Cercospora rodmanii Conway

Material Information

Cercospora rodmanii Conway a biocontrol agent for waterhyacinth
Series Title:
Bulletin Agricultural Experiment Stations, University of Florida
Freeman, T. E ( Thomas Edward ), 1930-
Charudattan, R
Place of Publication:
Gainesville [Fla]
Agricultural Experiment Stations, Institute of Food and Agricultural Sciences, University of Florida
Publication Date:
Physical Description:
ii, 18 p. : ill. ; 23 cm.


Subjects / Keywords:
Water hyacinth -- Diseases and pests ( lcsh )
Weeds -- Biological control ( lcsh )
Cercospora ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )


Bibliography: p. 17-18.
General Note:
"October 1984."
Bulletin (University of Florida. Agricultural Experiment Station)
Statement of Responsibility:
T.E. Freeman and R. Charudattan.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
022594607 ( ALEPH )
14158859 ( OCLC )
ADA2690 ( NOTIS )
0096-607X ; ( ISSN )

Full Text


The publications in this collection do
not reflect current scientific knowledge
or recommendations. These texts
represent the historic publishing
record of the Institute for Food and
Agricultural Sciences and should be
used only to trace the historic work of
the Institute and its staff. Current IFAS
research may be found on the
Electronic Data Information Source

site maintained by the Florida
Cooperative Extension Service.

Copyright 2005, Board of Trustees, University
of Florida

Bulletin 842 (technical)

Cercospora rodmanii Conway
A Biocontrol Agent for Waterhyacinth

T. E. Freeman and R. Charudattan

Agricultural Experiment Stations
Institute of Food and Agricultural Sciences
University of Florida, Gainesville
F. A. Wood, Dean for Research

October 1984

Cercospora rodmanii Conway
A Biocontrol Agent for Waterhyacinth

T. E. Freeman and R. Charudattan

Plant Pathology Department
University of Florida
Gainesville, FL 32611


Preface ................ ............................. i
Introduction ................. ...... ......................... 1
Disease Symptoms .................. .......... ................ 1
Morphology of Cercospora rodmanii ............. .......... ........ 4
Culturing and Cultural Characteristics ............................. 5
Infection and Pathological Histology .................... ....... 10
Epidemiology ..................... ......................... 11
Biocontrol Potential ............... ........... .............. 14
Safety Considerations ................. ..................... 15
Potential for Practical Use .. ......................... ...... 15
Acknowledgments .................. ....................... 16
Literature Cited ................... ....... ... ............... 17


In recent years, biological control methods for controlling aquatic
weeds have received considerable attention. Various species of her-
bivorous insects, fish, snails, birds, and mammals have been investi-
gated for their ability to exert some control pressure on noxious
aquatic plants. Some of them, such as the alligatorweed flea beetle,
have been reasonably effective, especially in an integrated control
program. Surprisingly, until our program was initiated in the early
1970's, plant pathogens had been rarely considered as biocontrol
agents. Plant pathogens have all the prerequisites of good biocontrol
agents and thus are potentially useful. They may be used either alone
or in integrated programs with other control agents and methods.
Research efforts reported herein have been aimed at the use of the
plant pathogen Cercospora rodmanii Conway in control programs for
waterhyacinth, Eichhornia crassipes (Mart.) Solms. The purpose of
this bulletin is to acquaint researchers and potential users with the
pathological, morphological, and biocontrol characteristics of C. rod-

