflieger, W. L. 1997. The fishes of Missouri. Revised edition. Missouri Department of 
Conservation, Jefferson City. Vi + 372 pp. 
Rashleigh, B. and D.L. DeAngelis. 2007. Conditions for coexistence of freshwater mussel 
species via partitioning of fish host resources. Ecological Modelling 201: 171-178. 
Rogers, C. L., and R. V. Dimock Jr. 2003. Acquired resistance of bluegill sunfish Lepomis 
 macrochirus to glochidia larvae of the freshwater mussel Utterbackia imbecillis 
 (Bivalvia: Unionidae) after multiple infections. Journal of Parasitology, 89(1): 51-56. 
Santucci, V.J. S.R. Gephard, and S.M. Pescitelli. 2005. Effects of Multiple Low-Head Dams on 
Fish, Macroinvertebrates, Habitat, and Water Quality in the Fox River, Illinois. North 
American Journal of Fisheries Management, 25(3): 975-992.  
50 
 
Schwalb, A.N., K. Cottenie, M.S. Poos, and J. D. Ackerman. 2011. Dispersal limitation of 
unionid mussels and implications for their conservation. Freshwater Biology 56: 1509-
1518. 
Schwalb, A.N., T. J. Morris, N.E. Mandrak, and K. Cottenie. 2013. Distribution of unionid 
freshwater mussels depends on the distribution of host fishes on a regional scale. 
Diversity and Distributions 19: 446-454.  
Sietman, B.E. 2003. Field Guide to the Freshwater Mussels of Minnesota. Minnesota 
Department of Natural Resources: St. Paul, Minnesota. 144 pp. 
Smith, B. May 15, 2014. Skipjack herring (Alosa chrysochloris). www.dnr.wi.gov. [Online] 
Available:http://dnr.wi.gov/topic/EndangeredResources/Animals.asp?mode=detail&Spec
Code=AFCFA01030, Accessed May 20, 2014 
Strayer, D. L. 2008. Freshwater mussel ecology: a multifactor approach to distribution and 
abundance. University of California Press, Los Angeles. 
Surber, T. 1912. Identification of the glochidia of freshwater mussels. Report and Special Papers 
 of the U.S. Bureau of Fisheries 1912:1-10 
Surber, T. 1913. "Notes on the Natural Hosts of Freshwater Mussels." Bulletin of the Bureau of 
Fisheries 3  101-116. 
Swanson, C., R. C. Mager, S. L. Doroshov, and J. J. Cech Jr. 1996. Use of Salts, Anesthetics, 
and Polymers to Minimize Handling and Transport Mortality in Delta Smelt. 
Transactions of the American Fisheries Society, 125:2 pp326-329. 
Vaughn, C.C. 2012. Life history traits and abundance can predict local colonization and 
extinction rates of freshwater mussels. Freshwater Biology 57: 982-992 
51 
 
Vaughn, C. C., and C. M. Taylor. 1999. Impoundments and the Decline of Freshwater Mussels: a 
Case Study of an Extinction Gradient. Conservation Biology Vol. 13, No. 4 pp912-920. 
Watters, G.T. 1992. Unionids, fishes, and the species-area curve. Journal of Biogeography 19: 
481-490. 
Watters, G.T. 1994. An annotated bibliography of the reproduction and propagation of the 
Unionoidea (Primarily of North America). Ohio Biological Survey Miscellaneous 
Contributions No. 1, 158 p.  
Watters, G. T. 1996. Small dams as barriers to freshwater mussels (Bivalvia, Unionoida) and 
their hosts. Biological conservation 75 pp79-85. 
Watters, G. T., and S.H. O?Dee. 1999. Glochidia of the freshwater mussel Lampsilis 
Overwintering on fish hosts. Journal of Molluscan Studies 56: 453-459. 
Williams, J. D., A. E. Bogan, and J. T. Garner. 2008. Freshwater Mussels of Alabama and the 
 Mobile Basin in Georgia, Mississippi, and Tennessee. University of Alabama Press, 
 Tuscaloosa. 
Williams, J. D., M. L. Warren, Jr., K. S. Cummings, J. L. Harris and R. J. Neves. 1993. 
Conservation status of the freshwater mussels of the United States and Canada. Fisheries 
18(9):6-22. 
Wilson, C. B., and H. W. Clark. 1914. The mussels of the Cumberland River and its tributaries. 
Report and Special Papers of the Commissioner of Fisheries 1912 (Document Number 
757): 1-63, figures 1-2.  
 
