Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00164
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1966
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00164
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text


Vol. 49, No. 3 September, 1966

ARBOGAST, RICHARD T.-Migration of Agraulis vanilla
(Lepidoptera, Nymphalidae) in Florida......----------.---..... 141
STEGMAIER, CARL E., JR.-Liriomyza commelinae, a Leaf
Miner on Commelina in Florida (Diptera, Agromyz-
idae) ..... ---------------------------....... ..........-----------....... 147
STEGMAIER, CARL E., JR.-A Leaf Miner on Lantana in
Florida, Ophiomyia camarae (Diptera, Agromyzidae) 151
KERR, S. H.-Susceptibility to Miticides of Some Florida
Tetranychus Populations ........... ------------ ----......---- .......... ..... 153
floridana n. sp. (Phycomycetes: Entomophthoraceae),
a Parasite of the Texas Citrus Mite, Eutetranychus
banksi -----... -----------.....---........------ -- 155
Entomophthora floridana Attacking Eutetranychus
banksi .........-------................................----------------------------......................... 161
Malathion, Naled, Fenthion, and Bayer 39007 Thermal
Fogs for Control of the Stable Fly (Dog Fly), Stom-
oxys calcitrans (Diptera: Muscidae) -------.------------ 169
ROBINSON, F. A., AND J. L. NATION-Artificial Diets for
Honey Bees, Apis mellifera-----------------................. ...-------.. .... 175
BROOKS, R. F., AND R. C. BULLOCK-Control of Yellow Scale,
Aonidiella citrina, on Florida Citrus--------....---------.-. 185
FROST, S. W.-Notes on Common Scarabaeidae Taken in
Light Traps at Archbold Biological Station, Florida...- 189
JOHNSON, ROGER B.-Hexachlorophene for Citrus Rust Mite
Control --- ---.--.-.-.....------ 195
MUMA, MARTIN H.-Feeding Behavior of North American
Solpugida (Arachnida) ... ..------------------------ 199
N notices .......................................................................( 145, ) 52, 173 ,
Book Note ----------------------------- 54

Published by The Florida Entomological Society^ ^ ^


President...........................................................................-------------------------------J. R. King
Vice-President.------------------......................................................... J. E. Brogdon
Secretary ....-----------.........---.............................------------------...........-....--....-.........---. S. H. Kerr
Treasurer........................................................................ -------------------------D. H. Habeck
J. B. Gahan
E. D. Harris, Jr.
Other Members of Executive Committee...-..- N. C. Hayslip
J. E. Porter
W. A. Simanton

Publications Committee
Thomas J. Walker..--...................----..----...--.....-------...........Editor
Stratton H. Kerr.............--......-........--------Associate Editor
Dale H. Habeck...........................-----------Business Manager

THE FLORIDA ENTOMOLOGIST is issued quarterly-March, June, Septem-
ber, and December. Subscription price to non-members $5.00 per year in
advance; $1.25 per copy. Entered as second class matter at the post office
at Gainesville, Florida.
Manuscripts and other editorial matter should be sent to the Editor,
Entomology Department, University of Florida, Gainesville. Subscriptions
and orders for back numbers are handled by the Business Manager, Box
12425, University Station, University of Florida, Gainesville. The Secre-
tary can be reached at the same address.
When preparing manuscripts, authors should consult Style Manual for
Biological Journals, 2nd Edition (American Institute of Biological Sciences,
Washington, D. C., 1964). For form of literature citations, see recent
issues of THE FLORIDA ENTOMOLOGIST. Further, authors are re-
ferred to "Suggestions for preparation of manuscripts for THE FLORIDA
ENTOMOLOGIST." Fla. Ent. 48 (2): 145-146. 1965.
One page of figures and/or tables is allowed free. An additional one-
fourth page of figures and/or tables is allowed free for each printed page
beyond the fourth printed page. Authors will be charged $2.50 for each
one-fourth page or less of tables and/or figures in excess of the above
allowances. An author may have his manuscript published as soon as it
has been reviewed and edited by paying the full costs of publication ($10-20
per printed page). Such manuscripts are published in addition to (rather
than instead of) those normally published. Twenty-five reprints of each
article are furnished free to authors. Additional reprints may be ordered
when the proofs are received for corrections.

Each additional
No. Pages 50 copies 100 copies 100 copies
1-4 ................................................ $ 6.80 $ 8.10 $ 1.30
5-8 ................................................ 11.40 14.00 2.70
9-16 ................................................ 16.90 22.20 3.90
17-20 ................................................ 25.00 30.50 5.20
More than 20 pages, per page.... 1.30 1.15 .25
Additional for covers with title and author's name,

First 50 ............................$6.80

Additional, each............$ .02


Department of Entomology, University of Florida, Gainesville

Migration has been defined by Schneider (1962) as a prolonged escape
movement in which there is a tendency to maintain a constant direction
and which results in the permanent or periodical abandonment of a hab-
itat. Many insects migrate in this sense, and numerous accounts of their
migrations can be found in the literature. Recent reviews of insect migra-
tion have been published by Schneider (1962) and Williams (1957, 19.58).
In spite of its widespread occurrence, the migratory habit has been studied
carefully in few species.
The gulf fritillary, Agraulis vanilla (L.), has been reported to migrate
northward through Florida in the spring and southward in the fall (Wil-
liams 1958). The field observations reported here were made as part of
a study of the biology and migratory behavior of this butterfly.


Migrations of Agraulis vanilla were observed during October 1963
and September, October, and November 1964 in a large field near Gaines-
ville, Florida. A circle 50 feet in diameter was laid out in this field and
marked by placing four-foot stakes at twenty-degree intervals around the
circumference. The stakes were labled with large black numerals begin-
ning with 1 at north and continuing clockwise to 18. Observations were
made from stations 8 feet outside the circle. There was one station di-
rectly outside each stake, and the stations were used in random order with
a different station for each day of observations.
Wind speed and direction were measured by means of a cup anemom-
eter and wind vane connected to remote wind speed and direction indicators
(Nassau Windmaster, Model No. 409, Science Associates, Princeton, New
Jersey). The anemometer and wind vane were mounted at a height of
5 feet on a tripod placed 25 feet outside the circle and 20 degrees clock-
wise from the observation station.
In describing the flight speed and direction of a migrating insect, it is
convenient to use the terminology of aircraft navigation as follows:
Track-the migrant's direction relative to the ground.
Ground speed-the migrant's speed relative to the ground.
Course-the direction in which the migrant is heading.
Air speed-the migrant's speed relative to the air.
Wind direction-the direction from which the wind is blowing.
Wind speed-the speed of the wind relative to the ground.
Tracks were determined by recording the numbers of the stakes be-
tween which each migrant entered and left the circle. It was assumed
that the points of entrance and exit were midway between the stakes

'Present mailing address: USDA, ARS, Market Quality Research Di-
vision, Stored-Product Insects Research and Development Laboratory, 3401
Edwin Ave., Savannah, Georgia.

142 The Florida Entomologist Vol. 49, No. 3

through which the migrant passed. A line through the center of the circle
and parallel to the line through these two points gives the track within
10 degrees. Ground speed was determined by measuring with a stop
watch the time required for a migrant to cross the circle and noting the
points of entrance and exit as described above. The wind speed and di-
rection read at the time the migrant left the circle were assumed to repre-
sent the wind speed and direction as it was crossing the circle. The ground
speed, g, in miles per hour is given by the expression

sin 1/2 a
g = 34.1
where t is the time in seconds required for the migrant to cross the circle,
and a is the angle subtending the chord of the circle which represents the
path of the butterfly through the circle.
In calculating air speed and course, angles were measured counterclock-
wise from 0 to 180 degrees and closewise from 0 to -180 degrees with
respect to the vector representing the track, using the tail of the vector as
the origin. The air speed, v, is given in miles per hour by the expression
v = g w cos 9,
where w is the wind speed in miles per hour, and 0 is the angle between
the wind direction and the track. The true bearing of the course, c, is
given by the expression
c =a .

where a is the true bearing of the track, 9 is the angle between the course
and the track, and
w sin 0
sin o =- ----
Since the crosswinds were light during this study, the calculated courses
differed only slightly from the corresponding tracks, and flight direction
will be discussed only in terms of the observed tracks.


Migrating gulf fritillaries fly at a height 3 to 6 feet over open terrain,
and upon encountering an obstacle such as a building or wooded area,
they fly up and over it without changing direction. In general, the flight
is very persistent, but occasionally migrants pause briefly to feed at
flowers. The flight direction of most individuals lies between 110 and 160
degrees (Fig. 1), and does not vary with time of day (Fig. 3). While
the path of the migrant over the earth may be influenced by the wind, the
migratory direction is not determined by this factor (Fig. 2). Instead,
the migratory direction appears to be under the control of the insect it-
The ground speed of a flying animal is a function of its energy output
and of the wind component along its course, while its air speed is a func-
tion only of its energy output. If the animal expends energy at a constant
rate, its ground speed will be less with a head wind than with a tail wind,
while its air speed will be the same in both cases. To maintain a constant


Migration of Agraulis





0--- -ST I031 5'IS 10 1 6- ---

A ........ I O. S




Fig. 1-3. Tracks of migrants. The numbers in the circles represent
the total number of migrants observed in each case. Fig. 1. Tracks ob-
served during the fall of 1963 (left) and 1964 (right). Fig. 2. Tracks and
mean wind observed between 1300 and 1400 EST on various days in the fall
of 1964. Wind direction is indicated in each case by the direction of the
arrow. Each full barb in the tail represents 2 miles per hour of wind
speed. Fig. 3. Tracks observed at various times of day on 23 Sept. 1964.
Fig. 4-6. Flight speed of migrants. Fig. 4. Ground speed observed
in calm air. Fig. 5. Ground speed observed with a head wind of 1-5 miles
per hour (above) and with a tail wind of 1-4 miles per hour (below).
Fig. 6. Air speed observed with a head wind of 1-5 miles per hour (above)
and with a tail wind of 1-4 miles per hour (below).

The Florida Entomologist

ground speed, the animal must vary its energy output to compensate for
the effect of wind, and this will produce a variation in its air speed.
Within the range of observed winds, it appears that each gulf fritillary
expends energy at a more or less constant rate, but this rate varies from
individual to individual. The distribution of ground speed observed in
calm air is presented in Fig. 4. In Fig. 5 the distribution of ground speed
observed with head winds of 1 to 5 miles per hour is compared with that
observed with tail winds of 1 to 4 miles per hour. The same comparison
is made for air speed in Fig. 6. The means are significantly different for
ground speed but not for air sped. It follows that the ground speed will
decrease as the head wind increases until the butterfly must increase its
energy output, land, or be carried backwards. Which of these alternatives
actually occurs was not determined. The head winds did not reach this
magnitude during this study, and they seldom do at the flight level of the
The migration density, as indicated by the number of migrants cross-
ing the circle in one hour, varies considerably from day to day, but in gen-
eral, the migration is heavier during the first half of the migratory period
(Table 1). The density does not appear to be influenced by cloud cover
unless the sky is overcast, and then the migration ceases.

1300-1400 EST (1964).

Number Cloud Cover
Date Crossing Circle Speed Direction

20 Sep. 9 Partly cloudy 00-15 360-060
23 Sep. 12 Clear 00
25 Sep. 7 Partly cloudy 00-07 030-110
29 Sep. 3 Partly cloudy 00-08 100-170
5 Oct. 0 Overcast 06-19 250-280
7 Oct. 1 Partly cloudy 05-12 330-060
9 Oct. 10 Partly cloudy 00-08 360-090
23 Oct. 2 Clear 03-07 340-110
25 Oct. 1 Partly cloudy 02-09 050-100
7 Nov. 3 Partly cloudy 00-08 090-190
18 Nov. 2 Partly cloudy 00-06 330-230

Of 43 migrants captured, 72% were females. Since there is no reason
to believe that females are more easily captured than males, it appears
that while on the basis of laboratory rearing the sex ratio is 1:1 in the total
population of Agraulis vanilla, the females outnumber the males among
the migrants. Of six migrant females dissected, five had spermatophores
in the bursa. Mature eggs were present in the oviducts of three, and in
some, the abdomen was partially empty, suggesting that they had already


Vol. 49, No. 3

Arbogast: Migration of Agraulis

The observations reported here show that Agraulis vanilla migrates
southeastward through Florida in the fall and that the migratory direction
is independent of time of day and wind direction. The following questions
remain to be answered by future research: (1) What is the nature of the
orientation mechanism underlying the unidirectional flight? (2) What
factor or combination of factors initiates the migration? (3) What is the
origin and destination of the migrants?
Answers to these questions will contribute much to our understanding
of insect migration in general.

Schneider, F. 1962. Dispersal and migration. Ann. Rev. Ent. 7: 223-
Williams, C. B. 1957. Insect migration. Ann. Rev. Ent. 2: 163-180.
Williams, C. B. 1958. Insect migration. The New Naturalist, Collins,
London. 235 p.

The Florida Entomologist 49(3) September 1966

In many issues of THE FLORIDA ENTOMOLOGIST a number of one-half-
page spaces are available for items of interest to subscribers. If you have
a news item or a request for information or material that you would like
to appear in THE FLORIDA ENTOMOLOGIST, please send it to the editor.
Such items should be no more than one double-spaced, type-written page.
They will be published at the discretion of the editor as space is available.

Articles requiring one or two printed pages in THE FLORIDA ENTOMOL-
OGIST often can be printed without the usual delay of from 9 to 12 months.
Each issue must contain a total number of pages that is evenly divisible
by four. Consequently in some issues the editor finds that he must either
supply copy for one, two, or three additional pages or publish blank pages
(that cost nearly as much to publish as printed pages!). In such instances
the policy is to use any manuscripts of suitable length regardless of date
of receipt. Manuscripts of four or fewer typed pages are presently in
short supply.



you'll find
All your pest control
needs under
the dependable
FASCO label...

Fumigation Covers
Soil Fungicides
and Insecticides
Foliar Fungicides
and Insecticides
Bulb,Tuberand Rizome
Fungicides and Dips

Delivered when you need them from one of
our warehouses that's near you.




11335 N. W. 59 Ave., Hialeah, Florida 33012

The species here discussed was described by Frost (1931) as Agromyza
commelinae from specimens reared from Commelina virginica L. in St.
Vincent, B.W.I., and from C. nudiflora L. in Cuba. Da Silva and De Oliveira
(1952) reared and described the species and its work in C. nudiflora in sev-
eral localities in Brazil and referred the species to the genus Liriomyza.
E. E. Blanchard (1954) described Liriomyza bahamondesi from C. virginica
in Cordoba, Argentina, but his species was considered to be a synonym of
L. commelinae by Frick (1959) and Spencer (1963). Frick also stated
that he had seen specimens from Trinidad and that the larvae form ser-
pentine mines in leaves of Commelina elegans H.B.K., C. longicaulis Jacq.,
and C. virginica. Spencer also cited personal collections from Rio de Ja-
neiro, Brazil, and from Jamaica, on C. nudiflora, and erroneously spelled
the name of Blanchard's species as "bahomondesi."
Commelina or dayflower, is a genus of the spiderwort family (Com-
melinaceae). About 100 widely distributed species of this genus grow in
warm and temperate zones in the world. Eight species are found in the
United States, along streams, in gardens, and in waste places especially
in the eastern half of the United States, Massachusetts southward to Flor-
ida and westward to Arizona. The plants are not of much economic value,
although in the Asiatic countries some are used for medicinal purposes;
the fleshy rhizomes of some species may be eaten when cooked, as they
contain much starch and mucilage.
The first Florida infestations were collected on Commelina, diffusa
Burm., Miami Springs, 18 Aug. 1962 (B.K.Dozier); the collection rep-
resented the first United States record and a new host record for this
leaf miner.
Observations by the author on the life history of L. commelinae in
Florida are as follows: The larvae form distinct serpentine mines on the
upper surface of the leaves of C. diffusa. Rarely more than one larva
may mine a single leaf providing the leaf is large enough to accommodate
two larvae. Da Silva and De Oliveira (1952) have also found more than
one larva mining a single leaf; they also observed that one larva's mine
sometimes unites (on a single leaf) with a mine of another and that the
two larvae may feed within a common mine. In the latter case, the mine
channel becomes a pronounced blotch mine for a short distance. I have
not made this observation in south Florida.
L. commelinae has always been observed, by the author, to pupate with-
in the mine channel; the large black forms in the leaf mines (Fig. 1) are
pupae and are under the epidermis of the leaf. This finding is in agree-
ment with Da Silva and De. Oliveira (1952) but in contradiction to Saska-
wa (1960), who stated, "All members of the Liriomyza always pupate out-

1 Contribution No. 79, Entomology Section, Div. Plant Industry, Fla.
Dep. Agr., Gainesville.
2 Collaborator, Fla. State Collection of Arthropods, Div. Plant Industry,
Fla. Dep. Agr.

The Florida Entomologist

side the mine." Liriomyza sorosis (Williston) has also been found repeat-
edly, by the author, to pupate within the leaf mine (under the epidermis)
especially on the leaves of Paspalum spp., in South Florida. L. com-
melinae has a characteristic and distinctive serpentine mine: the larvae
deposit a trail of fecal pellets (Fig. 1) at regular intervals throughout
their mines and pupate in the large mine terminals. The feces are the
small black specks within the mine channels. Da Silva and De Oliveira
(1952) do not illustrate the fecal pellets in their diagrammatic drawings
of the leaf mines; however they describe them as "accumulated in alternate
ranks and at the side opposite to that of the excavation." Blanchard
(1954) has a photograph of four leaves, but the pellets do not show. very

a, 4 f

Department of Agriculture, Mildred Eaddy, Photographer.

L. commelinae is presently known to infest Commelina diffuse and C.
spp. in Florida. The northernmost distribution for L. commelinae is
Courtenay, Florida, collected and reared on Commelina sp. (G.W.Desin).
I have found L. commelinae in south Florida all year infesting Commelina
with scattered infestations; severe infestations were noted from September
through December in the greater Miami area during 196i5.
The following rearing records are presented with the hope that they


Vol. 49, No. 3

Stegmaier: Liriomyza commelinae in Florida

will be of interest to entomologists and as a further contribution to the
life history and biology of the Agromyzidae in Florida.
Commelina diffuse: Miami Springs, 18 Aug. 1962 (B.K.Dozier); Hialeah,
27 Aug. 1962 (C.E.S.); Hialeah, 15 Nov. 1965 (C.E.S.); Hialeah, 27
Dec. 1965 (C.E.S.).
Commelina spp.: Courtenay, 3 Sep. 1964 (G.W.Desin); Fisher Island,
Miami, 28 Dec. 1965 (J.E.Porter).
One parasite was reared by the author from Liriomyza commelinae dur-
ing this research in south Florida. Dr. B.D.Burks (Entomology Re-
search Division, U. S. Department of Agriculture) determined the parasite
as Chrysocharis majoriani (Girault), a Neotropical eulophid, occurring in
the West Indies and Brazil. It has not been previously reported from
North America, according to Dr. Burks.