During the winter of 1973-1974, waterhyacinth plants in the
vicinity of Gainesville, Florida, were affected by a leaf-spotting dis-
ease not previously noticed. The disorder was caused by a species of
Cercospora, subsequently identified as C. piaropi Tharp (13). This
was only the second reported occurrence of this organism in the
United States; it was originally described from Texas in 1914 by
Tharp (17). The fungus did not appear to be causing appreciable
damage to the waterhyacinth plants when it was first noted.
In December of 1973, a Cercospora species, along with many other
fungi, was isolated from declining waterhyacinth plants in the Rod-
man Reservoir, Florida (10). The disease was characterized by root
rot, leaf spots, and leaf necrosis. Although the root rot at first was
thought to be the primary disease damage, it was later determined to
be of secondary origin. The leaf spots and the severe leaf necrosis
were regarded as the primary symptoms. In a preliminary test the
fungus was found to be pathogenic on waterhyacinth, and in later
tests the fungus inflicted considerable damage on this plant. Affected
plants eventually died and sank to the bottom of the test vats.
Microscopic examination revealed the fungus to be a Cercospora.
However, the spores conidiaa) were much longer than those recorded
by Freeman and Charudattan (13) for C. piaropi. In addition, symp-
toms caused by the Rodman fungus differed from those recorded for
C. piaropi. The former caused a general blighting symptom on the
foliage, whereas the latter produced more discrete leaf spots. How-
ever, the two symptoms may have been manifestations of the same
disease. When leaves infected by C. piaropi were incubated under
moist conditions, both small conidia and long conidia frequently
developed on the dead tissue. Therefore, the Rodman fungus could
have been either a long-spored variant of C. piaropi or a new species
of Cercospora. The conidia of the two fungi not only differed in size
but in the morphology of their basal cells. These differences in conid-
ial morphology and symptomatology prompted the description of the
Rodman fungus as a new species, C. rodmanii Conway (4).

Cercospora rodmanii causes small, punctate necrotic spots on the
leaves and petioles of the plant. The typical symptom is shown on the
cover and in Fig. 1. The spots are more numerous at the distal portion
of the leaf but can occur over the entire leaf surface and the upper

Figure 1. Typical leaf spot and confluent necrotic symptoms on
waterhyacinth leaves infected by Cercospora rodmanii.

portions of the petiole. Because of these spots, the leaves die from the
tip back, with death of tissue gradually spreading towards the base of
the leaf until the entire leaf is killed. The progress of these symptoms
is shown in the disease severity rating scale depicted in Fig. 2.
Plants with severely infected leaves become chlorotic and develop
spindly petioles. Newly emerged leaves remain small because of the
disease stress on the plant. Under severe disease conditions, leaves
are killed faster than they can be regenerated and the entire plant
eventually succumbs and sinks to the bottom of the water body. In the
latter stages of disease, root deterioration is frequently evident (Fig.
Disease stress within a population of waterhyacinth is manifested
initially as an overall chlorotic appearance of the plants. Numerous
severely spotted or dead leaves soon become evident on the plants. As
the disease progresses, the entire population of waterhyacinth has a
brownish appearance. At this stage the population begins to decline,
and open water may become evident in areas where previously there
had been dense stands of waterhyacinth. As the disease continues,

the mat of vegetation breaks up, and small clusters of plants under
severe stress float away from the mat. The plants in these clusters
have only one to three small chlorotic leaves with leaf spots and
numerous dead leaves still attached but sunken beneath the water
surface. Finally the entire cluster gradually sinks to the bottom. The
entire disease progression in a dense population of waterhyacinth
may take several weeks to months.
The symptoms described and depicted in Figs. 1, 2, and 3 are
typical and easy to diagnose, but the disease may not be easily
recognized in later stages without the benefit of either historical
knowledge or early observations. In such cases, it is necessary to

Numerical ratings of symptoms: 0-no spots on lamina or petiole; 1-1 to 4 spots on
lamina, no petiolar spotting; 2-less than 25 percent of lamina surface with spots, no
coalescence or petiolar spotting; 3-less than 50 percent of lamina surface with
spots, some coalescence, no petiolar spotting; 4-less than 50 percent of leaf
surface with spots, coalescence, some tip dieback, and petiolar spots; 5-less than
50 percent of leaf surface with spots, coalescence, 10 percent tip dieback, and
petiolar spotting; 6-less than 75 percent spots, coalescence, 30 percent tip
dieback, and petiolar spotting; 7-greater than 75 percent spots, coalescence, 60
percent tip dieback, coalescing spots on petiole; 8-dead lamina, petiole green, but
heavily spotted; and 9-dead lamina and petiole (submerged).

