White flies injurious to citrus in Florida


Material Information

White flies injurious to citrus in Florida
Physical Description:
Morrill, A. W ( Austin Winfield ), 1880-
Back, E. A ( Ernest Adna ), 1886-
U.S. Dept. of Agriculture, Bureau of Entomology ( Washington, D.C )
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 029691336
oclc - 01810124
System ID:

Full Text


*te, i



k* S

it 0,11 vil T- *w
1_ 41


i :v



All 4

44, __t

_j '4






"i A
61 ,A


'Wn-wI!IWWl .,.

Bul. 92,. Bureau of Entomology, U. S. Dept. of Agriculture.




, 4?!-A ": id, a.

L. 0. HOWARD. Eaolo6i and ChWa of Bumsu.




E. A. BACK, PH. D.

ISSUED JULY 12, 1911.





L. 0. HOWARD, Entomologist and Chief of Bureau.
C. L. MARLATT, Entomologist and Acting Chief in Absence of Chief.
R. S. CLIFTON, Executive Assistant.
W. F. TASTET, Chief Clerk.
F. H. CHITTENDEN, in charge of truck crop and stored product insect investigations.
A. D. HOPKINS, in charge of forest insect investigations.
W. D. HUNTER, in charge of southern field crop insect investigations.
F. M. WEBSTER, in charge of cereal and forage insect investigations.
A. L. QUAINTANCE, in charge of deciduous fruit insect investigations.
E. F. PHILLIPS, in charge of bee culture.
D. M. ROGERS, in charge of preventing spread of moths, field work.
ROLLA P. CURRIE, in charge of editorial work.
MABEL COLCORD, librarian.


Washington, D. C., March 2, 1911.
SIR: I have the honor to transmit herewith, for publication as
Bulletin 92 of the Bureau of Entomology, a manuscript prepared by
Drs. A. W. Morrill and E. A. Back, dealing with the life history of
the white flies injurious to citrus trees in Florida.
The investigation of the citrus white flies in Florida, under the
general direction of the assistant chief of this bureau, Mr. C. L. Marlatt,
was begun in 1906, and is now approaching completion. There has
already been published a bulletin (No. 76) dealing fully with the
general subject of fumigation with hydrocyanic-acid gas for the white
fly. A circular (No. 111) has also been issued, giving brief directions
for winter fumigation.
The present publication is a general account of the two species of
white flies which are of special economic importance to the citrus
grower in Florida. The publication includes the history of these
insects in the United States, their distribution and food plants, and
a very detailed study of the habits and life cycle of the two species.
A great deal of painstaking and minute work has been done, and the
information secured furnishes an accurate foundation for the develop-
ing of the best means of control.
Supplementing this publication, which deals largely with life his-
tory and habits, it is proposed to publish a bulletin on control by
sprays, fungi, and other enemies, and to supplement or reissue in
revised form the bulletin dealing with fumigation.
Entomologist and Chief of Bureau.
Secretary of Agriculture.




Introduction............... .................................
Species of white flies affecting citrus.........................
The citrus white fly (Aleyrodes citri R. and II.)..............
Historical review..................................
Origin............... ...............................
Earlv hit.nrv in t.h Tlnit.pd Stsateps


Nature of injury. ..................
Loss of sap........................
Sooty mold.......................
Extent of injury......................
Injury to fruit....................
Injury to trees....................
Summary of losses.................
Increased cost of maintenance.....
In the United States..................
In foreign countries...................
Food plants...............................
Authentic and questionable records...
Economic significance of food plants
food plants and insects.............
Citrus ..........................
China trees and umbrella trees ...
Cape Jessamine ................
Privet hedges..................
Japanese and wild persimmons...
Lilac ...........................
Prickly ash ....................
Coffee .........................


.. .......interrelationship between..
interrelationship between

. . . . . . . . . . .. ..=.=.

Occasionally infested food plants.

Spread in the United States.......................
Checks on successful establishment. ---.........
Flight of adults...............................
Winds.............. ............................
Vehicles, railroad trains, and boats.............
Citrus nursery stock and ornamental plants ....
Accidental spread by man......................
Life history and habits.............................
Methods of study..........................
The egg................... ......................
Duration of egg stage.......................



VL.y ALA ".&K; %JLALLIIUA "U"LOU0 - - - - - - - -
. . . . . . . . . . . . . . . . . . .







The citrus white fly-Continued.
Life history and habits-Continued.
The egg-Continued.
The larval and pupal stages............

Description of stages.............................
Duration of stages.............--.-...............
Locom otion.....................................
Growth............... ............................
Feeding habits of larvae and pupae-...............
The adult..........................................
Description................... ...................
Emergence..................-.- .....-..........
Duration of life.................................
M ating .........................................
Proportion of sexes.............................
Influence of weather conditions on activity of adul
Feeding habits of adults..........................
Length of life cycle.................................
Seasonal history...........................................
Generations of the citrus white fly...................
Seasonal fluctuations in the numbers of adults or so-cal
The cloudy-winged white fly (Aleyrodes nubifera Berger).......
Amount of injury by the cloudy-winged white fly. -.......
D distribution ............................................
Food plants..............................................
Spread .................................................
Life history and habits...................................
S The egg..............................................
The larval and pupal stages...........................
The adult ..........................................
Length of life cycle................................
Seasonal history..........................................
Generations of the cloudy-winged white fly............
Seasonal fluctuations in numbers of adults or so-called '



....... .. .......




.............. 57
.............. 57
...... ....... 58
- 58
.............. 62
... .......... 63
.............. 64
.------- .- ---.. 6 4
--- --- --- --- 65
.............. 65

.............. 66
.............. 67
..............- 71
............... 71
............... 72
S.............. 77
ts.............. 78
............... 78
............... 79
S.............. 80
............... 81
.............. 81
led "broods".. 84
.............. 86
.........- ..... 86
.............. 87
............... 89
............. 90
............-. 90
.............. 91
.............. 91
............... 94
............... 98
.............. 100
............... 101
.............. 101
"broods"...... 102
.............. 105

I L U S T R A T IO() N S.

PLATE I. Fig. 1.-Orange covered with sooty mold. Fig. 2.-Leaf of orange
coated with sooty mold................................ Fronti.ispiee.
II. Fig. 1.-Aleyrodes monri on mulberry. Fig. 2.-Woolly white fly
(Aleyrodes howardi) on orange. Fig. 3.-Paraleyrodes 7'erse;e on
orange. Fig. 4.-Woolly white fly (Aleyrodes howardi) showing
eggs in circles, and adults. Hig. 5.-Aleyrodes sp. on sea grape... 10
III. Fig. 1.-Sooty mold on orange following white-fly attack; broken and
falling from leaf. Fig. 2.-Sooty mold on cinnamon tree following
attacks by cinnamon scale ..................................... 18
IV. Fig. 1.-China tree defoliated during winter. Fig. 2.-Same tree in
full foliage in summer.......................................... 34
V. Fig. 1.-Leaflet of umbrella China tree showing infestation by A.
citri. Fig. 2.-Umbrella China tree infested by A4. citri, showing
orange trees in rear of house................................... 34
VI. Fig. 1.-Nursery citrus trees infested with white flies, set out in an
isolated noninfested grove without having leaves removed. Fig.
2.-Buggy in an orange grove; buggy-top full of adult white flies
ready to be carried to other groves. Fig. 3.-Train at station; adult
citrus white flies swarming from near-by umbrella China tree into
coaches ready) to be carried for miles down the Florida east coast.. 48
VII. Fig. 1.-Rearing cages in position on orange trees. Fig. 2.-En-
larged rearing cage............................................. 52
VIII. Fig. 1.-Leaf showing pupa cases of A. citri; also pupae and eggs.
Fig. 2.-Leaf showing heavy infestation by A. citri. Fig. 3.-
Leaf showing pupa cases of A. nubifera........................... 62
IX. Fig. 1.-Tender shoot swarming with adult citrus white flies. Fig.
2.-Leaf of same, enlarged...................................... 66
X. Fig. 1.-Adults of A. nubifera, the cloudy-winged white fly, and
many eggs scattered over leaf. Fig. 2.-Larvae and pupme of both
the citrus white fly and cloudy-winged white fly killed by fumi-
gation, and eggs of the citrus white fly along midrib.............. 98


FIG. 1. Map showing distribution of the citrus white fly (Aleyrodes citri) in
Florida......................................................... 26
2. Diagram of laboratory grounds at Orlando, Fla...................... 37
3. The citrus white fly: Eggs........................................ 53
4. The citrus white fly: Larva, first instar, dorsal view................. 59
5. The citrus white fly: Larva, first instar, ventral view................ 59
6. The citrus white fly: Antenna and left hind leg; first instar larva .... 59
7. The citrus white fly: Larva, second instar, ventral view.............. 60
8. The citrus white fly: Larva, third instar, ventral view................ 60


FIG. 9. The citrus white fly: Pupa and details.............................
10. The citrus white fly: Adult and details............................
11. Diagram showing annual generations of the citrus white fly...........
12. Diagram showing abundance of adults of the citrus white fly at Or-
lando, Fla., throughout 1909......................................
13. Map showing distribution of the cloudy-winged white fly (Aleyrodes
nubifera) in Florida............................................
14. The cloudy-winged white fly: Eggs.................................
15. The cloudy-winged white fly: Larva, first instar, ventral view .....
16. The cloudy-winged white fly: Larva, third instar, ventral view-......
17. The cloudy-winged white fly: Pupa and details......................
18. The cloudy-winged white fly: Dorsal view of pupa, showing adult
insects about to emerge.........................................
19. Diagram showing relative abundance of adults of the cloudy-winged
white fly and the citrus white fly, throughout 1909. at Orlando Fla..








The present bulletin includes the principal results of stuL(lies of the
two species of white flies most destructive to Citnrus in the United
States, commonly known as the citrus white fly (Aleyrodes citri R.
& H.) and the cloudy-winged white fly (Aleyrodes nubifera Berger).
With these pests successful control measures must be based on a
complete understanding of the insects themselves. On this account
the study of the insects, their life history, seasonal history, habits,
food-plant relationships, and related topics has occupied an unusually
important position in the white-fly investigations.
The authors have concluded that unless natural enemies capable
of controlling the two white-fly pests are existent and are secured,
control measures will require permanent expert supervision for the
most satisfactory and economical results-not supervision of work in
individual citrus groves, but supervision aimed principally to prop-
erly correlate individual efforts and to take full advantage of favoring
local conditions. For supervision of this nature, a good foundation
of extensive and reliable studies of the insects is necessary. While
the portion of the white-fly investigations herein reported is com-
paratively extensive, it is necessarily not exhaustive and in the
course of time certain features of this work can undoubtedly be con-
tinued with profit as an aid to the future improvement of control
The white-fly investigations now in progress were begun in July,
1906, by the senior author, who was in field charge up to the time of
his resignation from the bureau in August, 1909, The junior author's
connection with these investigations dated from June, 1907. The
life-history studies of the first two years have been largely superseded
by the more extensive work of the third year. Practically all of the
data presented under the subjects of the life history and habits and
the seasonal history of each species are based on studies by the junior
author and were written by him. The remainder of the bulletin was
written by the senior author.



Twelve species and one subspecies of the family Aleyrodida are
known to breed upon citrus. The list of these insects, the authority
for the original description, the recorded distribution, and the food-
plant records are given in Table I :

TABLE I.-Aleyrodidx that breed upon citrus.

Species. Described by- Occurrence. Food plants other than

Aleyrodes citri (syn. au- Riley and How- North and South America, See list, p. 29.
rantii).1 ard. Asia, Japan.
Aleyrodes floccosa.......... Maskell........... Mexico, Jamaica.......... Guaiacum officinale.
Aleyrodes floridensis....... Quaintance....... United States (Florida).... Persea gratissima (alligator
pear), Psidium guajava
Aleyrodes giffardi......... Kotinsky......... Hawaii................... None recorded.
Aleyrodes howardi......... Quaintance....... Cuba, United States (Flor- Do.
Aleyrodes marlatti............. do............ Japan..................... Do.
Aleyrodes mori.............. do............ United States (Florida).:. Do.
Ale vrodes mori arizonensis Cockerell .......... United States (Arizona)... Do.
Aleyrodes nubifera........ Berger ............ United States (Florida, Do.
Louisiana), Cuba.
Aleyrodes spinifera........ Quaintance....... Java..................... Rosa spp.
Aleyrodes struthanthi..... Hempel........... Brazil.................... Michelia flava Loranthus
(struthansus5 flexicaulis.
Aleyrodes vitrinellus 2...... Cockerell.......... Mexico................... None recorded.
Paraleyrodes persee........ Quaintance...... United States (Florida)... Do.

Mr. A. L. Quaintance, after careful comparison of material from Maskell's collection, evidently type
material, with A. citri, concluded that Maskell's aurantii was the same as Riley and Howard's citri. Through
the kindness of Mr. Quaintance the authors have had an opportunity to examine the material referred to
and agree with him in considering auranlii a synonym of ciri.
2 There seems to be some doubt as to the identity of the food plant of this species, for in connection with
the description the authority for it gives the following food plant record: "On the under side of leaves which
appear to be those of orange."

Of the Aleyrodidae referred to above, A. citri, A. giffardi, A. howardi
(P1. II, figs. 2, 4), and A. nubifera are known to be orange pests or
capable of becoming orange pests. A. floridensis, A. mori (PI. II,
fig. 1), A. mori arizonensis, and Paraleyrodes persex (P1. II, fig. 3)
apparently are not likely to cause injury to citrus, while the remainder
of those listed are doubtful in this respect.
Paraleyrodes persew is found in all sections of Florida and is fre-
quently quite abundant, but in only one instance has it been known
to cause blackening of the foliage of citrus trees. This was in the-
winter of 1906-7 and occurred in a pinery where in one section citrus
nursery trees were being grown. In the course of two or three months
after being first noticed the insects were reduced to the point of
scarcity through parasitism by a new species of Encarsia, which
Dr. L. 0. Howard has described under the name of Encarsia variegata.
Observations extending over three years indicate that this parasite
will effectively control P. persea and that it is unlikely that this
aleyrodid will ever cause noticeable injury under ordinary conditions.
It is, however, possible that the appearance of a prolific hyperparasite
of Encarsia variegata might seriously interfere with the present equi-
librium in nature.


- - --------------------------------- ------------------------------ ----

Bul. 92, Bur.,,u of E_,,m '',, U. S Dtpi of A'.!r PjtLjE'

Fig. 1.-A.lr,/J,,!. mori on mulberry. Fig. 2.-Woolly whlit tfl (.I i ',i,,,..i. l,,,,ii,,ii) on oninrn...
Fig. :.- 'nrlf li r',,ii t.- perscis on ',range. Fi -. 4.-Woolly while t Hy (.l1/i ,t,,i,' h/.. ,,,,rrli),show-
ing egg- in circles and adults. Fig. 5.-Ali,.41,',, .-" sp. onsea grnil'i. Urigiiiil.i


.' .. ." ... .


The woolly white fly (.'l4r//rodes. iowvrdi Quaiitance (I'l. II, figs. 2,
4)) was first discovered in this country at T'mpa, Fla., ib the jeinior
author in November, 1909. 1T110 iInsect alppeairs to be of receiitt iitri-
duction, since the infested area lias been under observation at inlter-
vals during thej past three an( a half years by the several men
connected with these investigations.'
Of the four species known to be (destructive ito )citrus, Alegroddes
citri and A. nubifera are iclu(lCded inll tlie investi.gationls herein reported.
(Alcyrodes cirri R. & II.)
The origin of the citrus white fly is by circumstances quite defi-
nitely indicate(l to be Asiatic. The present known occurrence of it
in Japan, China, and India will be referred to under the subject of
distribution. The list of food plants, showing as it does the natural
adaptations in this respect, indicates in itself that the fly is not
native to either North or South America, but to Asia. Moreover, if
the citrus white fly were a species native to the Gulf coast region of
North America, or if it had been introduced before 1850, it would
almost certainly have become a pest worthy of mention by Townend
Glover in his reports on the orange insects of Florida published in the
United States Agricultural reports for 1855 and 1858. According
to these reports orange growing was very extensive in proportion to
the population and very profitable in spite of the temporary check
due to the freeze of 1835. The principal orange-growing district in
Florida was, at the time of the reports of Glover, already mentioned,
the northeastern section of the State, along the St. Johns River and
at St. Augustine. Orange growing on a large scale gradually spread to
the south and southwest, the center of production being correspond-
ingly moved. To-day citrus fruits are generally grown in all the
counties of the peninsula of Florida, yet, according to the authors'
estimates, only about 40 per cent of the orange groves of the State are
infested by A. citri.2 These infestations in the different sections are
almost without exception readily traceable to the ordinary sources
of dissemination, with all the evidence strongly against the fly having
been a native species infesting uncultivated food plants. The same
may be said in regard to the occurrence of the citrus white fly in
orange-growing regions in Mississippi, Alabama. Louisiana, and Texas.
I The Woolly White Fly, a New Enemy of the Florida Orange. Bulletin 64, Part
VIII, Bureau of Entomology, U. S. Department of Agriculture, 1910.
2A. nubifera alone occurs in not more than 5 per cent of the groves. In 15 of the
40 per cent above mentioned both A. citri and A. nubifera occur.

.... -



Riley and Howard give the following account of the status of the
citrus white fly previous to 1893:
For many years an important and interesting species of the type genus has been
known to infest orange trees in Florida and in more northern greenhouses, and more
recently the same form has appeared in injurious numbers in the orange groves of
Louisiana. In the Florida Dispatch, new series, volume 11, November, 1885, this
species received the name of Aleyrodes citri at the hands of Mr. Ashmead. The Florida
Dispatch, however, is a local newspaper of no scientific pretensions, and the descrip- :
tion accompanying the name was entirely insufficient to enable recognition aside from
the food plant. We adopt'the name in connection with a full description, not with a
view of encouraging such mode of publication, which is not sanctioned by the canons
of nomenclature formulated and generally accepted, but as a manuscript name, satis-
factory in itself, the authority to be recognized for it being comparatively immaterial.
Our first acquaintance with the species was in June, 1878, when we found it occur-
ring in profuse abundance on the leaves of the citrus trees in the orangery of this
department. Some observations were made upon its life history during that summer,
and all of its stages were observed. During the following years we observed it in
Florida, and it was studied by two of our agents, Mr. H. G. Hubbard, at Crescent City,
and the late Joseph Voyle, at Gainesville. The species was not treated in Mr. Hub-
bard's report on the insects affecting the orange, as we wished to give it a fuller consid-
eration than could then have been given, and other duties prevented doing so in time.
Moreover, at the time when Mr. Hubbard's report was prepared the insect had not
become of especial economic importance.
Since that time many further notes have been made in Washington, and we have
received the species from Pass Christian, Miss.; New Orleans, La.; Baton Rouge, La.;
Raleigh, N. C.; and many Florida localities; and during the past year or two it has
become so multiplied in parts of Louisiana and Florida as to deserve immediate
The authors quoted above specifically recorded the occurrence- of C
the white fly in Florida only at Gainesville (Alachua County), Cres-
cent City (Putnam County),' and Manatee (Manatee County). Dr.
H. J. Webber in 1897 (basing his statement on records in 1893 and
1894) referred to the occurrence of the white fly at the following addi-
tional points: Evinston (Alachua County), Ocala and Citra (Marion-
County), Ormond (Volusia County), Panasoffkee (Sumter County),
Orlando (Orange County), Bartow (Polk County), and Fort Myers
(Lee County). Prof. H. A. Gossard in 1903 mentioned only the follow-
ing additional localities specifically: Tallahassee (Leon County), Lake
City (Columbia County), Jacksonville (Duval County), and Candler ,
(Marion County). In the same publication the following additional

Examination of the specimens of white flies in the collection of the Bureau of
Entomology, collected by Mr. H. G. Hubbard in 1895 and bearing the locality label
"Crescent City," indicate that this record with little doubt refers to Aleyrodes nubifera.
Circumstances known to the authors, but which need not be discussed here, show that
with little doubt the citrus white fly was the species present at Crescent City before
the freeze of the winter of 1894-5. The specimens collected by Mr. Hubbard probably
came from the Hubbard grove at Haw Creek, several miles southeast of Crescent City.


counties were reported more or less in F'e.sted wit iout rifc'reiie.v to
definite localities: Baiker, Jell''rsoii, lAe'Om, and Brevard.
Messrs. Riley a1111d (mowrid anmdl D)r. II. .1. Webber z1dvalice [o tlleio-
ries in regard to tlhe original Florida infestations. Prof. (;hss.ard
however, has the following to say in regard to ti(e matter:
The fly seems to have been first known throughout the' region c'imniris-d in V\'lusia,
Marion, Lake, Alachua, and (O)range ('ountiet, from whiiih I hl:'* liii l or [m d,4iiht
it was transferred to the Manatee country and to l1' al enters along t iv irthern
borders of the State.
According to reliable information received from lMr. MN. S. More-
man, of Switzerland, Fla. ; Mr. A. M. Terwilliger, of Minis, Fla., a1nd
Mr. T. V. Moore, of Miami, Fla., the citrus white fly appeared in tihe
northern part of St. Johns County at a date which indicates that thiis
section was one of the first or possibly the first to be infested in the
State of Florida. Mr. Terwilliger informs us that he first observed
the white fly at Fruit Cove on the St. Johns River in 1879 in a grove
of large seedling trees owned by Col. McGill. The McGill grove
adjoined the grove of the Rev. T. W. Moore, whose son, Mr. T. V.
Moore, corroborates Mr. Terwilliger on the point of thle occurrence
of the white fly in this section prior to 1880. According to Mr.
Moreman the white fly was known in the vicinity of Switzerland on
the St. Johns River in 1882, and was first discovered in his own grove
in 1888. The species concerned is with little doubt the citrus white
fly, A. citri, for the authors and Mr. W. W. Others have been unable
to find specimens of any other species at Switzerland or St. Augus-
tine, the two points visited in the northern part of St. Johns County,
or at Green Cove Springs, located a few miles below Switzerland on
the west side of the St. Johns River in Clay County. These early
reports of the citrus white fly in this section of the State are supported
by the fact that the earliest collected specimens of this species in the
collection of the Bureau of Entomology bear the d(late 1888 and the
locality label "St. Nicholas,'" a point located in Duval County about
15 miles north of Fruit Cove.
Interesting information concerning the early history of white-fly
infestations in Florida has been obtained from Messrs. Borland and
Kells, citrus growers at Buckingham, Lee County, Fla., formerly of
Citra, Marion County. According to these gentlemen, the presence
at or near Panasoffkee, in Sumter County, Fla., of a small white
insect which caused blackening of the foliage of orange trees became
known among orange growers around Citra, at that time in the
heart of the orange-growing district of Florida, in 1881 or 1882.
The grove of Bishop Young, of Panasoffkee, was one of the first
reported infested. It is believed that Bishop Young, after traveling
in Asia (Palestine?), brought back with him plants which he set out,
and in a year or two thereafter blackening of the foliage of near-by


citrus trees in association with a new insect pest first became notice-
able. The white fly affecting citrus trees at Paxasoffkee was exter- i
minated by the freeze of 1894-1895 and, so far as the authors can
learn, has not reappeared. There seems to be at present no means of
determining whether the report given above refers to the citrus I
white fly or to the cloudy-winged white fly. j
Mr. A. J. Pettigrew, of Manatee, Fla., a reliable observer who has :
been in the citrus nursery and orange-growing business in Manatee
County since 1884 and who has been familiar with the white fly since
its first discovery in that country, has furnished the authors with a
statement concerning the early history of the pest in that section of
Florida. According to Mr. Pettigrew, Messrs. C.- H. Foster and F. N.
Horton each received from Washington, D. C., 6 tangerine trees in
1886 or 1887-as near as can be determined at this time, although
possibly earlier by a year or two. A year or two after the trees were
received and planted, the fly was noted by Mr. Pettigrew as abundant |
on a rough lemon near one of these tangerines, and the following year
it was first noted as abundant in a seedling orange grove near by.
At Mr. Pettigrew's suggestion specimens were sent to the Department l
of Agriculture at Washington and identified as a white fly. These
specimens were probably sent to Washington in 1891, for a letter
from Mr. Foster, dated January 8 of that year, was published in
Insect Life1 with the reply. The oldest specimens of the citrus
white fly now in the collection of this bureau, which were collected
in Manatee County, Fla., bear the d(late of March 5, 1891, with "Man-
tee" as the locality record. These were probably sent in by Mr.
Foster in connection with later correspondence than that referred
to above. |
Concerning the history of the citrus white fly in Louisiana, Prof.
H. A. Morgan in 1893 made the following statement:
This pest, common from Baton Rouge to the Gulf, is known as the white fly. Orange
growers claim that it has been recently introduced-that is, within the last ten years- -
and it is supposed to have come in upon plants brought to the New Orleans exposition
in the year 1885. The present wide distribution of the white fly in the southeastern
United States is due to the lack of restrictions, until very recently, against shipments i|
of infested nursery stock and of privets and the Cape jessamine.