The author is indebted to the following persons for their help in mak-
ing this paper possible: Mr. Harold Denmark (Chief, Entomology Section;
Florida Department of Agriculture) for the photograph; the late Professor
Erdman West (Botany Department, University of Florida) for the plant
determinations; Mr. Kenneth A. Spencer (London, England) for deter-
minations of the leaf miner; Mr. B. K. Dozier (Plant Quarantine Division,
Miami), Dr. J. E. Porter (Quarantine, U. S. Public Health Service, Miami),
and Mr. G. W. Desin (Plant Pest Control Division, Sanford) for collections
of the leaf miner; and Dr. B. D. Burks for the parasite determination. I
especially wish to thank Mr. George Steyskal for reviewing this manu-
script and for his critical comments.

Blanchard, E. E. 1954. Sinopsis de. los Agromizidos Argentinos (Dip-
tera, Agromyzidae). Inst. Sanidad Veget., Min. Agric. Ganaderia,
Republ. Argentina, Publ. No. 56: 1-48.
Frick, K. E. 1959. Synopsis of the species of Agromyzid leaf-miners
described from North America (Diptera). Proc. U.S. Nat. Mus.
108: 347-465.
Frost, S. W. 1931. New species of West Indian Agromyzidae. Ent. News
42: 72-76.
Sasakawa, M. 1960. A study of the Japanese Agromyzidae (Diptera),
Part 1. Sci. Pep. Kyoto Prefec. Univ. Agr. 12: 76-82.
Silva, G. A., Da, and S. J. De Oliveira. 1952. Sobre um "Agromyzidae"
(Diptera) cujas larvas minam folhas de Trapoeiraba (Commelina-
ceae). Rev. Brazil. Biol., 12(3): 293-299.
Spencer, K. A. 1963. A synopsis of the Neotropical Agromyzidae (Dip-
tera). Trans. Roy. Ent. Soc. London. 115(12): 291-389.

The Florida Entomologist 49(3) September 1966





p p.



How can they serve you?

SHELL Chemical Company, in coopera-
tion with federal, state and local agri-
cultural, specialists, is continually striving
to help farmers reach higher agricultural
goals. Products such as aldrin, dieldrin,
endrin, methyl parathion, Phosdrin and
Vapona Insecticides have been of major
assistance to the farmer, homeowner and
industry. So have D-D and Nemagon
Soil Fumigants and Aqualin* herbicide,
slimicide, biocide.
These products have proved their effec-
tiveness and versatility by solving many
of the economic pest problems confront-
ing the farmer. Shell insecticides are prov-

ing equally useful in a growing number
of non-agricultural applications in indus-
try and the home.
The never-ending search for additional
uses of established Shell pesticides and for
new, improved products to help you, is a
welcome assignment at Shell Chemical
Company chemical partner of agricul-
ture and industry.
Get full details about the Shell pesti-
cide that fits your needs'at your nearest
Shell Chemical Co. District Office, or
write: Shell Chemical Co., Agricultural
Chemicals Division, 110 West 51st Street,
New York 20, N. Y.

*Thtoe are more than 130 species of nema-
tO(kes known to attack plants. Netnitgon and
D-D Soil Fumigants control most of these.

Agricultural Chemicals Division


11335 N. W. 59 Ave., Hialeah, Florida 33012

Spencer (1963) described Ophiomyia camarae from specimens, reared
by F. D. Bennett, infesting leaves of Lantana camara L. in Trinidad,
B.W.I., July, 1962. Spencer stated that the leaf mine pattern of Ophio-
myia camarae is a narrow channel running primarily along the midrib,
with offshoots along the lateral veins, and that pupation takes place with-
in the leafmine channel. The three leaves in Fig. 1 illustrate the varia-
tions in the form of the leaf mines.

Fig. 1. Leaf mine variations of Ophiomyia camarae in leaves of Lan-
tana camera. Center leaf shows processes or offshoots from the midrib.
Leaf on right shows the mine processes along the lateral veins. Photo-
graph courtesy of the Division of Plant Industry, Florida Department of
Agriculture, Mildred Eaddy, Photographer.

0. camarae is presently known from a single host, Lantana camera.
The species infests Lantana during the entire year in South Florida. Leaves
of this ornamental are blemished with numerous unsightly leaf mines;
some plants have every leaf with one or more leaf mines.
The author collected and reared 0. camarae from L. camera, Hialeah,
Fla., 5 April 1963 (first emergence of adults began 17 April 1963). Mined
leaves of L. camera were collected by H. V. Weems, Jr., Key West Botani-
cal Garden, Key West, Fla., 7 June 1964. Leaf mines in the photo were
collected by the author in South Miami, Fla., 16 Nov. 1965.

1 Contribution No. 80, Entomology Section, Div. Plant Industry, Fla.
Dept. Agr., Gainesville.
Collaborator, Fla. State Collection of Arthropods, Div. Plant Industry,
Fla. Dep. Agr.

The Florida Entomologist

Ophiomyia camarae occurs in Florida from greater Miami southward
to Key West wherever L. camera grows. Other species of Lantana are
potential hosts for 0. camarae; more information is needed on its biology,
distribution, and parasites.
I am indebted to Mr. Kenneth Spencer of London, England, for the de-
termination of 0. camarae and for his many helpful suggestions and to
Dr. Howard V. Weems, Jr., Florida Department of Agriculture, for his
assistance with regard to collections. I wish to thank Mr. Spencer and
Mr. George Steyskal for reviewing this paper.

Spencer, K. A. 1963. A synopsis of the Neotropical Agromyzidae (Dip-
tera). Trans. Roy. Ent. Soc. London 115(12): 291-389.

The Florida Entomologist 49(3) September 1966

In the article "Two larval rearing media for Ips bark beetles" by W. C.
Yearian and R. C. Wilkinson (Vol. 48, No. 1, p. 25-27) the amounts of
two of the constituents of the modified artificial medium (p. 25) need
correction. The correct amounts are as follows:

(1) Modified artificial medium
Alphacel (powdered cellulose)' 10.00 g


Vol. 49, No. 3


82.00 ml




Department of Entomology, University of Florida, Gainesville

Kerr (1964) reported on the first of a proposed series of studies with
economically important Florida arthropods to determine the present status
of susceptibility to pesticides and to establish base lines of susceptibility
for comparative purposes at later times. The present report is an exten-
sion of this work and includes data regarding the use of parathion and
ethion against populations of Tetranychus urticae Koch collected from
strawberries in Apopka, ligustrum in West Palm Beach, and roses in
Gainesville. The mites were collected during the fall of 1964, and the
testing done during the winter of 1964-65.
The mite colonies were reared and the laboratory tests were conducted
as in the earlier studies (Kerr 1964). The dosage series of parathion was:
0.075%, 0.0125%, 0.0175%, 0.0225%, 0.0275%, and 0.0325%. The ethion
dosages were: 0.005%, 0.01%, 0.015%, 0.02%, 0.025%, and 0.03%. The
probit analysis was performed by a computer using program number
UCRBL 50 written by the University of California, Riverside. Dr. F. G.
Martin, Department of Statistics, University of Florida, cooperated.
Table 1 presents the results. The LC50o figures for both chemicals
against all three populations were not significantly different. The slopes


Parathion Ethion
LC o. LCg, LC50 LC95

Apopka Population 0.0180% 0.0413% 0.0168% 0.0461%
lower limit* 0.0132% 0.0132% 0.0137% 0.0353%
upper limit 0.0215% 0.0766% 0.0195% 0.0794%
Gainesville Population 0.0162% 0.0351% 0.0137% 0.0406%
lower limit 0.0145% 0.0309% 0.0124% 0.0354%
upper limit 0.0177% 0.0422% 0.0149% 0.0491%
West Palm Beach Population 0.0212%0 0.0170% 0.0513%
lower limit 0.0129% 0.0364%
upper limit 0.0207% 0.1203%

*Fiducial Limit at 95% confidence level
fEye-fitted curve; incomplete data.

were very similar and the LC,, figures did not differ significantly either.
The LC50's for parathion were lower than those for the Fort Myers and
Stuart populations of T. urticae in the 1963-64 tests. The present LCs0's
of 0.016-0.021% would still be considered in the resistant range, using tests

The Florida Entomologist

in other states for comparison. There are no earlier base lines using Flor-
ida populations with which to compare these results.

Kerr, S. H. 1964. Some tetranychid mites on Florida ornamental crops,
and laboratory studies of their susceptibilities to miticides. Proc.
Fla. State Hort. Soc. 77: 481-484.

The Florida Entomologist 49(3) September 1966


INSECT BEHAVIOR. Symposium No. 3, Royal Entomological Society
of London, 41, Queen's Gate, London, S.W.7. Edited by P. T. Haskell.
1966. 113 p. illus. clothbound. $6.50.
This is the third volume in a praisworthy series that is known to too
few American entomologists. The first two volumes (Insect Polymorph-
ism, 1961; Insect Reproduction, 1964), like this one, contain valuable re-
view articles by eight or more world authorities on a single topic of in-
terest to nearly all entomologists.
Symposia are often organized with a single principal purpose-to pro-
mote exchange of ideas among specialists. When the results of such sym-
posia are published, they are often useful only to specialists, if at all.
The symposia of the Royal Entomological Society of London have an ad-
ditional principal purpose-to provide authoritative, up-to-date outlines of
some of the varied aspects of a subject for entomologists in general. The
published volumes do just that.
The present volume contains articles on the following aspects of insect
behavior: orientation (G. Birukow, Germany); rhythms (P. S. Corbet,
Canada); flight (P. T. Haskell, England); feeding behavior (V. G. Dethier,
U.S.A.); sexual behavior (A. Manning, Scotland); communication (J. D.
Carthy, England); social insects (E. 0. Wilson, U.S.A.); outstanding ques-
tions (J. S. Kennedy, England).-TJW


Vol. 49, No. 3


Pathologist with Pathologie Hmyzu, Entomologicky Ustav Ceskoslovenske
Akademie Ved, Praha, and Entomologist with University of Florida
Citrus Experiment Station, Lake Alfred

Recently, Selhime and Muma (1966) studied the biology of an entomoph-
thoraceous fungus attacking the Texas citrus mite, Eutetranychus banks
(McGregor). This study and a series of earlier publications including Fish-
er (1954), Muma (1955 and 1958) and Muma et al. (1961) have indicated
the unusual frequency of the fungus in the field and its potential impor-
tance in the biological control of the Texas Citrus mite. This fungus, de-
scribed in the present paper, is not the only Entomophthora known to infect
Acarina. Petch (1940) described Entomophthora acaricida from Halo-
tydeus destructor Tucker and later, Petch (1944) described E. acaridis from
other infected Acarina. Recently, Batko (1965) recorded Conidiobolus
brefeldianus Couch from Tyrophagus perniciosus Zachvatkin and other
tyroglyphid mites.

Stained fungus preparations were made from infected mites treated
with one percent cotton blue in Amman's lactophenol. Unstained prepara-
tions were made in Swan's fluid. Nuclei in hyphal smears were stained
with Heidenhain's iron hematoxylin.
Fungus infected mites were placed on cover slips in petri dishes with
wet filter paper to produce conidia and anadhesive spores.
The description given below was prepared by the senior author from
material collected by the junior author.

Entomophthora floridana, new species

Mycelio septato, fragmentario, hyphis bi-seu quadrinucleariis 3-4A latis
et 30-351 longis. Conidiophoris inflatis 30-35A longis et 6-8/ latis, simplicis,
conidio solitario in apice. Conidiis piriformis, hyalinis, quadrinucleariis, typo
papillato, subpapillato seu epapillato sensu Lakon (1919), 13-18u longis et
11-13g latis cumpapilla 5-6A lata. Columella persistens. Conidia non-
adhesiva clavata, superficie brunea et striata 15-201 long et 10-12,u lata,
apice papilloso, adhesive, 1.5A long. Sporis perdurantibus (Zygosporis)
subsphaericis seu sphaericus in gametangio hyalino, cum foramine rotundo
4g in diametro, exo-et endosporio tenui atque transparent, 20-23.51 latis et
22-26/ longis. Hospes typicus: Eutetranychus banksi (McGregor), Lake
Alfred, Florida, U.S.A.
Mycelium divided into short tubular or club-shaped hyphal bodies with
2-4 nuclei, growing into curved and obtuse, .mostly unbranched segments
(P1. I, Fig. 16). They are distributed throughout the host body, not in a
compact network (Pl. II, Fig. 1). Root-like hyphae grow through the
host cuticle and form slightly broadened, sometimes curved conidiophores

The Florida Entomologist

outside of the mite body (Pl. I, Fig. 8 and 9). Conidiophores are single,
30-35g X 6-8A, growing from a root-like hypha of the same length but
only 3-4/ wide. Conidiophore protoplasm with numerous refringent gran-
ules, fat droplets or starch. Hyphal protoplasm hyaline, without granules
in the septal region; septum formed at periphery (P1. I, Fig. 9).

Entomophthora floridana n. sp. Fig. 1-7, 10-15 and 17-19 conidia, in
Fig. 1 and 5 the nuclei. Fig. 8, 9 conidiophores, Fig. 16 hyphal body, Fig.
20, 21, 25-27 anadhesive spores. Fig. 22-24 and 28-30 resting spores.


Vol. 49, No. 3

Weiser: Entomophthora floridana n. sp.

Conidia are pyriform (Pl. I, Fig. 1-7, 10-15, 17-19), papillate, subpapil-
late, or epapillate type, and 13-18g X 11-13g averaging 12 X 15i with the
basal end 5-6, wide. Conidia with 4 spherical nuclei and refringent granules
of fat and starch. These structures are not stained by cotton blue in Am-
man's solution. Conidiophores have persistent columellae, not bursting
during the spore discharge. There is no gelatinous substance on the surface
of conidia. Prior to discharge, both the membranes of the columella and
that of the conidium are flat, not inflated. In fixed mounts, conidia on con-
idiophores are similar to the truncate type of Lakon's (1919) classification
(Pl. II, Fig. 2). Secondary conidia are formed from primary conidia by
a single, lateral, hyphal bud. Primary and secondary conidia are the same
size and shape. Microconidia have not been observed. Anadhesive spores
are produced by primary and secondary conidia at the ends of thin threads
1.5g X 50-60,u. These threads which are curved adjacent to the anadhesive
spore retain a remnant of the conidial membrane at the opposite end.
Mature anadhesive spores are claviform, 15-20g X 10-12A. At the narrow
end, is a knob-like apex 1.51 wide. Anadhesive spores have a brownish
striated cuticle; the apex seems to be adhesive because almost all anadhesive
spores become attached to mite setae and cuticle at this end (Pl. II, Fig.
3). Primary anadhesive spores produce secondary anadhesive spores at
the end of a capillary tube. These spores are apparently disseminated by
air currents.
Resting spores, seldom seen in field collected material, are spherical or
subspherical with a smooth thin three-layered wall. The upper wall of
the resting spore has a rounded foramen at the former connection with the
hypha. Resting spores have small refringent globules distributed through-
out the protoplasm but no oil globules or vacuoles. Resting spores are
20-23.5 X 22-261 in diameter, with the wall only 0.5/ thick and the foramen
4g in diameter (Pl. I, Fig. 22-24, 28-30).
Typical host: Eutetranychus banksi (McGregor) on citrus.
Type locality: Lake Alfred, Florida, U.S.A.
Type specimens: Deposited at Entomology Department, Czechoslovakia
Academy of Sciences, Praha.


This fungus differs from known Entomophthora in the size and shape
of the developmental stages, in the surface sculpturing of the anadhesive
spores, and in its host relationship. It is considered to be a new species
of the genus Entomophthora sensu lat.
The comparative morphology of the Entomophthoraceae is not suffi-
ciently well known for the placement of E. floridana among the genera
proposed by Batko (1964 and 1965). Further, many authors have not ac-
cepted this system of classification. For these reasons, the fungus de-
scribed here has been placed in the collective genus Entomophthora. It
should be noted, however, that quadrinuclear conidia are typical for Triplo-
sporium Thaxter, recognized as a genus by Batko (1964). Further, all
species of Triplosporium have conidia and anadhesive spores that are sim-
ilar to those of E. floridana. Anadhesive spores and microconidia are also
produced by Conidiobolus coronatus Constantin but microconidia are not
found in E. floridana. E. floridana differs from E. (T.) fresenii (Thaxter)
Nowakowski in size of conidia and sclerotization of resting spores. E. (T.)

The Florida Entomologist


Q. x

Vol. 49, No. 3

Weiser: Entomophthora floridana n. sp. 159

lageniformis (Thaxter) has much larger conidian than E. floridana. The
closely related E. (T.) fumosum (Thaxter) Speare occurs in Pseudococcus
citri Risso on citrus, but it has long elliptical to cone-shaped conidia and
dark brown to black resting spores. Owing to the scarcity of resting spores
of E. floridana from mites, it is impossible to compare and discuss the
formation of resting spores.


Batko, A. 1964. On the new genera:Zoophthora gen. nov., Triplosporium
(Thaxter) gen. nov. and Entomophaga gen. nov. (Phycomycetes:
Entomophthoraceae). Bull. Acad. Polon. Sci., S6r. Sci. Biol. 12:
Batko, A. 1965. Monograph of Polish Entomophthoraceae. Unpublished
Ph.D. thesis, Inst. Bot., Univ. 'Warsaw.
Fisher, F. E. 1954. Diseases of citrus insects. Fla. Agr. Exp. Sta. Ann.
Rep. for 1953, p. 191.
Lakon, G. 1919. Die Insektenfeinde aus der Familie der Entomophthora-
ceae. Z. Angew. Ent. 5: 161-216.
Muma, M. H. 1955. Factors contributing to the natural control of citrus
insects and mites in Florida. J. Econ. Ent. 48:432-438.
Muma, M. H. 1958. Predators and parasites of citrus mites in Florida.
Proc. Xth. Int. Congr. Ent. 4: 633-647.
Muma, M. H., A. G. Selhime, and H. A. Denmark. 1961. An annotated
list of predators and parasites associated with insects and mites on
Florida citrus. Fla. Agr. Exp. Sta. Tech. Bull. 634: 1-39.
Petch, T. 1940. An Empusa on a mite. Proc. Linnean Soc. N.S.Wales
65: 259-260.
Petch, T. 1944. Notes on entomogenous fungi. Trans. Brit. Mycol. Soc.
27: 81-93.
Selhime, A. G. and M. H. Muma. 1966. Biological studies on an Ento-
mophthora attacking Eutetranychus banksi (McGregor). Fla. Ent.