Figure. The progression of disease damage caused by Cercospora rodmanii
and the damage rating scale used in assessing disease severity.

2 3 4


0 1

Figure 3. A waterhyacinth plant showing signs of severe stress due to
Cercospora rodmanii. Note the necrotic laminae, spindly petioles, and the
lack of an extensive root system.

examine diseased tissue for the presence of the causal agent. To
accurately examine the tissue, a knowledge of the morphology of the
pathogen is essential.

The morphological characteristics of Cercospora rodmanii in and
on host tissues are shown in Figs. 4 and 5. As mentioned previously,
C. rodmanii can be distinguished from C. piaropi, which also occurs
on waterhyacinth, by differences in conidial size and morphology.
Conidia of C. piaropi seldom exceed 150 im in length (Fig. 4g). The
base of the C. rodmanii conidium is truncate (Fig. 4f); that of C.
piaropi is obconic. In addition, conidia of the former are frequently
flexuous whereas those of the latter are usually straight or only
slightly curved. Those of C. rodmanii are borne on dark-colored

conidiophores arising in fascicles through the stomata (Fig. 4h; Fig.
5a, b). Differences in conidial morphology, especially the form of the
basal end of the conidium, are dependable features on which species
of Cercospora are often based (3). Accordingly, Conway (4) used
conidial characteristics as a primary basis for the establishment of C.
rodmanii as a new species and described the disease and the fungus
as follows:
Leaf spots black, punctate to circular (1-3 mm diam.), leaf
and petiole chlorotic, tip of leaf necrotic, conidiophores
amphigenous, 3-12 in each fascicle, brown sympodial,
arising from a well-developed stroma, emerging through
the stoma, 84-(145)-284 x 4-(4.5)-5 |xm; conidia hya-
line, truncate at base, acicular, multiseptate, 66-(172)-
374 x 3-(4)-5pLm.
A stroma and an associated Asteromella pycnidial state also may
be formed in older cultures. Conway (4) described it as follows:
Asteromella pycnidia dark brown, ostiolate globose, 80-
95 x 80-110 xm, substomal, later erumpent, ostiole 30-
40 x 25-30 pLm; conidia hyaline, bacilliform 2-3.5 x
1-1.5 Rm.
The presence of this stage was a further criterion for establishing
C. rodmanii as a new species.

Cercospora rodmanii grows well between 20 and 30C, with opti-
mum growth occurring near 25C. It grows well on a variety of solid
and liquid media. However, it varies considerably in growth and
cultural characteristics on various media (Fig. 6 and Table 1). It grew
well on potato dextrose agar (PDA). The addition of 5 g/L of yeast
extract (Y) further enhanced growth on all media. The growth in-
crease with Y was especially evident on Difco Czapek-Dox agar (C-D).
To determine further the best growth medium, three types of PDA
were tested with and without Y. Freshly prepared PDA (broth from
200 g cooked potatoes, 20 g dextrose, and 17 g agar/L) was superior to
Difco brand of dehydrated PDA, which was superior to the BBL brand
of dehydrated PDA (Table 2). The same pattern of growth rates was
also evident after the addition of Y to each of the three types of PDA.
There was no appreciable difference in growth on Difco PDA when
compared to fresh PDA. Therefore, since the preparation of fresh
PDA is time-consuming, Difco PDA plus Y was adopted as the stan-
dard solid medium for growing C. rodmanii. For liquid culture, Difco
potato dextrose broth (PDB) plus Y was adopted for general use.
On petri plates with PDA plus Y, cultures are light to dark grey on
top and deep red on the bottom. A diffusible red pigment is present in

Figure4. Details of the morphology and pathological histology of Cercospora
rodmanii as observed with a compound microscope.
a. Mycelial infection through stomatal opening (25pm), 200 X,.
b. Conidiophores on leaf. (Note amphigenous fruiting on leaf (350 Rm
thick), 40 x).
c. Conidiophores, 200 x .
d. Conidiophores, 400 x.
Figure he agar surrounding the culture. An exudate frequently ofCerospora
rodmanii as observed with a compound microscope.