Winkler, P. 1987. A Method to Minimize Stress during Fish Transport.  The Progressive Fish-
Culturist, 49:2, pp 154-155.  
52 
 
Weirich, C. R. and J. R. Tomasso. 1991. Confinement- and Transport-Induced Stress on Red 
Drum Juveniles: Effect of Salinity. The Progressive Fish-Culturist, 53:3, pp 146-149. 
Wydoski, R. S., and G. A. Wedemeyer. 1976. Problems in the physiological monitoring of wild 
fish populations. Proceedings of the Annual Conference of the Western Association of 
Game and Fish Commission. 56: 200-214. 
Zale, A. V., and R. J. Neves. 1982. Fish hosts of four species of lampsiline mussels (Mollusca: 
Unionidae) in Big Moccasin Creek, Virginia. Canadian Journal of Zoology. Vol. 60, No. 
11 pp 2534-2542. 
53 
 
Table 1. Skipjack herring survival from collection site to lab.  
Collection Date 
 
# Collected # Surviving transport to lab 
3/18/2011                
3/24/2011 
6 
24 
4 
23 
3/29/2011 29 25 
 
54 
 
Table 2. Results of gill inspections for encapsulated glochidia of Reginaia ebenus on all fish taxa 
tested.  Y represents successful encapsulation, N represents no encapsulation.  
Fishes 
   
    
  
# Inspected 
Encapsulation 
(Y/N) Day after infection 
Cyprinidae 
   
 
Lythurus bellus 6 N 4-6 
 
Notropis ammophilus 2 N 4-6 
 
Campostoma oligolepis 3 N 4-6 
 
Luxilus chrysocephalus 3 N 4-6 
 
Pimephales promelas 6 N 4-6 
Fundulidae 
  
4-6 
 
Fundulus olivaceous 6 N 4-6 
Percidae 
   
4-6 
 
Etheostoma sp. and 
Percina sp. 6 N 4-6 
Centrarchidae 
  
4-6 
 
Lepomis macrochirus 6 N 4-6 
 
Micropterus salmoides 6 N 4-6 
Ictaluridae 
  
4-6 
 
Ictalurus punctatus 6 N 4-6 
Sciaenidae 
  
4-6 
 
Applodinotus grunniens 4 N 4-6 
Clupeidae 
  
4-6 
 Dorosoma cepedianum 4 N 4-6 
 
Alosa chrysochloris 1 Y 11 
 
Alosa chrysochloris 7 N 19 
 
 
55 
 
Table 3. Alabama shad and skipjack herring survival from collection site to lab.  
 
Fish species Collection Date 
 
# Collected # Surviving transport to lab 
Alosa alabamae 
Alosa chrysochloris 
3/22/2012                
3/20/2012 
30 
44 
29 
43 
 
     
56 
 
 
Table 4: Mussel length, percent viable glochidia, and viable brood size of mussels utilized for 
glochidial baths on two dates. 
Date used Mussel length 
(mm) 
% viable 
glochidia 
Brood size (# viable 
glochidia) 
4/3/2012 105 90.8 548,000 
4/3/2012 107 96.0 1,506,000 
4/3/2012 100 96.0 386,000 
4/4/2012 106.3 94.2 1,108,000 
4/4/2012 104.4 97.1 948,000 
 
57 
 
 
Table 5: Number of sloughed glochidia from Bluegill infected with elephantear. No juveniles 
were observed. 
 
 Days post infection 
Bluegill ID 2 6 8 10 12 14 16 23 
1 78 9 0 0 0 0 0 0  
2 146 0 0 1 0 0 0 0  
3 4 15 2 0 1 0 0 0  
4 51 0 0 0 0 0 0 0  
5 108 2 1 0 1 0 0 0  
 
58 
 
Table 6: Percent viable glochidia, and viable brood size of mussels utilized for glochidial bath. 
Date used % viable 
glochidia 
Brood size (# viable 
glochidia) 
4-15-2013 88.1 111,500 
4-15-2013 93.5 87,000 
4-15-2013 67.0 32,500 
59 
 
Figure 1. a) 5700 liter rectangular fiberglass tanks used to hold large river fish. b) Baffle, and 
airlifts used to generate circular water flow.  Arrows show direction of flow. 
a
b
  
 
 
60 
 
 
Figure 2. 8500 liter round fiberglass tank and cover used to hold large river fish.   
 