The Florida Entomologist 49(3) September 1966

Fig. 1. Eutetranychus banksi infected with Entomophthora floridana
n. sp., stained with cotton blue in Amman's solution. Fig. 2. Conidia of
E. floridana. Fig. 3. Anadhesive spores of E. floridana on leg of E. banksi.



qI ,-




Big DaddyTM


Banvel D
Banvel' M
Banvel D

Velsicol's continuing research and develop-
ment program, which has produced these
major pesticides, generates a good deal of in-
formation about insect, weed, and plant dis-
ease control. We would welcome the oppor-



Velsicol "2-1"
Memmi .BEC


Methyl Bromide

tunity to share this with you. May we send you
our technical and educational literature?
*1 341 E. Ohio Street Chicago, Illinois 60611

"C Cu corporation
L 0 vueet'r nal
Germ h O utI -elslco:l e

ADOD s--S .S p- --


Entomology Research Division, Agricultural Research Service,
U. S. Department of Agriculture, and University of Florida,
Citrus Experiment Station, Lake Alfred, Florida

During experiments designed to evaluate the biological control of the
Texas citrus mite, Eutetranychus banksi (McGregor), in citrus groves, it
was found that counts of these mites and the known biological-control
factors present did not adequately explain population increases and de-
creases. There are several species of predatory mites and a fungus, En-
tomophthora floridana Weiser and Muma, known to attack the Texas
citrus mite. Need to increase count accuracy by refinement of counting
technique was indicated. Since methods of counting predators in host mite
samples had been verified it was suspected that the error involved estimates
of diseased host mites. A laboratory investigation demonstrated that
this was true.
When Texas citrus mites were cleared in hot lactic acid, fungus tissues
and mite structures were sharply differentiated. This led to a rapid, ac-
curate determination of the number of fungus-infected and noninfected
mites. At this point, it became obvious that a more complete knowledge
of the life cycle and life stages of the fungus would be necessary to differ-
entiate it from adventitious and saprophytic fungi also found on and in
the mites.
Review of pertinent literature revealed the following: Fisher (1954)
first recorded an Entomophthora sp. from Texas citrus mites, noting that
it appeared to be different from the unidentified species that she (Fisher
1951) had reported attacking citrus red mites, Panonychus citri. In her
1951 paper, Fisher discussed gross manifestations of diseased mites and
presented a general discussion of the life cycle of Entomophthora taken
from Steinhaus (1949), Bessey (1950), and Fitzpatrick (1930). Other
publications on Entomophthora in citrus spider mites merely listed the
association (Muma, Selhime, and Denmark, 1961), or discussed and evalu-
ated fungus-caused epizootics in two species of spider mites (Muma 1955,
1958). The present study was initiated to determine the infective stage
and life cycle of, and the epizootiology of the disease caused by E. floridana
in Texas citrus mites. Susceptibility of the citrus red mites and six-spot-
ted mites, Eotetranychus sexmaculatus, to the fungus was also investigated.

Investigations of the infective stages and life cycle of E. floridana were
conducted in the laboratory with healthy and diseased mites collected from
citrus groves. Mites presumed to be diseased were cleared in hot lactic
acid and the presence of the fungus was verified. The same procedure was
used to verify grove-collected healthy mites. Diseased mites were confined
in petri dishes until the fungus produced conidia and the conidia had dis-
persed. Healthy mites were then exposed to the fungus and isolated in
petri dishes, on leaves or on fruit, for development of disease symptoms,
until they died or until conidia were produced. Conditions of exposure,

The Florida Entomologist

isolation, and development were varied to determine the phases of the
infective stage, the life cycle, and the interspecific infectivity of the fungus.
Grove studies of the epizootiology of the disease caused by E. floridana
involved weekly samples of leaves from commercial citrus groves. These
samples consisted of 120 randomly selected terminal leaves, 2 from each
of 6 terminal twigs from each of 10 trees.
Rainfall and temperature data came from U. S. Weather Bureau sum-
maries taken at four stations encompassing the study area.

The infective stage of the fungus was determined by exposing Texas
citrus mites, citrus red mites, and six-spotted mites to primary conidia and
anadhesive spores under conditions of high humidity, low humidity, and
condensed moisture.
Experimentation resulted in the production of haloess" of primary con-
idia several millimeters in diameter around dead, grove collected, diseased
Texas citrus mites. Haloes were not produced under low humidities. Wa-
ter films and high humidities resulted in the production of haloes but not
invariably. Haloes of primary conidia were produced in water films,
whereas primary conidia and anadhesive spores were both obtained with
high humidities. In the latter conditions haloes of primary conidia appar-
ently produced anadhesive spores under low or varying humidities that
existed at the time of examination.
Healthy Texas citrus mites were exposed by allowing them to walk
through conidial haloes. The mites were then isolated. At 24-hour in-
tervals, exposed mites were cleared, mounted, and compared with a sim-
ilar number of unexposed mites. Data from a series of infection tests
showed a positive correlation between humidity and infectivity (Table 1).
Condensed moisture on the haloes after anadhesive spores had been pro-
duced resulted in the highest incidence of infection.

TABLE 1. INCIDENCE OF Entomnophthora floridana IN TEXAS CITRUS

Incidence (%) of
Number of Conditions Entomophthora*
specimens of at the indicated hours
exposed exposure 24 48 72 96 120 144

120 Low humidity** 0 0 0
129 Low humidity** 3 5 5 11 -
26 High humidity** 50 80 20 -
75 Condensed moisture** 83 60 81 -
30 Low humidity 0 0 -
40 High humidity .05 -
35 Condensed moisture 0 .04 -

*There was no disease incidence in any control series
**Anadhesive spores
tPrimary conidia


Vol. 49, No. 3

Selhime: Biology of Entomophthora floridana 163

These infection tests did not rule out the possibility that primary
conidia may cause higher rates of infection if they make direct contact
and adhere to the healthy mites. Our observations indicate that the pri-
mary conidia do not adhere well to the mites. Thus it is probable that the
anadhesive spores are the principal source of infection.
Secondary and tertiary anadhesive spores were demonstrated in the
studies, but no test was devised to compare their infectivity with that of
primary anadhesive spores. Primary anadhesive spores are the infective
stage preponderately found on grove-collected mites, so it seems likely that
secondary and tertiary spores play a minor role in infection.

Infectivity tests which mites were mounted and examined immediately
after exposure to the fungus and at 24-hour intervals for 120 hours indi-
cated the following life cycle for this species of Entomophthora.

3 4

Fig. 1. Anadhesive spore attached to leg. Fig. 2. Anadhesive spore
attached to body; note germination. Fig. 3. Hyphal bodies. Fig. 4. Body
filled with hyphae. Fig. 5. Hyphae in leg. Fig. 6. Resting spores.
Fi.e.d wit yhaeig. 5.or Hyphaed i leg. Fig. 6. Retng es. spore

The Florida Entomologist

Vol. 49, No. 3

Anadhesive spores become attached to the legs and lower portions of
the body, most frequently the former (Fig. 1 and 2). They then germinate
and grow through the integument. Inside the mite, the fungus expands to
hyphal width (Fig. 2). From points of infection on the mite, the fungus
extends hyphae throughout the lower regions of the body. After an un-
determined length of time, the hyphae break up into hyphal bodies (Fig.
3). The hyphal bodies appear either to grow and divide until the body
cavity (Fig. 4), mouth parts, and legs (Fig. 5) of the mite are completely
filled, or to grow into hyphae that completely fill the mite.
Death of the mite appears in most instances to be associated with com-
pletion of internal fungus growth. Some male and a few female mites
die before internal growth of the fungus is complete.
When the mite body is completely filled with fungus hyphae or hyphal
bodies, conidiophores are produced. Two types of conidiophores were ob-

~( 0

.5 ~ Ni44A


A'.:?^ .-t.f>< ... -*. ** '* .: ..*..,--, t-.. l. t. -
.~1 1

.: s 4 t.t. > .:.,t: S< -.
A. : ;,, ::: J :.


Fig. 7. Hyphae flattened against body wall. Fig. 8. Short, subconical
conidiophores and conidia. Fig. 9. Elongate, slender conidiophores and
conidia. Fig. 10. Primary conidia. Fig. 11. Development of anadhesive
spores. Fig. 12. Development of secondary anadhesive spores.


Selhime: Biology of Entomophthora floridana

served under laboratory and grove conditions. Under conditions of total
submersion in water in the laboratory and a water film produced by rain
or heavy dew in groves, the conidiophore is the same width as the internal
hyphae and quite elongate (Fig. 9). This type of conidiophore is distinctly
smaller in diameter than the greatest width of the pear-shaped primary
conidium. Under laboratory conditions of high humidity, but without a
visible water film, and under the same conditions in groves, the conidiophore
is short and conical with the base as wide or wider than the greatest width
of the primary conidium (Fig. 8). In mites with short conidiophores, the
internal hyphae appear to have pressed against the inside surfaces of the
dorsal and lateral body wall of the mite and flattened into broad disks
from which the broad-based conidiophores develop (Fig. 7).
Primary conidia (Fig. 10), produced in a film of water, break free and
float away from the conidiophores. Those produced under high or fluctuat-
ing humidities are propelled varying distances of several millimeters away
from the dead mites. Primary conidia produced in a film of water did not
produce germ tubes or anadhesive spores. In several tests in which con-
ditions approached a relatively humidity of 100%, primary conidia were
disseminated but did not produce anadhesive spores. Under most condi-
tions, however, a primary conidium that fell on a dry surface germinated
a fine, erect tendril and produced a terminal anadhesive spore (Fig. 11).
Secondary and tertiary anadhesive spores were produced in the same man-
ner (Fig. 12), apparently under conditions of varying humidities. These
secondary and tertiary spores, each smaller than the former, were not pro-
duced abundantly. Most haloes contained a predominance of primary an-
adhesive spores.
Subspherical bodies (Fig. 6) were observed in several deutonymphs one
protonymph and one male. So far we have not observed them in females.
Because of the limited number of specimens that contained these bodies we
could not determine their biological function. In several tests, diseased
mites, dehydrated or chilled for short periods of time, did not produce
these bodies. It is not known whether these subspherical bodies (resting
spores) are chlamydospores, zygospores, or azygospores. They are, how-
ever, always found in mite infestations infected with E. floridana.


Hours Dead mites held
after for halo** Dead mites
exposure No halo Halo No Ent. Ent.

96 3 0 2 1
120 10 7 1 8
144 22 7 5 16
168 10 2 4 6
192t 2 4 0 2
216 7 1 3 3

*Mode cycle-5 to 6 days; mean cycle-6.3 days
**Dead mites held for halo production 24 hours
tEvidence of reinfection noted


The Florida Entomologist

At a constant temperature of 26C under laboratory conditions, com-
pletion of a life cycle from infection of a host to production of the infec-
tive conidia takes a modal period of time varying from 5 to 6 days (Table
2). The higher and fluctuating temperatures that prevail in Florida citrus
groves during the late spring and summer may appreciably alter this
cycle time. There is no explanation as to why only a third of the experi-
mentally diseased mites produced haloes to complete the cycle.


Since an Entomophthora has been reported from citrus red mite (Fish-
er 1951), it was felt that infectivity of the fungus under study should be
checked in other spider mites.
Two tests comparing rates of infection of Texas citrus mites and citrus
red mites alternately exposed to secondary conidia resulted in average
infections of 35.2 and 17.3 per cent, respectively. A single test comparing
infection rates of Texas citrus mites and six-spotted mites resulted in in-
infections of 60.0 and 20.0 percent, respectively. These limited compari-
sons indicated that the species of fungus that attacks Texas citrus mites
will also attack the two other common spider mites on Florida citrus but
at a much lower incidence.
The data on six-spotted mites are academic. This spider mite primarily
infests the under surfaces of leaves well beneath the tree canopy and would
not, under ordinary circumstances, be exposed to infection. There are no
records of an Entomophthora attacking six-spotted mites under grove con-
With the citrus red mite, the cross infectivity obtained indicated two
possibilities. The first and most obvious is that the two fungi reported
by Fisher (1951, 1954) may be conspecific. On the other hand, the much
lower rate of infection in citrus red mites may indicate a degree of host
specificity and the existence of two species of fungi.


Although laboratory studies were primarily an investigation of infec-
tion and life cycle of the fungus in adult females, under grove conditions
E. floridana occurred in protonymphs, deutonymphs, males, and females
of the Texas citrus mite. It was not found in eggs or larvae.
The following data for rate of infection from 31 samples collected from
10 groves between 20 April 1964 and 27 July 1964 were obtained. Only
samples that exhibited disease in one or more stages of the mite are in-
cluded. Protonymph infection ranged from 0 to 16.5 per cent with an
average of 1.9, that of deutonymphs from 0 to 20.2 percent with an aver-
age of 2.2, and that of adult females from 0 to 49.2 percent with an average
of 5.7. Although numerous males were diseased, an infection rate for this
sex was not calculated.
Since the above infection rates indicated that control potential was
variable, even when data were biased to incidence of disease, it was de-
cided to summarize the total data accumulated from the 10 groves. Inci-
dence of disease and rainfall are compared with the populations of healthy
and diseased mites in Table 3. These data indicate that a high incidence


Vol. 49, No. 3

Selhime: Biology of Entomophthora floridana

and rate of attack by E. floridana are dependent upon both high host densi-
ty and regular, though not necessarily heavy, rainfall.


INCIDENCE AND RATE OF ATTACK OF Entomophthora floridana

Female mites Rainfall

Sample No. Groves with Mean Mean
week Entomophthora* Total Diseased** inches days

April 27 1 287 2 0.55 1.5
May 4 0 99 0 4.70 4.8
11 1 225 2 trace 0.0
18 2 125 2 0.08 1.0
25 1 282 0 0.25 0.3
June 1 0 377 0 0.20 0.3
8 3 670 60 1.09 3.8
15 4 733 30 0.35 2.0
22 2 712 10 0.06 0.8
29 3 743 33 1.40 3.8
July 6 6 443 108 2.58 6.0
13 5 259 43 0.33 2.0
20 2 142 6 1.61 3.8
27 1 79 6 2.27 5.5

*A grove was considered infected if any specimen of any stage of the host-mite was infected.
**Based on subsamples of a maximum of 120 females per grove.
tAverage number of days with measurable rainfall at 4 local stations.

This knowledge of the limitations induced by host density and moisture
prompted a second study of infection rates. Sixteen samples of adult
female mites were collected from a moderately infested grove during a
rainy period between 4 and 10 August 1964. Infection rates from this
series ranged from a low of 12.1 to a high of 100 percent with a mean of
49.3 and a mode of 40 to 60 percent. These values are probably repre-
sentative of the infection rate of E. floridana under optimum conditions.
Muma (1958) and Simanton (196.0, 1965) reported seasonal fluctuations
of Texas citrus mites. Heavy populations normally occur in April, May,
and June but are greatly decreased in July, August, and September. Mod-
erate populations occur in October and November but are reduced in De-
cember and January. The above data offer a possible explanation for
these fluctuations. Epizootics of disease caused by E. floridana may de-
plete moderate to heavy mite infestations at the onset of regular summer
rains in July and August and with the increase in ground fog and dews in
December and January.

In laboratory and grove studies Entomophthora floridana infecting
Texas citrus mite, Eutetranychus banksi, was determined to have a 5- to
6-day life cycle. Anadhesive spores appeared to be the principal infective


168 The Florida Entomologist Vol. 49, No. 3

stage and death of the mite was associated with completion of internal
hyphal growth. The fungus also attacked citrus red mite and six-spotted
mite, but the rate of infection was much lower than in Texas citrus mite.
Grove studies showed that host density and moisture, as represented
by regular rains, were the factors limiting epizootics caused by E. flor-
idana. Analyses of fluctuations of Texas citrus mite populations indicated
that the fungus may act as a natural-control agent.

Bessey, E. A. 1950. Morphology and taxonomy of fungi. Blakiston Co.,
Philadelphia. 791 p.
Fisher, F. E. 1951. An Entomophthora attacking citrus red mite. Fla.
Ent. 34: 83-88.
Fisher, F. E. 1954. Diseases of citrus insects. Fla. Agr. Exp. Sta. Ann.
Rep. :for 1953: 191.
Fitzpatrick, H. M. 1930. The lower fungi-Phycomycetes. McGraw-Hill
Book Co, New York. 331 p.
Muma, M. H. 1955. Factors contributing to the natural control of citrus
insects and mites in Florida. J. Econ. Ent. 4: 432-438.
Muma, M. H. 1958. Predators and parasites on citrus mites in Florida.
Proc. 10th Int. Congr. Ent., Montreal, 1956. 4: 633-647.
Muma, M. H., A. G. Selhime and H. A. Denmark. 1961. An annotated
list of predators and parasites associated with insects and mites on
Florida citrus. Fla. Agr. Exp. Sta. Tech. Bull. 634: 1-39.
Simanton, W. A. 1960. Seasonal populations of citrus insects and mites
in commercial groves. Fla. Ent. 43: 49-57.
Simanton, W. A. 1965. Mite populations in Florida citrus groves before
and after severe freezes. Proc. 12th Int. Congr. Ent., London 1964.
p. 364-366.
Steinhaus, E. A. 1949. Principles of insect pathology. McGraw-Hill
Book Co, New York. 757 p.

The Florida Entomologist 49(3) September 1966


*~t~ t.
tk.. a~

*73. f






Entomology Research Division, Agr. Res. Serv., USDA, Gainesville, Fla.