the culture. In addelial infection through stomations radiate from the center to the
b. Coniouter edges of the cultures. Note amphigenous fruand nonsporulating sec-(350
thick), 40 x).
c. Conidiophores, 200 x.
d. Conidiophores, 400 x.

the agar surrounding the culture. An exudate frequently forms on
the culture. In addition, invaginations radiate from the center to the
outer edges of the cultures. Nonpigmented and nonsporulating sec-
tors sometimes form; such sectors are common for Cercospora spp. in
culture (2).

rN 79:!J

Figure 4. Details of the morphology and pathological histology of Cercospora
rodmanii as observed with a compound microscope.
e. Conidiophores (5 p.m wide) showing conidium attachment (arrow),
400 x.
f. Conidium showing truncate base, 400 x.
g. Conidia; arrow points to a 300-pLm-long conidium, 200 x.
h. Top view of fascicle of conidiophores on leaf surface, 200 x.
i. Subepidermal mycelium; arrow points to base of conidiophores, 400 x.

The diffusible pigment produced by C. rodmanii in most culture
media has been identified as the phytotoxin cercosporin (Cullen et al.,
unpublished). Its production in culture is enhanced by light. The
toxin causes necrosis ofwaterhyacinth leaves and may be involved in

Figure 5. Scanning electron micrographs of Cercospora rodmanii on
waterhyacinth leaf.
a. Conidiophores emerging from stomata. (Note conidia lying on the leaf
b. Single fascicle of conidiophores. (Note basal end of the conidium lying at
the lower left corner.)
c. Tip of conidiophores showing spore scar (enlargement of inset in 5b).


Figure 5. Scanning electron micrographs of Cercospora rodmanii on
waterhyacinth leaf.
d. Hyphae growing into stomata (infection process). Enlargement of inset
is on the right.
e. Hypha from germinating conidium growing over leaf surface towards

Figure 6. Comparative growth characteristics of Cercospora rodmanii on
various culture media after incubation at 25C for 14 days.
a. Difco potato dextrose agar plus yeast extract.
b. Difco potato dextrose agar.
c. Freshly prepared potato dextrose agar.
d. Nutrient agar.
e. Czapek-Dox agar plus yeast extract.
f. Czapek-Dox agar.
g. Cornmeal agar plus yeast extract.
h. Cornmeal agar.

Conway (4) reported that C. rodmanii sporulated well on V-8 agar
(200 mL V-8 juice, 3 g CaCO3, and 15 g agar/L). The fungus also will
sporulate on PDA plus Y. Sporulation is augmented by incubating
cultures under 12 hr darkness and 12 hr light. In culture, primary
conidia frequently produce secondary conidia that are shorter than
the primary ones.

Figures 4 and 5 show the infection process and the histology.
Infection can originate from either mycelium or conidia. In both
instances the hyphae grow into the stomata (Fig. 5d, e), ramify in the
substomatal cavity, and invade the surrounding tissue (Fig. 4i). A
stroma develops in the stomatal cavity and a fascicle of 3 to 12
conidiophores arises from it and emerges through the stomata (Figs.

4b, c, d, h; 5a, b). Primary and secondary conidia are produced on the
Since penetration occurs through the stoma, the number and dis-
tribution of stomata on the pseudolamina and the petiole (float and
subfloat) influence the development of infections. Stomata become
fewer in number from the tip of the lamina to the base of the petiole
(Fig. 7). Distribution of C. rodmanii leaf spots follows the stomatal
distribution pattern.

In nature, conidia produced on diseased tissue are spread by the
wind and serve to disseminate C. rodmanii between loci. Production
of conidia on diseased tissues is curtailed by temperatures below 10C
and reaches a maximum in the range of 20 to 30C (14). The speed and
intensity of the epidemic is directly related to the number of conidia
produced, which is related to the amount of diseased and dead tissue
that is available for sporulation. The amount of diseased and dead
tissue can be increased by well-timed spray applications of formula-
tions containing infective propagules of C. rodmanii.