The citrus white fly was first given a valid scientific name and
adequately described by Riley and Howard in an article published
in Insect Life2 in April, 1893. Following the account of the early j|
history heretofore quoted, these authors describe the different stages i
of the insect in detail, give an account of the habits and life history,
and give records with discussion of results obtained by a correspondent |
2 )..ii 2,,,9 3
Insect Life, vol. 4, p. 274. 1d., vol. 5, no. 4, pp. 219-226, 1893. i



in Manatee ('omnty, 1ia., whoi lilt([ unidraikei ii someI iil coioperulive
experinieiits iii s)rayinIg.
During te samlilie Near (1M1"S )' Pl'rof. II. A. .liorgaill, till(i e(n1toiiiolol4gist
of thlie Louisiania Agriculturlal Experinenti, Sttioncl, g'ave n il count of
the citrus white liv in ,llouisianail in ai llulhlt iii of tflilt stationn'
The 1)ivision of Vegetable Phiysiology an111( Platioigy of tl( I'lnit (
States Departmnient of Agriculture b)egln investa,.igati(l(is (of citrrus
diseases in Florida in 1893. These included invlestligaio tPIs (of tlie
"sooty mniold resulting from whlite-fly infestation, 1111(l tile first report
on the subject was published by Swingle and Webber in 18962 1111d t
more extended report by Dr. II. J. Webber in 1897.3 ConclusioIns
from a series of spraying experiments are included in this )uLbliCta-
tion and manyimniportant observations are recorded, particularly in
connection with thle two most useful fungous enemies of tlhe white fly
which were discovered by Dr. Webber in the course of Mis work.
Prof. H. A. Gossard, then entomologist of the Florida Agricultural
Experiment Station, published, in 1903,4 an account of the white fly
situation up to that time, witlih his conclusions from observations
extending over several years.
In a volume entitled "Citrus Fruits," published in 1904 by Prof.
H. H. Hume, four chapters are devoted to citrus insect pests and
methods of control, the whliite fly receiving due attention.
Since the present investigations by the Bureau of Entomology hliave
been in progress, Dr. Berger, entomologist of the Florida Experiment
Station, has published two bulletins5 which present a summary of
white-fly conditions with recommendations for control, particularly
with reference to the use of fungous enemies. In the later pub-
lished of the two mentioned, thlie specific distinctions are pointed out
and illustrated, separating from tlie common A. ciri thle form which
Dr. Berger has named A. nubifera.
Messrs. P. H. Rolfs and H. S. Fawcett, in a bulletin issued in July,
1908,6 discuss in a general way the use of fungous parasites of thle white
fly in Florida and give recommendations for the introduction of the
three most common species. The most important contribution to our
knowledge of the fungous parasites of thle citrus white fly is contained
in a paper by Prof. LI. S. Fawcett, published in 1909.7
The Orange and Other Citrus Fruits. By W. C. Stubbs and H. A. Morgan. Spec.
Bul. La. Agr. Exp. Sta., pp. 71-73, 1893.
2 The Principal Diseases of Citrus Fruits in Florida. By W. T. Swingle and H. J.
Webber. Bul. 8, Division of Vegetable Physiology and Pathology, pp. 25-28, 1896.
3 The Sooty Mold of the Orange and its Treatment. Bul. 13, Division of Vegetable
Physiology and Pathology, U. S. Department of Agriculture, 1897.
4 White Fly. Bul. 67, Fla. Agr. Exp. Sta., June, 1903.
5 White Fly Conditions in 1906, the Use of Fungi. Bul. 88, Fla. Agr. Exp. Sta., Jan-
uary, 1907; White Fly Studies in 1908, Bul. 97, Fla. Agr. Exp. Sta., February, 1909.
6Bul. 94, Fla. Agr. Exp. Sta., July, 1908.
7 Special Studies No. 1, University of State of Florida, 1909.



In Louisiana the demand for information concerning the citrus
white fly has resulted in a publication on this subject by Mr. A. H.
Rosenfeld in 1907.1 The discovery of the white fly in California in
the same year led to the publication, by Prof. C. W. Woodworth, of a
circular of general information,2 and of a second circular' dealing
with the methods of eradication that were being employed in that
State. A very complete account of the white-fly infestation in Cali-
fornia was given by Mr. C. L. Marlatt, assistant entomologist of the
Bureau of Entomology, before the Entomological Society of Wash-
The foregoing paragraphs refer to the principal publications in
which the citrus white fly is treated, exclusive of short papers in hor-
ticultural periodicals, press bulletins, experiment station reports, and
transactions of the Florida State Horticultural Society. Numerous
press bulletins have been issued by the Florida State Experiment Sta-
tion dealing within several phases of white-fly control and written from
time to time as the occasion demanded by Prof. Gossard, Dr. Sellards,
Dr. Berger, and Prof. Fawcett.
Reviews of the white-fly situation for the year, with notes on new
observations, have been included in their annual reports by each of
the first three named, who have served successively as entomologist
at thlie Florida Experiment Station. Many important papers and dis-
cussions on the white fly have been published in the Transactions of
the State Horticultural Society, but for the most part these have
been incorporated or the ground covered more fully in the regular
bulletins referred to.
Taken as a whole, the literature on the citrus white fly is quite
extensive, giving a fairly good idea of the status of the white fly and
progress in methods of control from year to year since the publi-
cation of the paper by Riley and Howard referred to in the opening
The description of the different stages and the account of the life
history and habits of the citrus white fly by Riley and Howard have
been followed quite closely by subsequent writers, few additional
records having been made up to the beginning of the present investi-
gations. Records of food plants, miscellaneous life-history records,
general results of field experiments, and conclusions from general
observations on the efficiency of spraying, fumigating, and natural
control by fungous diseases have been published by Messrs. H. J.
Webber, H. A. Gossard, E. H. Sellards, E. W. Berger, and H. S.
Fawcett. Comparatively little real data has been published so far in
Circular 18, State Crop Pest Commission of Louisiana, 1907.
2 Circular 30, California Agricultural Experiment Station, 1907.
3 Circular 32, California Agricultural Experiment Station, 1907.
4 Proceedings of the Entomological Society of Washington, vol. 9, pp. 121-123, 1908.



connection with experimental work with the white fly. A review ,f
all the literature to date shows that data have b)e on publlisMhedl on tie
effect of kerosene emulsion on wiite-fly eggs, lby Riley and Howard;
on the subject of effects of cold upon white-fly larv a and pupr', by
Prof. Gossard; on the percentage of trees infected by the spolre-
spraying method of introducing the fungous parasites, and on the
amount of honeydew secreted by the larvae of the insect, by Dr. E. W.
Berger; upon subjects related to fumigation,' by the senior author of
the present bulletin; and on laboratory experiments with the fungous
parasites, by Prof. H. S. Fawcett. Aside from the above, practically
no data have been heretofore published.


The direct injury by the citrus white fly may be included under
two main heads: (1) Injury by removal of sap from foliage, and (2)
injury from fungous growth known as sooty mold (Meliola), which
develops upon foliage and fruit on the excretions of the insects.
The direct injury is principally included as loss in value of trees,
extra expenses of maintenance, and losses from scale insects and dis-
eases, which more seriously affect white-fly infested trees.

The amount of sap extracted by the insects is not generally con-
sidered an item of great importance compared with the injury from
sooty mold. While the extraction of sap by itself probably would
not cause sufficient injury to make the white fly rank as an important
citrus pest, it is doubtless of considerable importance when combined
with the lowered assimilative powers of the foliage due to the sooty
mold. As mentioned more in detail under the subject of feeding
habits, it has been estimated that the loss of sap per day amounts to
about one-half of a pound for 1,000,000 larvae and pupae.


Sooty mold is the principal evidence of white-fly injury, and is the
most important element of damage, affecting both the foliage and fruit.
(See P1. I, frontispiece.) No special attention has been given by the
authors to its botanical aspects, but the following notes concerning it
are taken mainly from Dr. H. J. Webber's report on this subject:2
1 Fumigation for the Citrus White Fly as adapted to Florida Conditions. Bulletin
76, Bureau of Entomology, U. S. Department of Agriculture, Oct. 31, 1908.
2Bulletin 13, Division of Vegetable Physiology and Pathology, U. S. Department
of Agriculture, pp. 5-11, 1897.

86850-Bull. 92-11- 2


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . .. .. .i .... [ ~ i' ~i !i ,! .."..- ;
.. ... .: : :. .,. '. .:, !!!: i~ i ,, i .i:, i: .. ,,.": .. .... .. .'.
:: .. .:: ": ::m :. :. .. .......:...:.:;:. ":::i.


The sooty-mold fungus is a species of the genus Meliola of the
order Pyrenomycetes. Dr. Webber states that in Florida and Loui-
p *i In I

siana it is quite generally known as smut or black smut, but as the
fungus concerned is not a smut fungus these terms are erroneous,
and their use should be discontinued. When abundant on leaves and
fruit of citrus, this fungus forms a dark-brown or black membranous
coating composed of densely interwoven branched mycelial filaments.
At first this coating covers only limited spots or is not thick enough
to form a distinct membrane, but later, if the honeydew-secreting
insects are abundant, the coating becomes thick enough to be entirely
removed from the leaf and torn like paper. (P1. III, figs. 1, 2.)
Frequently the fungus membrane becomes detached at some point
and is caught by the wind and large fragments torn off. These
fungus fragments are found scattered about in badly infested groves
in the fall, being especially noticeable during the winter after a high
wind or after the trees have been sprayed.
Dr. Webber recognized several forms of reproductive agents, which
are easily distributed by various means, but principally by winds.
The fungus is entirely saprophytic in so far as known, deriving its
nourishment from the honeydew secreted by certain insects. As such
honeydew falls mostly on the upper surface of the leaves and on the
upper half or stem end of the fruit, the sooty mold develops most
densely in these places, but it is usually present to a greater or less
extent on the lower surface of the leaves, sometimes developing in
tufts on drops of honeydew which diseased insects fail to expel in a
normal manner. Sooty mold also develops on the twigs and in
some cases on the sides of buildings when heavily infested trees are
growing near by.
Seasonal history of sooty mold.-The sooty mold resulting from the
attacks of the citrus white fly is most abundant late in the season.
Very little sooty mold develops during the winter months, while the
films of blackish mycelium gradually become removed from the leaves
by winds and rains and much is knocked off in picking the fruit, in
spraying, pruning, fumigating, etc. The thicker the coating of sooty
mold, the more readily and thoroughly it is removed. By the time
of the appearance of the new spring growth the greater part of the
sooty mold on the old leaves has disappeared and from this time to
the 1st of May there is very little, if any, evidence of a new growth
of this fungus. Slight blackening of spring growth has been noted
as far north as Island Grove in Alachua County, Fla., as early as
May 20, the average number of live larvae and pupae per leaf being
estimated as about 50, not including old leaves which were practically
uninfested. By June 20, leaves from McIntosh, in the same county,
Generally referred to M. canmellia (Catt.) Sacc., but perhaps including more than
one species.


Bul Q.. Bur4iu of EIOIT'I ohigy.y U. 5 D.-pI or Ag ''u;'.'-'j

Fig. 1.-Sooty mold on orange leaf following white-fly attack: broken and falling frim leaf.
Fig. 2.-Sooty mold on cinnamon tree following attacks by cinnamon scale. (Original.)





with an average of about 11 live larva, and pupa) casesI per leaf, were
slightly blackened. In general, heavy coats of sooty m(old on leaves
are common in Florida by the 1st of June in groves heavily infested
by the citrus white fly.
effect of sooty miold on leaf functions.-Dr. Webber lhas (liscussed
the effect of sooty mold on leaf functions in the report already referred
to, and as there is nothing to add at this time, the following paragraph
(pp. 10-11) is quoted:
When it is remembered that various investigations have shown that the process of
phytosyntax 2is almost entirely checked in a plant placed in the back part of a living
room, opposite a window, where the light is fairly bright, but diffused, it can readily
be judged that the effect of the dark, compact mycelial membrane of the sooty mold
covering the leaves would be to almost wholly check the process of phytosyntax in the
orange tree. Quite bright or direct sunlight is necessary for the best results. The
injurious effects of sooty mold on the phytosyntax was clearly demonstrated by Busgen.
He removed the fungus membrane from a small portion of a leaf and exposed the leaf
to the sun. In the evening, after a sunny day, the leaf was plucked and the chlorophyll
extracted with alcohol. After this leaf was treated with iodine, the parts from which
the membrane had been removed in every case stained a dense blue, indicating the
formation of an abundance of starch, while the surrounding portions of the leaf, which
were protected from the sun by the fungus membrane, remained entirely uncolored,
showing that no starch was formed. The stomata, or breathing pores, are also to some
extent closed by the sooty mold, and in this way the passage of gas is more or less
hindered. In the orange leaf, however, the stomata are confined to the lower surface,
where generally there is but little sooty mold. In plants where the stomata are on the
upper surface of the leaf also, the damage resulting from the obstruction of the passage
of gases would probably be considerably greater.


In the following discussion the statements concerning injury and
the estimates of the extent of this injury by the citrus white fly refer
to groves in which the fly has become well established and in which
no remedial measures have been practiced.


Unless otherwise stated, oranges and tangerines are referred to.
These constitute more than 88 per cent of the citrus fruit crop of
Florida. The total injury to grapefruit by the citrus white fly is
rarely over 15 per cent and is frequently inappreciable.
Ripening retarded.-Ripening of fruit on heavily infested citrus trees
is greatly retarded, and in case of the formation of a very heavy coat-
ing of sooty mold on the upper half of the orange the rind underneath
it may remain green indefinitely while the lower half of the fruit is

'Some of the first generation had matured, but are properly included with the
insects responsible for the sooty mold present.
2 "Phytosyntax refers to the process of the formation of complex carbon compounds
out of simple ones under the influence of light; "photosynthesis" is a more common
term for this process of assimilation.



well colored. The retardation of ripening, delaying as it does in some
cases the time when the fruit is marketable and materially increasing
the percentage of culls, causes injury which is very conservatively
estimated to range from 2 to 5 per cent of the value of the crop. The
injury to grapefruit in retardation of ripening by the citrus white fly
is much less, varying from none at all to 2 or 3 per cent.
Number and size.-The greatest injury by the white fly is in the
reduction of the salable crop of fruit. Dr. Webber on this point
makes the following statement :1
The effect of the sooty mold on the orange is very noticeable, the growth being
usually greatly retarded and the blooming and fruiting light. In serious cases growth
is frequently entirely checked, and blooming and fruiting wholly suppressed until
relief is obtained.
Prof. Gossard has estimated 2 that during a six-year period the
reduction in yield due to the citrus white fly is from 25 to 40 per cent.
Replies to a circular letter of inquiry addressed to orange growers
and the observation of the authors in Florida indicate that the reduc-
tion in yield due to the citrus white fly amounts to 50 per cent, on
the average, when no artificial methods of control are practiced.
From information received from many growers and from personal
observation, the authors would estimate that with continued good care
and with the additional fertilizer usually given infested trees the
reduction in yield in different groves in a series of years amounts to
an average between 20 and 50 per cent.
The decrease in yield due to white-fly infestation ordinarily consists
of a decrease in the actual number of fruit produced and also in the
packing size. From information obtained it seems a conservative
estimate to consider that oranges and tangerines are reduced either
one or two packing sizes as a result of white-fly attack. For each
packing size, the number of reduced fruit remaining the same, the
reduction in the crop would average about 12.5 per cent.
Expense of cleaning.-Fruit noticeably affected with sooty mold
requires cleaning before marketing. One of the most economical
machines for washing fruit used in Florida is a California washer used
by Mr. F. D. Waite, of Palmetto, and Mr. F. L. Wills, of Sutherland.
The cost of washing with these machines ranges from 1.4 to 2.5 cents
per box. The cost of cleaning with the simplest machines is about
5 cents per box. Mr. E. H. Walker, of Orlando, Fla., estimates the
cost of hand cleaning oranges at 10 cents per box as a minimum and
7 cents a box for cleaning grapefruit. In consideration of the fore-
going it is estimated that the range in cost of cleaning the sooty
mold from fruit to be from 1 to 10 per cent of the value of the crop.
Shipping and keeping quality.-The sooty mold produced by the
white fly and other citrus pests does not, so far as known, affect the

2 Bul. 67, Fla. Agr. Exp. Sta., p. 617.


I Loc. cit., p. 9.


shipping quality of the fruit directly, but tlihe processes of cleaning
have been proved to be of considerable importance in this respect.
The subject of the deterioration in slippiIlg lualtity oIf citrus fruits
has been thoroughly investigated in California by agents of the
Bureau of Plant Industry under the direction of Mr. I. II. Powell.'
Their report shows in a conclusive manner that thle anmouint of ldectay
in shipment is very materially increased by brushing or washing the
fruit to remove the sooty nmhl. Table II, arranged from data pulb-
lished in the report referred to, shows tlie effect of dry Ibrushling and
washing fruit on the percentage of decay.
TABLE II.-Effect, on decay, of cleaning sooty mold from fruit.

It will be observed that dry brushing increased the amount of
decay to about two and one-half times the decay in the unbrushed in
record No. 1, and to about two and one-fifth times in record No. 2.
Washing increased the amount of decay to about six and two-thirds
tines in record No. 1, and to about five and one-fifth times in record
No. 2.
The injury from cleaning the fruit is due to the increased opportu-
nities for infection with spores of the blue mold and to mechanical
injuries in the process of cleaning. The chances of decay are still
further increased whenever the fruit is not thoroughly dried before
packing. Washing in constantly running water or by running the
fruits through brushes with water constantly sprayed over them is
considered much less objectionable than the ordinary systems.
Flavor.-The attack of the white fly is generally supposed to affect
the quality of the fruit in a marked degree. Dr. Webber and Prof.
Gossard describe the flavor as insipid as a result of heavy infestations.
The latter presents the results of chemical analyses of samples of the
fruit of tangerine trees in two adjoining groves. In one grove the
white fly was completely controlled by spraying; in the other the fly
was unchecked. The analyses showed that there was, in the samples
from the latter grove, 15 per cent less reducing sugar, 15 per cent
less sugar dextrose, and 5 per cent less citric acid. While oranges
and tangerines are frequently much affected in flavor, thoroughly
blackened groves in many cases produce as well flavored fruit as can
Bulletin 123, Bureau of Plant Industry, U. S. Department of Agriculture.

Unbrushed Dry Washed
fruit a brushes fruit ap-
Record fruit ap-
o parents parent ly parental
No sound. pound sound.

Per cent. Per cent. Per cent.
1..... 2.7 6.6 17.8
2.... 1.9 4.2 10.0
3--.....--- ...........------------. 1.8 2.6



be found in the market. When trees are supplied with as much
fertilizer as they can use to advantage the white fly does not ordi-
narily affect the flavor of the fruit to such a noticeable extent as is
commonly believed. It is suspected that a well-grounded prejudice
against the white fly rather than a discriminating taste is responsible
for a large part of the supposed effect on the flavor of the fruit in
infested groyes.
Increased injury from scale insects and from plant diseases.-The
number of culls is in some cases very much increased by diseases and
insect pests which thrive after the trees have been weakened by the
white fly. There are no data available showing the usual increase in
percentage of fruit injured by scales and by diseases of the trees as
a result of white-fly infestation, but this is a consequence observed
by many citrus growers and is properly considered a factor of white-fly
injury. As such it is conservatively estimated to vary from 1 to 5
per cent in groves thoroughly infested, although an instance of a
valuable crop being completely ruined by secondary scale attack has
come under the authors' observation.
Market value.-Imperfections in fruit rind due to diseases and
insect pests as followers of the white fly and to failure of fruit rind to
color up normally, in addition to the direct effect on the size of.the
marketable crop as heretofore discussed, usually lower the average
grade even after the fruit is cleaned by the most approved methods.
A few growers claim that after being cleaned their oranges and tanger-
ines bring as good prices as any, and leaving out of consideration
instances where it is claimed that most or all of the fruit is rendered
absolutely unsalable under any conditions, we may conservatively
estimate the depreciation in market value to range from none at all
to 10 per cent.
Sooty-mold-blackened oranges shipped without cleaning have a
market value ordinarily from 25 to 50 cents less per box than the
same fruit would have cleaned.1 Certain Florida brands of oranges
well advertised, carefully graded, and packed,would fail to bring within
a dollar a box of their average value if they appeared on the market
blackened by sooty mold.
Losses to growers estimated on basis of prices paid by orange buyers.-
The authors are indebted to Mr. E. H. Walker, of Orlando, for the
information that during the season of 1907-8 orange buyers in Florida
paid from $0.75 to $1.45 for oranges free from white-fly effects, and
from $0.50 to $1 per box for fruit blackened by white fly; during
the season of 1908-9 the price paid for clean fruit varied from $0.60
to $1 per box, and from $0.50 to $0.75 for fruit blackened by sooty
mold. The loss to the growers is not entirely represented by these
1 Statement based on information from Mr. E. H. Walker, Orlando, Fla.




figures, since, according to Mr. Walker, the best prices were not piid
for sooty-mold-blackened fruit until I itilInh intl1 season after thlie clean
fruit had nearly all been sliIppedl or dlisiosem i of. ('lelilan fruit It tis
time would have been proportionally inore proitalble.


Weakeningq j ritality.-It is doubtful if tile white fly is ever tlhe
direct cause of the killing of trees, limbs, or twigs inll well-fertilized
groves. It does, however, seriously stunt tile gro.wtIh of all heavily
infested trees, and may temporarily entirely check the growth of
young trees. Its greatest effect on the vitality of tlhe tree is an indli-
rect one. Infestation by the white fly appears to weaken tihe resist-
ance of orange and tangerine trees to foot rot, (lie back, melanose,
wither tip, and drought, and favors the multiplication of tlhe purple
and long scales which are second to the citrus white fly as citrus lpests
in the Gulf coast regions.
Depreciation in value.-The selling values of citrus groves are greatly
reduced by white-fly infestation, and citrus nurseries have their ter-
ritory for sales much restricted and values reduced. Concerning the
reduction in value of groves of bearing trees one of tlhe most experi-
enced dealers in orange groves in the State estimates that it is in
general about one-third. For years California has been closed to
Florida nurserymen as a field for the sale of citrus nursery trees, and
a similar quarantine regulation has recently gone into effect in Ari-
zona. In Florida and in citrus-growing sections of other Gulf coast
States a guaranty of freedom from white fly is generally required,
especially when the purchaser contemplates planting a more or less
isolated grove.

The estimates in the foregoing pages refer to ordinary losses where
the white fly is unchecked by natural enemies or by artificial methods
of control and not to exceptional or occasional losses. These esti-
mates, as applying to the fruit, are summarized in Table III.

TABLE III.-Estimates of losses to orange crops by while fly in uncontrolled condition.
'Maximum. Minimum. Mean.
SPir cent. Per cnt. Per cent.
Ripening retarded .................................................... 5 2 3
Number and size of fruits............................................. 5 20 35
Cost of cleaning ....................................................... 10 1 5.
Deterioration in shipping quality..................................... 6i 2 4
Indirect injury: Increased scale and disease effects on fruit............ 5 1 3
Loss in market value................................................. 10 0 5
Total-............................................................- 26 56


The mean of the total percentage of estimated loss is considered by
the authors to represent about the normal loss which the citrus white
fly is capable of causing in orange groves. It is estimated that the
condition is reduced to about 45 per cent loss in the average infested
grove as a result of net profits from spraying with contact insecticides
and of the natural efficiency of fungous diseases.
From extensive records obtained in the course of their investiga-
tions the writers estimate that the citrus white fly infests at present
45 per cent of the citrus groves in Florida. Of this, 5 per cent is a
sufficient allowance to represent the groves so recently infested that
normal abundance of the pest has not been reached. An injury of
45 per cent in 40 per cent of the groves is equal to about 18 percent
of the entire value of the crop as it presumably would have been if
the white fly were not present.
The latest Florida citrus crop concerning which statistics are avail-
able is that of 1907-8.1 The orange crop for that season is valued
at $3,835,000. With an estimated total loss of about 15 per cent
this represents 85 per cent of the value of the crop if not affected by
the white fly. Accordingly, the estimated loss in Florida is calcu-
lated to have been about $680,000 for oranges and similarly on the
basis of 10 per cent loss to grapefruit on a valuation of $469,700, the
percentage of infested groves the same as in the case of the orange
groves, a loss of $16,700 is estimated, making the total loss in valua-
tion of fruit about $696,700 for the crop of 1907-8. The crop of
1908-9 was doubtless affected to the extent of $750,000 by the citrus
white fly.
At present the spread of the fly into uninfested groves is undoubt-
edly faster than at the rate of 5 per cent new infestations per year.
Even on this basis, however, the annual increase in depreciation in the
value of Florida citrus groves due to white-fly infestation is more than
$200,000 per year.2 In addition, the citrus nursery business in Florida
is affected to an extent hard to estimate, but which would be only
nominally represented by $50,000 per year.
Figures are not available which would allow approximate estimates
to be made of the damage by the citrus white fly in the Gulf coast
citrus-growing sections outside of Florida, but the widespread occur-
rence of the white fly in those States indicates that the losses are

The items of expense of maintenance principally affected by the
white fly are fertilization, spraying, and fumigation. In Florida
1 Tenth Biennial Report of the Commissioner of Agriculture of the State of Florida.
2This is not shown by actual depreciation, for the number of groves coming into
bearing for the first time each year more than covers the loss.