The stable fly, Stomoxys calcitrans (L.), or dog fly, as it is commonly
called in northwest Florida, has been a serious nuisance in that area for
many years. Blakeslee (1945) found that DDT gave effective control when
it was applied as a residual spray to larval breeding areas in marine grass
deposits. However, in recent years this method has not provided the de-
gree of control necessary. As a result, large numbers of adult flies have
at times seriously affected the tourist industry of northwest Florida. On
the basis of our work, and that of others, with thermal fogs for mosquito
control, we felt that this method might be used effectively for rapid elim-
ination or reduction of adult stable fly populations. As a preliminary step
to evaluating this method, we determined the toxicity of a large number of
insecticides to stable flies in laboratory tests. From these we selected four
chemicals for thermal fog tests in the field with caged stable flies: mala-
thion, fenthion, naled, and Bayer 39007 (o-isopropoxyphenyl methylcarba-
mate). This paper presents the results of our laboratory and field tests
with these compounds.

The strain of stable flies used was established from larvae collected at
a livestock farm near Panama City, Florida, and reared in our laboratory
for more than a year.
In the laboratory tests, 3-7 day old adult stable flies were used. We
confined 25 females or 25 males in a cylindrical screen cage, and exposed
them to contact sprays in a wind tunnel 16-18 hours after they had been
fed citrated beef blood. One cage of males and one of females were used
per test, and three to six tests were conducted at each concentration. The
sprays were kerosene solutions of various concentrations of each insecticide
(selected to provide adequate dosage-mortality curves) atomized at a pres-
sure of 1 psi and drawn through the wind tunnel at a speed of 4 mph.
Knockdown counts were made 1 hour after exposure; mortality was deter-
mined after 24 hours.
In the field tests, 3-7 day old unsexed adult stable flies were used. We
confined 25 flies in a 16-mesh screen wire cage and exposed them to fogs
16-18 hours after they had fed on blood. The cages (4/test) were hung 5
feet above ground on stakes placed in two rows 125 feet apart and 125
and 250 feet downwind from the line of passage of the fogging vehicle.
The flies were protected from high temperatures after removal to the field
(except during fogging) by holding them in insulated chests containing
cans of ice. One to two hours after each series of tests the flies were re-
turned to the laboratory and transferred to clean screen cages in a cold
room (34F) and provided 10% honey solution absorbed on cotton pads.

The Florida Entomologist


40i *'







0i c M

tL- 0 L-
00 Mo Mo

o 0
o 0

01 00
0 C (

0 00

Vol. 49, No. 3




o .


o o



C .

Mount: Thermal Fogs for Stable Fly Control

Knockdown counts were made 3 hours after exposure to the fogs, and
mortality counts were made after 24 hours.
The fogs were applied between 9 AM and 4 PM with a Leco 120 fog
generator at an air field near Gainesville, Florida. The generator was cali-
brated to deliver 40 gallons of liquid per hour and was operated at a tem-
perature of 8500 F. Air temperatures ranged from 88 to 100 F, and aver-
aged about 940 F. Wind speeds ranged from 2 to 12 mph and averaged
about 5 mph. The vehicle was driven at a speed of 5 mph over a sufficient
distance to insure that the wind carried the fog past all the cages.
Technical malathion and 14- and 8-pounds-per-gallon oil soluble concen-
trates of naled and fenthion, respectively, were formulated in a fog oil
(molecular weight 300, specific gravity 0.92, and aniline point 145 F).
For satisfactory formulation of a 1.5 pound spray concentrate of Bayer
39007, we found it necessary to add a co-solvent, heavy aromatic naphtha,
at a concentration 21 times that of the insecticide, and a sludge inhibitor
(a mixed amide amine oleate from modified fatty acids and polyamines).
The Bayer 39007 settled out of the formulation after about 1 hour but
was easily resuspended or redissolved by proper agitation. Fog oil with-
out insecticide was used as a check. Three or more tests were conducted
with each concentration of insecticide.


Table 1 presents the results of the laboratory contact spray tests. We
averaged the results obtained with male and female stable flies since there
were no substantial differences between the susceptibility of the sexes to
the four insecticides. Naled and Bayer 39007 were the most effective com-
pounds, with LC5o's of 0.005%. Both of these insecticides gave 1-hour
knockdown counts that were about equal to or slightly better than their
24-hour mortality counts. Fenthion was the next most toxic insecticide
(LC5o of 0.013%), but produced little or no 1-hour knockdown. Malathion
was the least effective, with an LC50 of 0.094%.
Table 2 gives the results of the field tests with the thermal fogs. Bayer
39007 was the outstanding insecticide. At concentrations of 1-4%, it
caused 87-100% knockdown within 3 hours and 86-98% kill in 24 hours.
The heavy aromatic naphtha solvent used with this compound may have
contributed to the actual effectiveness of the compound by increasing the
toxicity of the formulation or the penetration through the screen wire
cages. Naled was the next most effective insecticide. It also produced a
high degree of knockdown in 3 hours and killed almost as many flies in 24
hours as Bayer 39007 at concentrations of 2-4% though it was less effec-
tive than Bayer 39007 at 1%. Fenthion caused very little knockdown in
3 hours; it killed 86% of the flies in 24 hours at a concentration of 4%
but only 46-58% at concentrations of 1% to 2%. Malathion caused little
mortality at concentrations as high as 12% and probably would be of lit-
tle value in fogs for the control of stable flies.

1 Mention of a trade name does not necessarily imply endorsement of
this product by the U.S.D.A.

The Florida Entomologist


Percent knockdown in Percent mortality in
Concen- 3 hours 24 hours
tration 125 250 125 250
Insecticide (%) ft ft Average ft ft Average

Bayer 39007 4 100 100 100 100 96 98
2 100 83 92 99 77 88
1 99 75 87 99 73 86

Naled 4 100 90 95 100 88 94
2 100 63 82 100 67 84
1 61 23 42 63 31 47

Fenthion 4 13 7 10 92 80 86
2 10 6 8 75 41 58
1 12 8 10 56 36 46

Malathion 12 19 3 11 35 20 28
8 11 2 7 23 14 19
4 7 11 9 21 13 17

Check (fog oil) 4 11


We recognize that in tests with caged insects the screen cage may in-
terfere with passage of the particles of insecticide. However, we think
that our technique was adequate to show the relative effectiveness of the
insecticides. Note that these tests were conducted under daytime condi-
tions when variable winds and thermal convection currents caused the fog
to drift erratically. The effectiveness of the insecticides might be increased
by application in the evening when environmental conditions are more
ideal for fogging; however, stable flies are most active during the day-
time, and we wanted to conduct the tests under conditions occurring then.
The results definitely showed that Bayer 39007, naled, and possibly fenthion
are promising insecticides for use in thermal fogs to control natural pop-
ulations of stable flies, but fairly high concentrations were required. Cost
of the chemicals will therefore be a determining factor in their practical
use. Our results, both in the laboratory and field tests, appeared to elim-
inate malathion as a potential insecticide for control of adult stable flies.

The authors gratefully acknowledge the assistance of E. D. Rockstein,
L. L. Roesler, and G. F. Sumner, of the Entomology Research Division,
Gainesville, Florida.

Vol. 49, No. 3

Mount: Thermal Fogs for Stable Fly Control

Baygon (Bayer 39007; o-isopropoxyphenyl methylcarbamate) was the
most effective of four insecticides evaluated as contact sprays in the lab-
oratory and as thermal fogs in the field against caged stable flies [Stomoxys
calcitrans (L.)]. As a thermal fog, Baygon caused 86% mortality at con-
centrations as low as 1%. Naled was the next most effective compound
with an average mortality of 84% at a contration of 2%. Fenthion pro-
duced more than 80% mortality only when fogged at a 4% concentration.
Malathion gave very low mortality at concentrations as high as 12%. Ill
laboratory tests, Baygon, naled, fenthion, and malathion had LC0o's of
0.005%, 0.005%, 0.013%, and 0.094%, respectively.
Blakeslee, E. B. 1945. DDT surface sprays for control of stable flies
breeding in shore deposits of marine grass. J. Econ. Ent. 38: 548-52.

The Florida Entomologist 49(3) September 1966



Dr. Alvah Peterson

Dr. Peterson has donated the entire remaining stock of this book
to the Florida Entomological Society. This 176 page hard-bound book
is sub-titled "An Angler's Guide to Useful and Interesting Informa-
tion about Many Common Insects and a Few Imitation Lures that
Fishermen Use for Bait." These are now available while the supply
lasts at only $2.50 each to members and $3.00 to non-members. Order
one for yourself and another to donate to your school or city library.
Address all orders to: Business Manager, Florida Entomological
Society, Box 12425, University Station, Gainesville, Florida 32601.



, '. ^. IP

For mite-free, productive groves

Kelthane MF

KELTHANE MF kills citrus rust, citrus red, six- selective miticide. Wide ranging because it kills
spotted and Texas citrus mites. It combines fast many species of mites on all citrus. Selective
initial kill and long residual action against nymphs because it does not harm mite predators, bees
and adults. KELTHANE MF is tenacious or other pollinating insects, when used
miticide. It withstands weathering and as recommended.
leaching thus providing extended pro- ROHM For a successful fight against mites,
tection from mites to foliage and fruit. IHRRS M insist on KELTHANE. See your farm
KELTHANE MF is a wide ranging yet HLAOELHA. PENNSYLVAA 19105 supply dealer today.


Department of Entomology and Zoology Department,
University of Florida, Gainesville

The search for a satisfactory substitute for pollen in the nutrition of
honey bee colonies has occupied the attention of apiculturists for many
years. Whitcomb and Wilson (1927) reported that colonies of bees could
rear brood when fed only sugar syrup. Haydak (1935) confirmed this
earlier report, but found that when a colony was fed a pure carbohydrate
diet, brood rearing continued only for a short period of time. He also
reported that the adult bees used materials from their own body tissues
to produce larval food. This conclusion was based on his finding that the
nitrogen content of the bodies of the adult bees decreased by an amount
equal to the nitrogen content of the bodies of the newly emerging bees.
Soudek (1929) reported that when newly emerged adult bees were fed
fresh egg white or dried yeast, their pharyngeal glands developed fully.
Ribbands (1953) cited Currie as reporting that dried milk and dried egg
white were the most promising pollen substitutes of the materials tested.
Both resulted in full gland development and some brood production. Hay-
dak (1933) reported better brood production by bees fed dried yeast than
with other substances tested. Later Haydak (1937) reported that brood
rearing resulted when caged colonies were fed soy flour.
Investigations comparing the nutritive value of pollen diets with non-
pollen diets containing proteins of animal or plant origin were reported by
Standifer et al (1960). Using the degree of pharyngeal gland develop-
ment and length of life of newly emerged worker bees as criteria for
evaluation, these investigators found that egg albumin was equal to pollen
in promoting gland development, and that skim milk powder, sesame seed
meal, and soy flour were nearly as effective. However, they were not
able to show a correlation between gland development and longevity among
bees on a particular diet and concluded that longevity alone could not be
used to evaluate food stuffs for growth or brood rearing.
Recently Haydak and Dietz (1965) reported that any appropriate pro-
tein source is satisfactory for the growth of emerging bees and develop-
ment of their pharyngeal glands, but that additional vitamins are indis-
pensable for brood rearing.
Most investigators have used the glandular development of the worker
bee and egg laying by the queen as criteria for measuring the suitability
of an artificial diet. While these factors are important, a diet cannot be
considered as a substitute for pollen unless several generations of brood
can complete their development and emerge as adult bees. The so-called
pollen substitutes actually should be called pollen extenders, as there is
no known replacement for pollen that will permit the maintenance, or
build-up, of a confined colony of bees for extended periods of time.
Recently, Weaver (1964) described an artificial diet on which caged
bees were maintained for long periods of time even though the diet was
greatly inferior to pollen as measured by brood production. Attempts to

1 Florida Agricultural Experiment Stations Journal Series No. 2284.

The Florida Entomologist

maintain bees on this diet at Gainesville, Florida, were unsuccessful, as the
bees would not feed on the dry portion of the diet.

In March 1965, investigations were initiated to study the nutritional
requirements of a diet that could be used to replace pollen in the diet of
adult honeybees. Small colonies of bees consisting of one frame of brood,
two empty combs, a laying queen, and three pounds of worker bees were
confined in 6x6x6 ft. screen cages. Each colony was supplied with distilled
water and 60% sucrose syrup ad libitum. The diets used in these tests
were formulated either as liquid or solid preparations. Liquid diets were
placed in pans outside the colonies. Solid diets were prepared by mixing
various fractions of bee collected pollen with the ingredients of a basic
artificial diet in the desired proportions. The mixture was moistened
with a small amount of filtered honey and formed into small cakes. Col-
onies usually were examined every day and a visual estimate of the amount
of food consumed was made. A fresh cake containing 20 g of the diet
was placed on top of the frames after removing any of the old food that
Separation of whole pollen into various fractions was accomplished by
the following standard technique. Lipid material was extracted by re-
fluxing for 30 minutes 250 g of pollen in 940 ml of a 3:1 mixture of chloro-
form and ethanol. The mixture was filtered and the procedure repeated a
second and third time. The defatted residue was air dried and the filtrate
was concentrated by evaporating the solvent at room temperature (25C).
The dried residue and the semi-solid lipid material were refrigerated until
Water soluble materials were removed by mixing the defatted residue
with water in a Waring blender for 15 minutes at high speed. The mixture
was centrifuged, and the liquid carefully decanted. The solid residue
was dried and the liquid portion concentrated to 1/5 its original volume
in a swiftly moving air stream at a temperature of 25C. Both portions
were refrigerated until used.
The final step in the fractionation of pollen was to reflux 100 g of the
defatted residue in 250 ml of 4N HC1 for eight hours at 1000C. The mix-
ture was filtered, and after repeated washing to remove the excess acidity,
the residue was air dried at 250C. The filtrate was concentrated under
reduced pressure, and both portions refrigerated until used.
The basic design of the experiments reported upon in this paper was
a sequential screening program much like that commonly used in testing
candidate insecticides. No attempt was made to compare various diet form-
ulations with respect to degree of effectiveness.
During most of the experimentation nine colonies of bees were used
simultaneously. Occasionally a sequence of inadequate diets would reduce
the population so severely in a colony that the colony had to be replaced
or at least a frame of brood ready to emerge added.
In Table 2 diets are listed in the sequence in which they were tested
in a particular colony. No attempt was made to achieve a standardized
condition in a colony before a new diet was introduced. New diets were
usually introduced the next day after discontinuing the previous test diet.
No colony was observed to store any of the solid portions of any diet;


Vol. 49, No. 3

Robinson: Artificial Diets for Honey Bees 177

therefore, any influence of a previous diet on the bees within a colony was
reduced to the immediate physiological state of the bees. For a discussion
of this effect, see the Discussion and Conclusions section.

Confining the colonies in the cages had no apparent effect on the be-
havior of the adult bees or on colony development, since normal brood de-
velopment occurred in the check colonies feeding on honey and pollen.
The first artificial diet tested (Diet 1) was a liquid-dry formulation
modified from the diet reported by Weaver (1964). It consisted of 400 ml
water containing in solution 600 g sucrose, 204 gg carnitine, 6000 ug inosi-
tol, 6000 gg choline, 600 ug glutathione, 204 gg p-amino benzoic acid, and
0.6 ml Deca-Vi-Sol vitamin mixture. The following quantities of vitamins
are supplied by 0.6 ml Deca-Vi-Sol: Vitamin A, 3000 U.S.P. units; Vita-
min D, 400 U.S.P. units; Vitamin C, 60 mg. B1, 1 mg; B2, 1.2 mg; B6,
1 mg; B12, 1 pg.; niacinamide, 8 mg; panthenol, 3 mg; and biotin, 30 gg.)
The liquid portion of Diet 1 was fed ad libitum. The dry portion of this
diet was a mixture containing 200 g vitamin-free casein, 10 g corn oil,
10 g RNA, 2.4 g tryptophan, 5.4 g histidine, 1.2 g cystine, 3.8 g glycine,
1.0 g cysteine, 10 g cholesterol, and 4 g Wesson's salt mixture.
Diet 1A contained the same components as Diet 1, but all vitamins
were incorporated into the dry portion of the diet. In both Diets 1 and
1A, the dry portions were presented to the bees in small aluminum pans.
The bees, however, failed to accept or feed on the dry food.
Diet 2 was a liquid formulation that contained the following nutrients
dissolved in 400 ml water: 10 g gelatin, 10 g casein, 1 g corn oil, 1 g
cholesterol, 0.1 g Wesson's salt mixture, 0.1 g RNA, 600 g sucrose, and
0.6 ml Deca-Vi-Sol mixture. A few drops of Triton X-100 emulsifier were
added and these ingredients mixed in a Waring blender. The bees ate
this preparation readily, but some difficulty was experienced with the gela-
tin and casein settling out on standing.
Diet 3 was the same as Diet 1 except that the dry ingredients were
moistened with filtered honey and formed into small cakes that were placed
inside the hive on top of the frames.
Diet 4 was an all solid formulation containing 10% gelatin, 10% casein,
5% corn oil, 5% Wesson's salt mixture, 1% RNA, 1% cholesterol, and
68% sucrose. Ten grams of Diet 4 were moistened with filtered honey and
0.6 ml of Deca-Vi-Sol, formed into a cake, and placed on top of the frames
every other day.
It was clear that consumption of the artificial diets was poor when
compared with consumption of control diets of honey and pollen. Various
modifications of Diet 4 were tried in an attempt to improve the palatability
to the bees. The addition of 1 drop of a mixture of the following essen-
tial oils resulted in slightly increased consumption but did not improve
nutritive value: anise star wormseed, cedar wood, geraniol, pennyroyal,
caraway, Bergamot, sassafras, and fennelseed. The addition of an ether
extract from 5 g of fresh citrus pollen did not improve either acceptance
or nutritive value of the diet. One modification, reducing the amount of
cholesterol from 1% to 0.5%, and subsequently to 0.25% by weight of dry
ingredients, seemed to significantly increase the acceptance of the food.
The diet was further modified and designated as Diet 15 (Table 1). Al-

The Florida Entomologist

though colonies feeding on Diet 15 were unable to mature larvae beyond
an age of 3-4 days, as shown in Fig. 1, it was selected for further testing
and is referred to as basic artificial diet in the later tests described in
this paper. Results obtained in tests with Diet 15 and subsequently mod-
ified diets are presented in Table 2.