Table 1. Average diameter of Cercospora rodmanii cultures grown at
25C on various media.
Diameter of Culture, mm
Medium 7 days 14 days
Fresh potato dextrose agar (PDA) 31.6 56.0
Difco PDA 29.5 55.0
Difco PDA+ yeast extract (Y) 39.3 67.0
Difco corn meal agar (CM) 24.0* 44.3*
Difco CM + Y 21.0 24.6
Difco Czapek-Dox agar (C-D) 19.6 30.3
Difco C-D + Y 38.0 66.3
Difco nutrient agar 20.3 23.6
*Very little aerial mycelium formed.

Table 2. Average diameter of Cercospora rodmanii cultures after 7 days'
growth at 25C on variations of PDA.

Diameter of Culture, mm
Medium With yeast extract Without yeast extract
Fresh PDA 34.6 31.3
Difco PDA 33.3 27.6
BBL PDA 27.8 23.5

In a study conducted in 1976,
conidia were trapped with a
8.7 Roto-Rod air sampler in Lake
Alice on the University of Flor-
95.2pseudo ida campus over a 12-month
lamina period. It was observed that
-83.3- sporulation reached a peak dur-
ing the fall and early winter
months (Fig. 8). High numbers
of conidia were associated with
35.7 -isthmus high disease intensity and low
numbers with low levels of dis-
ease, suggesting that the num-
bers of conidia trapped in-
50.5 -float creased as disease intensity
0 -subfloat Since its original discovery,
C. rodmanii has been found on
waterhyacinth in areas in Flor-
Figure 7. Mean number of stomata ida other than its type location
per mm2 on the adaxial surface of a in Rodman Reservoir. During
typical waterhyacinth leaf. surveys conducted in 1978, the
fungus was found at several
locations along the St. Johns River from Lake Poinsett near Cocoa
northward to Lake George. It was also found in Lake Rousseau on the
Withlacoochee River. However, it was not found in nearby Crystal
River, where C. piaropi was the predominant pathogen. In north-
central Florida, C. rodmanii was present on plants from Orange
Lake. It was not present on plants from either the Santa Fe River or
the Suwanee River. Neither was it present in the Aucilla River nor
the Wacissa River in west-central Florida. It has been found in
irrigation canal systems in Palm Beach and Broward counties (14).
Cercospora rodmanii was also confirmed in samples collected from
Louisiana at Hayes North, Pecan Island I, Pecan Island II, and
Manchac I by the personnel of the U.S. Army Corps of Engineers
Waterways Experiment Station. It was not present in samples from
Sorrento II, Hayes West, Centerville C, Bayou Lourse, and Manchac
IV (1). Infected plants from Lake Concordia had been disseminated in
all of the above sites by the Corps personnel.
In a study conducted in the Rodman Reservoir, under optimal
conditions the waterhyacinth plants were capable of producing one
new leaf every five to six days. This rate will vary with environmen-
tal conditions. During unfavorable periods, it may decline to less
than one leaf produced over a three-week period. The success of the

Figure 8. Seasonal variation in air-borne conidia of Cercospora rodmanii expressed as conidia per cubic meter of air over Lake

epidemic will depend upon the rate at which the pathogen can infect
and kill these new leaves. To determine if there was an optimal
concentration of inoculum necessary to initiate disease, an inoculum-
rate experiment was conducted in a small lake in Fish Prairie,
southeast of Gainesville, Florida, using a laboratory-grown inoculum
(7). Waterhyacinth plants were confined in 9 m2 frames and treated
at three mycelial inoculum concentrations: 48 g/m2, 96 g/m2, and 192
g/m2. In this study, regardless of the initial inoculum level, the rate of
disease spread finally became equalized because ofinoculum buildup
on the infected plants and cross infectivity between plots. The max-
imum amount of damage produced by C. rodmanii was assessed at
the 192 g/m2 inoculum level. This amount was not exceeded, even
with an additional application of inoculum later in the season. The
maximum rate of damage caused by C. rodmanii during this experi-
ment corresponded to the death of 1.0 to 1.3 leaves of the
waterhyacinth every three weeks. Therefore, when conditions exist
that favor disease development and limit leaf production to less than
one leaf per three weeks, C. rodmanii can infect and kill leaves faster
than the plant can produce new leaves. The plant then becomes
debilitated and dies unless conditions change to favor its regrowth.