I.' !


ordinarily the expense of tile fertilizer necessary to maintain orange
trees in good productive condit ion varies froint 10 to 20 cents lper box
of fruit prodlucedl. The wide range given is largely d, t(i io clilFerences
in soil conditions. Mr. 0(). Painter, in response) to an inquliVy ,on
the subject, writes that citrus trees infested with white fly in his
opinion, require at least 15 per cent increase in fertilizer for best
results under the circumstances. On tihe basis of cost of fertilizer
amounting to 10 to 20 cents per box and an increase of 15 per cent
due to white fly infestation, the extra expense wNhich ia2Y I1 be charged
as white fly injury amounts to 1.5 to 3 cents per bo)x.
Cost of control measures properly chargeable to increased cost of n1 in,
tenance.-Estimates based on the exirerience of the writers in fumi-
gating and in spraying for the white fly give the range in expense of
the former method of control as 5.5 to 14 cents per box of oranges
produced, and of the latter method 12.5 to 20 cents per box. These
estimates refer to thorough control, with the result that production is
entirely unaffected by the white fly. The minimum estimate on the
expense of fumigation refers to groves so located that the migrations
of adults from outside groves does not make treatment necessary more
than once in two years. The maximum estimate refers to conditions
where treatment every year is required to prevent loss. Increase in
production, due to destruction of scale-insect pests, is not taken into
consideration. In the estimates of expense of control by spraying
the minimum estimate refers to cases where three applications of in-
secticide per year have resulted in satisfactory control. This result
can be attained only after the insect has been brought into complete
subjection, such as referred to in the introductory paragraph of the
subject of artificial control. Insecticides costing more than 1- cents
per gallon when mixed ready for application have not been taken
into consideration.

As has been shown in the historical review, the citrus white-fly
at present is generally distributed in North America. In the north-
ern part of the United States it occurs in greenhouses, and in the
southern part, and in limited districts in California, it occurs on
citrus, China trees, privet, cape jessamines, and other food plants.
In the present publication we are concerned only with the distribu-
tion of the species in the citrus fruit-growing regions of the United
According to the statistics of the Florida commissioner of agricul-
ture, in 1905 there were 17 counties in the State reporting more
than 5,000 bearing citrus fruit trees. In all but two of these, Dade
and St. Lucie, the citrus white fly (Aleyrodes citri) occurs to a greater



or less extent. (See fig. 1.) The 17 counties referred to, arranged in
order of the number of bearing citrus trees, is as follows: Orange, Lake,
Volusia, Polk, Putnam, Brevard, Hillsboro, De Soto, Lee, Manatee,
Dade, Marion, St. Lucie, Osceola, Sumter, St. John, and Alachua.
Palm Beach as well as Dade and Monroe Counties are infested with the
cloudy-winged white fly, as hereafter noted, but so far as known the
citrus white fly does not occur there. In order of the percentage of
groves infested the foregoing counties which are known to be infested
would be arranged about in the following order, so far as our obser-
vations and records show: Marion, Alachua, St. John, Manatee,
Orange, Lee, Volusia, Polk, Putnam, Lake, Hillsboro, Sumter, De
Soto, Osceola, and Brevard. If the groves infested by the cloudy-
winged white-fly only were also taken into consideration, Hillsboro
and Lake Counties
_would be transposed
in the list, as would
Osceola and Brevard,
but aside from this
there would be no
change. The arrange-
Bment is only approxi-
mate, being based on
^ observations made
by the various men
connected with the
white-fly investiga-
-tions upon informa-
tion and samples of
infested leaves re-
FIG. 1.-Map showing distribution of the citrus white fly (Aleyrodes ceived from corre-
citri) in Florida. (Original.) spondents and upon
nearly 250 replies received in response to circular letters sent out in
the spring of 1907.
At the present time the writers estimate that throughout the State
of Florida about 40 per cent of the citrus groves are infested by the
citrus white fly, and that an additional 5 (or 10) per cent are infested
by the cloudy-winged white fly alone.
The citrus white fly occurs in nearly all the larger towns in north-
ern Florida, infesting the various food plants which are grown as
ornamentals as well as the citrus fruit trees which are grown to a
limited extent. The insect is of common occurrence, principally on
China trees, cape jessamines, and on privet and hedges of Citrus
trifoliata in South Carolina and in southern Georgia, Alabama, Mis-
sissippi, Louisiana, and Texas. In the last two States citrus fruits
are being grown quite extensively, and a large percentage of the
citrus-growing localities are infested.


Aside from the Gulf coast States, citrus frlits in tlV I'enitd Stat.s
are grown only in ('alifornia and Arizona. ThI' tcitirts wv!,itv fly dt ts
not occur in Arizona. Iln ('aClifornia the pst was first. dis,,cov'redl in
May, 1907. Mr. C. L. Marlatt, ltas given t .fn allowing acc(Junt of
the distribution of the white fly in that State in 1907:1
Marysville is situated a few miles north of Sacnrainmento, and th e first infesftation
seemed limited to this town, but toward the end of Ilh sinmmier Ihi' whiii, fly was
discovered well established at, Oroville, in ]utte C'ountv, some 26 miles to Ihe north
of Marysville. The Marysville infestation was confined to ie th town and it) yard treevs
or small garden orchards. Oroville lies in a considerable orange district, and the
white fly had been carried from the town into several of the adjacent orchards and
had become rather widely scattered. Shortly after the discovery of the fly at Marys-
ville it was found also to have established itself locally near Bakersfield,- in the soiuth-
era end of the San Joaquin Valley, and separated only by a mountain range from the
citrus district of southern California.

For years the citrus white fly has been supposed to be an intro-
duced species, and much interest has been attached to its occurrence
elsewhere than in North America. Prof. II. A. Gossard in 1903
stated that Mr. Alexander Craw, of the California State commission
of horticulture, had received this species on plants from Chile, where
it was reported to be a great pest. Mr. G. W. Kirkaldly, in his cata-
logue of the Aleyrodidae, in 1907, gives "Mexico, Brazil, and Chile
(?)" as the known habitats of the citrus white fly outside of the
United States. The writers are informed by Prof. A. L. Quaintance
that he was told in person by the late Prof. Rivera, of Santiago,
Chile, that the citrus white fly was abundant in that country. Prof.
Carlos Camacho, chief vegetable pathologist at Santiago, Chile, is
also, according to Prof. Quaintance, authority for the statement that
it occurs there.
The Bureau of Entomology received, in 1906, specimens of an
aleyrodid on orange leaves from China which Prof. Quaintance
determined as Aleyrodes citri,3 and still more recently it received,
through Mr. August Mayer, in charge of plant-introduction garden,
and through the California state commission of horticulture, speci-
mens of orange leaves infested with what Prof. Quaintance has
identified as this species from different parts o? China and Japan.
The occurrence of the citrus white fly in India (northwestern Hiima-
layas) has recently been established by Prof. Quaintance, who has
compared Maskell's A. aurantii, collected in the region mentioned

'Proceedings of the Entomological Society of Washington, vol. 9, pp. 121-122,
2 Specimens of the species present at Bakersfield were examined by the senior
author at the California State Insectary at Sacramento and found to be the cloudy-
winged white fly (A. nubifera).
3 Proceedings of the Entomological Society of Washington, vol. S, Nos. 3-4, p. 107.



above, with A. citri, and failing to find any differences in the egg and
pupal stages found it necessary to regard the name given by Maskell
as a synonym of that given by Riley and Howard.
The citrus white fly does not occur in Cuba, so far as known,
although it is not unlikely to be found there, since there have been
heavy shipments of nursery stock from infested citrus nurseries in
Florida to that country during the last few years.

The separation as distinct species of two forms formerly considered
as belonging to the species Aleyrodes citri makes it necessary that
all of the reported food plants of the citrus white fly be verified.
Nearly 60 species of the genus Aleyrodes have been recorded for
North America. Of these less than 20 have been described in the
first larval stage in a manner which distinguishes them, although
when carefully studied this stage has been found to have striking
specific characters. The second and third larval stages rarely possess
distinguishing characters. The fourth or pupal stage, or the empty
pupa case, is used as the basis of specific descriptions in the Aley-
rodidw, but even in this stage a careful microscopic examination is
usually necessary to positively determine the species. Good specific
distinctions in the adult stage have been found only in a few species,
and even those entomologists who have made a specialty of the Aley-
rodidt do not attempt to distinguish the different species in this
stage. It is obvious, therefore, that a list of food plants should
properly include only those verified by entomologists, with determin-
ations of the species made since the status of the two most abundant
citrus-infesting species of Aleyrodes has been fully recognized. Dr.
E. W. Berger has recently arranged the full list of food plants and
reported food plants in a graphic manner, separating the list into
two classes according to the degree of preference, and each class is
subdivided into native and introduced species. This method of
grouping the food plants is here adopted (see Table IV) with the
transposition of the lilac and coffee from class II to class I and omit-
ting certain reported food plants in order to restrict the list to include
only positive records, leaving the others for a separate discussion.
Dr. Berger has recently discovered the citrus white fly on wild olive,
and has also verified Prof. Gossard's report of the citrus white fly on
Viburnum nudum. Both of these food plants, together with the
green ash, will eventually be found to be subject to heavy infestation
and be placed in class I.


TA uLE IV.--- DejIfti ith known j'nild fluntlI ,f Mh, rriu infh /it..lf (. 1 bl/ruIs ,iriu/ .

1. CilruiM (till species (cult i vat d iii Ainirici).
2. (China tree ( lMrlia alzrdhir'h).
3. Umiubrella C('hina tree ( Mlia aze'dcaraih uimiln-irulfij''rra).
4. ('ape j(ssamilie ((Ghrdein i J'SII in oid',s).
5. Privets (Ligus'trum spp.).
6. Japan persimio)n (Diospyros kaki).
7. Lilac (Syringa sp.).
8. Coffee (Coffca arabica).
9. Prickly ash (Xanhoirylumn clavn-herrulis).
10. Wild persimmon (Diospyros virgin iunai').

11. Allamanda (Allamanda ncriifoliah.
12. Cultivated pear (Pyrus spp.).
13. Banana shrub (Magnolia fuscalum n).
14. Pomegranate (Punica granatumin).
15. Smilax (Smilax sp.).
16. Cherry laurel (Prunus laurocerasus).
17. Wild olive or devilwood (Osmnanthus americanus).
18. Viburnum ( Viburnum nudum).
19. Green ash (Fraxinus lanceolata).
In addition to those in the foregoing list 1 there are several plants
reported as food plants of the citrus white fly which, while probably
true food plants, can not consistently be included in the recognized
list until the observations have been repeated and the infesting spe-
cies positively identified. In some instances where eggs or larvae
have been found there is doubt as to whether the white fly could
- develop to maturity on the plants in question. Plants upon which
the insect is unable to develop to maturity can not properly be con-
si(lered true food plants. The following is the list of plants reported
as food plants, but which in each case require further observations
either as regards the ability of the insect to reach maturity thereon
or as regards the species of white fly concerned, in view of the recent
separation of A. citri and A. nubifera: Water oak, reported by Prof.
A. L. Quaintance; Ficus altissimna, Ficus sp. (from Costa Rica), and
scrub palmetto, reported by Prof. H. A. Gossard; honeysuckle and
blackberry, reported by Dr. E. H. Sellards; oleander, reported by

In addition to those already mentioned as being food plants in Florida, the follow-
ing plants are on record at the State insectary at Sacramento, Cal., as food plants of the
citrus white fly observed at Marysville and Oroville by agents of the State commission
of horticulture: English ivy (Jiedera helix), yellow jessamine (.Jasminum odoratissi-
mum), Ficus macrophiylla, bay (Laurus nobilis), tree of Heaven (A ilanthus glandulosa ),
and crape myrtle (Myrtus lagerstramia). Information concerning the authorities for
the plants listed is not available.


the senior author of the present publication; camellia, reported by
Dr. E. W. Berger. Iri the case of the last two plants mentioned
the uncertainty as to their proper standing is on the possibility of
the insect reaching maturity thereon and not on the identity of the
infesting species.
The present status of the plants which have heretofore been listed
by entomologists as food plants of the citrus white fly is shown in
ithe foregoing paragraphs. There are doubtless numerous additional
introduced species and a few additional native species of plants occur-
ring in the United States which serve or are capable of serving as
food plants of tlhe citrus white fly, but for the reasons connected with
the identification of the insects, stated in the opening paragraph under
the subject of food plants, reports of food plants other than those
included in classes I and II should never be credited unless verified
by or made by an entomologist. There are no important food plants
occurring in the Gulf coast region omitted from this list, and future
additions to the list probably will be of little significance economically
as affecting the control of the pest. There is a widespread belief that
many other conumnon trees, shrubs, and vines in Florida are food plants
of the citrus wliite fly, but the correctness or falsity of this belief
can be readily ascertained in the case of the individual plants sus-
pected by submitting specimens of the foliage and of the infesting
insect to the Bureau of Entomology or to theState experiment station.
There are three common causes for erroneous reports concerning
citrus white-fly food plants. The first is the presence of sooty mold
on many plants, due to other honeydew-secreting insects, such as
aphides, scale insects, and mealy bugs. The insects themselves are
not. seen in this case and the mistaken idea is due to ignorance of the
fact that other insects than the citrus white fly excrete honeydew on
which the same species of sooty mold fungus thrives. The second
cause for erroneous reports in this respect is the misidentification of
the insect concerned. The necessity for the identification of the in-
festing insect by an entomologist has been discussed. The third
cause is the frequent occurrence of the adult citrus white fly on the
foliage of plants upon whichli it does not breed and upon which it sel-
(dom or never deposits an egg. In the course of the present investi-
gation by the Bureau of Entomology several trees and shrubs have
been thoroughly tested as possible food plants by cage experiments,
and observations have been made on these and other plants, showing
that if it is possible for tlhe citrus white fly to develop on one of them,
it is, at the most, of too rare occurrence to be of any significance.
Cage tests have been made with oak (Quercus brevifolia), Magnolia
(Magnoliafrtida), blackberry (Rubus spp.), Laurel cherry or mock
olive (Prunus carolirtiana), and cultivated figs (Ficus carica) and crape
myrtle (Myrtus lagertrmenia). In each c-ase a rearing cage (Pl. VII)


was attached to li,,ti< ed (if at branli cov'r in,. nw gro th ai fro,
50 to 10( aliid lts oif .1. f ';tri w('rICh ('dlil ri tlir,'iil. lxc'.C lt ii lli' case
of tilhe blacklberry, il wnIi iii ,lo bshr'l, tionl 1 W Iwas 1 c1 I. f lic iniut
the adults wtre(' iiott'i a ts resting c,. i it'nii l llv IIIJl l,'ii pjtir I1'l fdi, oln
the ltea 'es ifor oilt' o()I two Vi) s a tiiltr t itrI h eg 'elirYii ii 'v ( 'I case,
hliowe'er, all t lt r ti t e adlts weerlli e ilIt'IdIi lv aite' conftiiz(lneiet
on thile i)JIits notedl, althoughlll check lots of adults co'llctcd ait tl,' .lsielie
time but .co il ic 0 ti o il ll cl it's ( 'itfUin t i 'l s Iiv'd(' fI in a nori iim i 'riC n.
No eggs were del)ositod iiii ally of t lie iteis, idllio lio tlie click uits
deposited eggs on the citrus leavti'es i1i a1i02ir l niiiienler.
Eachi of thle five plants tested witli tlit cage exJt'Piiiviits hliavc in
addition been subjected to very carful e xallminatio(11 h by tiet writers
under such circumstances thliat thle ol))ortuIlities for ilifestation l\by
the citrus white fly were at their best. In addition, particular attetn-
tion has been given to examinations of sl)ec(ies of oaks (Qu(ans spvl).)
and bays (Persia spl)p.), guavas (Psidiunl si).) and mulberries ( DMorus
spp.), when located near, and in some cases with branches intermin-
gling with infested citrus or other favorite food plants.

Entomologists familiar with the present white-fly situation agree
in their conclusion that a requisite for satisfactory control of tils
pest is proper attention to food plants othlier than citrus fruit trees.
Mr. H. G. Hubbard, whlo was a well-known authority on orange
insects, being a special agent of tile Bureau of Entomology, was a
strong advocate of destroying food plants of tlhe white fly that were
of no value. Dr. Sellards, formerly entomologist at tile Florida Ex-
periment Station, Dr. Berger, thle l)resent entomologist, Prof. P. II.
Rolfs, director of the Florid"a Expl)erimnent Station, and tlhe authors
have each emphasized the iml)ortance of thle relation of tlhe various
food plants to white-fly control.
The following paragraph from-the senior author's )bulletinl on tlhe
subject of fumigation for thle citrus wlhiite fly states in a general way
the situation in this respect as viewed by entomologists who have
investigated the whiite fly:
The presence of food plants of the white fly other than citrus trees, in citrus fruit-
growing sections, constitutes a serious menace and in itself ottn l)revents successful
results from remedial work. Fortunately the list of food plants is limited, and the
greater number of those thus far recorded is subject to infestation only when located
near or in the midst of heavily infested citrus groves. The food plants which are of
most importance in connection with the white-fly control are the chinaberry trees,
privets, and cape jessamine, and these-except for the last, in certain sections where
grown for commercial purposes-can be eradicated readily, or their infestation may
be prevented where community interests precede those of the individual in controlling
Bulletin 76, Bureau of Entomology, U. S. Department of Agriculture, pp. 9-10.

TIlE A 'IIl'TI l'.VS W 111TE .'1. '" : 1.l l) ( plantsNTS .


public sentiment. These food plants favor the rapid dissemination of the white fly
from centers of infestation and their successful establishment in uninfested localities.
They seriously interfere with the success of fumigation, as well as of all other remedial
measures, by furnishing a favored breeding place where the white fly can regain
its usual abundance in a much shorter time than would be the case if it were entirely
dependent upon citrus fruit trees for its food supply. The plants mentioned, together
with Citrus triJbliata (except where used in nurseries), and all abandoned and useless
citrus trees should be condemned as public nuisances and destroyed in all commum-
ties where citrus fruit growing is an important industry.
Not only is a knowledge of the relation of the various noncitrus
food plants to white-fly injury of great importance, but it is also of
considerable importance to growers to know the capability of the
insect for multiplying on the different citrus fruit trees in order that
advantage may be taken of it in the arrangement of new groves and
the improvement of old groves.


It is a matter of common observation that injury from the white
fly is most marked on citrus fruits of the Mandarin group. This
group includes the Tangerine, Satsuma, and King of Siam. The
sweet oranges are next to the mandarins in this respect, followed by
the kumquats and grapefruits.
The relatively less injury to grapefruit by the citrus white fly
(A. citri) is sometimes obscured by the presence of A. nubifera.
Blackening of foliage and fruit by the citrus white fly is more notice-
able on grapefruit trees when they are surrounded by or are otherwise
unfavorably located in respect to oranges or tangerines. Solid
blocks of grapefruit trees rarely show more than slight effects of
white-fly infestation when only the citrus white fly is present. An
example of this is the Manavista Grove at Manavista, Manatee
County, Fla. This grove consists of 22,000 grapefruit trees, and
appreciable blackening of the foliage is rarely seen except occasionally
where orange groves adjoin. Only one record, based on actual
examination of leaves, illustrating the difference in the degree of
infestation of adjoining blocks of grapefruit and orange trees is
available. Thle grapefruit block consisted of about 400 trees located
immediately north of a block of 200 or 300 orange trees and separated
on the west by a public road from a grove of about 800 orange trees.
On April 23, 1909, after practically all the overwintering pupa had
matured, an examination of 100 or more leaves collected at random
from each grove, counting the pupa cases, showed an average of 8
insects that had reached maturity on the grapefruit leaves, 27 on
the orange leaves of the block south, and 56 on the orange leaves of
the block west. No studies have been made to determine the differ-
ent degrees of susceptibility to white-fly injury among the different
varieties of grapefruit, but the Royal variety appears to be more




nearly immune tlhian any other of those o(iiiiiiilv g.rown. Thi.s \aS
first pointed tut by Mr. F'. ). Waite, of P'aligaett,, FIl. li tlhi.s co)l-
nection it shloul d be noted itlia the t1ioyal variety ill its g elieral
characteristics is not a tYvpical gralpefruit.
Tile reason for tlie t pIrtil ililitiniity' of gral)tefrliLit trees to W wite-llv
injury is us yet obscure. Several eriobervati)jons (ill graij)efrutit ajil
orange trees grovwinlg side b)y sidee give Imbasis for li sitipi ositloll
that it is a matter of food-plant prefeirecres of tl-i adluilt (1hliles. I
some cases itle dilfferences in thie ainouit of new growth i11.st lie
taken into consideration. ('ounts of adtilts, [ I)ii ca.u-t's, tid 11a iitched
eggs of tlhe citrus white fly on alternating grapefruit i!d( orange
trees, six in all, located on tlie laboratory grounds at ()rla2io, were
made on June 4, 1909, wlen no new growth% was pretselit on tlhe trees.
Thle leaves were selected at random andl, witi tlie exception( of a few
upon which adults were counted, they represented tile spring growth
of 1909. The difference between tlhe nunlber of thie adults oil 500
grapefruit and 500 orange leaves, 87 and 104, respectively, is not as
great as would be expected, considering thie nmuch greater number
of insects that liad matured on the orange up) to tlhe time of tlhe exam-
ination. There were about twenty times as many l)pupa cases on
the 100 orange leaves as on the 100 grapl)efruit leaves, or 6 and 120,
respectively. This was offset by tlhe presence of about three times
as many live pupme on 10 grapefruit as on 10 orange leaves, 41 and 14,
respectively, making the sum of thie pupa cases and live ) pupi) 4.16
per leaf in the case of tlhe grapefruit and 2.6 per leaf in tlhe case of tlhe
orange. This is about the same p)rol)ortion as tlhe number of hlatlched
eggs on the two food plants. The condition of tlhe leaves, as sliown
by this data, fails to indicate any cause for tlhe partial immunity of
grapefruit trees.
The examinations by Mr. W. W. Others of two leaves picked at
random from each tree in a small isolated grove consisting of 41
grapefruit and 28 tangerine trees gave rather striking figures, show-
ing more rapid multiplication of the citrus white fly on tlhe latter
than on the former. The first examination was made on November
4, 1908, and the second on June 8, 1909. On tlhe former date tlhe
average number of live and dead white-fly larve andl pu1)l),e l)per leaf
was 31.9 on the grapefruit and 96.2 on the tangerine, 16.6 and 80.9,
respectively, being alive. During the winter a series of fumigating
experiments reduced the numbers of the white fly so that at tlhe
second examination the number per leaf was 1.1 on tlhe grapefruit
and 2.25 on the tangerine. The arrangement of thie two kinds of
trees in the grove was such that they had equal chances of becoming
reinvested by the insects which escaped the effects of the experimental

86850-Bull. 92-11--3



The difference in the degree of injury between orange and tangerine
trees is less marked than between tangerine and grapefruit or orange
and grapefruit, but the difference is nevertheless usually quite notice-
able. The practical application of this difference in the degree of
adaptation of the citrus white fly to the various citrus food plants
will be discussed in a forthcoming bulletin dealing with the artificial
control of the white fly.
While China trees (PI. IV) and umbrella China trees (PI. V), when
grown for shade and ornamental purposes, are, as has been pointed
out, very injurious to citrus fruit-growing interests, the investigation
of the utility of these plants as trap foods gives an increased import-
ance to a definite knowledge concerning them as citrus white-fly food
plants. Their injuriousness to citrus growers is very clear to pro-
fessional entomologists, but not as generally appreciated by the citrus
growers themselves as is desirable.
The umbrella tree is recognized by botanists as a variety of the
China tree. This variety is the one most commonly grown except
in a few localities, and observations reported herein specifically refer
to it and not to the China tree. The latter tree has, however, been
under observation by the authors, and no noticeable difference has
been observed between the two trees in their relation to the citrus
white fly, and the data and observations are in the main fully as
applicable to the one as to the other.
The numbers of the white fly which mature on individual umbrella
trees have been estimated in three instances and found to range
between 25,000,000 and 50,000,000 where trees are favorably located
with respect to nondeciduous food plants. Examinations were made
by selecting 10 or more leaves at random and from each selecting
a leaflet which appeared to represent the average condition of all
the leaflets composing the leaf. In two instances it was found that
the infestation was fully as great toward the top of the tree as on the
lower parts. In one instance an extensive examination of different
parts of an infested umbrella tree showed a decrease from lower
branches to top branches of 50 per cent. In order to be fully con-
servative, this percentage has been used as the basis of the calcula-
tions, making the average infestation throughout the tree 75 per
cent of the infestation of the leaves of the lower branches. Full-
grown leaves were found to consist of about eighty-two leaflets.
Complete records were made of eggs and of live and dead larvae and
pupte, but only a part of this data will be presented. The estimates
and counts of both leaves and insects in the case of the first treewere
made by the senior author; in the cases of the second and third trees
the estimates of the number of leaves per tree represent the average

Bul Y r-' Bur.au .. I 'E .' *,r gy,. U 5 L'-i. of A .- ,,. PTrL.

Fig. 1.-China tree defoliated during winter. Fi-z. 2.--Same tree in full fi.liage in summer.



.q '







Fi 1.-L- allet (f umbrella (lii .i tree. -'l} \ in-' infestatioii by A. citri. Fig. 2.--Um brella 'liiii;i tree with ii:; -..e iir e in rvar ,,i hoHM4. {, ,i '.. ,



of three estimates one by t.e, sentio r aiutholr, 1one b th1w jullnior aiitlior,
and onle bN. Mr. W. W. Yotlers. 'iu'll,' counts al d estiiiia tes-of insects
were madIe )y thN e t senior a'utlhor ini tilte sec'rol id.nsta nlce atid b I Iy te
junior author in tlhe tlird. 'TllIe data. obtazi,,ed in tllLhseh examinllIat,1011s
bearing on the numiiber of insects the umbrella trees are capable of
maturing are given in Table V.
TABLE V.-Aumbtr of ritris irite' jlius deiv'lopiiiq on zunbril i '(iina tre's.
,~'', in ,
i'l Kli nl iliii~r |
Estiiaitd N. iil r Niut iil'r riiii'r i, idrv ofI ,
Dr.e .atl ofexamn- nu iher of of p Iupa' of live ,', f l i',1 '
e main tion. leaves Oln C:S.t's pup; iL' t .i., I io :I1M l ll" '; ,f
tree. per leaf. p' r leaf. tree. x.riiu-

I ONt. 28.1l04 20.000 2,47s 4. 1 4 01.. l). (10ili( iI
2 Aug. 19. IWIS 25,00) 1.910 1 l 2 47. 7' .2. 04i 3 i 31) (i 10
3 Aug. 25, 190y 12,000 2,2.10 4. (I 26i.710, (KMll .'1, sooJ

The tliree trees examined are not in any way exceptional as regards
the degree of infestation, but may be considered as representative of
the condition of China trees and umbrella China trees in localities
where the citrus white fly is established. Tree No. 1 was located by
the roadside near a 5-acre grove of newly bearing budded orange and
grapefruit trees which were lightly infested by the white fly and on
which it is estimated that not over 100,000 insects could have matured
on any one tree during the season. Tree No. 2 was located most
unfavorably for a heavy infestation, standing in a vacant lot in the
business section of Orlando and having its source of citrus whlite-fly
infestation in the spring almost entirely restricted to two neglected
and worthliless orange trees of small size growing within a radius of
100 feet. Tree No. 3 was located in front of thle laboratory at
Orlando, with 36 orange and grapefruit trees on thlie grounds. The
least conservative of the authors' estimates would place thle numbl)er
of white flies which matured on any one of these citrus trees during
the year 1908 as not over 500,000, with the average of the 36 trees at
about one-half this number. It is estimated, therefore, that thle one
umbrella tree produced upwvardl of tliree times as many adult citrus
white flies during thle year 1908 as the 36 citrus trees on the labora-
tory grounds combined. The important relation of thlie remarkable
multiplication of the citrus white fly onil China and umbrella trees to
the spread of thle pest will be discussed under the heading "Spread.''
Two new points of importance have been established by the present
investigations in regard to umbrella China trees as citrus white fly
food plants. First, this insect shows in one respect a greater degree
of adaptation to this food plant than to citrus plants, as shown by
the very low rate of mortality in the immature stages. Table VI
gives the data obtained by five counts made at Orlando, Fla., d(luring
these investigations.


..............: :.. ...


TABLE VI.-Morlality of citrus white fly on umbrella China tree leaves.

tpma Live Dead
Date. casspa larvae larve Mortality.
and pupae, andpupwe.

Per cent.
Oct. 28, 1906... 806 1 111 12.0
July -S, 'US.... 497 338 192 18.7
July 21, 190US... 113 256 51 12.1
Aug. 19, 190S.. 232 49 152 35.1
Aug. 25. 19M .. 312 169 70 13.4
Total.. 1.960 S13 596 IS.2

The record made on August 19, 1908, showing the highest per-
centage of dead stages of the white fly, was based upon 10 leaflets
selected from a single leaf and is not considered so typical of the con-
dition throughout the tree examined as is the case in the other
records. In contrast to the low mortality records as shown by the
insect forms present on the leaves of the umbrella trees, 26 records of
mortality in citrus groves gave an average of 57.9 per cent dead on
the leaves. These records were based on the examination of about
2,000 leaves and over 100,000 white-fly forms. It should be noted
that the mortality in the above records is based on the number of
live and dead larve and.pupe, and of pupa cases present on the
leaves at the times of the examinations. The actual mortality would
be represented by the difference between the total live larvae, live pupae,
and pupa cases and the number of hatched eggs. On umbrella China
tree leaves this difference is slight and represented for the most part
by the number of d(lead larvar and pupa? found on the leaves. In the
case of the citrus trees, on the other hand, the number of citrus
white-fly forms on the leaves ordinarily represents only from 25 to
30 per cent of the total number of eggs deposited. This disappearance
from the leaves is discussed elsewhere. Its significance in this con-
nection is that the actual mortality on citrus leaves is much higher
than the average per leaf of 57.9 would indicate. The citrus white
fly forms in the leaves show a mortality on the umbrella tree amount-
ing to only one-third of the mortality on citrus trees. The considera-
tion of the number of hatched eggs as a basis for mortality estimates
would reduce this to about one-fifth. The figures refer to citrus groves
where the citrus white fly is well established. In newly infested
groves the rate of mortality is much smaller as a rule.
The second important point established in the course of the investi-
gations reported herein is that adult citrus white flies are so strongly
attracted by growing leaves of umbrella trees that under certain
conditions with umbrella and citrus trees growing side by side more
adults collect on three or four umbrella leaves than are present on
entire citrus trees of medium size.

..... . i



It has been lfreq(lientlIv o' .d eve 'd thi lti I wi l.ll lie c'irlis while. fly is
first, becoming estbisihedl iil a gr 've, iiF ('liiii t reoies or insiibrlln ('liiai
trees are Inear aa(d ilt.s ofte'i c l 13 '(iniel <1 I ld o (il,' Ii 1 wn1131 '11 1 ca li e
found on s LrrounLjlIg or inter'oeing 'cit! ins lre-. In ,1'er tl o 0)1,1ti11
a more dlelinite idea ll 4 the relative i("0lt ric(i\e'ness %oIif i b l II II('lijn,
trees and citruils trees, 4. re1 c ,r'Is wv r, 11wen, 1 Il I1,1lie senior inllut i ",r ,
thle lal)ornltorv groun.ids ( ig. 2) it Orlilind II oll)svi,\'tlii,1 Nil. I, I e
count on citrus wits
malde on 4 t rees. viz, 0 = O NGE, = GR4PE FRUIT7,
4 A, 4 B, 4 (, and 5 = UMTBRELL ,rn7cEc's,
S = Z= 1 OAE HD A60R,4r'OR-1
C, and the observa-
tions on umbrella
China trees were
made on 2 small OPAAVNGE TREES
trees located about
6 and 20 feet, re- /1 0 0 0 0 n
spectively, south- ORANGE
west of 4 A. These /0 0 0 0 TREE
umbrella China trees W ,.
were slender 2-year-90
old growths about 4 8 0 0 0
and 5 feet high and
together bearing 7 0:. t.o
about 40 leaves. 6 0 0 0 0 0
Observation No. 2
was made on grape- 5 0 0
fruit and orange 40 @ K
trees E 6, E5,E4, 0
D5, D4, and F3 F 3% 0
N3 0 k
and two steins of
the umbrella China 2 Q
tree cluster in space 0 0
F 7 nearest to tree -----
E 6. Observation A B C a E F
No. 3 was made onil Fi. 2.-Diaranm of the laboratory eroun',ls t (rl rrlo. Flai.
tree A 4 and the (Original.)
nearest umbrella China tree sapling. This latter hald been defoliated
since observation No. 1 was made. Observation No. 4 was made on
citrus trees D 7, E 6, E 5, E 4, E 3, and F 3 and on two sterns of
the umbrella cluster which had been defoliated since observation
No. 2. Tie data obtained by tlhe four observations are given in
Table VII.


TABLE VII.-Relative attractiveness to the citrus white fly of foliage of umbrella China
trees and citrus trees.

Citrus trees. Umbrella China-trees.

Observa N Approx-
lion No. Date. Number Number Notber Number Number mate
um oftreesnumber
of of adults oxam-trees of of adults of leaves
minutes, counted. Ile minutes, counted. ea
mned. exam-

1....... May 18 5 257 4 5 508 25
2....... ...do .... 5 19 6 5 615 30
3...... June 11 3 34 1 3 477 6
4. ..... ...do .... 5 52 6 5 830 15
S Total 18 312 17 18 2,427 76

In all, 2,789 specimens were counted, of which 88 per cent were on
umbrella ('China tree leaves. It was estimated that in each record on
a citrus tree approximately 2,000 leaves were examined, maldking
34,000 in all. The individual leaflets composing the 76 umbrella-
tree leaves numbered appl)l)roximately 6,000. For practical purposes
these leaflets are more comparable to the citrus leaves although the
latter have on the average fully twice as much surface. With this
basis for comparison it can be figured from the above data that there
was about one adult white fly per 100 leaves on the citrus trees while
there were about 40 adults per 100 leaflets on the umbrella trees.
As has been indicated, the difference between the number of China-
tree leaves in numbers 3 and 1 and between 4 and 2, respectively,
represents the oldest spring growth, which was removed on May 24,
leaving only a few growing leaves. No direct comparison was made
between the attractiveness of the older growth of citrus and umbrella
trees but apparently there is no striking difference between the two
food plants in this respect. New watershoots were present on the
citrus trees on both dates when observations were made but only in
the case of one tree, 4 A, were many adults found on this growth.
In observation No. 1 on the tree mentioned (4 A) 200 adults were
counted on two watershoots. Except for watershoots there was no
new growth on any of the citrus trees.
At Orlando the umbrella trees usually start to put on new foliage
in the spring before new growth appears on citrus trees. As a con-
sequence (China and umbrella trees located near infested citrus trees
receive large numbers of adults of the citrus white fly which migrate
in search of attractive food. On February 22, 1909, the authors
noted on the laboratory grounds that the shoots of the umbrella tree
were beginning to put out new growth, the leaves not fully unfolded.
The citrus white fly was found scatteringly on the umbrella leaves
but on citrus trees specimens could be found only after careful



On March L 7, 110)7, i tl" t LI.ti,,y ent('' ll,.tin il tl( grove
of Mr. J. M.. ('hllney, a striking exaiiiple 1 te attractiviene'ss ,f tile
umbrella i tree was ohlservNe(I. Tie tree referred to) was aIl)nut 25 feet
high and! the leaves whicl were on (lie average only ail)olt lalf
developed were estimated to ntn eill" r 5,000)). Ten lhaIves were.
selected at ran(Ioi. witthini 10 fee't (I' tlie groindl a(1nd tlie number of
adults an11(id eggs was o('Oulite(l, tlie formerr tlnutlnll)erng 5.;3 per le.f
on the average andl tlie latter 160 per leaf. Tilhe tree was cut down
andantan examination o()I tlie t)oplmlost leaves sliowe(Il an average of ISO6
eggs per leaf; thle adults, being d(listiu rbed, were not coulinte(l, but
judging from tlhe number of eggs present they evi(ientIly were no()re
rather than less numerous than on leaves near tlie ground ('on-
sidering the average of 5.:) per leaf, however, thle total numl)er (if
adults on lthe tree would 1)be estimated at 26,500, anti at 160 eggs per
leaf the number of eggs depositedd would be estimated at 800,000.
At the time of this observation about 50 per cent of the insects which
overwintered on the citrus leaves lihad matured. Thle citrus wliite
fly had been much reduced throughout the grove, in some sections
by unexplained influences, in others by these influences and fumi-
gation experiments combined, and on a few tangerine trees by a
fungus parasite, red Aschliersonia. The location of the umbrella tree
did not seem to be a favorable one as regards opportunities for
white-fly infestation, but examination showed the infestation to be
at least 100 times greater, as regards the number of adults present,
than on any citrus tree in the grove. There were, in fact, too few eggs
deposited on the leaves of tlhe citrus trees to allow of sufficient multi-
plication of the white fly during the season to cause any blackening
of foliage or fruit.

The cape jessamine lhas long been recognized not only as a favorite
food plant of thle citrus white fly, but as especially important eco-
nomically on account of its retaining its foliage throughout, tlie year.
From a statement by Riley and Howard 1 concerning observations
by Mr. H. G. Hubbard and statements by Dr. H. J. Webber, Dr.
Montgomery, and others in tlie discussions on thle citrus white fly at
a meeting of thie Florida State Horticultural Society,"2 it appears that
the freezes of December, 1904, and February, 1905, wlhich completely
defoliated citrus trees when not especially protected, failed to defo-
liate cape jessamrnines. In many localities it is probable that this food
plant was responsible for tlhe survival of tlhe white fly at the time
referred to. According to Dr. Sellards,3 temperatures as low as
'Insect Life, vol. 7, p. 282.
2 Proceedings of the Florida State Horticultural Society, 1896, p. 78.
3 Press Bulletin 56, Florida Agricultural Experiment Station, p. 2.


S '. :.........


16 above zero at Lake City, between January 26 and January 29,
1905, failed to defoliate cape jessamine.
Except where grown for commercial purposes, as is the case at
Alvin, Tex., where the blooms are shipped to northern markets, or
where grown in nurseries, cape jessamines have not been observed
growing in sufficient abundance to materially affect near-by citrus
trees in sections where the white fly is already established. If over-
looked in connection with the fumigation of citrus groves or defolia-
tion of citrus trees-by cold, cape jessamines might become a serious
hindrance in the control of the white fly. The greatest economic
importance of the cape jessamine as a food plant lies in the great dan-
ger it presents as a distributer of the white fly. This will be referred
to again under the subject of methods of spread.
The subject of the adoption of the cape jessamine by the citrus
white fly is not of sufficient importance to have been given more
than incidental consideration. In general the degree of adoption
seems to be less than is the case with the umbrella and China trees.
On November 17, 1907, an examination made of 30 leaves picked at
random from both old and new growth of a cape jessamine which
appeared to be in an ordinary condition of infestation as observed
when growing near infested citrus trees showed that there existed an
average of 45.1 forms per leaf.
The extensive growth of cape jessamines, or gardenias, as the
blooms are sometimes called, for commercial purposes is known to
the authors and occasions a conflict of interests only in Alvin, Tex.
From the orange grower's standpoint this, at the most, applies to a
location adjoining an orange grove where the citrus white fly is uncon-
trolled. Fortunately, however, for the citrus growers, it is of great
importance to the success of the florist's business that the white fly
be kept in subjection in gardenias.

Privet hedges are not uncommon in citrus-growing sections, and
heavy infestations by the citrus white fly occur in parts of Georgia
and South Carolina, where no citrus trees are grown. As a food
plant the privets are of economic interest in the same respects as is
the cape jessamine, but they are more extensively grown and of pro-
portionally greater importance. No studies have been made of the
degree of adaptation and attractiveness, but the several species of
privet observed in infested localities have shown the propriety of
classing them with citrus, -China trees, umbrella China trees, cape
jessamine, and other preferred food plants. The senior author' ob-
served a migration of adults from privet hedges in Victoria, Tex., in
the summer of 1904, which indicated that a hedge of this material


might well be colipared(l in its inljuriotus influence on citrus-growing
interests to one or more um11l)reilla or ('Chiia tr(es. 'Ilie privet, like
the calpe jessaminiIt, is llarlyv, anltli(f (l disadlvalItages of tlie former in
this connectitont are tlie same11 as those enctionl1wd inl discussing the
latter food plant.

Japanese and wild persimmons are attractive to tlie citrus white
fly early in thie season, but appear to be very little or not at all so
late in the season. Being (deciduolous, their economicc importance as
white fly food plants is prol)ortionally small. Unii(er niormial con(li-
tions the Japanese persimmons appear much more attractive to tlie
citrus white fly than citrus trees. Trlhese conditions have not been
investigated, but tlhecy are prob)ab)ly dependent upon tihe appearance
! of new growth in the spring a little earlier on persimmon than on
citrus. On June 16, 1909, an examination of a large bearing per-
simmon tree surrounded by citrus nursery trees and bearing citrus
trees of different kinds showed that thle first spring growth of tile
persimmon was much more attractive to tlhe first brood of adults
than were thle citrus trees. Tile second brood of adults, however,
found the persimmon comparatively unattractive and shliowed a
marked preference for the citrus trees. The earliest citrus growth of
the spring had become fully matured, and no new growth appeared

The comparative condition of infestation is shown by counts made
on leaves picked at random from the persimmon tree and from the
surrounding citrus trees, including the sweet orange, sour orange,
tangerine, and grapefruit. The average infestation within first-
generation forms of the citrus white fly on 25 leaves each of per-
simmon and citrus was in the ratio of 10.9 to 1.3, while that of the
same number of leaves by the second generation wa's in the ratio of
no forms on the persimmon leaves as compared with 191 on the citrus,
Sthus showing thle great preference of thlie second generation of adults
for citrus growth.
Neither tlhe Japanese nor the wild persimmons are usually infested
by the citrus white fly to the extent of causing noticeable blackening
from sooty mold. The infestation, however, might be between from
five to ten times as great as on the leaves from the trees referred to above
without producing this result. Small wild persimmon bushes hliave
been observed in a growing condition at thle time the adults of the sec-
ond brood are on the wing, and at such times they sometimes appear to
be very attractive as food plants. Mr. W. W. Others has observed
near Hawthorn, Fla., on April 29, 1900. the citrus white fly on wild
persimmon bushes growing in pine woods at distances upward to one-


fourth of a mile from any citrus grove, and the junior author has
made similar observations along roadsides near Orlando, Fla., in
June, 1909, the insects being in the adult stage only in this latter
case. On the other hand, the senior author noted on June 18, 1909,
that wild persimmon bushes growing in a vacant lot with China trees
andl abandoned citrus trees were only very slightly infested, although
the citrus trees and tlhe China trees were heavily infested. The
wild persimmon hadl male vigorous growth, but its white-fly infesta-
tion consisted of less than 100 eggs per leaf antd an occasional adult.
The examination of leaves of the China tree showed hundreds of
pupae and pupa cases per leaf, with a few adults and newly deposited
eggs. The old1 citrus leaves bore many larve, pupae, and pupa
cases, and thlie new leaves bore hundreds of unhatchedl eggs. The wild
persimmon 1)bush was as favorably located with respect to citrus trees
as was the China tree. Notwithstanding the exceptions noted in
degree of attractiveness, the Japanese and the wild persimmons very
evidently rank well below citrus trees, China trees, and umbrella
China trees.
In so far as observed the perslnnimons have little effect on the con-
trol of the citrus white fly, but in special cases they may rank as
important food plants. The fact that the Japanese persimmon is a
producer of fruit of some commercial value makes its ordinary light-
ness of infestation a matter of gratification. The wild persimmon, on
the other hand, is of practically no value either for shade or fruit, and
can easily be destroyed where advisable.

Lilac is not commonly grown in the citrus-growing regions of the
Gulf States, and on thi account, so far as observed, presents no
element of menace to orange groves. In company with all of the
ornamental plants listed as preferred food plants this one must be
considered, however, as undesirable for introduction and growing
in citrrus-growing regions.

Belonging to the family Rutacee, to which the genus Citrus also
belongs, it is not strange that thle prickly ash is a favorite food plant
of the citrus white fly. This plant seems to be highly attractive to
the adult flies, frequently being observed infested with more adults
than many near-by citrus trees combined. The prickly ash is common
in Florida and in some localities, where growing in abundance along
roadsides, it constitutes a distinct menace to citrus groves through its
coiuection with the spread of the white fly from city and town to
country and from grove to grove.


T1 IK 'lT'Rl'' S WIIITE I'L : S''IiAl)D.


Dr. E. IV. Berger lhas report Irt I11 ha vig obse1''ved I Itoffvie tree tlor
ouglily iN'fest cw'ithl 1 as zait% eggs o n its leh(ves s c'itr's li eavCs liaVy
have. Thlis h')l plant is to0) 'amely grown ill tih( ('lf Stiates inll
orange-grow'vin'g ir.etions to be ot' aillv inp)ortatI .nce econt ically us a
wltite-Ily V food l plant

The plants listed in ('Class 11 as a whole are ,of vmerv little illplortance
as retar(Is the ir bvl i enai 0g oH wiute-fly conitinol. Banana sill-ill), chieri
laurel, al l( cult ivatet pear miitl it, well I1e cons( sIleredl in a thlird ciass
for rarely inietsted plants. Altlouglh not utnco.mIllon, thleir attmra'tiomn
for the citrus -white fly is so slight as, to lmake it safe to iglnorle them
except in the matter Ol int rodtucing thle flyi on them into nonin'rested(
districts. In untlpublislhed notes Dr. Berger lhas re(orlded( tlie will
olive as a foodl plant. Ile lhas obsl)ervedl thle wild olive infested in conlm-
paratively isolate(i places. TIhe junior author has observed wild olive
heavily infested i ('lCIarleston, S. (C., and in several places in Orange
County, Fla. Tlite wild olive, being an evergreen, if neglected may
prove to be of considerable importance as a food plant when growing
in abundance near a fumigated grove or when citrus trees have been
defoliated by cold.
Dr. Berger has recorded pomegranate, allamanda, and smilax as
food plants, and lhas verified Prof. Gossard's record of }'iburnvum
nudum as a food plant of tlee citrus white fly. Tlhe positions of
these plants as regards their attractiveness to the citrus wltite fly has
not been fully ldeterminiIed, andl further observations will p)erhlaps
show one or more of them to be of fully as high if not of higher rank in
this respect than the persimmons. In general, however, like thle coffee
and lilac of (Class I, they are not of sufficiently common occurrence in
the Gulf coast citrus-growing regions to be of much economic impor-
tance as citrus white-fly food plants.

There is seldom )positive evidence in regard to the means by which
the citrus white fly hlias become established in a previously nlon-
infested grove or locality. Such (lirect observations, however, as it is
possible to make, aided by strong circumstantial evidence, give us a
sufficient knowledge of the methods of spread to show the advisa-
bility of certain restrictive measures.
Fortunately the chances are greatly against the successful estab-
lishment of thie citrus white fly in a previously uninfested locality,
which is outside the limits affected by large numbers of migrating


adults. If this were not so the pest would have become established
in every grove of the State long before the present time. Except for
spread by direct flight and on nursery trees and ornamental plants,
the chances are against more than a few insects being introduced into
a particular grove by any of the other methods discussed hereafter.
In the case of a single adult there are two chances in three that it
would be of the. reproductive sex. If, as would be probable, the
specimen were a female, there would be about one chance in three
that it would not have been fertilized. In this case the second
generation of adults would all be males, as shown by the observations
recorded under the subject of Parthenogenesis. This would, of course,
'end the infestation -directly due to the single specimen introduced,
as the original female would have died several weeks before the first
male matured. In case the originally introduced specimen were a
female and fertile the chances of a male appearing among the second
generation are not definitely known, but are with little doubt only
small. The chances of such a male appearing at a favorable time to
meet with and to fertilize a female of the same parentage are practi-
cally negligible, though possible as a result of the great variation in the
length of the life cycle as recorded under life history. The third
generation would, therefore, in all probability, be all males, and the
infestation ended. The chances that a single adult specimen intro-
duced into an isolated grove or into a previously noninfested com-
munity would successfully establish a permanent infestation are
extremely small. The chances are only slightly increased by an
increase to 5 or even 10 in the number of adults originally introduced
into a single grove.
From the foregoing considerations it is evident that two or more
distinct introductions of even a few individuals at proper intervals
during a single season might greatly increase the chances for the suc-
cessful establishment of the pest.

The flight of adults is the most important method of local distribu-
tion and is also an important element in its association with spread by
means of winds and vehicles, railroad trains, and boats.
The distance to which the insect is capable of flying.-It would be
almost impossible to obtain positive records on the distance the adult
citrus white fly is capable of flying. Mr. W. W. Others, on April 29,
1908, found on wild persimmon first and second generations of this
species of fly at a distance of one-fourth of a mile from the nearest
orange grove, which was also the nearest point of the occurrence of a
food plant upon which the insect could have passed the winter. The
infested persimmons were in pine woods and the insects were in such
numbers that it was evident that spread through pine woods might


... .... ..




easily greatly exceetl one-fo rlt ,,(If if a tmile.'. lri. W ('. 'i lj, ., ,of
W inter lPark, lFla., states tl 1t hat 1,ILs blserv(Ild adiltls illigratiing itl.,
OineI of ii.s gro es o il L1ake l' itL n d laan ttd tl 'ic'lli[ stlC lIls4ill1yv iidIi-
cat.i ag tl Jt i !thTev iittl travel IVd C'er inl1' watVIr for 1 illi's. ) )r.
Berer lias recortdedi anl itstaneII wliu'lh )rese,1nts .l is'i ,.ig S('vi (denUce., tlint
adult white flies have heavily infested citrus treesI tlIirougli 11glits o(f it
mile or tire. Onl the other landl, t1,here ItreI or- ,,g', grove's wit lin
tliree-fotlrthis of a mlile of the city limits lof ()rlaiuIIId, Fl., adIl witn11111
2 miles of tlie courthouse which have only so recently I.ecoI( ei infested
that no blackening of the foliage t lias taken l)lace, altlougih t (hle citrus
white lyv has occurred at Orlando for more than 10 years wit It nigrat-
ing adults in summer about as abundant as in any town in tlhe State.
As regards the capability of flight of the citrus white fly, it may be
said to be undoubtedly more than a mile and perhaps several miles
when aided by a gentle breeze. Distances of even a mile, however,
are not usually attained except under certain circumLstances which
are largely preventable and which are discussed in tihe following
Cause of extensive migrations by flight.-Overpopulatio)n of food
plants, usually associated with the emergence of adults in large num-
bers at seasons when the new and attractive growth is scarce or en-
tirely wanting, is the main cause for migrations from citrus, cape
jessamnine, and privet. Migrations from China trees and umbrella
trees, probably the most potent factors in the spread of the pest, are
not due directly to overpopulation, so far as observed, since leaves are
never overcrowded in a manner comparable to the overcrowding
on citrus leaves. An average of 25 live larvae or pupae and pupa
cases per square inch of lower leaf surface would represent an un-
usually heavy infestation of a Chlina or umbrella tree and is rarely
exceeded, whereas an average of 50 or 60 per square inch is not un-
usual for citrus leaves. In the case of China and umbrella trees, imi-
grations are evidently due to lack of attractiveness of the foliage to tlie
adult white flies at the times when the migrations occur. There are
comparatively few live larve and pupae on the foliage after thle middle
of August at Orlando. The greater part of these represent delayedd
emergence from the second generation of white flies and not tlhe result
of eggs deposited by the third brood of adults. This supports direct
observations to the effect that the third brood of adults, which is con-
cerned in the most extensive migrations, deposits practically no eggs
on the China and umbrella trees.
China. and umbrella trees as a factor -in dissem(Ititi)n.-Umbrella
and China trees are extensively grown throughout the Gulf coast
citrus-growing regions, and they are almost entirely responsible for
the hundreds of millions of adults which in midsummer appear on
the wing throughout most of the towns where the citrus white fly


occurs. That these are principally those which have bred upon
China and umbrella trees is shown clearly by the fact that at Gaines-
ville, Lake City, Tallahassee, and other points in the northern part of
Florida, where other food plants are too few to produce noticeable
numbers of migrating adults, the numbers are apparently not less
than where both citrus trees and China and umbrella trees are exten-
sively grown, as at Orlando. On this point, Dr. Berger states:'
"The principal food plants in Gainesville and north Florida are China
and umbrella trees, there being only enough citrus, privet, and other
evergreen food plants to bring about the restocking of the deciduous
trees every spring." These considerations indicate very positively
thle main source of the enormous number of migrating adult flies on
trees in midsummer, sometimes observed between the middle of
May and the middle of June. These adults are the second brood of
the season and the first to mature on the food plants mentioned. The
newer growth of these trees is, as has been shown, very attractive to
the adult flies, and if there is an abundance of it comparatively few
migrate. The third brood, composed mainly of individuals of the
second and third generations, matures over a more extended period,
in general covering the months of July and August in different sec-
tions of Florida.
Estimates of the number of adult citrus white flies breeding on
umbrella trees and on citrus trees as given under the subject of food
plants have shown that a single umbrella tree of medium size may
Produce as many adult white flies by midsummer as could be pro-
duced on 7 acres of orange trees. The maturity of so many adults
on single trees, and their migration therefrom in search of a more
desirablee food supply than China and umbrella trees afford in mid-
summer, cause the rapid spread of the pest throughout the towns,
directly by flight of the adults and by mediums hereinafter dis-
cussed into the surrounding cotintry and from town to town along
railway lines and wvatercourses.
Dissemination by flight uwhen citrus trees only are concerned.-It has
been shown under the subject of food plants that the citrus white fly
does not ordinarily increase to the point of overcrowding on grape-
fruit. Migrations of adults in noticeable numbers from solid blocks
of these trees probably never occur under ordinary circumstances,
and spread through such blocks or groves from the first point of infes-
tation is very slow if no other food plants are concerned. The spread
in groves of orange or tangerine trees or of both is more rapid, but
not as much so as ordinarily considered. The white fly is rarely
observed during its first year's appearance in a citrus grove. Atten-
tion is usually first attracted to its presence through the blackening
of foliage on one or a few trees. This blackening of foliage in itself
Press Bulletin 108, Florida Agriculture Experimental Station, February 13, 1909.


TIlE (cIIuIi's \wIII-rl t -I",Y: .si'ittI;.\I.4

is ialml(ost lpositive' t'v ild'eiHe 0 I ie Jt)sn'('iIe' oIf tli 1 fly in litp i -,\
tdulriig lit. least tile i pilre'diig two years leilielllsN ti' ilifest.ti1ti1)11 i- i dlle
to InIli' I iitiit 'i'o II ( fina.o or u 1bll ii IIrclla tI'N'S o f o mii II tI 1k ,,
lieigill)oring 'itNI rls gIll l's. In l iils i c Shc is s il If'sl. liiit 1 V I, E 'iii'
q il'te general tirilouigliol.it t'V tCl'li ai s ill oliiie ea1s41., nI11111 d xtrl ,si.ve,
blackeninig of tite foiliaie Uiv result. elly iii lite ix ,t season, o iii
about oi Vea'iLr aIft"er the first, iltrodulictio). Ii til case iof li ew lW ilif.S-
tation in ianly ilocality, however, tif' bl(giiining d(1(iiiItle.ss i, usual ly
thile intlrolductlionl of a few insects. by soie (ioic of ifte iiea lns l ii ertIl'n-l-
after disctus.sell. In itl mixed grove of ta iligerilie i ol(l ralil(ge tlie pest
is discoveried first as a rle on thingel'ritn, and iin a g"rovc of' sC(li1,
trees witl Ii a few bllddetd trees interinixed usually til, latter airt irst
discovered to be infested. Many citrus grower's wiS)i have g'ro1es,
sucli as tliose mentioned, and wlio lihave watched cal crefull forl tlie
appearance of thle pest in their groves, have iniailly foutlinl it well
estabIishd ion a single tangerine or budded oriangeL tree before any
evidence of thle presence of thle insect was observed I elsewhere.
Through the hindrances to successful establishment and tle checks
on multiplication, principally tliose discussed in connection witli
parthenogenesis and natural mortality, thle lite fly frequently
develops so slowly after its first introduction that it 1may not increase
to the point where it. is usually first observed for three or four years.
It is a common error to consider tliat thle first discovery of tile whliite
fly in a. grove is an indication of its very recent iltrodulctioni. 'Thiis
may or may not be the case. Usually it is not thle case. It should](
be borne in mind in this connection that in tlhe most careful insp)ec-
tion, even by a competent entomologist, thle failure to discover al
single specimen of thlie whiite fly is not positive )roof tliat it is nlot
present. The foregoing generalizations are based uponU many
observations by tlhe agents of tillis bureau whio have been enigageirL in
these investigations, 1 ore lC particularly the authors of this bulletin
anti Mr. W. IV. Others.
The rapidity of spread into a citrus grove from neighlboring infested
groves is a subject which becomes temporarily iml)ortant when a
nonisolated grove becomes infested for the first time. It is a. subject
of more far-reaching importance in connection witli fumigation, anid it
is in this connection tliat the most extensive studies inll this line have
been made. The result of these studies will be published inl a final
report on fumigation.
The slowness with which the citrus white fly increases ill numbers
and spreads from thle first point of infestation has been noted by
many citrus growers who have been observant enough to discover
the white fly soon after its introduction into their groves. When
the rate of spread of the white fly through the grove is affected by
the presence of migrating adults from China or umbrella trees, the


..' ... .... .. .


difficulties in effectively utilizing artificial checks, spraying and fumi-
gation, are greatly increased. As the infested area in a newly infested
grove or locality becomes larger the rate of spread by flight increases,
aided by secondary centers of infestation which become established
by various means. .....
Light winds are an important adjunct to flight in the local distri-
bution of adult white flies, but strong winds are ordinarily of slight
consequence. The effect of light winds is shown by the influence
of almost imperceptible movements of the air on the direction of
migrations. This is especially noticeable in the vicinity of China
and umbrella trees during a season when adults are emerging in abun-
dance. The principal effect of the movement of the air under such
conditions is not in carrying the insects, but in causing the flight
energy of the insect to be expended in one general direction rather
than to be wasted in zigzag lines with comparatively little real pro-
gression. Other conditions being equal, the adult white flies migrate
in greatest abundance when the atmosphere is calmest, and con-
versely show the least tendency to migrate in strong winds. It.
is possible that isolated infestations may sometimes result from
spread of adults by strong winds, but it is seldom that there is not
a more plausible explanation obtainable. With the white fly pres-
ent in abundance for many years in Orlando, Fla., and other towns
and cities in important orange-growing sections of Florida, the fact
that there are still many noninfested citrus groves within a radius
of 5 miles of nearly all such centers of infestation is in itself an indi-
cation of the minor influence of winds in this connection. Strong
breezes or winds exert some check on the spread of adults by causing
them to cling tenaciously to their support, as pointed out by Prof.
H. A. Gossard.1
In towns in Florida where the citrus white fly occurs and China
trees and umbrella trees are abundant it is a matter of common
observation that during the periods of migration large numbers of
adults alight upon automobiles, carriages, wagons, and railroad coaches.
The authors have seen covered carriages with more than 100 adults
resting on the inside of the top and sides. In driving through a
heavily infested citrus grove in late afternoon at certain seasons,
hundreds of adults may be observed on the carriage (P1. VI, fig. 2).
Newly infested groves show the first infestation so frequently on trees
close to a driveway or road that conveyance of the citrus white fly by
means of carriages, wagons, and automobiles must be considered one of
the most important methods of spread from town to surrounding
'Bulletin 67, Florida Agricultural Experiment Station, p. 13.

BuJ '-'. Burvau it E nFroLrAii, iiy



Fig. 1.-Nursery citrus tree- infested with white flies set out in an isolated noninf.sted grove
without having l';ives removed. Fig. 2.-Buggy in an oriItg- grove: Ii.ggv top full of adult
white flies ready to be c(-irried to other groves. Fig. 3.-Tnr in at st Ulio: adult citrus white
flie, swarming from near-by umbrella China trees into coaches ready to be carried for miles
down the Florida east ,_oa-t. (Original.)

U+ 5 Drp!, of Ag, i ...



country or from grove to grove. At ( )rlaIni o, iill .Ily, I1(1H;, adiiltl
citrus white flies were oblservecd late iii thle aaft eill I ,g lillt ig ,o t lie
sides of coaches anti living, into thile wiidiows nIi ,lors of ,icii's
of a passenger train standing u.t a railroad station (11. \f, fig. 3).
Hundreds of adults were carried, west owadI \Vildwood sI through
points which, so fair as knoiw\\i, were oti tfiestedi at Oet t tile. Tile
presence of ('iina, uiibriella, or citIrus trees iea. i1a railrv mid statioiis
increases the chances for successful iltr odlu1ctionl by) railroad, trains.
In this connection tlie recent action of tle Atlantiic (oast Lilne iail-
road and Seaboard Air Line Railway i destroying suchl trees along
their right of way is to be commendled. A map1) of Florida i,
the distribution of the citrus white fly plainly indicates tlie 'relation
between the railroads and the main lines of dlisseIliilatiin. This is
shown in an incomplete way by figure 2, in which are giv-en tle points
infested by the citrus white fly in Florida accordlinr to thle records
made in connection with the present investigation and stich other
records as are undoubtedly correct or which have been verified. Tlhe
infestation at Arcadia, Fla., first discovered in Janutiary, 1907, but
which probably resulted from an introduction of citrus white flies
in 1905, was with little doubt due to the introduction of adult flies
by means of railroad trains. An examination of tlhe situation in
February, 1907, by the senior author showed the center of infestation
to be located near railroad stations, and careful inquiry concerning
other possible sources showed that railroad trains were thle most
likely means of introduction. North of Arcadia no nearer infested
point was known than Bartow and toward thie south no nearer
infested point than Fort Myers. The distance in each case was al)ott
40 miles. So far as known there were at that time no intermediate
points infested between Arcadia andl tlhe two )points mientitiolned.
Here again the factors unfavorable to tile successful establishment
of the pest in a previously uninfested locality play an important role,
as shown by the fact that even at tile plresenlt writing thle citrus white
fly is not generally distributed between Bartow and Fort Myers.
As we have no record and have heard no report of the occurrence of
the fly at any other point than ArcatLia, it is unlikely that other
infested points exist.
Steamboats are used quite extensively on thle rivers and along the
coast of Florida in transporting citrus fruits and have in ai degree a
S similar status to railroad trains in tranlsportiig thle citrits wlhite fly.


The carriage of the citrus white fly in its egg, larval, and pupal
stages by means of citrus nursery stock (Pl. VI, fig. 1) and ornamental
plants has always been an important factor in the spread of the
86850-Bull. 92-11----4


, ." ,...: ,-.. .., ,^
I -


insect. The citrus white fly was without doubt introduced into the
United States and distributed to the most important centers of
infestation by this means. In Florida the white fly was probably
introduced first on citrus nursery stock into some citrus grove on the
St. Johns River in St. Johns County, and later by the same means
into Manatee and Fort Myers. Gainesville, Ocala, Orlando, and
Bartow were probably among the points to which the white fly was
introduced on nursery stock. The distribution of the citrus white
fly along the Gulf coast citrus-growing regions west of Florida has
been largely due to shipments of infested citrus nursery stock,
umbrella trees, privets, and cape jessamines. Of all methods of
spread which are operative over greater distances than the flight of
adults, introductions of live immature stages on trees or shrubs for
transplanting purposes are by far the most certain to result in the
successful establishment of the species. Fortunately it is practicable
to prevent spread by this method by defoliating the trees as they leave
the nursery. Much has been accomplished in the past by individual
citrus growers, but more attention should be given to this matter in
communities not now infested by both of the white flies treated in this

Carriage oj the adult white flies on human beings.-Man is doubtless
responsible to a limited extent for the spread of adult white flies.
During migrating periods, when in heavily infested orange groves or
in towns where there are infested China and umbrella trees, adults are
frequently observed on the clothing. Prof. H. A. Gossard states
that he has carried adult white flies for nearly half a mile on his
clothing after standing beneath a heavily infested tree.
Introduction in pickers' outfits.-In some instances the citrus
white fly is believed to have been introduced into previously unin-
fested localities by orange pickers. In this case the principal danger
lies in introducing live pup.e on citrus leaves accidentally brought
in with picking sacks and field boxes. The authors consider that there
is practically no danger of the carriage of adults of the citrus white
fly by pickers' outfits between December 1 and March 1. The few
adults present in citrus groves during this period would rarely result
in their transference to uninfested groves by such means, and the
unfavorable factors heretofore discussed would almost certainly
prevent the successful establishment of the pest. It would be
almost impossible to conceive of any likely method by which a suc-
cessful introduction of the citrus white fly into a noninfested grove
could be accomplished by the carrying of leaves infested by eggs or
larvae. Leaves infested with live pupae, however, particularly about
the time of the beginning of emergence of the first spring brood,



might readtlily produce t siuflicient number of ad tilts to su(''(ess.flly
estal)isIL the pest. SuchI leaves, after iltrod (Itctiol, w11ould need to
have at favored locattion, for ex)postlre to lmucli sunliglit or to too
ulinchi moisture would sooll destroy tihe insects.
Introduction on. l(e'e.a inffx'td u'ith pjnrviaifi' fiPigi. Trl iti atter of
sp)reIal of the white fly in coinnectionI witIL tilt atteiipt to initrid()ll'ce
parasitic fungi is a subject of considerable imI)ortaILce. 7'I'I d(iinger
here is due to the failure to recognize thle distiniciont betwee'i tlie
citrus white fly (Aleyrodes citri) and( tie clouly-winged wlitite ly
(Aleyrodes nubifera). The owner of a grove infested I)y tie latter
species only, would provide a very favorable o0)I)ortulity for tlhe
introduction of the first and most destructive species if in introduc-
ing parasitic fungi lie should obtaLin Iis supply of leaves from certain
sections of Florida. The spread of the clou(ldy-winged wlite fly li.s
been encouraged in a similar manner. Thle tree-p)lanting metlho(d of
introducing the fungi, especially the brown fungus, is tihe most (dan-
gerous practice in this connection. Of somewhat less (langer in the
individual cases, but of far greater danger on account of the more
frequent opportunities presented, is the introduction of fungus-
infected leaves for pinning or for spraying the spores. Thie pinning
of leaves as a means of introducing the parasitic fungi ihas little more
to recommend it than the tree-planting method, but it lhas without
doubt been the means of introducing tlhe citrus white fly on many
occasions. Leaves introduced for tihe spore-spraying method of
spreading the fungus parasites are an element of much danger under
certain conditions. Some sections of Florida in which only tihe cloudy-
winged white fly occurs are in more danger of having the citrus white
fly introduced by some uninformed person in this way tlian they are
S of its introduction in any other manner. Specific examples might be
cited where the introduction of either A. citri or A. biJ'ijini was
with little doubt (lue to introducing fungus-infected leaves or trees,
i but the danger is too obvious to require further discussion in this plaice.

The eggs of the citrus white fly (fig. 3) are laid scatteringly, with
few exceptions, on the underside of the leaves of the various food
plants, and hatch in from 8 to 24 days, according to the season.
During ordinary summer weather from 75 to 100 per cent hatch
on the tenth to twelfth day. Infertile eggs hatch as readily as fertile
eggs and produce adults of the male sex only. After hatching, the
young larva (figs. 4-6) actively crawls about for several hours, when
it ceases to crawl, settles upon the underside of the leaf, and begins
to feed by sucking the plant juices. It molts three times before


becoming a pupa. After the first molt (see fig. 7) the legs become l
vestigial; hence thereafter it is impossible for it to materially change
its location upon the leaf. Larval life averages in length from 23 to
30 clays. The pupa (fig. 9) closely resembles the grown larva (fig. 8)
and requires from 13 to 304 days for development. The adult fly
(fig. 10) has an average life of about 10 days, although several females
have been known to live 27 days. Females may begin depositing
eggs as soon as 6 hours after emergence and continue ovipositing
throughout life. The maximum egg-laying capacity is about 250
eggs, although 150 more nearly represents the number laid under
grove conditions. Unfertilized females deposit as many eggs as
fertile females.
The entire life cycle from egg to adult requires from 41 to 333
days; the variation in the number of days required from eggs laid
on the same leaf on the same day is very remarkable. During the
course of the year the fly may pass through a minimum of two gen-
erations and a maximum of six generations. The generation started
by the few adults that emerge during the winter is entirely dependent
upon weather conditions and may or may not occur. Each genera-
tion except those started after the middle of August is more or less
distinctly two-brooded.
As it is impossible to rear citrus white flies through their entire
life cycle on detached leaves, a gauze-wire cage was devised by the
senior author which has proved of great value and convenience in
carrying on life-history studies under conditions as nearly normal' as
it is possible to get them. This cage (P1. VII), which is cylindrical
in shape and open at one end, may be made any size, but one 64
inches long by 31 inches in diameter has proved most convenient.
It can easily be made by fashioning two rings of heavy wire to which
is soldered the wire gauze, as shown in the illustration. To the open
end is attached a piece of closely woven cheesecloth long enough
to extend about 4 inches beyond the cage. After the leaf, or leaves,
to be caged have been cleaned of all stages of the white fly by means
of a hand lens and cloth, the cage is slipped over the foliage. The
adult flies are then introduced, if desired, and the cloth attached to
the cage wrapped around the stein of the shoot or petiole of the leaf,
as the case may be, in such a manner that the flies can not escape
nor the ants and other predaceous insects enter. To keep the entire
weight of the cage from falling on the petiole of the leaf or its short
stem, and to regulate the position of the leaf within the cage, a cord
is tiedl around the outer end of the cage and attached by the loose
end to a convenient branch.

B,' 92 Burou of Ent mrnilniyi U S Dept ti Ajer, ji.u'iP L

Fiig. 1.--rLarill >';age' in position on orantiv trees. Fig. 2.-Eiflairgc.l rerinm cage. (Ori-inmil.)


T11 l, (I r'liI'S HvIIITE FLY : I.II : E I IST() ilY .ANI I iIrsITS.


A v m-rv :ati f .ct wr m ett l () of d'fl lilitciv milal 'killn I, ,",\ ,, i, ,l)l ,.l-I,
tliait io m mistake ) limy I bi W J a1 dejl itit ild'1tifyilntg fildI n tll s wit1 the1'
ititliviluill IaIrva. t o whiicili they, 'f1. 'er, is to s,.rat.l lii,,htl" ,l f tIll.
e'pitierillis of tliI leafl, \vitli a thorn llor pin, i llrcket or ote mwli ui'rk
011d o tsildl( tiiis a nu e hr t at slall 'co T -espo d witl tli]H t 111 .i.,i ill
ti* nott bmook. Ill iniarkilg lrv';e can slionll Ieb tinkcii ill ,.'t a ,liintl ,.
tih I'lif to allow\" 'ol" tile l' tti' g ro\ tli ll' ti l' i sc 1)(,t '( ni ol to 1 i .lj11 '1,
tlit' tepiliT lnis of ( l tl ]leaf tmoo s. vel' v. Ill tli.s ia- n ,i ;i I ar, g.. iii tit il.-
of lrva' *' were, inni 'k ul a s oo Us tl it sett l u ii' tI eii"r gr,.oi\ iit lIi tedaI
by tlzily tOl)S(vi'vatiO)ls.
In deteminiilhung tlu s" lln.s oh fictive tc1(i)ipcriltii'e, 4:)'- F. IniUs I ,nii
takeiins a hls)s in iiacc)ro ncoe witlIi %r. IM\erraini's g(riu'r*Ila blw
altlhong!i this ha.is l l to certain ina'ccriiacies o \if liIi i tin', atillit)F.s
are awire. lThe dletenrmination 2 tif thie ef'ctive
temperature in tlhe case of tlhe white filv would
require a special stEdiiy which it hlus blen illipruc-
ticable to indlertake. /"
Tufn.: E'cu. -

The eggs of the citrmis white fiv (fig. 3) are s,
rl^ Pr .-- :{ The ,i'rim biile
small that. tlhey appeal r to thlie iunailded( eve as [ly,.i/l,,,,,r, /.,, i: .'s.
fine particles of w liutisli (lust on ttlie unldeir str- i<.,,,I-ni .-".i. i,,rii-
face of the leaves. Their minute size is emplia-
sized by the fact that 11S jplace(l eind tto eiid woiild measure iut anll
inch, while about 35,164 could be p)laIce( sid 1 e by si le in.( ole sitquare
inch. Vndler the magnifying lens they appear as smooth, polislheld,
greenish-yellow objects shiapedl much like a kernel of wlieat. Fol-
lowing is a more minute description:
Length, 0.2-0.23 un.; width, 0.08-).09 mm. Surface highly polished, without
sculpt tiring, color pale yellow with faint g-reeniis-h tine when first (depisited, pile'r than
the under surface of the leaf. Eg. elongate, subellisl slipi, slghtly widler l 'eb (lnd the
middle or at al)out the point, where the eves of thee embryo Iub-,equently appear ,1,rne
at end of a comparatively ,slender brownish petiole or fbotstalk, slightly shorter than
the width of the egg, and somewhat knibbl)l)ed at base.
As the embroyo approaches maturity its purple eyes may be seen
showing distinctly through the egg meml)ranies at a 1)point beyond tile
middle of the egg. At about tilis time, also, the hitherto uniformly
colored egg contents become orange or golden at the proximal endl and
whitish translucent on the distal three-fourths. rThe egg, surface
sometimes assumes a white pruinose appearance, due to tine presentice
of wax rubbed from the o)(lies of the adults while crawling over tile
eggs. ggs ti(leposite(d oin leaves from which tlie adul2.ts have been
excluded after egg deposition do not show thlis pruinose condition.


Eggs in which the embryonic development is normal do not turn dark
in color, but those killed through attack by thrips or by other agency
frequently become bronze colored, thus resembling the eggs of A.
nubifera from which the waxy sculpturings have been rubbed.
That no doubt might arise concerning the exact age of the eggs
used in obtaining the data incorporated into Table VIII, suitable
leaves were selected from which all eggs previously deposited were
carefully removed by the aid of the hand lens and a cloth. Similar
attention was given the leaf petiole and the stalk, and wads of cotton
were tied about the latter both above and below the leaf to prevent
crawling young from reaching the leaf along the petiole (PI. VII, fig.
1). These preliminary steps completed, a rearing cage containing
adult white flies was placed over the leaf and allowed to remain the
length of time desired, usually from 1 to 24 hours, with preference
shown the latter number. The cage was then removed and an
empty one put in its place. By this method all doubt was removed
as to the period of time over which deposition took place. As there
is scarcely a leaf in a grove infested with the citrus white fly that
does not bear from a few to many eggs, this becomes an important
point and failure in its recognition has led in the past to statements
greatly underestimating the minimum duration of the egg stage dur-
ing the warmer months.

TABLE V1II.-Duration of egg stage of the citrus whitefly.

Record number.


1 ............................... F r. 23
2 . .. .. .. .. .. .. .. .. .. .. .. ....... . %I al'. :1
3 ............................... .\ )r. 211
4 ............................. June 16

5 .............................. July 17
S.............................. J3 ly 19
7 ............................... A uig. 1
8 ............................ ... Se p. 1] -21
9 ............................... Sept. 21

First egg


3' ;0


26 0

JilyI 21i
July :Ii
Aug. 11

Sept. 26
Oct. 2

Last egg

I)er ofi
(I ays.

I Out)

Mar. 24
M:iv I;
June :30

A. ig. 5
,iJilv 31



n11111r -
btr ol

...... 21'
] i



Daily rate of hatching in percentages with degrees of tucim'ilalIive effTt'- iv it'rnlIer:itvIrPs.

OF. I 9
I F. |days.


OF. 10

0 ....... .i 0
0 ......i 0
0 ...... 1 0

0.1 371 24.3
0 .... .; 0

1. 9 318 1.9
0 0







11 0 Ff.l l 1 S.'
days. ,ala's. la's.
I -
Ii I







i 424

S .. ..
51.0 4 "7

3'., 4. . 4
.'); I ,5. 9

U ... ...
6. 2 4 12
11.7 4,2

o I
0 .. . .

1 ar





:1). 4 4,41 4.2
i. 4 15 9.5

Daily rate of liitctiinn 111i percentlaget with Idetrees ofaceiirniiltairveelTectIi ve lem pera ttiirer.

RI record number.

7 .... ......... ...... ... .... ....
8 .. ...... ......... ..... .. ...
8 . .. .. ..... .. .. .. .. .. . ...






" 117









" 0"

I F.




OF. 211 F. 21 o-F. 22 OF.
Sdays. days. lays. 0

li .... .. ...... .... ..
51S 0.7 .541 .......S S ,y( l
0 ......
:. .. . 0. . . .. . ... I . . .. . .. .
I......... ..
77 ...... ...... ... ..... ..........

23 O. I 24
flays. tl'dys

I,.m j.


1 The thrips AI.turodothrilipsfsciu/tclinnis Franklin is responsible for d(.aitli of 9.5 per ccnIt of the eggs.


oJ.- .*

4: ":

=H-1 i
eLI 7


0F. 7



Daily rale of hatchintz in i)ert'entage, with de,.z, rees of at.cttmulat i\e etlect i\'e temper.' ture. .



I t-trdl.- inchi-le eg.:,, hul llhin;i mt i1- n' i ,o tih v.


The conclusions presented herewith are drawn from the data pre-
sented in Table VIII, based upon daily observations of over 5,000
eggs (lelposite(l at intervals from February to October. From these
and other data not included, it can be stated that the eggs hatch dur-
ing a period of from 8 to 24 days after deposition, according to the
season of the year. While there are no data regarding the length of
the incubation period for eggs deposited by the few female flies occa-
sionally seen during the winter months, as noted under seasonal his-
tory, it is probable that hatching extends over even a greater number
of (lays during the winter season. The deposition of such eggs is,
however, a comparatively rare occurrence and will receive no further
mention here. In no instance have eggs been seen to hatch before
the eighth day from deposition, even during the months July and
August, 1907, when the average mean temperatures were slightly
above normal, while under the most favorable summer weather con-
ditions from 75 to 100 per cent of the eggs hatch during the period
from the tenth (lay to the twelfth (lay from deposition.
In general, the warmer the season the shorter and more nearly uni-
form is the period of ecgg 'development or incubation. During the
months of July and August, when the normal monthly temperatures
at Orlando range from about 72 F. as the mean of the minimums to
about 93 F. as the mean of the maximum records, practically all the
eggs hatch from the tenth day to the twelfth day. Even at this most
favorable season, in one instance hatching was delayed for 19 days.
During the somewhat cooler weather of late September and early
October an(l the decidedly cooler months of February, March, and
April, hatelling is more or less delayed according to the prevailing
temperature and is scattered over a larger number of days. This
same result is brought about, only in a less degree, by a cool period
occurring in an otherwise warmn season, as shown under record 4
(Table VIII).
Reference to thle daily rate of hatching in Table VIII, and. to the
accompanying degrees of accumulated effective temperatures, shows
that regar(lless of the time of year deposited and the number of days
required for incubation, over 90 per cent of the eggs, on an average,
hatch between the accumulation of from 375 to .475 of effective
temnpera tuire.
Exception to this statement must be taken in records 1 and 2
(Table VIII). The number of degrees of effective temperature
required seems to be greater at this season, although this might not
prove to be the case if, as is probable, an error has arisen from using
43 F. as tihe basis for calculating the effective temperature.
Reference to the two preceding tables shows that considerable
variation exists in thlie length of the egg stage among eggs deposited
on the same day, or even within the same hour, and subsequently



sul)jected ti idcillitial ('<)ii(litiIm.s of IIIai ait d ,1i.sti re. ,vei wlilhn
ha1'tchiing wa.V I m st concen('011 tra0tedl dlrin'il,. tlhe ]he1 at o1f s('11,r" iiinn r 1,11 1.!)J.s
lper centlt ol the eggs hatched ol tihe t'iillt 111 id Iventhl days from
date of dlepisition, liatchring extended over a itpe !(ri (d f frn to 19
days. 1 latcjting over a period of from (n to 7 day-s after thle first
crawling young aj appears is 1an ordinary occurrence dimuring the .oo, Ier
portions of tlie se,,lason of activity. I t is respect vli(\, i em* s a1re
markedly different .from the eggs of llmo(st, t)tlher insects delposited in
batches which usually hatch within one or, at the most, a few hours
of each other.
The existence of parthenogenesis among aleyro(lids was first rec-
ognized by tlie senior atithor 1 in connection with his investigations
of the greenhouse white fly (A. vaporariorutzm). His prediction at that
time that this method of rel)roduction would ultimately be proved to
occur among many if not all the species of AleyVrodles has been strength-
ened by the results of the l)resent investigations. While there are no
definite data to the effect that parthenogenetic eggs are (le)posite(l
under natural conditions, there is practically no doubt that such (del))-
sition. does occur, especially by females not yet mated or by females
appearing at unseasonable times or when males are decidledlv in the
minority. Scatteredl females emerging during the winter, or resulting
from the comparatively few pupe surviving fumigation, either never
have the opportunity to mate or deposit many of their eggs before
such opportunity presents itself.
That virgin females of A. citri, emerging from pup"e kept separately
in vials, and later confined in rearing cages under normal grove con-
ditions, except for the exclusion of males, will readily deposit the nor-
mal number of eggs, and that these eggs will develop normally and
will produce adults of the male sex, has been thoroughly demonstrated.
Of the five separate cage experiments starte(l with parthenogenetic
eggs, all of 111 adults emerging in four of the cages were males, while
of 208 more adults emerging from tlhe fifth cage, all but 4 individuals
were males; the 4 females emerging under such conditions as to lead
to the sul)pposition that thtey came from fertile eggs overlooked in
preparing the leaf for the experiment.
In hatching, the egg membranes rupture at the end opposite the
pedicel, and then split down each side sufficiently to permit the
young larva to crawl out. The glistening eggshell, somewhat resem-
bling in appearance a bivalve shell, eventually becomes shriveled
and loses its original form.
Notes on Some Aleyrodes from Massachusetts, with Descriptions of New Species.
Psyche, April, 1903, p. 81. Technical Bulletin No. 1, Mass. Agr. Exp. Sta., pp. 31-33.



Proportion of eggs that hatch.-Observations covering many thou-
sands of eggs, both in the cage experiments and in the grove, have
demonstrated that the number of eggs that fail to hatch is too insig-
nificant and has too little practical bearing to warrant the collection
of data on this point. It is safe to say that considerably less than 1
per cent do not hatch. In fact, it seems evident that no egg would
fail to hatch except owing to the dropping of the leaf or unless sub-
jected to attack from without. In many instances failure to hatch
can be directly traced to attack by several species of insects and a
fungous parasite.
Effect of drying of leaves on hatching.-In 10 instances leaves bearing
many thousand eggs were so placed that the eggs were exposed to
direct sunlight or to partial shade, and although frequent observa-
tions were made none of the eggs were known to hatch. In general
the (Idrying of leaves to which eggs are attached prevents hatching of
all except those eggs containing nearly mature embryos. This fea-
ture is probably common to all aleyrodids, since the senior author
has noted a similar occurrence in the case of the greenhouse white fly
(A. vaporariorun ).
The larva-e are thin, translucent, elliptical, scalelike objects, found
usually on the underside of the leaves, though more rarely upon the
upper surface. When normally attached to the leaf they are so-
nearly transparent as to be seen with difficulty. They readily
become visible, however, by either bending or rubbing the fingers
along the opposite side of the leaf, thus loosening them and allowing
the air to get beneath them. They then appear whitish (P1. X, fig. 2).
So very inconspicuous are the live larvi and their attack so unac-
companied by any visible effects on the leaves, aside from the black-
ening of the foliage, that their presence is very frequently overlooked
by the casual observer. A detailed description follows:
First instar larva (figs. 4-6). Length, 0.3 to 0.37 mm.; width, 0.182 to 0.22 mm.
Body flat, scalelike, somewhat swollen ventrally, especially in the cephalothoracic
region; margin entire, with 30 small tubercles, each bearing a horizontally directed
spine of which 6 cephalic -and 4 anal are proportionately longer. Spines of second
pair, counting from anterior end of body, arising from tubercles not on, but slightly
posterior to, margin on ventral surface. Relative lengths of the 15 pairs of spines as
Pair 1 2 3 4 5 6 7 10 11 12 13 14 15
Spaces 1-1' 9.-8' H11' 6.5' 5.4' 5-.5' 6' 4 P' 4' 18' 5.-4' 18.5'
'The larvae and pupae are frequently called by many growers the "egg" of the
white fiv. This misapplication of terms should be discouraged as it leads to unde-
sirable confusion when referring to the various stages through which the white fly
passes during its growth from egg to adult.


No marginal wax fringm appears lIefiore ior after crawliiL' y,',tug -e.li1e4. 1' e. ,al,,-
thoraci and Ihianrucic artitilatliiins invisible; ,S (ir Iiss.ilblv : :il,9lliiiiinal ei.L t,IInl.lt are.
seen wilth lit Ie d illIlIv. Seginents at peisterrior end f id,,,y iimilied li" va.if,,rn
orifice. latter inarlv seni.irc'lar in outline, .mm)IIvinal
longer than wide, lirdIrend laterally l, chitiniius thicken-
ing,- which tip not meet peisteriirly; l. urcullli, seni'irc-u- j
lar, nearly eq tiualing in size tIhe vasifirm onrificei itself f. (.4 ver- --
ing the' lii'la and hearing ,n its median Ixsttri,,r iargin I'I.
what appear to I)e two pairs if small spins, the pe'nulti- / \,
matte pair (if which is about twice as long a. the ultiinmate.
ligusla darker in cohir and broadly L
Screscentiic in shape. ( in either side L
of, and slight ly anterior to, the vasi- -_ i.
form orifice is a short backwardly A -"
S directed spine arising from a small \
!i tubercle. Thetwopairs if rounded, ) ..
simple, reddish-brown eye(s, hle'ss
than 0.01 mm. in diamneterand 0.096
--- mm. al)art-a dorsal pair and a yen-
j tral pair- are situated mesad and
slightly anterior to the fifth pair of FiG. 4. Theciirus white fly:
Smarginal spines, the dorsal pair be- C'rumling young: firsi in-
Snea r t a d stardorsal view. G;really
S ing nearer the margin and slightly enlarged. i Original.)
_. \anterior to the ventral pair.
/, iAntennap, legs, and mouth-parts on the venter. Antennae
/ anterior and mesad to the anterior pair of legs, 0.1 mm. lona,
FIG. 5.-The citrus white very slender; apparently 4-segmented, articulations between
fly: Crawlingyoung; first the segments seen with difficulty and frequently that be-
instar, ventral view. tween the third and fourth entirely wanting, while in a few
Greatlyenlarged. (Orig- ?n
al specimens the second segment appears to be divided into
two parts: Segment 1 short, stout, fleshy` segment 2 one-half
as wide and twice as long as segment 1; segment, 3 narrower than segment 2 and abnut
four times as long; segment 4 very slender, less than one-h?!f as long as segment 3. and
bearing on its proximal posterior side a minute spine, and distally a ,un.g spine. Legs
short, moderately stout, where ex-
tended about one-third the width
of the body; coxme very short and r
stout, the two posterior pairs on /t
the posterior inner side with a
moderately stout spine about
equal in length to the diameter of /
the coxie and directed backward
and inward; trochanters distin-
guished with difficulty, about / )
one-third as long as wide and
collar-shaped; femora more elon- /
gate. slightly tapering distally. /
about four times as long as tr,,-
en FIG. 6.-The citrus white flyv: .Antenne and left hind leg,
chanters; tibie much narrower, first instar. Highly magnified. (Original.)
somewhat longer than the femora.
with numerous short bristles, two on the outer proximal portion longer and more easily
seen. on the outer distal portion with a long bristle forwardly directed and curving
inward toward the tip of the tarsi; tarsi short, ending distally in an enlarged disk-
like process.

.... ..


Midway between the anterior pairs of legs in the middle of the body is the fleshy
mouth papilla from which arise the mouth setae, at first when bent backward reaching
only to slightly beyond the posterior coxae, but later becoming more elongate. Ante-
rior to the mouth papilla is the semiovate prostomal plate, extending anteriorly as
far as a line connecting the antennae, and divided longitudinally by two curved
sutures into one elongate median and two shorter lateral pieces. At the anterior end
of the prostomal plate is a pair of small papillae, each papilla
bearing a small forwardly directed spine.
On the venter beneath and to the side of the vasiform
-:-* / .orifice is a pair of spines arising from small tubercles,
normally directed backward and outward, equal in length
"''.- / to the distal tibial spine.
Second instar larva (fig. 7).-Length, 0.37 to 0.43 mm.;
4' ^"-Jwidth, 0.24 to 0.29 mm. Broadly ovate, dorsum densely
\__ Erugose, all marginal tubercles and spines wanting except 2
Scephalic and 4 anal, the three pairs, counting from the
cephalic region, giving the relative lengths: 1 2 3
In,. i 9.5' 4.5' 10.5'
Eyes smaller and less regular in outline than in the first
FIG. 7.-The citrus white instar, but distinctly evident. Antennae greatly reduced,
fly: Second larval instar, unsegmented, directed backward and slightly outward,
ventral view. reallyly en-
larged. (Original.) tapering, reaching nearly to base of first pair of legs; on
inside near base with a distinct spinelike projection, and on
basal portion with numerous roughenings; legs almost rudimentary, reduced to short,
stout, fleshy processes without distinct segments, composed of a very stout, tapering
basal portion, and a comparatively small, rounded, thick terminal disc; the second
and third pairs of legs on the inner side at the base with a minute spine. Mouth
parts as in previous stage; prostomal plate anteriorly indis-
tinct and its pair of spines wanting. Spines on either side
of vasiform orifice, both on dorsum and venter, as in first
instar. A marginal pore, on either side of body opposite
base of first pair of legs, and formed by an upward fold of the
integument, becomes very evident in this instar. &
Third instar larva (fig. 8).-Length, 0.62 to 0.78 mm.; width,
0.43 to 0.58 mm. Very similar to second instar but larger;
the most striking difference presented by the antennae, which
have migrated backward so as to arise from a tubercle
slightly anterior to base of first pair of legs. Antennae im-
movable, directed mesa for about two-thirds of their length, FIG. 8.-The citrus white
and then suddenly doubled backward so that the distal third' fly Thirdlarvalinstar,
S1ventral view. Greatly
lies in the same plane as the basal portion. Legs smaller in enlarged. (Original.)
proportion than in second instar and prostomal plate less de-
veloped, but the marginal pores and anal cleft more fully developed. A waxen
rod is seen often protruding from the marginal pores. Relative lengths of the mar-
1 2 3
ginal spines: .
3-4' 2.5' 4.5

The introductory remarks regarding the general appearance of the
larva apply with equal force to the young pupa (fig. 9, a, b, and c),
with the exception that the pupa is larger, being nearly one-sixteenth
of an inch long, is more easily seen, and on either side of the thoracic |
region 3 distinct curved lines representing the outlines of the legs



art vrv dist inct. As tll puptI ItIec ,lues oiidir it I ituiii's tli'ker,
o11M 1' rottidledl a d ( (l lI ILI tl doiittlies. tf' lih eigs iIi' obsc'ell ,d
lv tlle )IlittIlt' ts ()r tlIIv l d)(Iv. At tlint' Il [Ij111AIlI of aIn ltI ril I bIriglt
retl (Ir t raI.izit' spot devells nt l' I( back, anld fri- 'II llin.I io .i l(''it ,lays
hlfIe, ,viIirgtiictt' lite eyeTs ol tlhe atult lbCniIt, e 'i.lsilId,. A dli'tailel'i
descriptitI1 is as follows:
L.iglh, 1.10 InInmm to 1.40 nIm ; width, 0.(0 Tnn ti 1.0 n1in inboly Ir,,: llv 'llili,'al,
thlin. iit raised from lnIaf on vertical wax friniig, ',;I,,'r pialv' vellwi,.-g .re, 1n c miing
mntre yellowish and thicker onm approaching maturil ty; thiciraic 'l,.s, n,.jr.,.c, irig ,iii 111-
lines otf the three pairs of legs, and a linte exlendin.,, frim nI btwwe.n first lw, pair. i-f
legs and from the vasifiorrn orifice to edge of lvly distlinl'ly l inure Vllu-i.-li, as are also
the lines representing the union of the lidy segmeni.t altlioigih lh.04-m lasI arv r',ini-
nent. As body thickens thoracic lobes beomine less dislinci r td I' I,,l (i" ',netnl.ts, a
bright orange or red medio-dorsal spot devel ips
at anterior end of abdomen, and later, a few .
days before emergence, the purple eyes of adult
become very distinct, as also do the white devel-
oping wing pads; rimni of vasiform orifice brown or "" '""'
yellowish. All marginal bristles lost, except one
anterior and one posterior pair of minute bris- A. -
tles. A low medio-dorsal ridge or carina and
corresponding depressions on each side extend "I -" j
from the head to the anal ring, traversed by short 1 J'
transverse ridges on the thorax and abdomen. 4t:'\ P "' .,"'
terminating in a low subdorsal ridge hardly per-
ceptible; from these last numerous very fine -
granulated strike radiate all around the body to FI. 9.'-The citrus while Ily: a, ventral
the lateral margin. A short transverse ridge aspect of pupa; b. v-aiforni orifice of
appears near posterior margin of head with a same; c, margin of body of same. a.
i .i <'reafilv enlarved; b, r. higihly magnified.
curved impressed line in front. A minute brown really enlarged b, r highly magnified.
tubercle at the anterior end of the subdursal ( )
carina is sometimes to be seen. From a pore at the edge of the body, between head
and thorax and top of anal slit, issues a very fine, glistening-while, curled thread
of waxen secretion. These so-called "pores" in margin if the cephalo-thoracic
region are formed by a slight upfolding of the body which extends from margin to
cephalo-thoracic spiracle and forms an outlet for secretions from same. Location of
spiracles and respiratory system as already described for alevrodids. Legs and
antennae easily seen with high-power lens. Antennae located as shown in fig. 9,
partially concealing front pair of legs, apparently 3-segmented but division into seg-
ments not distinct; last segment as long as other two combined, with quite a number
of irregular annulations; tip provided with a stout spine. I.egs short, very stout,
especially the two posterior pairs; front legs projected forward; all without distinct
segmentation; tarsus very short, stout, anti rounded. Vasiform orifice nearly semi-
circular (for details and shape see fig. 9, b).
Pupa cose.-White, firm, retaining definite shape, and remains firmly attached to
leaf unless forcibly detached. (See Pl. VIII, fig. 1.)

T Ill': fi'l 'l \V IIIT I,- I,'1 I.I 1"1.. lii IN A MI I.\ll .

*:: t ",... .. ' lii i"i~i "i: '" ". ::? : ::... : :: " *!
,,, m




Data upon the duration of the larval instars nave been secured by

daily observations of over 300 specimens marked as soon as the

young larvae had settled, supplemented by frequent counts of several

thousand specimens in various life-history cages.

From these records those included in Table IX have been chosen

as representative. A study of thes6 will give a very accurate knowl-

edge of this subject, and will impress upon one the considerable

variation in the duration of the several instars of larvae hatching*

at the same time, feeding upon the same leaf, and consequently sub-

ject to the same weather conditions. The data also emphasize the

retarding effect of cool spring and fall weather upon the length of

larval life, although this has not been found to be as great as many

have thought. The. period of larval growth ranges from an average

of 23 days during the warmest months to an average of 30 days

during the cooler months.

TABLE IX.-Duration of larval instars of the citrus whitefly.

Period of growth.

Mar. 22-Apr. 26..
.....do ...........
... .do...........
... ..do ...........
.... .do ...........
.... .do ...........
....June 2do.-July...
June 26-July 18..
June 26-July 19..
June26-July 16..
June 26-July 18..
June 26-July 20..
June 26-July 18..
Junme 26-July29..
June 26-July 18..
June 26-July 21.
June26-July 18.
June 26-July 16..
June26-July 18...
June 27-July 19...
June 27-July 20..
..... do ............
June 27-July 16...
June 28-July 18..
J une 28_J u Iy28 --
June 28-J uly 18 --
June 29-July 27...
July 28-Aug. 18...
Sept.28-Nov. 14.
Sept. 30-Oct. 25..
Sept. 30-Oct. 23..
Oct. l-Oct.21.....

Number of
days in-





--- 7'











Sum of




age. -

Period of growth.

Oct. 3-Oct. 27...
Oct. 3-Oct. 30...-
Oct. 3-Nov. 11....
Oct. 3-Oct. 29....
Oct. 3-Dec. 4....
Oct. 3-Nov. 4....-
Oct. 3-Nov. 1....
Oct. 3-Nov. 2....
Oct. 3-Nov. 3....
... do ...........
Oct. 3-Nov. 1....
Oct. 3-Nov. 10..-.
Oct. 3-Oct. 31....
Oct. 3-Nov. 4....
Oct. 3-Nov. 8....
.... .do ..........
Oct. 3-Nov. 2....
Oct. 5-Nov. 1....
Oct. 5-Nov. 7....
Oct. 5-Oct. 31....
Oct. 5-Nov. 8-...
Oct. 5-Nov. 5....
Oct. 5-Nov. 1....
Oct. 5-Nov. 3...--
Oct. 5-Nov. 5...
Oct. .5-Nov. 1....
Oct. 5-Nov. 19..
Oct. 5--Oct. 31...

Number of

Number of
days in-



4June26-Aug.18... 7.2
1Sept. 28-Dec. 4... '7.8






Sum of


10.5 1945.7
14.3 903.4

2 Does not include No. 59.



Does not include No. 37.

1 -11

Bul 92. Buimj ,,f E,.,. ,.-.:y U S D (,.p. ,, ALA T .V ,.,

Fig. 1.-Leaf showing pupa cases of -le:/,r.,hs ;,'tri: also a f.w puiipa ,tn- egrs. Fig. 2.-Under
surface of orange ivif. showing heavy infestation by citrur while fly. Fig. 3.-Leaf showing
pupa casesof . ubif'rrr. Note delicatie tructureascomparedwiththoseuW A.citri. (Original.)



* :

TIE 'Ii:lIL'S WV11TI FrI,"Y" I,-I'F1 II1STOI)IYA AN 11li.11Ts. (6

One of tile most interesting pli ase(s of life-Itistorv stAllis Is Illbeen
the wi'Ide range in till' dluration of tile tiipail stage ; a ani',ge of frlI)n 1{ 83to
:304 ldas. (Considlerinig tilt-e relatively slight variation in tin, length of
ti i larval'life, this range ailiong specill(,ens passing into ti l ))ii(pal
stage at. practtically tihe, same time is remarkable. Ii view ,of the it fact
that. the' effect of this variation uponi the d duration of life idii nuuiii r
tif annual generations will I)e fulfly discussed iiunder th ,ose .'headinigs. aiitl
brought out in Tables XV Iand XVII Itilid figure 12 ,l2. ly I (a vew of the
large number of records on file are given in Table X to illustrate this
range in pupal life during different parts oI the year.
TABLE X..-D)urutiOn oJ pu"pal stage (f thi: 'ci/rus whilc /l/.

N I- I Ni S-l Sum ; of Sp1eci- Nuin- Slliiof
cii Period olg-ro Ihl. 1 Ir of 1'1'01 il'i Pe'riud ofgro %ih. lIr of liliii .g
Nu.th days. tIre No. days. rs.

1 Apr. 30-May 13.... 13 410 11 A-g. 15-Sept. 6.... 22 s
2 Apr. 30-June '2)..... 51 1,833 12 Aug. li-Mar. 1..... 214 .5, 414
3 Apr. 30-ALg. 3 ...... ti4 2,5fi4 13 Aug. l--Mar. 2t).... 21i 5. 479
4 May i,-JuneoL 5 (&A i4 14 Aug. 17-Mar. 25 .... 220 5. 574
5 May 1S-Jily ..... 74 2,866 15 Aug. lS,-SUpl. 10... 23 931
ti May IS-Mar. IS..... 304 ........... If- i AI l g .IS-Apr. 1 .... 22i 5.752
7 Jill\- 15-July 30..... 15 W12 17 Sopt. 310-Mar. 3!... 1s- 2 1 473
8 July" 15-Aug. 4.....I 20 S(S IS Oct. 2.-.Apr. 19.... 17:3 4. 17
9 Aug. 15-Aug. 27.... 1 12 479 19 Nov. I-Apr. 17.... 1;7 4,i',9
10 Ailg. i 5-Aug. 28.. 13 521 20 Nov. S-Mar. 25.. 137 3, 25,

It will be noticed that pupe pass either a comparatively few or a
comparatively large number of (lays in this stage and that ordinary
temperatures and humidity (do not have thie ower to determiine whichi
it shall be.

On hatching from the egg the young larva is provided with well-
developed legs, as shown in figure 5, bv the aid of which it crawls
about the leaf for several hours and then settles and begins to feed.
Because of the aimless way in which it crawls, frequently doubling
on its own course and turning aside for tlhe least'obstacle, it travels
over a very limited area. It is therefore improbable that tlhe crawling
larvae ever leave thie leaf upon which they were hatched, unless carried
on the feet of birds or insects or blown or dropped from one leaf to
another. After settling, the larva does not change its position on tlie
leaf, while with the first molt its legs become vestigial (see fig. 7) and
unfit for locomotion. Larvwe frequently move slightly, especially
directly after or during molting when they merely describe an arc
of 180, using their mouth parts as a pivot. The larva passes into the
pupal stage without materially changing its position on the leaf. The
only time, then, during the life cycle when the white fly is capable
of moving about from place to place is during the winged adult stage
and the crawling larval stage.


Pronounced and striking growth in size occurs only at molting,
when thle soft flexible skin of the larva or pupa is able to stretch before
assuming its normal rigid condition. With each successive molt the
larva greatly increases its horizontal dimensions, until by the time it
reaches the pupal stage these are about eighteen times as great as in the
newly hatched larva. When first settled after molting the larva is very
thin, papery, and transparent, being seen with difficulty except with
the aid of a lens, but after feeding several days it slowly becomes
thickened until, from two to five days, sometimes longer, before molt-
ing into the next instar, it is decidedly plump and whitish opaque in
color. Oftentimes before molting the larva becomes very much
swollen as though gorged with liquid. This appears to be an abnor-
mal condition, since many that become thus unduly enlarged either
fall or d(lie without molting. During the increase in thickness follow-
ing feeding, there is no increase in the horizontal dimensions. On
the contrary, increase in the former is secured at a slight expense of
the latter.

Daily observations on over 300 marked individuals from time of
settling to emergence of adult have conclusively demonstrated that the
larva passes through but three instars before reaching the pupal
stage, instead of four as has been previously supposed. Each larva,
then, molts or casts its skin three times before becoming a pupa. The
process of molting was first described by Riley and Howard 2 and as
observed by the authors is as follows:
In preparing for a molt the insect curves the abdomen upwards at considerably
more than a right angle, moving it also occasionally up and down. The margin of J
the abdomen has at the same time a slightly undulating motion. During these move-
ments the insect is shrinking away from the lateral margin until it eventually occupies
only about one-third of the original lateral space, causing a distinct dorsal and ventral
median ridge. The skin then splits, not on the dorsum, as would be expected, but
either at the anterior end or underneath the head. The head and prothorax are then
pushed out and the skin is gradually worked backwards by means of the abdominal 1
motions, the portion already out swelling as soon as it is free.
As the insect flattens after molting it appears milky white, the head,
thoracic lobes, and abdominal segments being more greenish. At
this time the legs, which resemble much the prolegs of a caterpillar,
are very active, and there appears a pair of fleshy protuberances more
or less movable, not as large as the legs, but apparently of the same
'.This agrees with the senior author's observations on the greenhouse white fly
(A. vaporariorum) and the strawberry white fly (A. packardi), which are the two species
of the genus which have previously been studied in greatest detail. Tech. Bul. 1,
Mass. Exp. Sta. and Can. Ent., vol. 35, pp. 25-35.
2 Insect Life, vol. 5, p. 223, 1893.

'i :"* .
..... ...


TI1 E L'I'CIt'I'VS W1 'rI : 1'L\ A 1.'E 1F11 I V AI IiJ i.lA ill' S. (65

strulctuilre, whiicl ina t its stcking disks to uid taie isi.Lcti itll i''Iiag' ng
itself. Tlltse 1 )'rot iie'1lra e lied' iM 1 1' (ittr wit i1ri'Iwnl SOt 1111 il. 1l Inti)', Olf
lt'in remains. Wihi le e'icoming at ticllied to till, e1,f theli ill,''t 1iiiY
be se''i toccisioilliyiIv to rot lite itself t hiNo' ghli alil iil' f l'2'70", iM the l ieali, -
wlhile fr'q tl raising liii lo)weIrili (r he 1 3bdo11 ii 'I'LI' iist .,kinsl
tiare usually bllowini iWiy ibY the l)'eez' or t'fall inj'i31 tlie leaf as s ill as
ml ti (l, l)Lit l1ot, infrii it ciily aIV Ni* '1 il l oirtillly l)i io ii31(lI'1d I -elt'a li
the body of tie insects. .Molti., 'occ's 11104 a1cti iv (alylrii liwi ours
of lhighli liIniiitv. Newly lllote(1 larvt' i' ILlar a)i111Uida iit rig tlie curly
morning when thie iumiditv ranige.s lietween 1000 anld 1J0'.
i.'EDIiN HAI 'Trs (O- LAIV.i: ANI) I'UI'.I.
As the wliite flies., ()or Ale(hvrodida'v, b)elonli t ile II('miit('ril, (r
sucking insects, tile larva' aid plipaL (do) inot eat tl li i tissiuI(' f llI, 1 lea1f,
but insert their thliread-like iiotithpirts id stick tli plaLit, juices I)y
the aid of a suction apparatus located in thie lieadt. T"h 'ir ravages arcil
not accompanied by any visible effect upon tile leaf itself, bit t llmay
be detected by means of tie sooty mold wilhic develops after thle fly
becomes very abundant. Our only means of estimating tile amount
of sap taken up by tlie insect is by tile allmounlt of waste material, or
honeydew, ejected by it. A first-instar larva, on ljbeinlg watched under
the compound mniscroscope for 20 consecutive minutes witi tl he teim-
perature at 90 F., was seen to eject ho)neyd(ew 4S times, or ian average
of about 10 times every 5 minutes. A pulpa with well-developed eve-
spots, in March, with the temperature at 05' F., ejected hioneydew
4 times in 5 minutes. This difference inll thie .amount of lIhoneydew\
secreted is due in l)art to the different temperatures at wlich tie
observations were made as well as to thle difference inll tihe degree of
A very interesting observation on tlie amount of sap extracted by
larvie and pupae of the white fly lias been nade by Dr. Berter
Leaves with live larvae and piupte were placed between glass plates so
that the ejected honeydew was collected on tlhe glass. By weighing
it was found that each live insect had excreted about 0.0005 gltraim in
48 hours. At this rate a tree infested with 1,)00,000 white-fly larvae
anti pupae would lose one-hlialf pound of sap) per dLiy.
The adult citrus white fly is very small, measuring only about one-
sixteenth of an inchli in length, and with a wing ex)pallse of less than
one-eighth of an inch. The natural color of the body, anltennail, legs,
and wings is entirely obscured by secretions of delicate whiite wax par-
ticles, so that the insect appears snow)y white (Pl. IX; text fig. 10, a-i)
Bulletin 97, Florida Agricultural Experiment Station, pp. 63-64, 1909.
86S50-Bull. 92-11-- 5


without spots or traces of darker shades upon the wings. Only the
purple eyes are free from the white wax, and are in sharp contrast to
the color of the rest of the body. A detailed description of the adult,
by Riley and Howard, follows:
9 .-Length, 1.4 mm.; expanse, 2.8 mm.; four-jointed rostrum about as stout as legs;
joint 1 shortest, joint 2 longest, and about as long as 3 and 4 together; joint 3 some-
what longer than joint land a little shorter than 4. Joint 1 of the 7-jointed antenna very
short, as broad as long, subcylindrical, slightly wider distally; joint 2 twice as long asl, 1,
strongly clavate, and at tip somewhat broader than 1, bearing or 4 shorthairm arising
from small tubercles; joint 3 longest, about twice as long as 2, slenderer than this and with
a very narrow insertion, rather abruptly stouter at apical third, corrugated and ter-
minating above in a small callosity resembling a similar organ in Phylloxera; joints 4

-^ >---- G^.t, ;:

: [: a

FIG. 10.-The citrus white fly. Adult. a, Male; 6, claspers of male; c, female; d, ovipositor of female;
e, sidle viu.\ of head of female; f, antenna; g, enlarged margin of wing; h, tarsus and claws; i, tibia.
a, c, Greatly enlarged; b, d-i, more enlarged. (Adapted from Riley and Howard.)
and 5 subequal in length, each nearly as long as 2, joint 5 bearing a short spine ante-
riorly near apex; joints 6 and 7 subequal in length, each somewhat longer than 2, 7 with
a stout spine at, tip); joints 4 and 7 somewhat corrugate or annulate but less so than 4
api.al third of 3. The 2-jointed tarsi about half the length of the tibia, joint 1 of the
hind tarsus bearing 6 rather stout spines on each side; joint 2 supporting at base 3
rather prominent claws, the middle one longest. Ovipositor short, acute, and retrac-
tile. Eyes divided into two by a curved pointed projection from middle of cheek,
the upper portion being smaller than the lower portion. Wings clear, colorless; costa
delicately serrate. General color, light orange yellow, tip of rostrum black, tarsi and j
part of tibia orange.
S.-The male resembles the female in all important respects except in being
smaller. C'laspers about as long as preceding abdominal joint, or one-fifth the length
Riley and Howard, Insect Life, vol. 5, p. 222, 1893.

r- ... 4"

Bu 1 9 BLIurr,-.u of Entlir' g.l.'.y U. S. Dpt ,f A" u Iu'nPI,..

Fig. l.-Tender growth swairniin.g with aduliit. Fig. 2.-Leaf of same unillargd. (Original.)


.~~~~ ~ ~ ~ ...... ...

TILE CITRUS WlII'l, l'1Y" I lFI.: II HIST(ulY .\NDI) lHAlltS. 67

ot the abdomen, curved gently upward an1 I inwardl, ac'i hliarinl, .1 .ir .5 e'pihilit.itajit
KmiiZuto cylindrical pililfertis tiil.rrrcrs oin upptij r alnld niilir ,dg' ; nIm 1v alIii,.-il :. Ig
as chla tpers, rally r islulter at lb.tse, more shleinler iwaril fill, IH-'rilliilaliInI. im atill s.pi,.
at upper end. head and abtdmenli wilh l 'heavy tufls otf wax mni, afller i..,tiii g friiin
Examination of a large numb11e Of111 ante unnta' sllhows Ilin1t till- '''lalive
lentitli of ti t' anten t il1 seg i'1t lt' ts is .il c't to sliiglit V\1'ritti tis. T 'Ite
avet'ratge r'ltive Igtlils are alboutt, a s fIollows"
Segment 1 2 :3 -* 5 6i 7 Sp)in'.
spaCes l' )10, 21, .!3, 1, li IS, 22, 3
Altlhourgl tlt'e liave examtiined tliousatids of males tli at. and for
some time after emergence and as tliy ocu'tr at aIll tiliis tllrouglount
the grove, tlhe authors lave never been able to ol)serve males \witli tlice
tufts of wax on lead and adl)omen mehtitniid in t.lie above ldescrip-
tion anti illustrated in conlllect.io)l witlh its original publications.
The emergence of the adult occurs soon after its purple eves and
folded whitish wings can be seen tlistinctly through the pupal skin.
About 20 minutes before the putlpal skin is uIl)turled the hotdy of tihe
adult shrinks gradually away from it and assumes its natural shape.
This gradual shrinking away from the edges of the pupa, andl tlhe
accompanying thickening of tlhe body, brings a pressure to bear on
the pupal skin which causes it to split from margin to margin between
the thorax and abdomen and along the median line from this
rupture to the anterior margin. Through the T-shaped opening
thus formed the insect first pushes its thorax, then its head, with
little apparent exertion. The body now projects almost lperpendtIicu-
larly from the pupa case, as the pupal skin is called, with the an-
tennae, legs, and abdomen still in their pl)upal envelopes. By a series
of backward and forward movements the antenna' and legs are freed
from their membranes and are in constant motion. Tlihe abdomen is
now so nearly out of the pupl)al case that thle ly is practically free,
holding on only by means of thie end of tlhe abdomen. With a sudden
forward bend of the body the legs are brought in contact with the leaf,
and with their aid the fly frees the rest of its abdomen and crawls
away rapidly.
The period covered between the time the insect ruptures the 1)upal
skin antd the time it becomes entirely free from tie case and is crawl-
S ing is from 7 to 10 minutes. Not infrequently flies dlie touring
1 These represent the spaces read on eyepliece rnmiromter when 1-inch eyepiece
and t-inch objective are used, and the miscroscupe tube is drawn to 160.




Immediately after emergence from the pupa case the adult differs
from the more mature individuals in that the lemon-yellow color of
the body is not obscured by tlhe white waxy secretion that subse-
quently appears. Also the wings, which appeared as crumpled
whitish pads when the thorax was first protruded from the pupa
case, have had time only to partially expand. As the fly crawls away
from the case thie wings are held perpendicularly above the back, but
as the wings gradually unfold and assume their normal shape they
are lowered to their natural position. It requires about 7 minutes
for the wings to become straightened after the fly leaves the case,
and from about 14 to 17 minutes from the time they first begin to
expand. When fully expanded, the wings are colorless and trans-
parent, with the costa pale yellowish. The powdeIry whiteness so
characteristic of the flies as seen in the grove gradually appears as
the wax glands secrete their particles of wax. In about one and three-
fourths hours the wings and body have become perfectly white.
Aside from that inherent influence affecting the development of the
citrus white fly and determining whether thle adult shliall emerge
during the first or second general emergence period, as hereinafter
described under "Seasonal history," many field observations.made at
all seasons during the past three years, supplemented by laboratory
experiments, have emphasized the great influence which temperature
hlias on emergence. While a normal amount of humidity is necessary
for emergence to occur, it is not so controlling a factor as temperature
during or(linary Florida weather, as will be shown later. Light also
seems to affect emergence under certain conditions.
Conclusions drawn from field notes, supplemented by laboratory
experiments, show that emergence seldom occurs outside the range of
62 F. to 85 F., with preference to temperatures ranging from 70
to 85. During the winter months of December, January, and
February, when the average monthly mean is about 60 F., no emer-
gence occurs except to a slight degree during warm spells of several
days' duration. In January, 1906, when the average monthly mean
temperature was 59.6, or practically normal, no flies were noted on
wing at Orlando, Fla., except in small trees beneath pinery sheds
whIere the temperatures averaged several degrees higher than outside.
During late December, 1908, and early January, 1909, the tempera-
ture had been sufficiently high to cause a limited amount of new
growth to appear on some trees in Orlando, and on January 4 a com-
paratively large number of adult white flies were seen feeding and



depositing. eggs oI. new 'rowt 1 in a very shell t.r. pi'. in1 v.riv,
mean teinpeat, tare of llit (i tnys J)reedli Ig this IIsiV'rtioil W. was 4 16
F., while, for thie 6 tda'nys piredci 'li|I|. liI1.,,,, W t 1 jli t, wliit, lit.s \MV ,'
on wingp, tileo vtI* i aw s al \'A).sI btI ;S.) Ail NIlt. wlit fli,-s
were not st'ii inll tine laboratory g'vc ii lI'lnrIuI, I90)), iintil ,lIit
the 2,0thl, or unItil tthe I," emnInra reortds foI'r tlin grovy skioww', Iil
ave-rage daily 1mean1 of al)utt (64.5'. It is fromn tlihe aovt facts'; tliat
the lowest telp)eriatlir. a"t wlitNilch emieirivi'1, occurs li;ts Ien dtel.r-
mined to be about 62'0 F. I'his cmnclusion, IIiwrawn i',|n fm genni.l tic.]I
observations, is streigthenmed bly en iergeict'i'e rIT',rdIs kept, in .,o-
ne('tion w\it li (are lii[c-liistor\" work dIturitng tle periodp at' ti'* e sj iiin
emergence of MNarclI, S .1)8, when I tlu' n IItlil inen i Wis 71. lI'f1wr-
ence to tlhe data contained in TIalle XI l)rings ou(t tlie 'ic.t tlnt while

TA TBLE X I.-- lt'lotion oJ" 1rmi'1 rnraI r' IrE1 o rmr v'i't'ertqf" Ilbc t'ir ii fbit /;.l.

Range in A average
Date. tempera-r;-, im'an eit-
Sluire. perature.

1908. 0F. OF.
Mar. 19 622-S9 75.5
Mar. 20 63-NH) 7o).5
Mar. 21 54-i-45 59.5
Mar. 22 diO-Sl 70 0
Mar. 23 Co.-!K) 7S.0

1ni'Trenct''Q rovr Is.

No. 1. No. 2. No. 3.

4 .........
5 18 7S
0 0) 1
10 15 6;9
17 35 71

emergence had been going on actively two (days before and after March
21, when the average mean temperature was about 75 F., a drop) in
the mean temperature on the 21st to 59.5 F. practically prevented
any white flies from emerging. The 1 white fly that is recorded
under Nos. 3 and 4 may have emerged on the 20th after thle (daily
record had been taken. Such eniergences are not rare at tis season
of the year, as will be shown later. Of 2 lots of abIout 100 p1l)lpV
each, from which adults were nearly ready to emerge, 1 was l)lacedl
in a refrigerator at about 56 F, and the other kept at, room tempera-
ture which ranged between 70 F. and So0 F. while emergence was
taking place. Of those kept on ice, but 1 white fly emerged
during the first 12 hours, as compared within 17 from pupaw kept
at room temperature. White flies continued to emerize on 3
consecutive (lays from pupw kept at the latter temperature. Noll mo(re
Emerged from the refrigerated )ti)a'. It is thlerlefore evident thlat
emergence may occur at as low a temperature as 56 F., though very
That white flies seldom emerge after the temperature reaches 85
F. may be concluded from the following facts: During the months of
July and August, when the average daily mean is about 82 F., a
newly emerged adult is rarely seen in the grove after 8 a. m. Prac-

. :.. .
H :

No. 4.


tically all adults at this season emerge between 4 and 7 a. m. Thisis
true both in the laboratory and in the grove. Of 233 white flies
emerging separately in vials in the laboratory during August, 1907,
212 emerged between 3.30 and 8 a. m., and the remaining 21, with
one exception, emerged between 8 and 9.30 a. m. In the grove over
95 per cent of the white flies emerge before 7 a. m. At this time of
day the temperature ranges between 70 F. and 85 F. During the
early spring, when the daily maximum temperature does not usually
exceed 85, emergence is not restricted to the early morning as during
the heat of summer, but occurs at all times of the d(lay. It may also
be added that like conditions exist in October and November, but
because of differencee in seasonal history, they affect chiefly the spotted-
wing white fly.
Under normal Florida conditions at Orlando, at any season of the
year, the relative humidity rises to nearly or quite 100 per cent by
from 6 to 10 p. m., and there remains until about 6 a. m., when it
normally drops rapidly, sometimes to as low as 19 per cent, though
more often to from 35 to 60 per cent. It has already been stated that
over 95 per cent of the white flies will have emerged before 7 a. m. or
before the humidity has fallen far from the saturation point. That
temperature and not humidity is the more important factor governing
emergence in Florida, can be inferred by a comparison of the humidity
and temperature records of Table XIII. It so happened that the
col(l wave of March 21, 1908, was accompanied by a higher average
humidity; but the temperature and not the humidity prevented
adults from emerging. Again, during the spring, when the daily
maximum temperature is seldom above 85-usually less-emergence
goes on even at midday when the humidity has dropped to as low as
330. In this connection attention should be called to the fact that the
humidity in the corked vials mentioned under the preceding heading
remained at about 100 per cent throughout the greater part of the
There are, however, times of abnormal weather conditions when
lack of humidity seems to play an important part in preventing
emergence. During the month of March there sometimes occur dry
win(Is of several clays' duration, accompanied by more or less heat,
which seriously check emergence, and, as far as can be determined,
cause many pupe from which adults are about to emerge to die.
Two such periods occurred during March, 1909, from the 3d to the
6th, and from the 25th to the 27th, respectively. During these
periods the relative humidity was extremely low, on one clay dropping
to 19 per cent. For 42 hours during the latter period the humidity
ranged below 50 per cent and for 36 hours above 50 per cent. During


TilE CITRUS WVII IT I-N .vIAFE iiI.: iistl MYiv .\ NIP IIA:';:. iw71

these periodtls emerl.tvnce wai s ,%otc-,iI to 1)be sevioilj- I i'yli .kI jini ,l ,ti.
end of the latter ll\marl of 31 i0 per (Tilt 4O liii' ilpWe' \'re dIedI, ip;-
ently from no other catise.
-:Fv .'rc 'r i ." I.I ir i:N | :\t (r,' N, 1 .
During the summer n uontlis light seems. to hlave :n;i infIi,11iu',. ,,ii
emergence. At tilhis seisoi. e ellrgel( 'e ill t1 I1:1) 'i11*1,% l "a,\r \'
begins at about laireak. Ob1serv',tions mae.al, t l1IitiiIl\ iiti.iv:l.
on the emergence ofl 233 a iltils, fro 3.30 aI itld -1 :1. In. s.io\ tl it I I. I% ir,,
flies rarely emler.,ge lt')fore this time. II oIIe i i.t..w1 4.11Cn l( :111Y (IM.-
'third as many1 white flies ('1nergeI from )pula,' kept in 'rII dI:'< ;i 'r,1in
those kept inl thle open, an1d their emergence wvas itNictc;lly dl*(.1: -\Ed.
During thle cooler montlis thle low morning tillne'ratlires J)rE','(x1t tI li
white flies from respondling,. to tills apparent stimizlatioi dle to li..dir,
and they emerge at various times after tlie' tenplieraitl' re lias isell
sufficiently high.
WithoWutfood.-In none of the experiments con(ludted to determine
the length of adult life without food have white flies lived l,,I.,nger tl;,i1
30 hours, and a very large percentage has died 1)efore th enild "i) 24
hours. When confined on leaves of plants other titan tl).hose, r(,)ecognizedl
as food plants, life is usually longer than this, lhut never appiro.ies
the normal length. White flies confined on crape myrtle in July lied
as soon as those kept in empty cages, hut flies caged on oak, in lMa rNhil,
lived as long as 4 days; those on fig, in August, 3 days; anl ,,hn an;min
shrub, in July, 2 to 3 days. In all these tests flies were placed only
on the tenderest growth.
With food.-Adult. life under normal outdoor conditions averages
about 10 days, although individual white flies kept in cages have l evin
known to live as long as 27 days. Adults are so fragile and so easily
killed by winds and heavy-v showers and by numerous species of sp)iders,
and ants that their duration of life is at most very uncertain. (Cage
experiments (Iuring March, April, July, August, and Septenl)er show
that, in the cages at least, there is little difference in the length ( f
life at various times of the year.
The courtship of the citrus white fly has been observed to beginl
within 2 hours after emergence, and in one instance even before tle
wings of either male or female had become whitened. There is no
time in the clay when the males can not be seen courting the females.
The male appears unable to locate the female at a distance much
greater than one-fourth of an inch, according to Prof. II. A. Gossard.
Observations made during the present investigations show that when
males and females are placedl in separate receptacles and separated


only by a very porous cheesecloth they show absolutely no attraction
to each other. Mating, therefore is not so likely to occur when the
adults are scarce, as it seems to be the result of chance meeting upon
the leaves rather than to such a definite attraction as exists between
males and females of many moths.
Upon detecting the female, the male approaches her nervously,
stopping at intervals, especially as the distance lessens, and swinging
his body about excitedly in a semicircle, the head being ilsed as a pivot,
his wings in the meanwhile opening and closing spasmodically. While
no movement is made by the female, she is repeatedly approached
from many directions before coition occurs. More often the male lies'
alongsi(le the female and courts her in this position, raising and low-
ering his wings as above described, and raising and swinging his abdo-
men from side to side. During these antics of the male the female
remains quiet, only occasionally flittering her wings. While males
may be seen courting females at all times of the day, it is seldom that
one sees a pair in coitu excepl)t late in the afternoon and evening. Be-
cause of the uniformity of color and the ease with which adults are
disturbedd and made restless the duration of copulation can not be
stated with certainty, but it probably lasts but a short time. Experi-
ments to determine the duration of fertility have thus far proved
Virgin females in confinement have deposited eggs within 6 hours
after emergence. In one instance 35 virgins deposited 58 eggs between
5 and 91 hours after emergence during summer weather, with the
temperature ranging from 80 to 92 F. However, even at this tem-
perature single females occasionally did not deposit eggs for over 24
hours. Prof. H. A. Gossard 1 states that egg laying begins at from
IS to 30 hours after the emergence when the temperature ranges from
65 to 75 F. Laboratory tests have shown that lack of fertilization
does not prevent a female from depositing eggs, but that she will
readily deposit infertile eggs until opportunity for mating presents
If not numerous, the females deposit almost exclusively on the under
surface of the leaves, laying over 75 per cent of their eggs on the half
of the leaf bordering the midrib. It is only when very abundant
an(l pressed for room that they deposit eggs thickly over the entire
lower surface and more sparingly on the upper surface, the petioles,
and the stems of twigs. Next to the portion bordering the midrib,
the natural depressions and the curled margins of the leaf, especially
'Bul. 67, Fla. Agr. Exp. Sta., p. 609, 1903.


of the tender growtll, lre fa'Voriite plae's fi ,oviplsition. ai, lnot
infrequently as high as 4(0 petr ,,celit 4f tilt' eggs are there liid, i ve,
when the adulults are not very ailt ut Iuat. .AIIlt Iu nugh egs mnny beI,
deposited along thle leaf nanrgill, it is seldoI1Ml tliLat tllhe" a'e ii otI tIM,
margin itself, aIs is tle case within tlie (loudI-winge!d \hl1it< pir. 1ven
when not crowded for leaf splac(', ti(e adults sotiet imi1's settle iilpoi tlIi,
under side of VtNU g fruit, where tle" de entil feed.

Previous to tliese investigations lno data liavc, y eell I)iI lishiHle(I oni
the dailyy rate oif oviposition. In obtainin'g tlie (latia giiven iln Tble
XII, tihe females recordedtl were collected at rancd(mn tl iiriogilout tie
grove, without regard to age, and, together with males lo)t meltioitnedI,
were caged over leaves cleaned of all eggs aid larva' of the w1ite
fly and allowed to remain tlie recorded time, when tlhe atlult s were
removed and the eggs counted.

TABLE XII.-Daily rate of oriposition of the citrus i-hit e./lq.

A average
Number Duiration Nilmlher linihr \Average
Record oreJ~stdof Io^'\0 egrss ofegp er "lle.111
o Date deposited. oegs of eggs per tiean
No.of I of egg I depos- female Itempera-
No. females. laying. ia fe tiire.
I per 24

Hours, o "*.
1 Feb. 23-24, 1909....... 14 21, 197 13 74. 5
2 Apr. 11-15, 1907....... 3 103 9S 7. 6 iCo. 2
3 Apr. 20-21, 1909....... .50 241 454 9 1 77.2
4 Apr. 21-22, 1909....... 40 24 405 1(. 1 71. 2
5 June 1(-17, 1909....... 30 24 360 12 ,s2
6 July 17-1S, 1907....... 255 21 2.533 11.3 '2
7 July 22-23, 1907....... 105 24 1,211_ 11. i i
8S July 24-26, 1907....... 50 46 1,331 13. s i %4
9 Aug. 17-18, 1907 ....... 70 24 5 11.5 "1
10 Sept. 21-22, 1908 ....... 35 24 405 11. 1; 7'1

Number eggs per day per female, grand average. 11.2.

The generally uniform results obtaine(l in tle nine recor(ls wlen
the average mean temperature was abl)out 75 F. or above, together
with the grand average daily rate of oviposition for individual
females whose age was definitely known, as shown in Table XIII,
indicate that each female normally deposits on an average 10 or 11
eggs a day. Varying degrees of temperature above a daily mean
of 75 F. do not correspondingly increase the number of eggs
deposited. However, temperatures below an average mean of 723 F.
(estimated) have a distinct checking effect upon oviposition, as shown
by record No. 2.
Notwithstanding the general average number of eggs per day
deposited by the females of all ages in Table XII, and the same for
the females of known ages for the total number of days they lived,
in Table XIII, reference to tle (laily opposition records in tie latter


shows that as many as 14, 19, 27, or even 33 eggs may be deposited
by a single female in one day. It will also be noted that much
variation exists between the number of eggs deposited by several
different females on the same day and by the same female on suc-
cessive days without any apparent reason, and that there exists no
appreciable difference between the rate of deposition by virgin and
by fertilized females.

TABLE X I I I.-Nuinber of eggs deposited by single females of the citrus white fly.

Fly Dale of first
No. daily record.

1 Aug. 8, 1907..
2 ... .do.......
3 ... ..-do .......
4 ... ..do.......
5 Aug. 15,1907.
6 ....do ......
7 ... do......
8 do......

Condition of female.

+males 12 hours old.........
..... do ........................
..... do.------------------------
Virgin 12 hours old-............
Virgin S hours old............-
.....do.d .................... ....
9 +males 8 hours old .........
.. ...do ........................

Daily rate of oviposition by ii
Fl\" ___
No 13 14 15 16 17 iS 19 20

2 :-- . ......--. .,... -... ... ...-----
3 13 9 5 14 i 10 S I 6 9
4 5 10 2 2! 3 0 8 2,
5 5 20 1 1 2 (3,1 . . ......
6 19 .... 11 6 4 (4 ; .... .....
7 3 .14 12 2 ........
S 1 27 3 i .... .... ... .

Daily rate of oviposition )by individual females.

2 3 4 5 6 7 8 9 10 11 12

8 19 9 22 11 2 15 6 (')...........
5 19 14 12 12 15 13 14 (1...) ...........
3 9 13 5 13 9 16 8 1 15 19 14
5 12 17 4 3 20 5 1 0 6 3 6
5 13 15 14 12 12 15 13 14 10 9 9
5 16 17 5 33 13 23 16 15 14 10 3
8 8 16 14 17 11 16 7 1 4 2 4
2 18 16 14 9 2 13 15 6 9 9 1

individual females.

21 22 23
1 -


"" ("T ""i '(I) ,. ,
.1i: : ::i,:: :i:.::
..... I.. .

2 Dead; 13 eggs in abdomen.
3 Dead; 18 eggs in abdomen.
Numbler of eggs per day per female, grand average, 10. 3.

4 Dead; 11 eggs in abdomen.
' Dead; 7 eggs in abdomen.

The eight records in Table XIII are selected from about forty
similar records on file and are considered as representing an average
condition of oviposition. Although the general average of 10.3 eggs
per day throughout life for the 8 females recorded in Table XIII
agrees very closely with the similar average obtained in Table XII,
there is sufficient evidence in the data in Table XIII to warrant
the statement that the daily rate of oviposition for individual females
is usually greater during the early part of the insect's life and de-
creases with each successive week of existence. Leaving out of con-
sideration the first day, when the flies had not reached their normal
egg-laying capacity, a little calculation shows that the average daily
deposition for the three successive weeks is 12.8, 8.5, and 6.1, respec-
tively. This same decrease in the number of eggs deposited with
increase in age is perhaps better brought out by a study of the number
of eggs deposited by the individual females over 5-day periods.
Thus No. 6 averaged 16.2 eggs per day for the first 10 days, but for


I THE CITRUS WHITE FLY: LIFEJ: lllS'l'CItY ..AN IA l.ll',s. 7;5

the next 5 dropped to an av-evrge (I" 7.S; Nos..I .1,i, 7 .li,,,l ,I
falling off duringg tihe se'jc l 5 dlays. No,. :8 i is ill 14'111 ia;ii, i t111
above stattientIlt, maintaining an average l" ['ruin i.. t, lI. ',s (,
per day for thle first three )lri(iI.s ()f 5 fourth period of 5 days (de.,posited as manlly egs as 'iing tIj le first
5 d ays. It will be noticed[, Jjhwv('er, tliti No. :) deplsited rc I h IIli I'l-
tively few eggs miluring tile early palit of 1er I, li'e. In iew I Ilie fuct
that. the average adult. lift'e is Inly almbiot 10 days, ell. highe'r i'e 1'
deposition during early lile, llbs an influence oil ,n ltipicat ion.
First mention of tlie egg-laying capacity of liet' cit- is white Ilyi
was made by Riley and I toward,' wlho 1)hsevd tlieii c'inclusions o1n
the number of eggs thliat couIld be counted in thle adl)(biioiei t'f tlite
females when mounted in balsam, anti not iiup)on (ldail\ (olut.s of eggs
deposited by the females throughout life. Their estimate (of about
25 eggs as the probable total number of eggs dlel)positedi Iy a single
female during life hlias been generally accepted by slubl)seqltient writers,
none of whom has ever placed the maximum nuiml)er deposited
above this figure. The present investigations, however, :have delmon-
strated that this estimate is far too low and that tlie numlber'of well-
developed eggs to be found in the abdomen of tle female at aly- one
time is not indicative of the number of eggs 1eposited throughout
her life. Females have been known to (del)osit more than this number
of eggs in a single (lay. As will be seen by reference to Table XIII, as
many as 211 eggs have been actually deposited by -one female, and
should the 11 well-developed eggs found in her abdomen at de't i be
should ~ ~ ~ th II"eldeeo"o e
added a total of 222 eggs would be obtained. As this female, No. 6,
lived but 17 days and others have been known to live( 2, days, it is
even probable that as many as 250 eggs more nearly represent tlhe
maximum egg-laying capacity under most favorable con dit ions. How-
ever, it is seldom that a female lives sufficiently long to deposit her
full quota of eggs. With the average length of adult life curtailed to
about 10 days, the average of 149.2 eggs per female, as shown in
Table XIII, is beyond (loubt high. An a average of 125 eggs per female
is nearer the number of eggs (deposite(d (luring life i tlie grove.
In order to determine that, portion of thle dav y when eggs are most
freely deposited by females during summer weather, a(ldults were
inclosed in a rearing cage over leaves from which all previouslyy
deposited eggs had been removed, and allowed to remain for a period
of two hours, when the cage with adults was removed to another leaf
and the deposited eggs counted, with results shown in Table XIV.

SInsect Life, vol. 5, p. 222, 1893.

:: ..::' +t .


TABLE XIV. -Activity of the citrus whitefly in oviposition during different parts of the day.

Mean tern- Number Per cent of
Time of day. perature of eggs total eggs
for period, deposited, deposited.

6a. m.-8a. m.... 82 20 2.6
8a.rm.-10a. m... 89 55 7.3
10a.nm.-12m..... 91 35 4.7
12m.-2p. m..... 92 65 8.7
2p.mrn.-4p.mrn.... 90 30 4
4p.mrn.-6p.mrn.... 85 197 26.1
6p.mrn.-8p.ni.... 81 248 32.9
8p.mrn.-6a.rn.... 74 103 13.7
6a.rn.-6. a. m... 'S1.6 753 100
1 Average temrnperature for entire day of 24 hours; not the average of the 8 periods.

From the data it will be seen that while oviposition occurs at all
times of the day, nearly 60 per cent of the eggs are deposited between
4 p. m. and 8 p. m., and that oviposition does not cease on the ap-
proach of darkness. The variation in the number of eggs deposited
during the periods from 6 a. m. to 4 p. m. has little significance. It
was noted that the least number of eggs were deposited when the
bright sun fell directly upon the cage.
In further evidence of the greater activity of oviposition during
the latter part of the day, two other cages were started on August 1,
1909. One cage placed repeatedly over the tenderest growth resulted
in 698 eggs being laid between 10.15 a. m. and 4.15 p. m., as com-
pared with 895 eggs laid between 4.30 p. m. and 7.30 p. m. The
second cage, covering spring growth, gave 115 as compared with 786
eggs deposited during the same periods.
Relation between oviposition and food supply.-As the egg-laying
capacity of a single female is close to 250 eggs and but 25 well-
developed eggs have ever been seen in her abdomen at any one time,
it is necessary that slhe obtain nourishment sufficient to mature her
numerous "potential" eggs. There remain many interesting obser-
vations and experiments to be made on the relation between oviposi-
tion and food supply. That females deposit fewer eggs when feeding
upon many of the recognized food plants other than citrus than they
do on the latter is a subject of considerable interest. While adults
feed apparently as contentedly upon new growth of China trees and
umbrella China trees, they do not appear to deposit as many eggs
per female on these host plants as on citrus. The extremely small
number of eggs laid by females swarming over new growth of wild
persimmon in June at Orlando is even more astonishing considering
the marked preference shown by the females for this growth over the
spring growth of orange. Even on citrus itself oviposition is influ-
enced by the ages and corresponding toughness of the leaves, though
not. as markedly as is that of the cloudy-winged white fly. In one
instance equal numbers of adults were confined on a tender and an


ohl leaf of orange fo W ( i 1ours, wI.iIi I t -it adtInl t s wen.e r1,IM ve I itI,<|
576 eggs weret found to hav l)('ll ai o1 li te 'l1d1cr leaif 1i1d bitlt 25
on the old leaf. Again, uinider jracticallv the saie co.,,ilions, I1 I
eggs were deposited ul)On tender groiwtll o ,(1i d but 2 < e(Ili (l' 4( growthl.
The difference bt'tweet ovib)siliolln oI tender Alugust gro\,\lh anid
spring growth is lot as great as this, t1o13ugh v1ry mai rked, is aibot
90 per cent of the third-brood adults fly to tolie new gowti put ,II
by the trees late i, July and early in August.
From the foregoing it is evident that tlie (lIufmer (if eggs dip 3o1sit d
is strongly influenced by the nature of tlie insect's focdl. lF'emiles
confined in empty cages never deposit eggs, neither (1) tliose resting
upon thick bark, ladders, or picking boxes, and, as ]ias beell siat. under "Food plants," oviposition is entirely cllecked1 when fealies
are confined with leaves of nonfood plants. This difference ill tile
number of eggs deposited on various )lants may lrove of value fro'l
the standpoint of trap foods, and become a factor in tle control of
this pest.
Examination of thousands of adult citrus white flies at all seasons
of the year has shown that after a grove has become well infested an
equilibrium between thle proportion of males and females is estab-
lished from which there is under ordinary conditions little variation.
In such groves it has been found that from 60 to 75 )per cent of tlhe
adults are females. Of the records on file, about 66 )per (cenTt. give
percentages of 60 and over for females, while 66 to 76 per cent are
more frequent percentages where adults are abundant.
In groves where the progress of the white fly has been very seri-
ously and suddenly checked by natural or artificial causes, tile pro-
portion of sexes is subject to a much wider variation and there follow
for a time fluctuations between a predominance of males and females.
In one such grove where the wlmte fly had been greatly reduced ill
numbers because of the scarcity of adults of thle first brood, there was
Sa very large percentage of males appearing with thle second brood,
which in turn resulted in the third brood of 90.5 per cent females.
In a second grove, where over 99 per cent of tile white fly were killed
by fumigation, the few females of the first brood, because of their
isolation due to scarcity in numbers, were forced to deposit mostly
infertile eggs, which resulted, in tlhe second brood, in a reduction of
females to 18.6 per cent.
Dependence of sex upon. parthenogenesis.-The proportion between
the sexes is largely and evidently entirely dependent upon p)artheno-
genesis. It has been shown that infertile females deposit eggs in as
'Three hundred adults of A. citri confined on the tenderest spring growth of oak
for three days deposited 1 egg.




large numbers and as frequently when males are not given access to
them as do fertile females, and that the adults developing from these
eggs are all males. Whether the adults from fertile eggs are invari-
ably females has not been proved, although the evidence leaves little
doubt that they are. If otherwise, it would be difficult to account
for the fluctuations in sexes mentioned under the preceding heading,
or to explain the great predominance of females over males after the
species lihas become well established.
During the cooler portions of the year, when adults are present on
the trees, verve few are seen flying about from tree to tree unless
abundant. The morning and evening temperatures easily chill them;
hlience their activities are confined to the warmer part of the day.
IHowvever, after summer weather has become established the white
flies rest very quietly on the under surface of the leaves during the
greater part of the day. They shun the bright sunshine and prefer
leaves in shaded 1)laces. When exposed to the sun without protec-
tion they soon (lie. As the temperature falls during the late after-
noon, and especially after afternoon showers when the humidity has
risen to 90 or even to 100, they become very active, and about 4
o'clock begin to fly about from leaf to leaf and from tree to tree, and,
when very abundant, swarm in such large numbers about the groves
and town streets as to arrest the attention of pedestrians, to whom
they become at times a source of much aggravation, becoming en-
tangled in the hair, crushed upon the clothing, breathed in with the
air and causing choking, and flying into the eyes.
The adult insects, having well-developed sucking mouth parts, feed
upon the plant juices in the same maiuer as do the larve and pupa,
but with tile advantages of not being confined to the same location.
They do not leave any external evidence of the feeding except on
very young growth, when the feeding of a large number of adults
frequently produces a crinkling of the foliage.
It is difficult to determine positively whether or not an adult citrus
wliite fly is feeding when it is resting on a leaf or stem. Adults rest
contentedly during tihe warm portions of the day upon the underside
of leaves of plants upon %which they have never been known to de-
posit eggs. Under these circumstances they even appear to mate,
and it seems probable that they feed to a limited extent. When on
one of the principal food plants of the species, however, it is safe to
consider that adults feed wherever eggs are deposited in noticeable
numbers. It is because of this indiscrimnlinate settling upon vege-
tation upon which they are not able to subsist, and upon which they

..,.,; 1 .''


never bret, Il, tliat, tit. llit' vh, t rli i.s reiv ul .iiIi I ii ifuirtiliill'a I widev
ciriiutltitniio i'i. amon ., iatlg o r elglW1tI.nt ti CitI'lls SW iiIt \ ltI IIIb .II -,e ,in
all kitiids of 1aiiiniiock tretl'Ms, sNlitrus1), dil( grasses. .a l' rlh.z I s of tlIe
food plait1 tlite adtl iltS f't'tI iIliitost e.XCltINiv l iVi1 uipo tlile ii1il'i"r 1,:1il'fa'l
of tlhet loaves, inor1 'rarely Lu oii tile riuit, and (l 1 ii|t'l Wli l tint' \\d, \
portions of" tle ttre. tWhen iw growlti is 4'erv o*tng :11!1 tli t h-':\,
have not expaitdel adtiults often feediItl U ipoin othi sii'l'a's tof t1I. I'il1',
the petiole, luild evenl tlhe tentdei shoots, but tills lasts oniill 'o' a slm it
time. At, till seasons the newest growth is preferred, as iidli.at'
by the datait under thle caption of tilie relattion o1'f footS! slill)ji to 4Wilim-
sitioni, and thle portion of thle plant selecte(l coiIlcitles wit li t lit
already discussed for oviposition. It should be notelid here tllit tli
decided peerference of tlie adults for tlie Inew rogw\\th li is a cleckijn,
effect, as noted elsewhere, u 111)011on multiplicatn1011, US tlhey are clitirilv
lacking in instincts preventing over-ovipositionl

The relation of multiplication to food supply aindi tlthe restrictions
upon multiplication due to overcrowding, natural mNortality, drop-
ping of leaves after freezes, parthenogenesis, andt attacks by insects
ant other predaceous enemies and fungi will be found treated c.lse-
where. It has been estimated that not more than 5 per cent, at tlhe
most, of the eggs deposited throughout tlhe State result in tlie (levelop-
ment of mature insects. If each female deposited her full number
of eggs anti all thle forms lived, it has been estimated, tle ,i'gproey of
a single pair of wliite flies emlerging in January would amlloulint to
about 55,000,000,000 in one year.


..... . .... .... ..



Data concerning the duration of the egg, larval, and pupal instars
of the citrus white fly have already been given, but not in a form readily
showing the relation to the complete life cycle. From some of the
more important and complete of the life-history studies the data in
Table XV have been arranged to illustrate the important points in
this connection:

TABLE XV.-Length of life cycle of the citrus white fly at Orlando, Fla.

Eggs de-' First fly Last fly First fly Last fly
Lot No. flyd.'megd emerged emergedem ge
poied. emerged.e g emerge&
in fall. in spring.

No. 1 ................................................ Feb. 23 Apr. 30 ..... ............ ........
No. 2 ............................................... Mar. 3 May 9 .................... .........
No. ;i ............................................... Apr. 3 May 30 ................... ........
No. 4............................................... Apr. 20 June 7 .-.-.-...- .-------.. Mar. 18
No. 5 ............................................... June 16 July 30 ....... ...... .............
No. U6 ............................................... July 17 Aug. 27 Sept. 10 Mar. 16 Apr. 16
No. 7.................................................. do..... Sept. 4 Sept. 17 Mar. 17 May 4
No. 8............................................... July 19 Sept. 2 Sept. 21 ...do.... May 1
No. 9............................................... July 26 Sept. 6 Sept. 20 Mar. 18 May 10
No. 10.............................................. Aug. 1 Sept. 19 Sept. 26 Mar. 24 Apr. 6
No. 11 ............................................... Aug. 3 Sept. 25 ...do..... Mar. 23 Apr. 15
No. 12.............................................. Aug. 8 Sept. 19 Sept. 27 .......... ..........
No. 13.............................................. Aug. 9 Mar. 30 0 Mar. 20 May" 12
No. 14.............................................. Sept. 18 Mar. 16 0 Mar. 16 Apr. 28
No. 15.............................................. Sept. 21 Mar. 12 0 Mar. 12 May 10

Smallest Degrees
Least Largest P e Percent number adcrs iu-
uminherof nunm herof er vinter- egrese lating laccn
Lot No. days for days for emerging ing over effective before lasting
before tern befrre before
develop- develop- winter to emerge ture ora spring
witr tr o last fly
ment. meant. in spring. d emer- a
develop- g e. emerged.
ment. gence.

No. 1 .......................... 67 .......... 100.0 0 1,783 ...... ........
No.2 .................. .......... 67 ........ 67.. .. . 100.0 0 1,885 ....................
No. 3 ........................... 57 .......... 100.0 0 1,888 .............. ....
No.4 ....... .................... 48 333 .................... 1,712 -...................
N o. 5 ........................... 44 ................... .......... 1,725 .......... ..........
No. 6 .......................... 41 273 51.9 43.1 1,641 6,632 7,619
No. 7 .......................... 49 291 30.8 69.2 1,972 6,665 8,253
No. .......................... 45 286 12.7 87.3 1,815 6,504 8,059
No. 9 .......................... 42 288 29.8 70.2 1,703 6,322 7,981
No. 10 ......................... 49 2,18 5.5 94.5 2,015 6,255 6,654
No. 11 ......................... 53 255 3.6 96.4 2,153 6,107 6,858
No. 12 .......................... .4.2 2.7 97.3 1,735 ........-- ...........
No. 13 .......................... 223 276 0 100.0 5,825 5,825 7,545
No. 14 .......................... 179 222 0 100.0 4,552 4,552 5,836
No. 15 .......................... 172 2.31 0 100.0 4,289 4,289 6,100

From this table it will be seen that the period of development for
individuals hatching from eggs laid upon the same leaf within a few

hours of eachll other is subject to an astonishing variation, ranging
from 41 to 333 days. This variation is absolutely independent of.
both temperature and humidity influences. It will be noted that
the sums of effective temperatures required for the minimum dura-
tion of immature stages for individuals developing from eggs de-
posite(l between February 23 and August 8 vary from 1,641 to
2,153, with an average of 1,846, which may be regarded as very
nearly the normal for minimum development up to the time when