Percent by Weight

Gelatin 5.00
Casein, Vitamin Free 5.00
Zein 5.00
Egg Albumin 5.00
Wesson's Salt Mix 5.00
Mazola Corn Oil 5.00
Cholesterol 0.2.5
RNA 1.00
Sucrose 45.00
Cellulose 23.75
Vitamin Mixture (Amount/20 gms. of above mixture)
Riboflavin 1.2 mg Inositol 12.0 mg
Pyridoxine 1.0 mg Carnitine 0.46 mg
Niacinamide 8.0 mg p-amino benzoic Acid 0.48 mg
Thiamine 1.0 mg Glutathione 1.18 mg
Ascorbic Acid 60.0 mg Biotin 0.0298 mg
Panthenol 3.0 mg Vitamin B13 0.00064 mg
Choline 12.0 mg

(Enough filtered honey added to allow the molding of 20 g cakes)

It was evident that the basic artificial diet was deficient in some way
that was critical for development of older larval stages, and that this
deficiency could be satisfied by the addition of small amounts of whole
pollen. As shown by the data in Table 2, and in Fig. 2, addition of 10%
by weight, or even as little as 7.5% by weight of whole pollen (Diets 28
and 40) to the basic artificial diet resulted in complete brood development
by colonies 5 and 8, but when only 5% whole pollen was added (Diet 19)
brood development was not possible in Colony 5.
The addition of 10% or 20% of pollen lipids (Diets 17 and 27) to the
basic artificial diet did not promote brood development in colonies 3 and 1,
but colony 7 reared all stages of brood throughout the entire test period
while feeding only on the residue remaining after the lipids had been ex-
tracted from whole pollen (Diet 21) and the 60% sucrose syrup. Normal,
complete brood development also occurred in Colonies 4 and 9 when they

Fig. 1 (above) Inability of larvae in Colony 1 receiving Diet 15 to
mature beyond the 3-4 day stage. Fig. 2 (below) Normal brood in Col-
ony 5 receiving Diet 28 plus 10% whole pollen.


Vol. 49, No. 3



180 The Florida Entomologist Vol. 49, No. 3

4 0
rf s a i l r1

> 4

o o~

0 4 6 -


4- U) ~
z -~ 3- ^

vi~c c8c q



o -

E q4 0
00 --
^ i.12 ^

C1 2 01 C) C SlC

Q .Q .^ i i l

^ iI i~i*-
o8g a S a )t
4- pq Cl 1 cq Cql
Cl '4 Cl Cl -4 0

-4Cl 0

Robinson: Artificial Diets for Honey Bees

M mU

p p

En.S E.S

-M En -r- El)

V2 to U)

0 P n 0W
En E En

Ci) ho C

*3 C) '*4) U23C)

cU UdC) U2o
- En -B E1 n E
&C Sf -cb


CC 10 Cl
'l CC '-


*- C)


CC C - 4 C ) 0 C
ad a4

0 00 0 I 0 C N

LO mol- o 0



0) )

- En

E0 o

















- 0

C.) -4-

Ed f


. En





C) a-4
cit -



The Florida Entomologist

were given the basic artificial diet fortified with 33% by weight of the resi-
due remaining after the lipid and water soluble materials were removed
from whole pollen (Diet 32).
The addition of 20% by weight of the acid hydrolysate of pollen to the
basic diet (Diet 36) resulted in a mixture that the bees in Colony 3 would
not eat. Although the queen continued to lay eggs, the adult bees did not
feed the young larvae after the eggs hatched. Colony 5 was able to rear
brood through all stages while feeding on the basic artificial diet plus
33% by weight of the residue remaining from the acid hydrolysis of lipid
free pollen (Diet 37). A microscopic examination of the residue from the
hydrolysis process revealed that some pollen grains were still intact and
it is possible that the acid solution had not penetrated those grains. Fur-
ther testing of this diet is needed to determine whether or not the essen-
tial principle of pollen can actually withstand such acid treatments, or if
the satisfactory results obtained with Diet 37 were due to the addition
of whole pollen in grains that were not broken.
As shown in Fig. 3, there was no significant difference between the
rate of development of brood in an uncaged colony and that in caged
colonies feeding on Diets 32, 21, and 28. In each case, brood was sealed
on the fifth day after the eggs hatched, and there was no noticeable differ-
ence in the size or appearance of larvae of the same age.

Fig. 3 Larval development at 24 hr.
colonies: (top to bottom) uncaged colony,
Table 2 for diet formulae.

intervals in uncaged and caged
Diet 32, Diet 21, Diet 28. See


Vol. 49, No. 3

Robinson: Artificial Diets for Honey Bees

One possible source of error in diet studies conducted in a sequential
manner as we have done is the ability of bees to draw upon their own
body reserves to provide larval food for developing brood. Some estimate
of the time such reserves might last is important. On 29 June colonies
containing all stages of larvae, some sealed pupae, and a few cells of
pollen were confined in cages and given cakes of artificial food. When
checked on 6 July, only colony 6 (feeding on a wax moth diet, Haydak
1936) still had larvae of all ages. On 9 July no larvae over 4 days old
were present in any of the colonies. On 12 July the diet of colony 6 was
changed and it received Diet 25 for 5 days, but larval development still
did not progress beyond the fourth day. On 17 July colony 6 was given
Diet 34, and by 20 July, 5-6 day old larvae were present. Similar responses
were observed in other colonies during these tests. These observations in-
dicate that for 7 to 10 days confined bees may be able to supplement a
deficient diet with reserves from their bodies. The response of depleted
bees to a good diet is apparent in 1 to 3 days.
Although it is not possible to identify the deficiencies in the basic arti-
ficial diet, several conclusions can be drawn from these tests. First, it is
evident that the particular lipids in pollen are not essential. The fact that
Colony 7 was able to maintain apparently normal brood development for
more than ,50 days on a diet consisting of the residue after lipid extraction
of whole pollen (Diet 21, Table 2) indicates that very little, if any, lipid
material is needed in the diet. The amount of lipid material extracted from
the mixed pollens used in our experiments amounted to 15-20% by weight
of whole pollen. This is higher than some previously reported values
(Vivino and Palmer 1944).
Since normal, complete brood development was obtained by, fortifying
the basic diet with relatively small amounts of pollen, we believe that the
basic diet contained most of the necessary nutrients. It is reasonable to
hypothesize that the essential missing material is in pollen in good quanti-
ties. Analyses show pollen to be rich in protein (Vivino aiid Palmer
1944), with some pollen containing 20% or more protein. Assuming that
the pollen used in our tests contained 20% protein, the addition of 2 g
of pollen to 18 g of artificial diet would supply 0.4 g of pollen protein/20 g
of food consumed (Diet 28). The artificial diet in this case would supply
3.6 g of protein/20 g of food consumed. Hence, while it is possible that pol-
len proteins could be critical, it seems improbable to us that this is the
case. Some pollens have been found to be unusually rich in the vitamin
inositol; as much as 40 mg inositol/g of pollen may be present, (Augustin
and Nixon 1957). At this level, 2 g of whole pollen would supply 80 mg
inositol/20 g of food when mixed with 18 g of the basic artificial diet.
The basic diet supplies only 12 mg of inositol/20 g of diet; thus adding
whole pollen at the 10% level could possibly increase the inositol content
more than six fold. Inositol is known to be an essential vitamin for some
insects (Dadd 1963) and the possibility of the level of this vitamin being
critically low in our artificial diet must be checked.


Augustin, R., and D. A. Nixon. 1957. Grass pollen constituents: the
meso-inositol content. Nature, Lond. 179: 530-531.


184 The Florida Entomologist Vol. 49, No. 3

Dadd, R. H. 1963. Feeding behavior and nutrition in grasshoppers and
locusts. p. 47-109. In J. W. L. Beament, J. E. Treherne, and V. B.
Wigglesworth [ed] Advances In Insect Physiology I. Acad. Press,
N. Y.
Haydak, M. H. 1933. Der Nahrvert von Pollenerstazstoffen bei Bienen.
Arch. Bienenk. 14: 185-219.
Haydak, M. H. 1935. Brood rearing by honeybees confined to a pure
carbohydrate diet. J. Econ. Ent. 28: 657-660.
Haydak, M. H. 1936. Is wax a necessary constituent of the diet of the
wax moth larvae? Ann. Ent. Soc. Amer. 29: 581-8.
Haydak, M. H. 1937. Further contribution to the study of pollen sub-
stitutes. J. Econ. Ent. 30: 637-642.
Haydak, M. H., and Alfred Dietz. 1965. Influence of the diet on the
development and brood rearing of honeybees. Minn. Agri. Exp. Sta.
Journal Series 55101.
Ribbands, C. R. 1953. The behavior and social life of the honeybee. Bee
Res. Ass. Ltd., London. 352 p.
Soudek, S. 1929. Pollen substitutes. Bee World. 10: 8-9.
Standifer, L. N., W. F. McCaughey, F. E. Todd, and A. R. Kennerer. 1960.
Relative availability of various proteins to the honey bee. Ann. Ent.
Soc. Amer. 53: 618-625.
Vivino, A. E., and L. S. Palmer. 1944. The chemical composition and
nutritional value of pollens collected by bees. Arch. Biochem. 4:
Weaver, N. 1964. A pollen substitute for honeybee colonies. Glean. Bee
Culture 62: 550-553.
Whitcomb, W. W., and H. F. Wilson. 1927. A suggested explanation of
why bees appear to use pollen substitutes for brood rearing. The
Amer. Honey Prod. 1: 36-38.

The Florida Entomologist 49(3) September 1966

4 -


~ ~ 1'
/ F,

,ii-' .E .4-

Ce. :J.

PO ,,-


University of Florida, Citrus Experiment Station, Lake Alfred, and
University of Florida, Indian River Field Laboratory, Fort Pierce

Yellow scale, Aonidiella citrina (Coquillett), in the past was only occa-
sionally reported on Florida citrus and was found most commonly in the
coastal areas, particularly in Pinellas County (Griffiths and Thompson
1957). However, in recent years, it has become established in most of the
citrus areas of Florida and is an important pest in numerous groves (Siman-
ton 1965). Although the presence of yellow scale on Florida citrus has
long been noted, its control was not investigated.
The experiments reported here were designed to measure the effec-
tiveness of scalicides currently recommended on Florida citrus and of some
of the newer insecticides against yellow scale. These experiments were
conducted in 1963 and 1964.

Two different experimental designs were used in the 3 experiments re-
ported here. Experiment 1 was conducted on 3-year-old 'Valencia' orange
trees, 6 to 8-ft high, located near Brighton. A randomized block design
with 2-tree plots replicated 6 times was used in this experiment. Experi-
ments 2 and 3 were conducted at the Indian River Field Laboratory located
near Fort Pierce. A 10 X 10 Latin square design was used in these tests.
Two-tree plots of 8 to 10-ft high 'Temple' orange trees were used in Ex-
periment 2 while 1-tree plots of 'Marsh' grapefruit, 15 to 20-ft high, were
used in Experiment 3. Yellow scale populations were determined by re-
cording the number of live third-stage female yellow scale on 50 leaves
picked at random from around each tree. The data were subjected to anal-
ysis of variance and Duncan's New Multiple Range Test at the 5% level.
The treatments were applied with a truck-mounted high pressure spray-
er equipped with double Boyce handguns. Treatments in Experiment 1
were applied on 27 August and in Experiment 2 on 3 October 1963. In
Experiment 3, applications were made postbloom and repeated in the sum-
mer of 1964.
In Experiment 1, all treatments had significantly fewer yellow scale
than the plots receiving no scalicide (Table 1). Azinphosmethyl, dimetho-
ate, and Methyl-Ethyl Guthion, although not significantly better than
ethion-oil, malathion-oil, parathion, parathion-oil, 1.3% emulsified oil, Shell
SD 8448, and diazinon, were superior to Bayer 41831, malathion, Stauffer
N-2404, Union Carbide 21149, and Shell SD 8447.
In Experiment 2, as in Experiment 1, all treatments had significantly
fewer yellow scale than the plots receiving no scalicide (Table 2). While
azinphosmethyl was not superior to Shell SD 8448 or parathion, it was
significantly better than all other treatments. However, Shell SD 8448

1 Florida Agricultural Experiment Stations Journal Series No. 2362.

The Florida Entomologist


No. of 3rd stage female yel-
Lb low scale per 100 leaves-
actual/ days after spraying*
Scalicide and formulation 100 gal 60 112 164

Azinphosmethyl (2 lb/gal EC) 0.250 0.7a 0.2a 1.2a
Dimethoate (4 lb/gal EC) 0.250 1.0ab 0.3a 1.3a
Methyl-Ethyl Guthion (2 lb/gal EC) 0.250 0.7a 0.5a 1.7a
Ethion (4 lb/gal EC) + 0.5% oil 0.375 2.8ab 3.3abcd 3.5ab
Malathion (5 lb/gal EC) + 0.5% oil 0.625 2.5ab 1.8ab 4.7ab
Parathion (25% WP) 0.250 3.0abc 4.7bcde 4.8ab
Parathion (25% WP) + 0.5% oil 0.125 4.5abcd 3.7abcde 5.5ab
Emulsified Oil, 1.3% 8.8def 6.2cde 5.5ab
Shell SD 8448 (2 lb/gal EC) 0.250 2.2ab 2.5abc 6.8abc
Diazinon (4 lb/gal EC) 1.000 6.2bcde 2.5abc 7.3abc
Bayer 41831 (4 lb/gal EC) 0.250 8.2cdef 5.0bcde 10.0bc
Malathion (5 lb/gal EC) 1.250 3.3abc 2.0ab 10.2bc
Stauffer N-2404 (4 lb/gal EC) 0.250 10.8ef 6.5de 10.3bc
Union Carbide 21149 (1 lb/gal EC) 0.250 10.7ef 6.2cde 13.3c
Shell SD 8447 (1 lb/gal EC) 0.250 12.2b 7.5ef 13.7c
No scalicide 17.7g 10.8f 22.2d

*Results of Duncan's Test: Treatment means followed by the same letters are not con-
sidered significantly different at the 5% level.



No. of 3rd stage female
yellow scale per 100 leaves-
days after spraying*

Scalicide and formulation 100 gal 67 130 Z04

Azinphosmethyl (2 lb/gal EC) 0.25 6.7a 3.9a 3.7a
Shell SD 8448 (2 lb/gal EC) 0.25 5.8a 4.4a 6.1ab
Parathion (15% WP) 0.25 6.4a 4.7a 6.1ab
Methyl-Ethyl Guthion (2 lb/gal EC) 0.25 8.1ab 5.6ab 8.5bc
Emulsified Oil 1.0% -- 8.2ab 9.6bc 8.9bcd
Perthane (4 lb/gal EC) 1.00 13.7cd 13.0cd 9.5bcd
Union Carbide 21149 (1 lb/gal EC) 0.25 17.4d 14.1de 9.9bcd
Niagara 9203 (25% WP) 0.25 11.0bc 14.4de 12.5cd
Stauffer N-2404 (4 lb/gal EC) 0.25 11.5bc 17.7e 12.9cd
No scalicide 12.5c 17.0de 17.7e

*Results of Duncan's Test: Treatment means followed by the same letters are not con-
sidered significantly different at the 5% level.


Vol. 49, No. 3

Brooks: Control of Yellow Scale

and parathion were superior only to Niagara 9203 and Stauffer N-2404.
No significant difference occurred among the remaining treatments.
The results of Experiment 3 are shown in Table 3 and again the un-
treated plots had significantly more yellow scale than the treated plots.
Azinphosmethyl and Polybutene LS-0811 were superior only to oxydemeton-
methyl. There was no significant difference among the other treatments.


No. of 3rd stage female yel-
Lb low scale per 100 leaves*-
actual/ days after 2nd application
Scalicide and formulation 100 gal 34 91 172

Azinphosmethyl (2 lb/gal EC) 0.25 0.4a 1.2a 1.1a
Polybutene LS-0811, 1% 1.4a 3.0a 1.2a
Dimethoate (4 lb/gal EC) 0.25 2.0ab 3.8ab 1.4ab
Ethion (4 lb/gal EC) + 0.5% oil 0.25 2.2ab 4.4ab 2.2ab
Parathion (15% WP) 0.25 0.9a 2.8a 2.4ab
Mobil Oil MCA-600 (50% WP) 1.00 0.7a 2.0a 3.0ab
Union Carbide 21149 (50% S)** 0.25 2.4ab 4.8ab 3.4ab
Emulsified Oil 98%, 1% 4.0b 4.2ab 3.6ab
Oxydemetonmethyl (2 lb/gal EC) 0.25 3.7b 7.2b 4.2b
No scalicide 9.6c 16.8c 12.6c

*Results of Duncan's Test: Treatment means followed
sidered significantly different at the 5% level.

by the same letters are not con-

**Formulated as a 50% soluble material and packaged in water soluble bags.


Although yellow scale has increased in importance as a pest to Florida
citrus over the past few years, most of the materials currently recom-
mended as scalicides are adequate for its control. The results of the 3
experiments reported here show that azinphosmethyl, parathion, 1.3%
emulsified oil, and the combinations of emulsified oil with parathion, ethion,
and malathion effectively control this pest. However, azinphosmethyl pro-
vided significantly better control than either malathion or 1.0% emulsified
oil. Some of the experimental materials that also provided good control
were dimethoate, Methyl-Ethyl Guthion, Shell SD 8448, diazinon, Polybu-
tene LS-0811, Mobil Oil MCA-600, and Union Carbide 21149.
The chemical names of proprietary materials mentioned in this article
Bayer 41831-0,0-dimethyl O-4-nitro-m-totyl phosphorothioate
Methyl-Ethyl Guthion-equal parts of azinphosmethyl and azinphos-
Mobil Oil MCA-600-4-benzothienyl-N-methyl carbamate
Niagara 9203-0,0-dimethyl S- [(benzoxazolin-2-on-3-yl)methyl] phos-

The Florida Entomologist

Perthane-mixture of 1,1-dichloro-2,2-bis (p-ethylphenyl) ethane (95%)
and related reaction products (5%)
Polybutene LS-0811-American Oil Co. emulsifiable concentrate of H-120
(viscosity 1163 SSU at 2100 F.)
Shell Development 8447-2-chloro-1-(2,4,5,trichlorophenyl) vinyl
dimethyl phosphate
Shell Development 8448-2-chloro-1-(2,4,5-trichlorophenyl)vinyl diethyl
Stauffer N-2404-0-(2-chloro-4-nitrophenyl) O-isopropyl ethyl-phos-
Union Carbide 21149-2 methyl-2-(methylthio) propionaldehyde 0-
(methylcarbomoyl) oxime
Griffiths, J. T., and W. L. Thompson. 1957. Insects and mites found on
Florida citrus. Fla. Agr. Exp. Sta. Bull. 591: p. 15.
Simanton, W. A. 1965. Citrus insect situation in Florida-end of May.
USDA Coop. Econ. Insect Rep. 15(24): 606.
The Florida Entomologist 49(3) September 1966




Carefully Executed

Delivered on Time




_ __1 ~II~I Y


Vol. 49, No. 3



The Pennsylvania State University, University Park, Pa.