Because of the biocontrol potential exhibited by C. rodmanii, pre-
liminary small-scale field evaluations were conducted in Lake Alice,
Gainesville, Florida; Rodman Reservoir near Orange Springs, Flor-
ida; and Lake Concordia in east-central Louisiana. In all of these
tests, C. rodmanii, either alone or in combination with other biotic
agents, was capable of exerting a high degree of stress on
waterhyacinth (1, 5, 8, 11, 14, 15). Large-scale evaluation has since
been conducted in Lake Theriot, Louisiana (18, 19).
Conway et al. (7, 11) reported that C. rodmanii can severely affect
waterhyacinth growth, especially under conditions that favor a re-
duced growth rate of the plant. Although the greatest effect of C.
rodmanii was on the height of the waterhyacinth plant, Conway et al.
(11) have reported the death of waterhyacinth and the appearance of
open water following the application of C. rodmanii to dense
waterhyacinth mats. In a field study conducted in 1982, a combina-
tion of C. rodmanii and waterhyacinth insects produced a 99% control
seven months after the initial application of the fungus (Charudattan
et al., unpublished).

Any organism used as a biocontrol agent must be safe to use in the
environment. A plant pathogenic biocontrol agent must be screened
against several plants to determine its specificity to the target host
and safety towards nontarget plants. Therefore, the host range of C.
rodmanii was determined in the greenhouse and the field (9). Eighty-
five plant species in 22 families were screened and only water-
hyacinth was susceptible. Therefore, C. rodmanii is host specific to
waterhyacinth (9).
One of the concerns in the use of C. rodmanii as a control agent for
waterhyacinth is whether it will harm fish. To determine this, Gam-
busia affinis (mosquito fish) was exposed to C. rodmanii in a standard
96-hr bioassay (6). Blended mycelium and spores of C. rodmanii were
placed in the fish containers at rates ranging from 0.4 to 6.34 g/L. The
lowest rate corresponded to an inoculum level of 48 g/m2, which was
the inoculum rate used in Lake Concordia (1) and the lowest rate
used in Fish Prairie (7). The highest rate is equivalent to a surface
area rate of 800 g/m2, which is approximately four times higher than
the highest rate used in the Fish Prairie study.
None of the fish in any of the treatments was adversely affected. In
fact, fish in the tank receiving the highest rate of inoculum fed on the
fungus, which was subsequently isolated from their feces. Based on
this limited test, it appears that C. rodmanii will pose no threat to

The use ofC. rodmanii as a biocontrol agent for waterhyacinth has
been patented by the University of Florida (12). The University has
signed an agreement with Abbott Laboratories, Chicago, Illinois, to
produce a commercial product of the fungus. Abbott Laboratories
have developed wettable powder formulations of the fungus and have
obtained an EPA Experimental Use Permit (EUP) to evaluate it as a
microbial herbicide.
The Abbott formulations have been tested in the greenhouse and
the field with successful results (Charudattan et al., unpublished; 18,
19, 20, 21). An Abbott formulation was also successfully applied to
large (1.8 to 2.5 ha) infestations of waterhyacinth with an aircraft,
the disease was established, and significant reductions in biomass
were achieved two years after the initial spraying (19).
A number of spray additive substances and spray equipment have
been tested to select suitable ones for field application of the C.
rodmanii formulation (14, 16, 20). Some of the available additives