61 
 
Figure 3. 1900 liter round fish hauler with oxygen tanks used to haul fish from collection site to 
lab.
 
 
 
 
 
62 
 
Figure 4. Picture and schematic diagram showing collection of sloughed glochidia and juveniles 
from AHAB tanks.  Arrows show direction of water flow. 
In t ernal 
ba f fle
Fil t e r cup
  
63 
 
 
Figure 5.  Schematic diagram showing collection of sloughed glochidia and juveniles from cone 
tanks.  Arrows show direction of water flow. 
Double 
s t and pipe
Fil t er 
bag
 
 
 
 
 
 
 
 
 
64 
 
Figure 6. Photo of infected skipjack herring gills under cross polarized lighting.  Light colored 
?discs? are encapsulated glochidia. 
 
 
 
65 
 
 
Fig. 7.  Number of juveniles produced from skipjack herring gills on each day following 
excision.   
 
Da ys p o st e x cisi o n
0 1 2 3 4 6 8 T ot al
N
u
m
b
er
 
of
 
Ju
v
en
il
es
0
200
400
600
800
1000
1200
 
0 
 
183 
 145  
614 
 
12 
 
114 92 
1160 
 
66 
 
Figure 8: Total juvenile production (bars) and transformation efficiency (circles) from Alabama 
shad gills excised on different days.  ?X? symbols denote gills from which no juveniles were 
produced. Numbers above bars denote total # juveniles produced from each gill set.  
Da y of exci sio n
5 10 15 20 25 30 35 40
Total
 Juve
ni
le
s Prod
uce
d
0
50
100
150
200
250
300
350
Transfo
rmatio
n e
fficien
cy (%)
0
10
20
30
40
50
60
70
  
 
326 328 
293 
9 4 2 X X X X X X X 
67 
 
 
Figure 9: Daily juvenile production from Alabama shad gill arches excised on: a) Day 10, b) Day 
13, c) Day 14, d) Day 17, e) Day 20, and f) Day 23 after infection. 
0
50
100
150
200
250
300
350 a c t i v e  j u v e n i l e s
q u e s t i o n a b l e
d e a d
0 10 20 30 40
0
50
100
150
200
250
300
350
Days  pos t  ex c i s i on
0 10 20 30 40
0
50
100
150
200
250
300
350
Num
ber 
pr
oduc
ed 
f
r
om
 
ex
c
i
s
ed 
gi
l
l
s
0
50
100
150
200
250
300
350
0
200
400
600
1000
1200
1400
1600
1800
0
50
100
150
200
250
300
350
 
 
a 
c
a 
d 
e f
a 
b 
68 
 
 
Figure 10: Total juvenile production (bars) and transformation efficiency (circles) from skipjack 
herring gills excised on different days.  ?X? symbols denote gills from which no juveniles were 
produced. Number above bar denotes quantity of juveniles produced from that gill set. 
Da y  o f  e x c is ion
4 6 8 10 12 14 16 18
T
o
t
a
l 
J
u
v
e
n
il
e
s
 
P
r
o
d
u
c
e
d
0
20
40
60
80
100
T
r
a
n
s
f
o
r
m
a
t
ion
 
e
f
f
ic
ien
c
y
 
(
%
)
0
5
10
15
20
25
30
 
 
 
85 
X X X X X X X 
69 
 
 
Figure 11: Daily juvenile production from skipjack herring gill arches excised on Day 16 after 
infection. 
 
Days  pos t  ex c i s i on
0 10 20
Num
ber
 
pr
oduc
ed 
f
ro
m
 
ex
c
is
ed 
gil
ls
0
20
40
60
80
100 a c t i v e  j uv e n i l e s
que s t i o n a b l e
de a d
 
70 
 
Figure 12: a) Image showing representation of a ?gill unit? analyzed in this study.  Red ovals 
demonstrate general areas where individual filaments were selected for lamellae counts. b) Close 
up image of filament on a gill unit. Yellow oval shows individual lamellae that were counted.  
 
 
 
 
 
 
71 
 
Figure 13. Images showing decomposition of gill filaments on gill units across time treatments.  
Day 6 
Day 0 Day 3 
Day 12 
72 
 
Figure 14. Average number of lamellae per gill unit in infected (white) and uninfected (black) 
units for 12 days following excision.  Red x?s denote days in which lamellae were unidentifiable.