Insect light traps have been operated at the Archbold Biological Sta-
tion, Highlands County, Florida for seven winters from 1959 to and includ-
ing 1965. Generally the traps were operated every night. This was espe-
cially true of 1960 and 1965. In all cases the traps were run at least 85%
of the nights.
Fifty-three species of Scarabaeidae have been taken in these light traps.
Forty-seven were listed by Frost (1964). Only nine species can be consid-
ered common: Dyscinetus morator Fabricius, Phyllophaga, elizoria Say-
lor, P. glaberrima Blanchard, P. prununculina Burmeister, Cyclocephala
parallel Casey, Diplotaxis bidentata Leconte, Anomala nigropicta Casey,
Serica errans Blatchley, and Bothynus neglectus Leconte. Species of
Aphodius and Ataenius were very common, but it was impossible to iden-
tify them at the time they were collected.
Dyscinetus morator (Fab.)2 was a conspicuous visitor to the traps dur-
ing all seven years but especially so during 1962. This species is some-
what common in the eastern United States from Connecticut to Florida
and Texas, but apparently is more common in the southern states. Little
is known concerning the life history of this species. It is freely attracted
to light and has been found beneath debris, in muck about the margins of
streams and lakes, in compost, and near pig pens. It should be noted that
a large compost pile and chicken yards were adjacent to the area where
the traps were operated and may account for the abundance of these beetles
during certain years.
Table 1 illustrates the relative abundance of Dyscinetus morator during
seven winters. Records are given only for the months of January, Febru-
ary, March, and April. Although traps were operated during November
and December of 1958 and 1959, relatively few beetles were taken during
these months. In November 1958, 126 specimens were taken, 70 of these
on 9 November. The same year 226 specimens were taken during Decem-
ber, 178 of these on 9 and 10 December when the average night tempera-
tures were relatively high, 64 and 65 F. In November 1959, only 15 speci-
mens were taken, and in December the same year only 16 specimens were
taken. Counts were not made during April of 1959, 1960, and 1961. It
is evident that there is great yearly variation in the activity of these
Fig. 1 shows the abundance of Dyscinetus orator during the winter
of 1962. The traps were operated every night from 6 Jan. to 25 Feb.,
with the exception of 12-13 Jan. From 25 Feb. to 21 March they were
discontinued because of the abundance of these beetles, which made it
almost impossible to obtain other specimens in good condition. The re-

Authorized for publication 2 Dec. 1965 as paper No. 3080 in the jour-
nal series of the Pennsylvania Agricultural Experiment Station.
2 (== trachypygus Burm., discedens Casey, and borealis Casey).

The Florida Entomologist

duction in catches shown in Fig. 1 are due to low temperatures. The
average minimum night temperature of 12-14 Jan. was 38 F.; of 29 Jan.
through 4 Feb., 41 F.; and of 11-13 Feb., 43 F. It is evident that these
beetles are most abundant during January, February and March.

Dyscinetus morator Fab.

Year operated January February March April

1959 74 7 87 25
1960 84 11 11 19
1961 37 2 14 25
1962 61 1317 1632 563 334
1963 63 0 28 108 26
1964 95 15 21 83 106
1965 101 30 76 60 48

Counts of Dyscinetus morator taken during different hours on 21 nights
during January and February clearly show that these beetles started com-
ing to light shortly after dusk (between 6 and 7 PM), that the majority
came between 7 and 9 PM, and that a relatively few were taken between
9 PM and 7 AM. At times the abundance of these beetles interfered with
the operation of the light traps. Discontinuing operation between 7 and
9 PM, or replacing killing jars at one-half-hour periods between 7 and 9



Fig. 1. Light trap collections

25 21
of Dyscinetus morator, 1962.






Vol. 49, No. 3

Frost: Scarabaeidae Taken in Light Traps 191

PM, on such nights helped to solve the problem. The data presented in
Table 2 were taken on nights when the number of beetles was not the
greatest; nevertheless a trend is indicated.


Number of
Hours Specimens

6-7 PM 3619
7-9 PM 30225
9 PM to 7 AM 1770

Several species of Phyllophaga were taken in noticeable numbers during
these winters. Blatchley (1929), Luginbill and Painter (1953), Sim (1928),
and Young and Thames (1948) give notes and distribution records for
these species.
Phyllophaga elizoria Saylor was the only species of this genus that came
to lights in sufficient numbers to plot a curve. Young and Thames (1948)
state that this species is apparently rare, although it is locally abundant
in dry sandy areas. They found it at Indian River attacking young orange
trees. Luginbill and Painter (1953) state that the female is unknown.
This species is known only from Florida and Texas. During 1964 it was
not common at the Archbold Biological Station. Only 56 males and 49
females were taken. The following year it appeared in appreciable num-
bers. Fig. 2 shows the abundance of the beetles from 5 Mar. to 10 Apr.
Although traps were operated during January and February, only one
male was taken during this period. The peak of activity occurs during
the latter part of March, and the flight of this species is limited chiefly
to March and April.
Phyllophaga glaberrima (Blanch.) was second in abundance. Fifty-
four males and 3 females were taken between 23 Feb. and 10 Apr. 1965,
five days later the operation of the traps was discontinued. This species
is known chiefly from the southeastern United States, although specimens
have been taken in New Jersey and on Long Island. Young and Thames
(1948) state that it is locally abundant throughout the state of Florida
and especially common in the northern portion.
Phyllophaga prununculina (Burm.) was not taken in appreciable num-
bers at The Archbold Biological Station. Although light traps were op-
erated from 1 Jan. to 15 Apr. 1965, only 11 males and 8 females were
taken. This species is restricted more or less to the southeastern United
States, although it has also been taken in New Jersey. Young and Thames
(1948) report that it is locally abundant throughout Florida and at times
appeared in damaging numbers on Pinus taeda at Gainesville. Dozier
(1920) observed hundreds of these beetles eating the foliage of Pinus
Other species of Phyllophaga were taken in light traps less frequently.
These include P. dispar (Burm.), a species limited to the southeastern
United States; P. latifrons Lec., known chiefly from the southeastern United
States although taken at times in New Jersey; and P. subpruinosa Casey, a

The Florida Entomologist

Vol. 49, No. 3

species occurring in the extreme southeastern United States. No more
than 10 specimens of any of these species were taken during the years the
traps were operated.



20 -

15 -


-------- Females

a I I

I I+ !, a

ii I


I I ,i i


Fig. 2. Light trap collections of Phyllophaga elizoria, 1965.

Anomala nigropicta Casey occurs chiefly in the southeastern United
States. It was taken in light traps in noticeable numbers from 1 Jan. to
15 Apr. Table 3 gives the number taken each year. Blanks indicate
periods when no counts were made. The largest catches were usually made
during February.


Year January February March April

1959 74 310
1960 60 334 122
1961 137 30
1962 188 304
1963 1093* 832 52 2
1964 106 .546 143 4
1965 116 109 108 13

*All were taken on the last 5 days of January


I I I I I I I I i I



Frost: Scarabaeidae Taken in Light Traps

Diplotaxis bidentata Leconte is rated by Blatchley (1929) as one of the
most common scarabaeids of Florida; however, the writer did not take it
in light traps as frequently as many other Scarabaeidae. This species
occurs chiefly in the southeastern United States, although it has been taken
in New Jersey. Dozier (1920) found large numbers feeding at night on
chinquapin, presumably Castanea pumila. The records in Table 4 were
taken at irregular intervals; however, the data are sufficient to show that
the beetles were abundant during November, scarce during January and
February, and abundant again during March.


Year Nov. Dec. Jan. Feb. March April

1958-59 654 107 76 317 759
1959-60 489 107 51 48 305
1961 2 130
1962 40 173
1963 171 218 126
1964 138 54 754 317
196,5 25 53 630 243

Cyclocephala puberula Leconte is known only from Georgia and Florida.
Blatchley records this species from St. Augustine, Seven Oaks, Lake City,
Enterprize, and La Belle. The writer took specimens at light traps from
30 Mar. to 9 Apr. but never in noticeable numbers.
Cyclocephala paralella Casey, a Florida species, was taken occasionally
by Blatchley. It was a somewhat common visitor to the light traps at the
Archbold Biological Station from 31 Mar. to 11 Apr. 1965. During this
period 88 males and 18 females were taken.
Serica errans Blatchley, a relatively small species, was described in
1929 and recorded as somewhat rare. Blatchley took 3 males and 2 females
at a porch light, and other specimens were beaten from Spanish Moss at
Ocala. This species has been observed at lights for several years. At first
its identity was unknown, but during the years 1963 to 1965 counts were
made which are summarized in Table 5.


Year No. nights February March April

1963 23 106 13
1964 50 73 398 40
1965 71 210 484 52

Bothynus neglectus Leconte, a Florida species, was somewhat abundant.
Specimens were taken at light traps during November, March, and April
and were most common during March and April.
Copris minutus (Drury), known from Canada to Florida, was also some-
what common. Specimens were taken from November to March.


The Florida Entomologist

Pachystethus marginata (Fabricius), chiefly a southern species, was
taken in considerable numbers during March and April.
Five species of Trox were taken in light traps but only T. suberosus
Fabricius and T. terrestris Say were noticeably abundant. Both were
taken from 1 Nov. to 15 Apr.
Frost, S. W. 1964. Insects taken in light traps at the Archbold Biologi-
cal Station, Highlands County, Florida. Fla. Ent. 47(2): 129-161.
Blatchley, W. S. 1929. The Scarabaeidae of Florida. Fla. Ent. 13(3):
52-56, (4): 69-70.
Dozier, H. L. 1920. An ecological study of hammock and piny woods
insects in Florida. Ann. Ent. Soc. Amer. 13: 325-380.
Luginbill, P., and H. R. Painter. 1953. May beetles of the United States
and Canada. USDA Tech. Bull. 1060: 1-102. 73 plates.
Sim, R. J. 1928. Phyllophaga (Scarabaeidae) of the United States and
Canada. N. J. Dep. Agr. Circ. 145: 1-60. 12 plates.
Young, F. N., and W. H. Thames. 1948. A preliminary list of the Phyl-
lophaga of Florida. Fla. Ent. 34(4): 125-130.
The Florida Entomologist 49(3) September 1966


Complete Line of Insecticides, Fungicides and
Weed Killers
Ortho Division

Located at Fairvilla on Route 441 North
P. 0. Box 7067 ORLANDO Phone 295-0451


Vol. 49, No. 3


University of Florida, Citrus Experiment Station, Lake Alfred

The word "Nabac" is a trademark of the Nationwide Chemical Corpo-
ration. It is used to designate formulations of hexachlorophene [2,2'meth-
ylene-bis (3,4,6 trichlorophenol)] that are intended for agricultural use.
Nabac was first tested against citrus rust mite, Phyllocoptruta oleivora
(Ashm.) in a routine trial of new materials conducted during July and
August of 1963. Since then, Nabac has been tested in a dozen other field
trials. The results of these experiments are summarized here.

The freeze of December 1962 killed so much wood on citrus trees at the
Florida Citrus Experiment Station that only 1 or 2 limbs remained alive
on the lower portions of most trees. In 1963, these lower limbs were
sprayed with 2- and 3-gallon compressed air sprayers and were used as
individual plots. Tops of most trees were in fairly good condition and
supported large undisturbed populations of citrus rust mite. These mites
apparently migrated to the lower limbs, for reinfestation of the plots was
very rapid and revealed differences in residual activity between acaricides
in days rather than weeks (Table 1). The density of rust mite populations
in this type of experiment and in experiments on young trees (Table 2)
was determined by counting the number of mites on 10 cm2 of the lower
surface of 10 leaves per plot.


Active ingredient of Number of mites per cm2 of leaf*
miticides 16 July
(oz per 100 gal) (pre- 19 July 22 July 3 August
applied 17 July 1963 spray) (+2 days) (+5 days) (+17 days)

Nabac 25EC, 6.0 59.28 .12 a .52 a .58 a
Union Carbide-21149
1EC**, 6.0 58.42 .06 a .28 a .76 a
Ethion 4 Miscible, 6.0 60.58 .46 a 2.56 a 4.04 b
Chlorobenzilate 25E, 2.0 60.66 1.30 a 2.90 a 5.28 b
Unsprayed 60.60 47.64 b 39.58 b 11.42 c

*Numbers of mites followed by different letters are significantly different at the 5% level.
**2-methyl-2-(methylthio) propionaldehyde O-(methylcarbamoyl) oxime.

Experiments were conducted in 1964 and in 1965 on plots of 1 to 4 trees
sprayed with a conventional hydraulic grove sprayer with double Boyce

1 Florida Agricultural Experiment Stations Journal Series No. 2361.

196 The Florida Entomologist Vol. 49, No. 3

guns. The percentage of leaves and fruit infested with citrus rust mite
before spraying and at intervals after spraying was determined for each
plot by examination of 25 leaves and 25 fruit per tree with a 10X hand lens.
Chlorobenzilate was the standard acaricide in all experiments.


Number of mites per cm2 of leaf**
Miticide applied 16 Septem- 19 Septem-
17 September ber ber 4 October 28 October
1963* (prespray) (+2 days) (+17 days) (+41 days)

Nabac 25EC 6.39 .02 a .03 a .25 a
Morestan@ 25Wt 5.20 .08 a .09 a .46 a
Chlorobenzilate 25E 5.48 1.07 b .81 b 1.92 b
Azinphosmethyl 7.56 .31 a 1.27 c 5.62 c
Imidan 3E$ 8.58 .83 b 1.36 c 6.82 c

*All of the acaricides were applied at dosages of 2.0, 4.0, and 6.0 ounces of active ingredi-
ent per 100 gallons. Numbers of mites are thus averages of 3 dosages.
"Numbers of mites followed by different letters are significantly different at the 5% level.
t6-Methyl-2,3-quinoxalinedithiol cyclic S,S-dithiocarbonate.
$O,0-dimethyl (N-phthalimidomethyl) phosphorodithioate.

Nabac was included in a routine field screening of 9 new compounds in
1963. In this experiment, Nabac was 1 of 2 outstanding treatments (Table
1). Nabac not only produced a good initial mortality among dense popu-
lations of citrus rust mite, it also produced comparatively long control.
Similar results were obtained in 1 other test; and in still 2 others, there
was no difference between 3 formulations of Nabac.
Preliminary experiments conducted in 1963 showed Nabac to be highly
effective against citrus rust mite at the dosage of 6.0 oz. of hexachloro-
phene per 100 gallons of spray, but there was no information about the
effectiveness of lower dosages. Accordingly, an additional experiment to
determine the optimum dosage of Nabac and 4 other miticides was con-
ducted in September and October of 1963. In this experiment, the plots
were arranged in a split-plot design with dosages for main treatments and
materials for sub-treatments. No differences were found between the
dosages of 2.0, 4.0, and 6.0 oz. of active ingredient of any of the miticides;
but differences between materials were pronounced (Table 2) and again
showed the superiority of Nabac as a miticide for citrus rust mite.

One experiment was conducted in 1964 in a second attempt to deter-
mine the optimum dosage of Nabac for control of citrus rust mite. In this
experiment, 5 dosages of a new formulation of hexachlorophene (Nabac
50ECX) from 0.5 to 4.2 oz. per 100 gallons were applied after bloom and
again in midsummer. These 2 applications did not produce identical results

Johnson: Hexachlorophene for Mite Control

(Table 3), but were sufficiently similar to suggest that the optimum dosage
of Nabac for control of citrus rust mite is about 3.0 oz. of hexachloro-
phene per 100 gallons of spray. However, dosages of 2.1 to 4.2 oz. of
hexachlorophene were too phytotoxic to be acceptable. Furthermore, the
low dosages of 0.5 and 1.0 oz. of hexachlorophene, while not phytotoxic,
were significantly poorer than chlorobenzilate against citrus rust mite.


Active ingredient of miti-
cides (oz per 100 gal) % of leaves and fruit infested % burned
applied 31 March and with citrus rust mite* fruit
27 July 1964 21 July 14 September 15 October

Nabac 50ECX, 0.5 20.5 b 30.2 c 2.6 a
Nabac 50ECX, 1.0 16.0 b 18.2 b 4.6 ab
Nabac ,50ECX, 2.1 17.2 b 6.5 a 14.0 bc
Nabac 50ECX, 3.1 13.2 ab 5.5 a 21.8 c
Nabac 50ECX, 4.2 10.8 a 3.0 a 31.8 d
Chlorobenzilate 25E, 2.0 8.5 a 2.2 a 1.8 a

*Percentages of infested leaves and fruit are significantly different at the 5% level when
followed by different letters.

Tests conducted in 1964 demonstrated that dosages of hexachlorophene
low enough to be non-phytotoxic are of little value against citrus rust
mite. Therefore, if non-phytotoxic amounts are to be useful, they must
be supplemented with another material.. Zineb (derived from Dithane Z-
78) was selected for this purpose and was included in 3 tests in 1965.
The first of these tests was started on 4 May and included 0.5, 1.0, and
2.0 oz. of hexachlorophene applied separately and in combination with 6.0
oz. of zineb per 100 gallons. None of these treatments were phytotoxic to
'Pineapple' orange trees and all controlled citrus rust mite for about 2
The remaining 2 experiments (Tests A and B) compared the dosage
of 1.9 oz. of hexachlorophene (Table 4) with one-half of this amount, with
6.0 oz. of zineb, and with 2 mixtures of hexachlorophene and zineb. One
of these mixtures was a formulation called Zinabac (50% zineb + 9%
hexachlorophene) while the other was a mixture of zineb and hexachloro-
phene prepared in the spray tank. All of these treatments were combined
with 1% oil in Test B, but not in Test A.
Hexachlorophene at 1.9 oz. was the most effective treatment in these 2
experiments, especially in Test A, (Table 4). The one-half dosage of
hexachlorophene and the dosage of 6.0 oz. of zineb were both significantly
poorer. These results make it possible to determine whether zineb in mix-
tures of zineb and hexachlorophene improves the control of citrus rust
mite. In 1 such mixture, Zinabac, there was no significant improvement.
The tank-mixed combination, however, was more effective than either in-


The Florida Entomologist

gredient but still not as effective as 1.9 oz. of hexachlorophene without


% infested leaves and
Active ingredient of miticides Test A- Test B-
(oz per 100 gal) No oil with oil
applied 7 July, 1965 22 July** 23 Julyt

Zineb, 6.0 56.3 d 6.5 b
Zineb, 6.0 + Nabac Citrus Special, .95 30.7 b 4.0 a
Zinabac (Zineb, 5.6 + Hexachlorophene, 1.0) 36.7 c 6.0 b
Nabac Citrus Special, .95 41.3 c 6.5 b
Nabac Citrus Special, 1.9 22.0 a 3.0 a
Chlorobenzilate 4E, 2.0 42.3 c -
Check 100.0 e 7.5 b

*Percentages in each test are significantly different at the 5% level
different letters.
**Prespray populations ranged from 79 to 85% on 6 July.
tPrespray populations ranged from 53 to 68% on 6 July.

when followed by


Nabac from as little as 2.0 oz. to as much as 6.0 oz. of hexachlorophene
per 100 gallons of spray has produced control of citrus rust mite as good
or better than such standard materials as chlorobenzilate, ethion, and azin-
phosmethyl. On the other hand, dosages of hexachlorophene as low as 2.1
oz. have been phytotoxic. This situation has led to attempts to fortify non-
phytotoxic dosages of Nabac with zineb. This procedure has met with some
success, but needs further study.