and spray equipment were satisfactory for effective delivery of the
formulation. Application rates also have been determined. The
Abbott Laboratories have placed the effective application rate for a
formulation containing 1 x 106 viable propagules per g to be between
1.1 and 2.2 kg in 225 to 1870 liters of water per hectare (Abbott EUP
label). The efficacy of this formulation was further confirmed in 1982
field tests in which a 99% control was achieved.
A summary of the steps in the testing and development of C.
rodmanii as a biocontrol agent is listed in Table 3. The tests con-
ducted thus far show that C. rodmanii has considerable promise as a
practical biocontrol agent for waterhyacinth. Moreover, the fungus
can be successfully mass-produced. The Abbott formulation is under-
going rigorous tests for efficacy. The results of these tests will deter-
mine the desirability of registration of C. rodmanii as a microbial
herbicide for waterhyacinth.

Table 3. Steps in the development of Cercospora rodmanii as a biocontrol

Step Year
Discovery 1973
Identification and description 1975
Preliminary greenhouse and field tests
of efficacy 1973-1976
Host range testing in the greenhouse and
the field 1973-1977
Advanced field studies in Florida and Louisiana 1973-1976
A C. rodmanii use patent obtained 1978
Industrial production of C. rodmanii initiated 1978
An EPA Experimental Use Permit obtained 1979
Successful large-scale aerial application
of Abbott formulation 1980
EUP studies continued 1980-1983

This work was supported by grants from the Florida Department of
Natural Resources; U.S. Department of Interior, Office of Water
Resources; and U.S. Army Corps of Engineers. Support was also
provided by the Center of Aquatic Weeds, IFAS, University of
Technical assistance was provided by Richard Cullen, B. M. Reber,
John Dennis, Susan Broos, Carole Hennan, Elizabeth Shepeck,
Meredith Chester, Patty Hill, Caprice Simmonds, Frank Hofmeister,

and several student assistants. Dr. K. E. Conway was a postdoctoral
research associate between 1973 and 1978. To all, thanks. Special
thanks are due Clark Allen for the scanning electron micrographs of
C. rodmanii; Richard Cullen, who worked as a biologist on this
project for several years, for the cover photograph; and Thomas
Freeman for the line drawings of Figs. 2, 7, and 8.


1. Addor, E. E. 1977. A Field Test of Selected Insects and Pathogens for
Control of Waterhyacinths. Report 1. Preliminary Results for the 1975-76
Season. Misc. Pap. A-77-2. U.S. Army Waterways Experiment Station,
Vicksburg, MS. 44 pp.
2. Berger, R. D., and E. W. Hanson. 1963. Pathogenicity, host-parasite
relationships, and morphology of some forage legume Cercosporae, and fac-
tors related to disease development. Phytopathology 53:500-508.
3. Chupp, C. 1953. A Monograph of the Fungus Genus Cercospora. Pub-
lished by the author. Ithaca, NY. 667 pp.
4. Conway, K. E. 1976. Cercospora rodmanii, a new pathogen of
waterhyacinth with biological control potential. Can. J. Bot. 54:1079-1083.
5. Conway, K. E. 1976. Evaluation of Cercospora rodmanii as a biological
control of waterhyacinths. Phytopathology 66:914-917.
6. Conway, K. E., and R. E. Cullen. 1978. The effect ofCercospora rodma-
nii, a biological control for waterhyacinth, on the fish, Gambusia affinis.
Mycopathologia 66:113-116.
7. Conway, K. E., R. E. Cullen, T. E. Freeman, and J. A. Cornell. 1979.
Field Evaluation of Cercospora rodmanii as a Biological Control of
Waterhyacinth. Inoculum Rate Studies. Misc. Pap. A-79-6. U.S. Army
Waterways Experiment Station, Vicksburg, MS. 46 pp.
8. Conway, K. E., and T. E. Freeman. 1976. The potential of Cercospora
rodmanii as a biological control for waterhyacinths. Pages 207-209, in T. E.
Freeman, ed., Proc. IV Int. Symp. Biol. Control Weeds, University of Florida,
9. Conway, K. E., and T. E. Freeman. 1977. Host specificity of Cercospora
rodmanii, a potential biological control of waterhyacinth. Plant Dis. Rep.
10. Conway, K. E., T. E. Freeman, and R. Charudattan. 1974. The Fungal
Flora of the Waterhyacinth in Florida. Water Resources Research Center,
University of Florida, Gainesville. Publ. No. 30. 11 pp.
11. Conway, K. E., T. E. Freeman, and R. Charudattan. 1978. Develop-
ment of Cercospora rodmanii as a biological control for Eichhornia crassipes.
Pages 225-230, in Proc. EWRS 5th Symp. on Aquatic Weeds. P.O. Box 14,
Wageningen, The Netherlands.
12. Conway, K. E., T. E. Freeman, and R. Charudattan. 1978. Method and
compositions for controlling waterhyacinth. U.S. Patent 4,097,261. 7 pp.
13. Freeman, T. E., and R. Charudattan. 1974. Occurrence of Cercospora
piaropi on waterhyacinth in Florida. Plant Dis. Rep. 58:277-278.
14. Freeman, T. E., R. Charudattan, K. E. Conway, R. E. Cullen, R. D.
Martyn, D. E. McKinney, M. T. Olexa, and D. F. Reese. 1981. Biological
Control of Aquatic Plants with Pathogenic Fungi. Tech. Rep. A-81-1. U.S.
Army Waterways Experiment Station, Vicksburg MS. 47 pp.

15. Freeman, T. E., R. Charudattan, R. E. Cullen, and D. S. Kenney. 1982.
Biological control of waterhyacinth with the fungal pathogen Cercospora
rodmanii. Pages 239-240, in R. Charudattan and H. L. Walker, eds., Biolog-
ical Control of Weeds with Plant Pathogens, John Wiley, New York (Abstr.).
16. Pennington, J. C., and E. A. Theriot. 1982. Compatibility of a Cercos-
pora rodmanii Formulation with Selected Herbicide Spray Additives. Misc.
Pap. A-82-6. U.S. Army Waterways Experiment Station, Vicksburg, MS. 14
17. Tharp, B. C. 1917. Texas parasitic fungi. Mycologia 9:105-124.
18. Theriot, E. A. 1980. Large-scale Field Application of Cercospora rod-
manii. Pages 491-492, in Proc. 15th Annu. Meet., Aquatic Plant Control Res.
Plan. Operations Rev., Misc. Pap. A-81-3. U.S. Army Waterways Experiment
Station, Vicksburg, MS.
19. Theriot, E. A. 1982. Large-scale Operations Management Test with
Insects and Pathogens for the Control of Waterhyacinth in Louisiana. Pages
187-192, in Proc. 16th Annu. Meet., Aquatic Plant Control Res. Plan. Opera-
tions Rev., Misc. Pap. A-82-3. U.S. Army Waterways Experiment Station,
Vicksburg, MS.
20. Theriot, E. A., R. F. Theriot, and D. R. Sanders, Sr. 1981. Evaluation of
the Infectivity of a Cercospora rodmanii Formulation Using Two Application
Systems. Misc. Pap. A-81-2. U.S. Army Waterways Experiment Station,
Vicksburg, MS. 23 pp.
21. Theriot, E. A., R. F. Theriot, and D. R. Sanders, Sr. 1981. Evaluation of
a Formulation of Cercospora rodmanii for Infectivity and Pathogenicity of
Waterhyacinth. Misc. Pap. A-81-5. U.S. Army Waterways Experiment Sta-
tion, Vicksburg, MS. 34 pp.

This publication was promulgated at an annual cost of $1553
or a cost of 78 per copy to provide information on a biocontrol
agent for waterhyacinth.

All programs and related activities sponsored or assisted by the Florida
Agricultural Experiment Stations are open to all persons regardless of race,
color, national origin, age, sex, or handicap.

ISSN 0096-607X