Vol. 49, No.. 3


University of Florida Citrus Experiment Station, Lake Alfred

Solpugids are reported to be exclusively predatory arachnids with an
extraordinary voracity. Several workers have recorded solpugids feeding
until their abdomens were so distended that they could scarcely move
(Hutton 1843, Pocock 1898, Turner 1916, Hingston 1925, Fichter 1940,
Lawrence 1949, and Cloudsley-Thompson 1958). Prey capture, feeding
activity, drinking, and food variety have also been recorded several times
(Hutton 1843, Pocock 1898, Turner 1916, Hingston 1925, Fichter 1940,
Lawrence 1949, Bolwig 1952, and Cloudsley-Thompson 1958). These studies
have been conducted on single specimens or species. The recorded feeding
habits have been in general agreement but have varied in detail indicating
a need for a systematic investigation of solpugid feeding. With this need
in mind, a comparative study has been made on the feeding habits of two
families, seven genera, and 18 species of North American solpugids.

Much of the field work was made possible through the facilities of
the American Museum of Natural History's Southwestern Research Station
during August, 1963, June, 1964, and July, 1965.
Mr. Allen G. Selhime, USDA, ARS, ENT, Orlando, Florida, assisted
with photography.
I would also like to acknowledge the many tests and observations con-
ducted by my principal technician, Mrs. Thelma G. Kanavel.

Male, female, and immature solpugids of each species were studied.
When only one sex or immatures of a species were available for observa-
tion, this is so indicated in the text and tables. Most observations were
made in holding cages or terraria in the laboratory, but feeding under nat-
ural or field conditions was also recorded.
Holding cages or terraria varied from nine-cm petri dishes used for
early instar specimens to one-pint food-preservation jars and 8" x 12"
battery jars used for late instars and mature specimens. Prolonged ob-
servations of capturing and feeding techniques were made on specimens
contained in 8" x 12" battery jars or a deep-sided 23-cm petri dish.
In addition to careful observations of individual specimens under both
field and laboratory conditions, a number of laboratory feeding tests were
also conducted. These tests involved comparable series of five to 20 speci-
mens of one sex or stadium confined in one type of terrarium under uni-
form conditions and subjected to varying food regimens. For example,
five males of Therobates bilobatus Muma were watered daily and compared
for activity with five males receiving no water. Another example, five
second-instar nymphs of Eremobates durangonus Roewer were fed two ter-

1 Partial report on studies supported by National Science Foundation
Grant GB-496.
2 Florida Agricultural Experiment Stations Journal Series No. 2390.

The Florida Entomologist

mites a day and compared for developmental time with five nymphs fed
one termite a day and five nymphs fed one termite every other day. Such
tests are not specifically reported below, but the observations and results
are incorporated in the discussions.
Because of the wide variety of solpugid reactions to offered foods, it
has been necessary to stabilize the feeding behavior terminology as used
in this paper. If a species did not eat even after repeated or prolonged
food exposure, the term refused is used. When the offered food was eaten
only after several or prolonged exposures in the absence of other food,
the term rarely accepted is used. If the food was eaten one or more times
along with other foods, the term accepted is used. When the food was
eaten every time it was offered, the term accepted readily is used.
In most instances, foods were living, uninjured arthropods which were
placed in the terraria with the study specimens. Wings of moths were
trimmed to reduce flight. Foods, especially termites, offered to early in-
stars were frequently injured to prevent solpugid-termite combats. Some
living and dead foods were offered in forceps or on a small brush. Arthro-
pod prey were identified to genus where possible.
In the laboratory, all feeding observations were conducted at 80F.
and 70 percent relative humidity. Because most solpugids are nocturnal,
subdued lighting was maintained during feeding studies.
Solpugid identifications follow Muma (1951, 1962, and 1963).


The general facies of feeding are quite similar for all species of sol-
pugids. For this reason, the several food-related activities are presented
and discussed in this section. Literature citations usually refer to particu-
lar species. Only studies dealing with reactions common to two or more
species, as indicated by the literature and the present study, are included
in this section.
Solpugids will not feed under certain conditions. Larvae and first
nymphs do not feed; they apparently survive on residual egg nutrition.
Individuals do not feed for several days prior to, during, and succeeding
ecdysis. Further, nocturnal species do not feed during the day, diurnal
species during the night, and engorged individuals for a variable length
of time following engorgement. With the exception of these situations
and the males during mate searching and mating, North American solpu-
gids spend much of their lives in food-associated activities.
Food Search.-Little is known concerning the behavior of solpugids
while searching for food. Several papers have described rapid, apparently
random, running by solpugids seemingly in search of food (Putnam 1883,
Pocock 1897, Bolwig 1952, and Lawrence 1963). Hutton (1843) and Po-
cock (1897) have reported solpugids collecting food at lights. Several
species have been commonly collected from houses which they apparently
invaded in search of prey (Putnam 1883, Pocock 1897, Fichter 1940, and
Cloudsley-Thompson 1961a). Specific prey searching techniques have also
been reported (Marx 1892, Pocock 1897, and Lawrence 1963).
Data on North American species seem to confirm each of these four
searching techniques. Eremorhax magnus (Hancock), E. titania Muma,
Eremobates palpisetulosus Fichter, E. durangonus, Ammotrechula peninsu-
lana (Banks), Branchia brevis Muma, and B. potens Muma have been ob-


Vol. 49, No,. 3

Muma: Feeding Behavior of Solpugida

served with incandescent and infrared light running at random over the
ground at night, apparently in search of food. E. magnus, Eremobates
nodularis Muma, E. durangonus, E. palpisetulosus, T. bilobatus, and A.
peninsulana have been observed searching and feeding at incandescent
lights. E. durangonus, A. peninsulana, and Ammotrechella stimpsoni (Put-
nam) have been collected so regularly from houses that their accidental
occurrence in artificial structures seems unlikely. E. durangonus and A.
stimpsoni have also been observed while apparently involved in specific prey
searching techniques.
Present information indicates that a solpugid's food searching technique
may vary with ecological conditions, availability of food, and inherent be-
havior. For example, a solpugid normally following a random cursorial
prey searching pattern might abandon such a habit in the presence of abun-
dant food at a light. Similarly, a termitophilous species might regularly
enter termite-infested buildings. On the other hand, a normally subter-
ranean or limited-host solpugid would probably become cursorial or less
food-specific when food was scarce.
Prey Location.-Only Bolwig (1952) has specifically reported upon the
location of prey. He concluded that Solpugyla globicornis Kraepelin lo-
cated and oriented to prey in answer to tactile stimuli applied to the long
palpal and leg setae. Other workers have also commented on solpugid
sensitivity and response to prey touch. A few papers have inferred visual
sensitivity (Pocock 1898, Turner 1916, and Hingston 1925). Recently,
Cloudsley-Thompson (1961) has demonstrated that Galeodes granti Pocock
has a rather high degree of sensitivity to substrate vibrations.
Present data indicate that touch, sight, and vibrations are all utilized
in food location by North American solpugids. Most species respond rap-
idly to tactile stimuli, whether applied to the long or short palpal, leg,
and body setae, and orient for attack as described by Bolwig (1952).
Eremorhax magnus, E. titania Muma, and A. peninsulana have sufficient
sight to permit stalking of or orientation to an attack of prey several cen-
timeters distant. It also seems safe to assume that the diurnal Hemero-
trecha californica Banks can also see prey. Several species have been
demonstrated to orient to and approach sources of artificial and natural
substrate vibrations.
Two or more senses are probably involved in prey location. For in-
stance, a species may orient to substrate vibration but attack only when
touched, orient to vibration and attack when the prey comes into view, or
orient to touch and attack by sight.
Prey Capture.-Three general types of prey capture have been recorded
in the literature; these are chase, stalk, and ambush. Pocock (1898),
Turner (1916), and Hingston (1925) all reported prey capture by a high-
speed rush. Turner and Hingston also recorded solpugids lying in ambush
for prey. Pocock (1898) wrote that a Mr. A. Carter had watched a Gale-
odes stalk flies. To date, this is the only record of stalking.
In the present study, capture by chase and by ambush have been ob-
served for the Eremobatidae and Ammotrechidae. Both types of capture
have been recorded for Eremorhax striatus (Putnam), E. magnus, E. ti-
tania, E. durangonus, E. palpisetulosus, E. nodularis, Therobates imperialism
Muma, T. bilobatus, Hemerotrecha serrata Muma, and A. peninsulana. One
species, E. magnus, has been observed apparently stalking scarabaeid
beetles of the genus Diplotaxis.

202 The Florida Entomologist Vol. 49, No. 3

Function of the palpi and the chelicerae in the actual contact with prey
has been open to question. Turner (1916), Lawrence (1949), and Clouds-
ley-Thompson (1961b) have reported solpugids striking with the chelicerae.
Cloudsley-Thompson stated that the palpal organs sometimes assisted with
this action. Pocock (1897) and Fichter (1940) claimed the palpi contacted
and pulled the prey into the chelicerae; Pocock, however, stated that the
action was so rapid it was almost impossible to observe. Bolwig (1952)
observed that his species snatched small prey with the chelicerae; bigger
prey being gripped and pulled into the chelicerae with the palpal organs.
Present studies indicate that each above cited worker may have been
correct. Certain species appear to use the chelicerae for the primary prey
strike, some species assisting such action with the palpal organs. Other
species contacted prey with the palpal organs and then used the chelicerae,
but the action was so rapid that both actions occurred almost simulta-
neously. Still other species captured some prey with the palpi alone, other
prey with the chelicerae alone.
When both facets of prey capture are considered, solpugid size, prey
size, prey sclerotization, and inherent solpugid behavior seem to be the
controlling factors. For instance, a large, long-legged solpugid such as
H. serrata rushing to attack would probably strike large or small prey
with the chelicerae. A small or short-legged solpugid such as B. potens
in ambush for small prey might strike with either the palpi or the chelicerae.
Any solpugid in an area of high, small-prey density such as an ant or ter-
mite colony might use the palpi to lift food to the chelicerae. E. titania
and E. palpisetulosus have been observed to strike with the chelicerae,
T. bilobatus and E. magnus with the chelicerae and palpi and E. duranonus
with the palpi.
Food Ingestion.-Ingestion of food has been reported in detail by sev-
eral workers (Hutton 1843, Turner 1916, Hingston 1925, Fichter 1940, and
Lawrence 1949). All have been relatively careful observers and their re-
ports essentially agree on feeding fundaments. Some disagreement exists,
however, concerning the ingestion of .solid as well as liquid materials, and
none of the reports more than mentions the manipulation of prey car-
casses during feeding.
In the present observations, ingestion involved laceration by a vertical
movement of the lower or movable cheliceral finger against the upper or
fixed finger and the fondal teeth (Fig. 1). This operation was accom-
plished alternately by each chelicera during the rotary process of grind-
ing the prey tissues between the mesally, longitudinally ridged, basal seg-
ments of the chelicerae. Manipulation or movement of the prey tissue
through the cheliceral mill involved a slight lateral separation of the
chelicerae when one open chelicera was lifted and thrust forward to bite
into unmasticated prey while the alternate, closed chelicera was pulled
downward and backward. The combined deliberate processes of lacerating,
grinding, and manipulating the prey was accompanied by a rhythmic pump-
ing of the pharnyx which moved the liquefied food into the alimentary
canal of the solpugid.
The only observed variations of ingestion seemed to be intra- as well as
interspecific and were apparently governed by prey form, size, and scleroti-
Long, narrow prey such as caterpillars, mealworms, wireworms, or
millipeds were moved from end to end through the cheliceral mill (Fig. 2).

y~.' '




* I'


Fig. 1. Eremobates durangonus female cheliceral mill and palpi, 2.9X
magnification. Fig. 2. Eremorhax magnus young feeding on larva of
Tenebrio sp., 1.8X. Fig. 3. Branchia brevis female feeding on a single ter-
mite, 5.2X.


4 -
Q^ "'

. i

The Florida Entomologist

Large solpugids accomplished complete ingestion by one or two transverse
passages of such prey. Small solpugids completed ingestion on less than
one prey or passed the food several times through the chelicerae before
completing ingestion. Short, thick prey such as scarabaeid beetles, flies,
bugs, or moths were rotated through the cheliceral mill. Large solpugids
crushed such prey into an unrecognizable mass whereas small solpugids
frequently ingested only the softer inner tissues leaving a recognizable
hollow carcass. Small, soft-bodied prey such as termites, small grubs,
maggots, or gnats were consumed in rotated, multiple prey masses by
large solpugids; and one or two at a time by small solpugids (Fig. 3).
Small or immature solpugids and males frequently failed to wound heavily
sclerotized prey. Such prey were, however, eaten by the more powerful
females or larger species of solpugids.
Several engorged solpugids were dissected and the stomach contents
microscopically examined for particulate food material. Fecal matter was
also examined microscopically. As stated by Hutton (1843), Turner (1916),
Hingston (1925), and Lawrence (1949), solpugids ingest solid as well as
liquid food. Solid materials were, however, finely ground and not digested.
Exoskeletal and setal fragments were frequently found in both stomach
contents and fecal matter.
Engorgement.-Gluttonous feeding by solpugids has been reported by
most workers. Reports all claim engorgement until the abdomen was so
distended that the solpugid could scarcely move.
Females and immatures of all species observed during the present
study engorged until the abdomen was distended several times its original
size. Such engorgement did not, however, seriously hamper quick orienta-
tion, running, or burrowing. Engorged females and young were com-
monly seen under natural conditions. Males were never observed eating
to engorgement. Further, engorged males were never observed under
natural conditions even in the presence of abundant food.
Starvation.-Starvation has not been previously reported. In the pres-
ent study, it was noted that although solpugids normally fed ravenously
and to the state of engorgement, they also survived long periods of time
with no food. Several conditions under which solpugids do not feed were
mentioned previously. Other conditions are discussed below.
In the laboratory many species refused to feed in the presence of food.
A. stimpsoni, A. peninsulana, and E. magnus were observed to kill and
drop termites. In one instance, A. stimpsoni killed and buried a mass
of termites. Similar behavior has been reported by Hutton (1843) for a
species of Galeodes; but the uneaten prey were warm-blooded, a young
sparrow and young muskrats, and must be considered abnormal.
In the present study, many species underwent prolonged periods of
such unexplained fasting in the presence of arthropod food. Specimens
fed daily for life-cycle studies would suddenly stop eating and finally
die of starvation or dehydration with adequate food available. An imma-
ture A. stimpsoni died after 60 days without feeding; a sub-adult Eremo-
bates sp. after 55 days; and a juvenile of A. peninsulana after 70 days,
even though food was offered three to five times a week.
Several immatures and males of E. durangonus, Eremobates nodularis
Muma, E. palpisetulosus, and T. bilobatus were experimentally starved for
one, two, and three weeks and survived.


Vol. 49, No. 3

Muma: Feeding Behavior of Solpugida 205

The limited data obtained on starvation in the presence and absence of
food indicate that solpugids are gluttonous predators capable of surviving
for long periods of time without eating.
Water Ingestion.-Drinking by solpugids has been recorded several
times. Pocock (1898) quoted a Sudan war correspondent as stating that
a large solpugid "sucked water from the side of the (porous water) jar."
Hingston (1925) stated that Galeodes arabs Pocock used the palpi for drink-
ing, thrust the chelicerae into water, or even climbed into a watch glass
containing water. Fichter (1940) reported that Eremobates pallipes (Say)
drank from standing droplets of water.
Several North American solpugids have been observed drinking from
droplets of water on glass microscope slides. The water was imbibed by
lifting it with the palpi or thrusting the chelicerae into it. Several speci-
mens of different species have apparently obtained water by chewing the
moistened cotton plugs of collection tubes. On several occasions, speci-
mens .apparently near death after air mail shipment have been revived by
water and cricket hemolymph dropped between the chelicerae.
On the other hand, water sprayed over a terraria or individual speci-
mens greatly excited solpugids, causing them to run rapidly or begin bur-
rowing activities. Rain caused the same reaction under natural conditions.
It seems likely that solpugids obtain necessary water from their prey as
has been suggested by Pocock (1898) and Cloudsley-Thompson (1961a).
Communal Feeding.-This behavior, which seems to be restricted to
second instar nymphs, has been reported by Hingston (1925). In his dis-
cussion of the early stages of G. arabs, Hingston twice mentions two or
three "jerrymunglums" feeding on a fragment of lizard or on the carcass
of another solpugid.
In the present study, second instar nymphs of E. durangonus were
frequently observed feeding communally. As many as five were recorded
feeding on a single worker, Reticulitermes hageni (Banks).
Cannibalism.-Several workers have reported cannibalism among sol-
pugids, Hutton (1843), Hingston (1925), Fichter (1940), Cloudsley-Thomp-
son (1961a), and Muma (1966). In most instances, these records referred
to two or more specimens confined in the same terraria, two specimens
that contacted each other while food searching, or females that ate males
after mating. However, Hutton (1843) described a stag-fight between two
males, Hingston (1925) mentioned egg-cannibalism and communal-canni-
balism, and Muma (1966) reported males eating females while in the proc-
ess of mating. Cannibalism among solpugids apparently takes several
forms: egg-cannibalism, communal-cannibalism, mating-cannibalism, and
In the present study, egg-cannibalism was observed for E. durangonus,
E. palpisetulosus, E. nodularis, T. bilobatus, A. stimpsoni, A. peninsulana,
and Branchia brevis. Under natural conditions, egg-cannibalism is prob-
ably minimal as the females lay egg masses in burrows and then abandon
Communal-cannibalism was observed for second-instar nymphs of E.
durangonus on several occasions. This type of cannibalism is probably
also minimal under natural conditions wherein second-instar nymphs can
achieve a greater dispersion than in a laboratory terrarium.
Mating-cannibalism involving males took the form of stag-fights. Three
species, E. durangonus, E. palpisetulosus, and T. bilobatus were observed

206 The Florida Entomologist Vol. 49, No. 3

in stag-fights. In the first two species, the antagonists assumed a threat-
ening attitude with the chelicerae open, and the palpi and first legs flexed.
In this position, the males rocked back and forth on the remaining legs
until one or both leapt forward in an attempt to bite the other. If one
made contact, the fight was over and the conquered male was eaten. If
neither made contact, both males leapt backward and the fight continued,
or one fled ending the fight. T. bilobatus males conducted a much more
stylized stag-figth. It closely resembled that reported by Hutton (1843)
for Galeodes sp. After a brief threat, both males leapt toward each other
and thrust the palpi forward and locked them in a flexed position. The
two males then pushed, pulled, jerked, and twisted in a manner remarkably
similar to the stylized contests of judo wrestlers. If neither male achieved
an advantage, they sprung apart, threatened and locked palpi again. If
one was smaller, he turned and fled but resumed the fight if a second
contact was made. If one male achieved an advantage, he buried his cheli-
cerae in the other and the fight was over.
Mating-cannibalism involving males and females of the genus Eremo-
bates took several forms. A female that rejected a male leapt forward,
over-powered, and ate him before the mating procedure began. A female
that rejected a male continued to fight during the mating procedure and
sometimes over-powered and ate him. A mated female that recovered
from the mating trance during the mating procedure turned, if possible,
and ate the male after insemination. Occasionally, a male during the
mating procedure wounded the female; when this happened, he frequently
ate her. In one instance, a male of E. dulrangowns wounded a female, com-
pleted mating, and then ate her.
Contact-cannibalism involved females or immnatures which on contact
simply fought until one was overcome and eaten.
Frequency of cannibalism under natural conditions was not investigated.
It has been noted, however, that contacts frequently resulted in both in-
dividuals turning and fleeing. This indicates that cannibalism may be in-
frequent except possibly under conditions of stress such as inadequate food.
Food Preference.-Except for references by Pocock (1898) and Law-
rence (1963) to termitophilous species, most workers have indicated the
apparent wide food tolerance of solpugids. Fischer (1910) and Hingston
(1925) reported that Galeodes spp. when confined would even feed on in-
sects seemingly repellent to them on initial contact.
Food presence data on North American solpugids is presented in
Tables 1 to 4. Some solpugids refused certain foods and readily accepted
others. For example, E. magnus refused meadow grasshoppers but readily
accepted tenebrionids, small scarabaeiids, and termites; E. durangonus
refused roaches, earwigs, and velvet mites but readily accepted termites
and pardosid spiders; and T. bilobatus refused crickets but readily ac-
cepted termites. Other solpugids accepted or readily accepted most offered
foods, but occasionally immatures or one sex refused certain foods.
Certain facets of food preference are obscured by tabular presentation.
It is not apparent, for instance, that a large species such as E. titania or
E. palpisetulosus fed more readily on large prey such as grasshoppers and
less readily on small prey such as termites. On the other hand, small
species such as T. bilobatus and A. peninsulana avoided, when possible,
large prey and accepted small prey. Limiting or special factors influencing
food preference are discussed under specific feeding.

Muma: Feeding Behavior of Solpugida


pi1 :



ce P
0 ,

C) C
















43 4-
C) 43
* 3

I3 '


I +2

C) a


td 1


The Florida Entomologist

Vol. 49, No. 3

~t1~ ~


4, o
Z' 6 0



4. 14 i



Muma: Feeding Behavior of Solpugida





0 p<
0) 5

a 2
s a

C)) .



** 0

1 C)


* 0

) c~
a 3
4 4

S )

C) C)
C) Ca

0 t



C -
', 0
>> 1s

3.. a
C) )*


The Florida Entomologist

P 4 0

0 0a


i4 U -(







u 9s 4 -
CC I "
CC> CCri

94 C0 C|5
o- ;, 2

Vol. 49, No. 3


-- 0+


S C"

CC ^

*3 0

0 a

g C)
EQ +)

ID 0


43 0





d v



>p .

<1 K1

Muma: Feeding Behavior of Solpugida


M 0

o ^^
WS 41 ;
f Pi ",



O -q 0
ft a. 9
P- ^ c8

CO ~ C CO4-





I a)




a) ;

a) '





0 0

s 's


- 0
E 0

The Florida Entomologist

Certain species, because they were included in feeding experiments,
because they were available at observation times, or because they were
specifically selected, have yielded behavior information that may be spe-
cific in nature. There is no literature dealing with these species.
Eremorhax magnus.-This large, short-legged species accepted a wide
variety of insects as food but fed most readily on small beetles. Because
it burrowed more extensively than other species observed, it was also of-
fered earthworms, which were also accepted readily. The species was
observed twice in random cursorial searching under field conditions. It
was also observed several times at night lights both stalking and ambush-
ing Diplotaxis spp. Capture was made either with the palpi or with the
chelicerae although the latter seemed to be used most frequently. Inges-
tion was accomplished entirely with the chelicerae (Fig. 2). The palpi were
flexed and held above the propeltidium during feeding. When E. magnus
fed on termites, it did so clumsily, frequently dropping the partially
chewed mass. Termites were frequently killed and dropped uneaten. Only
females and immatures were available for observation.
Eremorhax striatus.-This large, long-legged species accepted or accept-
ed readily almost any insect offered as food. Size seemed to be the only
limiting factor; the only food refused at any time was termites. The spe-
cies has not been observed searching for or capturing food in the field.
In the laboratory, large prey seemed to be captured most frequently with
the chelicerae; but captures were accomplished so quickly that observation
was difficult. Small prey were lifted to the chelicerae with the palpi. Fe-
males and immatures located prey only by tactile stimuli. Only one male
was available for observation.
Eremorhax titania.-This large, long-legged species accepted or accept-
ed readily all food offered except a large mymeleontid adult that was ex-
posed to a male for two days. It readily accepted termites, which were
occasionally refused by E. striatus. With the exception of one female
collected from a can trap, all males and females were observed in random
cursorial searching under field conditions. Capture of large prey was
accomplished primarily with the chelicerae at the end of a lunge. Ter-
mites were lifted to the chelicerae with the palpi, then supported by the
first pair of legs. The palpi were held to the side in a flexed position dur-
ing feedings. Males located moving prey by sight; they apparently saw
both moving shadows and objects at distances of 10 to 20 cms. Females
seemed to respond only to tactile stimuli and males did not see non-mov-
ing prey.
Eremobates durangonus.-This moderate-sized, long-legged species (Fig.
1) accepted or accepted readily a wide variety of insects offered as food.
Although termites were accepted more readily than other prey, the criti-
cal factor seemed to be degree of sclerotization of the particular insect
offered. Roaches, earwigs, and velvet mites were, however, consistently
refused. The species was frequently observed in random cursorial search-
ing under field conditions. Specimens were also regularly found in houses
and at night lights capturing small noctuid moths. E. durangonus was
found most commonly among and under termite-infested "cow-pies." On
three such occasions, at night, the species was observed walking on the
under side of the "pie" actively feeding on termites. Food location seemed


Vol. 49, No. 3

Muma: Feeding Behavior of Solpugida 213

to be entirely by touch. The species did not see termites or other prey
until the prey touched a seta or appendage. Males, females, and young all
responded to tactile stimuli. Prey capture was accomplished primarily
with the palpi. Large, well-sclerotized prey were sometimes captured with
the chelicerae following a lunge. Termites were lifted to the chelicerae
with the palpi. A loose mass of 20 to 40 termites was frequently manipu-
lated by the chelicerae at one time. The palpi and first pair of legs en-
circled and apparently supported such masses of termites during inges-
tion. Termite masses were seldom dropped by this species; instead ex-
perimentally added termites that fell were promptly picked up with the
palpi and returned to the mass.
Eremobates palpisetulosus.-This large, long-legged species readily ac-
cepted nearly every prey offered. It seemed to feed easily on heavily as
well as lightly-sclerotized food. It apparently has a wide food range.
Males, females, and immatures were all observed in random cursorial
searching. Males and females were also seen frequently at night lights
but were never observed feeding under natural conditions. In the labora-
tory, the species seemed to locate prey only by tactile stimuli. Large, well-
sclerotized prey were captured entirely with the chelicerae and small,
lightly-sclerotized prey such as termites entirely with the palpi. With
most foods, however, the capture was so rapid that both chelicerae and
palpi appeared to be utilized. During ingestion, the palpi were held aloft
and to the side. Males, females, and young all seemed to locate their prey
by tactile stimuli. Both males and females, but especially the males,
cleaned their chelicerae of food particles by standing high on their tarsi
and doing a jittering dance while sliding the deflected chelicerae back and
forth on the sand.
Eremobates nodularis.-This moderate-sized, short-legged species ac-
cepted or readily accepted every prey offered. Males and young readily
accepted lightly-sclerotized food but would attack heavily-sclerotized food
and accepted only food wounded by such attack. Females seemed to readily
accept heavily-sclerotized as well as lightly-sclerotized prey. The food
range is probably wide but limited by degree of prey sclerotization. Males,
females, and young were all observed in random cursorial searching and at
night lights, but were not observed feeding under natural conditions. Lab-
oratory specimens were also observed apparently in ambush. In the lab-
oratory, the species, particularly males, apparently saw moving prey such as
termites at a distance of one or two cms but walked over non-moving prey.
Moving prey were, however, sensed at a distance from the rear as well as
from the front, suggesting the operation of some sense other than sight,
perhaps substrate vibration reception. Although small prey such as ter-
mites were captured with the palpi, large prey were captured with the
chelicerae. During ingestion, the palpi were held down and to the side.
This species was also observed cleaning the chelicerae in the manner as
described above for E. palpisetulosus.
Therobates bilobatus.-This small, fragile, long-legged species refused
or only accepted most offered prey. Only small moths and small, lightly-
sclerotized foods were accepted readily. Palpal position while feeding was
not adequately determined. Numerous males were observed at night lights
apparently in random cursorial searching and feeding on crambid moths.
Therobates cameronensis Muma.-A single female of this tiny, long-
legged species was offered two prey, a coccinellid larva and a calliphorid

The Florida Entomologist

fly. The larva was refused; the female seemed to be repelled as she
frantically cleaned her chelicerae after each contact with the larva. The
fly was captured immediately with the chelicerae and the soft inner organs
were eaten; the exoskeleton, wings, and legs were uneaten. During feed-
ing, the palpi were held back and to the side.
Ammotrechula, peninsulana.-This small, long-legged species accepted
or accepted readily most prey offered. Only earwigs and pardosid spiders
were refused. Because of its small size, the species was not offered heavily-
sclerotized prey. Males, females, and young were all observed feeding at
night lights, but prey were too macerated to be recognizable. A number
of specimens were found in bathrooms in houses. In the laboratory, prey
were captured with the palpi, and during ingestion the palpi were held
aloft in a flexed position. Prey location in the laboratory seemed to pre-
clude tactile response as specimens immediately oriented to distant prey,
indicating sight or some other sense.
Ammotrechella stimpsoni.-This small, long-legged species was fed pri-
marily on termites, Reticulitermes flavipes Banks. The species was not
observed searching and feeding under natural conditions, but most speci-
mens were taken on or in houses and in termite-infested logs and tree
limbs. Laboratory specimens, males, females, and young, apparently lo-
cated prey at a distance of several centimeters by sight. Prey capture
was accomplished with the palpi which were held aloft and to the side
during feeding.
Behavior discussed in this section encompasses reactions to stimuli that
would not normally occur under natural conditions. Turner (1916) manip-
ulated a dead cricket and induced feeding by Eremc-bates formicaria
(Koch) but found that the solpugid would not eat a dead, motionless
cricket. Hingston (1925) soaked a grasshopper in and filled a locust with
quinine and induced a Galeodes arabs Pocock to eat them. Lawrence
(1949) stated that Solpuga caffra Pocock ate rejected cricket legs if they
were left in the terrarium overnight. Cloudsley-Thompson (1961b) claimed
that Galeodes granti Pocock would pick up a dead insect and begin eat-
ing it.
In the present study, the following abnormal feeding was induced.
Eremorhax magnus, E. pulcher Muma, E. striatus, E. titania, Eremobates
durangonus, E. nodularis, E. palpisetulosus, E. zinni Muma, Therobates
plicatus Muma, T. imperialis, Hemerotrecha serrata, Ammotrechula pen-
insulana, Ammotrechella stimpsoni, and Branchia potens were all forced
to imbibe water. The water was dropped with a capillary eye-dropper
between and at the bases of the chelicerae. The solpugids slid the cheli-
cerae back and forth until the droplet was ingested and repeated the pro-
cedure for several droplets. With the same methods and results, E. mag-
nus, E. durangonus, E. palpisetulosus, A. peninsulana, and B. potens were
forced to imbibe a 1.0% aqueous yeast hydrolysate solution. The same
methods and results were used to revive a moribund male and immature
of E. durangonus with cricket hemolymph.
Dead insects or cricket legs were manipulated in forceps and accepted
by E. magnus, E. striatus, E. durangous, and E. palpisetulosus. Lean
ground beef was offered on a small camel's-hair brush to two females of
E. palpisetulosus, and both accepted several small fragments.


Vol. 49, No. 3

Muma: Feeding Behavior of Solpugida

Since solpugids fed on unnatural foods in unnatural conditions, they
might be reared on artificial nutrient solutions and gels.

This investigation of the feeding behavior of North American solpugids
has demonstrated the following:
Food searching behavior involves random running and congregating
in areas of high prey density at night lights, in houses, or at prey nests.
It is suggested that ecological conditions, food availability, and inherent
behavior may alter the method utilized.
Prey are located by orientation to tactile, visual, and substrate vibra-
tional stimuli. There are indications that different genera and species
vary in reactions.
Prey capture is of three types: chase, stalk, and ambush. Most species
appear to chase and ambush, but certain species also stalk. Either palpi
or chelicerae may make first prey contact, depending on prey size, prey
sclerotization, and inherent behavior.
The process of ingestion is similar for all species with variations ap-
parently caused by prey size, form, and sclerotization. Females and im-
matures feed gluttonously to extreme engorgement, but males do not en-
gorge. Most species can survive in the absence of food for one to three
weeks. Starvation in the presence of food also occurred in the laboratory
with survival for 55 to 70 days.
All species tested drank readily from standing droplets of water. On
the other hand, solpugids shunned rain or simulated rain.
Second-instar nymphs of E. durangonus practiced communal feeding,
up to five feeding at one time on a single termite.
North American solpugids practice cannibalism of eggs, during com-
munal feeding, prior to, during, and after mating, and on chance contacts.
Food preference studies indicated a degree of prey specificity governed
principally by prey size and sclerotization. Studies on feeding behavior
of nine species revealed differences in search, location, capture, and in-
gestion. The feeding of solpugids on abnormal foods suggests the possible
development of artificial foods.

Bolwig, Niels. 1952. Observations on the behavior and mode of orienta-
tion of hunting solifugae, J. Ent. Soc. South Africa. 15(2): 239-
Cloudsley-Thompson, J. L. 1958. Spiders, scorpions, centipedes and mites.
Pergamon Press, London.
Cloudsley-Thompson, J. L. 1961a. Observations on the natural history of
the "camel-spider," Galeodes arabs C. L. Koch (Solifugae: Galeodi-
dae) in the Sudan. Ent. Monthly Mag. 97: 145-152.
Cloudsley-Thompson, J. L. 1961b. Some aspects of the physiology and
behavior of Galeodes arabs Pocock. Ent. Exp. Appl. 4:257-263.
Fichter, Edson. 1940. Studies of North American solpugida, I. The true
identity of Eremobates pallipes (Say). Amer. Midland Natur. 24
(2): 351-360.
Fischer, C. E. C. 1910. Observations on the spider Galeodes indicus, J.
Bombay Natur. Hist. Soc. 20: 886-887.


216 The Florida Entomologist Vol. 49, No. 3

Hingston, R. W. 1925. Nature at the desert's edge, Chapters 11, 12.
Witherby, London p. 230-279.
Hutton, Thos. 1843. Captain Thos. Hutton on Galeodes (vorax?). Ann.
Mag. Natur. Hist. 75:81-85.
Lawrence, R. F. 1949. Observations on the habits of a female solifuge,
Solpuga caffra Pocock. Ann. Transvaal Mus. 21(2): 197-200.
Lawrence, R. F. 1963. The solifugae of South West Africa. Cimbebasia
8: 1-27.
Marx, Geo. 1892. Contributions to the knowledge of the life history of
Arachnida. Proc. Ent. Soc. Wash. 2(2): 252.
Muma, M. H. 1951. The Arachnid order Solpugida in the United States.
Bull. Amer. Mus. Natur. Hist. 97(2): 35-141.
Muma, M. H. 1962. The Arachnid order Solpugida in the United States,
Supplement I. Amer. Mus. Novitates 2092: 1-44.
Muma, M. H. 1963. Solpugida of the Nevada Test Site. Brigham Young
Univ. Sci. Bull. Biol. Ser. 3(2): 1-15.
Muma, M. H. 1966. Mating behavior of Eremobates Banks. Anim.
Behav. (in press).
Pocock, R. I. 1897. On the genera and species of tropical African Arach-
nida of the order Solifugae, with notes upon the taxonomy and
habits of the group. Ann. Mag. Natur. Hist. 20(6): 249-272.
Pocock, R. I. 1898. On nature and habits of Pliny's solpuga. Nature.
57(1487): 618-620.
Putnam, J. D. 1883. In memorial Joseph Duncan Putnam, The Solpugi-
dae of America. Proc. Davenport Acad. Natur. Sci. 3(3): 195-
Turner, C. H. 1916. Notes on the feeding behavior and oviposition of a
captive American false spider (Eremobates formicaria Koch). Anim.
Behav. 6: 160-168.

The Florida Entomologist 49(3) September 1966

.-"* ,***
-: 2:.-

r~ -


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs