Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices

U.S. DEPARTMENT OF AGRICULTURE,

BUREAU OF ENTOMOLOGY BULLETIN No. 79.

L. O. HOWARD. Entomologist and Chief of Bureau.

FUMIGATION INVESTIG ATIONS
IN CALIFORNIA.

By R. S. W<>< iTAIM,

Spec ia I Field Agent.

[CHEMICAL WORK PERFORMED BY THE MISCELLANEOUS DIVISION OF niii
BUREAU OF CHEMISTRY.!

Issued June 11, 1909.

^^K.U'S'D""'

*r»cuKu

WASHINGTON:

GOVERNMENT PRINTING OPBflCE.
1909.

BUREA U OF ENTOMOLOGY.

L. O. Howard, Entomologist and Chief of Bvreau.

C. L. Marlatt, Entomologist and Acting Chief in Absence of Chief.

R. S. Clifton, E unit ire Assistant.

C. J. (Iilliss, Chkf 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 and tick in vestigations.

F. M. Webster, in charge of cereal and forage plant insect in vestigations.

A. L. Quaintance, in charge of deciduous fruit insect investigations.

E. F. Phillips, in charge of apiculture.

I). M. Rogers, in charge of gipsy moth and brown-tail moth field work.
A. W. Morrill, in charge of white fly investigations.
W. F. Fiske, in charge of gipsy moth laboratory.

F. C. Bishopp, in charge of cattle tick life history investigations.
A. C. Morgan, in charge of tobacco insect investigations.

R. S. "Woglum, in charge of hydrocyanic acid gas investigations.
R. P. Currie, in charge of editorial work.
Mabel Colcord, librarian.

U. : DEPARTMENT Of AGRICULTURE,

BUREAU OF ENTOMOLOGY-BULLETIN No. 79.

L. O. HOWARD. I ntomoloyist and Chirf of Bureau.

FUMIGATION INVESTIGATIONS
IN CALIFORNIA.

By I?, S. WOGLUM,

Special /'<< Id Agent .

[CHEMICAL WORK PERFORMED BY THE MISCELLANEOUS DIVISION OF THE
BUREAU OF CHEMISTRY.]

Issued Junk 11, 1909.

WASHINGTON:

GOYK R N M E N T PRINTING ( ) F PIC E

1 0 0!).

LETTER OF TRANSMITTAL

U. S. Department of Agriculture,

Bureau of Entomology,
Washington, D. C, February 6, 1909.
Sir: I have the honor to transmit herewith a manuscript entitled
1 ' Fumigation Investigations in California. " It is a preliminary report
on the subject, and contains much information which will be of direct
practical value to citrus growers. The report is one of progress, and
much of the information has already been made available to the public
by means of lectures and field demonstrations, but it is important
that it should be put in published form so as to give it wider cur-
rency. The subject is one that requires abundant illustration, and
the figures and plates submitted are all deemed necessary for the
full understanding of the text.

Respectfully, L. O. Howard,

Entomologist and Chief of B u rea u.
Hon. James Wilson,

Secretary of Agriculture.

REFACE

Fumigation under tents with hydrocyanic-acid gas has been the

principal moans of controlling scale-insects on citrus fruit trees in

California for many years. Most of the commercial orchards in the
State are fumigated at intervals of one or two years, at a cost rang-
ing from 25 cents to $1.50 a tree, or a probable total annual expendi-
ture of about $1,500,000, on the basis of fumigation of 50 per cent
of the trees each year. It becomes, therefore, a matter of very great
importance to conduct the operation of fumigation in the most effect-
ive and economical manner. The work being done on the subject in
California by this Bureau is aimed to thoroughly standardize the proc-
ess. It was undertaken in response to urgent demands from the hor-
ticultural commissioners of the principal citrus-fruit-producing coun-
ties of California, and of many prominent growers. The need of this
investigation was most urgently championed by Mr. J. W. Jeffrey, for-
mer secretary of the Los Angeles County horticultural commission and
now State commissioner of horticulture of California. Recognizing
the general usefulness of the process of fumigation, Mr. Jeffrey called
attention strongly to the unevenness of results in the work of differ-
ent manipulators and against different scale pests, and that the whole
practice had grown up experimentally without ever having been given
thorough scientific examination. He urged that such an examina-
tion necessitated carefully conducted and recorded field work, sup-
plemented by chemical tests of ingredients and the determination of
reactions, and expert study of the physiological effect of the gas on
the trees and fruit; in other words, to remove the process from the
mere guesswork of the field man and to place it on an exact scientific
basis.

This investigation has been under the direct charge of the writer,
who made a personal study of the situation in southern California in
September and October, 1907, and planned the work to cover the fol-
lowing subjects:

(1) Dosage, or the amount of gas and duration of exposure neces-
sary for different purposes. The strength of gas necessary to effect-
ively control the three prominent scale pests of citrus trees in Cali-
fornia, namely, the red scale, the purple scale, and the black scale,
under different climatic conditions, as in the drier foothills regions and

3

4 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

the moister coastal strip; for different seasons of the year: and for the
different growing conditions of the tree, as whether in fresh leafage, or
in bloom, or with different stages of the developing fruit.

(2) Physiological effect on the tree and fruit. There is some evi-
dence to show that the gas may have a stimulating effect on the tree.

(3) Mechanical equipment. An important economical considera-
tion in gassing is the employment of the most suitable tent cloths, and
their treatment to give durability and imperviousness; also, the best
mechanical means of hoisting tents over the trees. To be determined
under this heading also are the most economical methods of generating
the gas, and an indication of the quality of chemicals best suited for
the purpose.

In connection with this experimental work the scale species them-
selves are being given a careful study in the field to determine their
exact life history as a basis for the intelligent application of the
remedy.

This investigation was started in July, 1907, under the field charge
of Mr. R. S. Woglum, who first made himself thoroughly familiar
with the problem by a personal examination of conditions throughout
the citrus-growing regions of southern California. The direct work
of investigation began as soon as the fumigation season opened, and
later Mr. Frederick Maskew was employed to assist in the work. The
experimental work as planned has been conducted on a commercial
scale, so that the conditions and results will be those normal to the
ordinary care of citrus groves. To carry out all the lines of experi-
ment indicated above, and the subsidiary ones which have developed
in the course of the investigation, takes a good deal of time, and will
probably occupy two or three years with the money and force now-
available. Nevertheless, very considerable progress has been made,
and the preliminary report herewith submitted covers the general
features of fumigation procedure.

Improved methods have been devised, and these are being very
rapidly adopted throughout southern California. These improved
methods make it possible to do much more uniform work and greatly
simplify the method of estimating the proper dosage. Full advantage
has been taken of the fumigation work conducted in Florida again-t
the white fly under the field direction of Dr. A. W. Morrill, and the
Morrill system of marking tents for the ready determination of dosage
has been introduced, with modifications, into California.

C. L. Maklatt.
Entomologist and Acting Chief of Bureau in Absence of Chief.

CONTENTS

Page.

Introduction 9

Extent of citrus orchards 10

Citrus pests K)

Insect enemies of citrus fruits, and their distribution LO

Injury resulting to scale-infested trees II

Method of propagation of the more injurious scale pests L6

Methods used in the control of scale pests of citrus trees L6

Fumigation 17

Sheel tents 17

Points on procedure 18

Dosage schedules of the more important writers on fumigation 19

The present system of scheduling dosage 23

The initial problem confronting this investigation 24

Method of computing volume and dosage for tented trees 25

Methods for obtaining the measurements and dosage of trees 26

The chemicals required in fumigation 30

Potassium eyanid 30

Sulphuric acid 30

Proportion of chemicals used by fumigators 32

The amount of sulphuric acid necessary 32

The effect of too great an excess of acid 34

Water as a factor in fumigation _ 34

The effect of different proportions of water on the temperature of the

gas 35

The temperature of the gas where large and small dosages are used 36

The effect of different proportions of water on the amount of available

hydrocyanic-acid gas 37

The correct proportion of water 38

The most economical proportion of chemicals to use in generating

hydrocyanic-acid gas 39

Mixing the chemicals 39

Purple-scale fumigation 40

Preliminary experiments for the control of the purple scale 40

The leakage of gas in fumigating small trees 13

The length of exposure 44

Kradieation of the purple scale 46

Difficulty of destroying the scale on the fruit 46

( reneral considerations 47

Leakage of gas during operations 47

Time of the year for fumigation 48

Fumigation during the blossoming period 49

Fumigation while I he fruit is of .-mall size 51

5

6 CONTEXTS.

General considerations — Continued. page.

Simple method of removing acid from drums and carboys 53

The protection of cyanid 55

Hydrocyanic-acid gas in drums 55

The marking of touts 55

A device for covering fumigation generators 56

An improved system of fumigation

Supply carl 59

Procedure 61

Advantages under this system 63

I >i eage schedule 64

The improved system in use

Fumigation simplified 68

Index G9

ILLUSTRATIONS

Fio. I. Map Bhowing principal localities in southern California where citrus

Eruita are produced id

l'. Leaves and branch of orange infested with purple scale l Lepidoaaphea

It ckii » II

:;. Fruit of orange infested with purple scale L2

i. Branches of orange infested with black scale (Sawetia 6U x) L2

5. Leaves and branch of orange infested with red Bcale (ChryaompJialua

aurontii) 13

6. ( tarange tree almost destroyed by red scale 11

7. Orange tree showing branches at center partly destroyed and Btripped

of leaves by purple scale L5

B. Tray commonly used for carrying the chemicals of Eumigation from

tree to tree L8

!). Man carrying tray and water bucket I!)

LO. A typical California lemon orchard with row of fumigation generators

placed ready for use the following night 20

11. Outline of a fumigation lent marked according to the Morrill system. . 28

12. A fumigation tent marked after the Morrill system 29

13. Chart showing total amount of gas evolved when different propor-

tions of water are used 38

14. ( Grange blossoms at an early stage of development 50

15. Lead-lined tank used in San Bernardino County for removing sul-

phuric acid from drums and for filling jugs 53

16. An improved pipe for removing acid from drums 54

17. Siphoning acid from a drum by means of a rubber hose 54

18. Carboy resting against a heap of dirt to facilitate pouring the acid 55

19. Carboy with handles attached to facilitate pouring the acid and carry-

ing the carboy 56

20. Zinc-covered top for protecting cyanid in the field 57

21. A cover device attached to a fumigation generator 58

22. Oifference in the direction taken by gas escaping from an open gen-

erator and from one covered with the corrugated lid 59

23. Cart used with the improved system of fumigation 60

24. Karthenware acid jar with attachments for field use 61

25. A table which can be used instead of a cart in fumigation over very

r< »ugh ground 62

A row of tented tree3, and cart at one end ready for dosing 62

27. Dosing a tree 63

28. Dosage schedule No. 1 65

7

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

INTRODUCTION.

Early in July, 1907, the writer received a commission to investi-
gate the use of hydrocyanic-acid gas in the control of insect pests
of citrus trees in southern California. Acting under detailed instruc-
tions from Mr. C. L. Marlatt, the Assistant Chief of the Bureau of
Entomology of the United States Department of Agriculture, be
spent the latter part of July and the following three mouths in a
thorough field investigation to acquaint himself with the condi-
tions of citrus culture throughout southern California, the distribu-
tion of the different citrus pests and the damage caused by them,
the existing methods for their control, and the status of fumigation
as then practiced in the various citrus districts. During this period
all the important citrus-producing sections south of Santa Barbara
were visited, and local conditions were carefully examined. This
work was greatly facilitated by the hearty cooperation of the differ-
ent county horticultural commissioners, who gave their time1 freely
and greatly assisted the writer in becoming familiar with all the
features of the problem in their respective counties.

The writer desires to acknowledge his indebtedness to the many
people who have assisted him during this investigation and facilitated
the progress which has been made. To Mr. C. L. Marlatt, Assistant
Chief of the Bureau of Entomology, he is especially indebted for
valuable assistance and advice. It is also the writer's great pleasure
to acknowledge his appreciation of the work of Mr. Frederick Maskew,
who has most capably assisted him in the performance of many of
his experiments. Mr. Maskew also prepared several of the illus-
trations used. To the Hon. J. W. Jeffrey, State commissioner of hor-
ticulture of California, credit is due not only for his activity in paving
the way for this investigation but also for the able support given
since field work was commenced. To Mr. William Wood, of Whittier,
CaL, the writer acknowledges his indebtedness for assistance in intro-
ducing the improved system of fumigation in the region adjacent to
Whittier, as well as for practical advice with regard to citrus insects
and their control, a subject about which he is especially well informed.
This occasion is also taken to thank the various horticultural officers
of southern California, packing-house managers, and the many citrus
growers who have assisted and supported this investigation.

9

10

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

EXTENT OF CITRUS ORCHARDS."

The production of citrus fruits in southern California is confined
to the narrow stretch of land south and west of the Sierra. Madre
Range, extending from Santa Barbara on the north to the Mexican
border. Although citrus plantings are located here and there through-
out this territory, in reality only a small proportion of the land
capable of cultivation is devoted to this industry. The most promi-
nent centers of production (see fig. 1) are in the foothills region and
lower land of the San Gabriel Valley; the corresponding regions of
the San Bernardino Valley, including the Redlands-Highland, the
Riverside, and the Corona districts; and the coast region of Orange

FIG. 1,— Map showing principal localities in southern California where citrus fruits are produced.

(Original.)

and Los Angeles counties. Regions of smaller production are found
in southern Santa Barbara and Ventura counties, in the San Fer-
nando Valley, and in western San Diego County.

CITRUS PESTS.

INSECT ENEMIES OF CITRUS FRUITS, AXD THEIR DISTRIBUTION*.

The larger number of the pests most injurious to citrus fruits in
southern California belong to the Coccidae, a group of insects popu-
larly known as scale-insects. Among the scale-insects which are

a For a general description of the California citrus-fruit industry, see Bulletin 123,
Bureau of Plant Industry, United States Department of Agriculture1, which may be
obtained for 20 cents from the Superintendent of Document.-. Government Printing
Office, Washington. D. C.

l.\>i ( I in BMIES OF CI nils i RUITS.

11

generally so destructive as to require extended efforts for their
control are the purple scale (Lepidosaphes beckii Newm.), the red
scale ( ChrysompTialus aurantii Mask.), and the Mack Bcale {Saisseiia
olese Bern.). The yellow scale {Chry8(ymphalus citrinus Coq.), con-
sidered a variety of the red scale, is much Less destructive generally,
though sufficiently troublesome in some Localities to be considered a
pest of primary importance. Other scale-insects attacking citrus
trees, which are so perfectly held in control by their natural enemies
and other causes as seldom to become very destructive, are the soft
brown scale ( Coccus Kesperidum \..\ the hemispherical scale (Saissetia
h mispha. rica Targ.), the oleander scale {Aspidiotus hederse Val.),

Fig. 2.— Leaves and branch of orange infested with purple scale (Lcpidosaphes beckii). (Original.)

and the cottony cushion scale (Icerya purchasi Mask.). Mealy bugs
(Pseudococcus spp.) are quite generally prevalent.

The most important pests other than scale-insects are to be found
among the mites, of which the rust mite of the orange or silver mite
of the lemon (Phyllocoptes oleivorus Ashm.) and the citrus red spider
{Tetranychus mytilaspidis Riley) an4 highly injurious. The orange
aphis (Aphis gossypii Glov.) becomes very numerous during some
seasons but is soon attacked by its natural enemies and held in con-
trol. A species of thrips worked quite extensively in some localities
on ripe oranges during the first months of 1908, removing the coloring
matter from beneath the epidermis, thus giving to the fruit a spotted
appearance which lowered its market grade.

The purple scale (figs. 2 and 3) appears to prefer the more moisl
dons in the vicinitv of the ocean. It is found in Santa Barbara

12

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

and Ventura counties; in Los Angeles County, inward from the coast
as far as Hollywood and Whittier, and in the lower part of the

San Gabriel Valley at
Covin a and Duarte;
throughout Orange
County; and in San
Diego County in the
region about San
Diego city. This insect
confines its attack to
citrus trees.

The black scale (fig.
4) is also considered as
partial to the more
moist regions, and with-
out doubt is able to
mature more freely here

Fig. 3.— Fruit of orange infested with purple scale. (Oridnal.) n ,-i r ,. • ,

than in the hot interior
country. It is distributed, however, throughout the citrus-growing
localities with the exception of the Rialto-Highland-Redlands region
of San Bernardino County. At Redlands this scale is found on olives
and some ornamental
plants; yet, to the best
of the writer's knowl-
edge, it has not been
reported from citrus
orchards. Even as far
inland as Ontario and
Riverside the black
scale is capable of
breeding freely during
some parts of the year,
but the hot days of
summer destroy a
large percentage of the
eggs and especially
those young scales
which are exposed to
the sun's ra}rs. This
destruction was espe-
cially noticeable in the
summer season of
1907, when the writer
was engaged in an examination of different localities. During the
first part of July occurred a few days of very hot weather. About

Fig. 4.— Branches of orange infested with black scale (Saisxttia
olex). (Original.)

[NSECT ENEMIES OF CITRUS FRUITS.

13

a month later inspections were made throughout the lower San Gabriel
Valley, at Pomona, Ontario, and Riverside, and in Orange County.
Throughout this valley a large majority of the young insects \\ hich had
hatched were dead at this time while fully 50 per cent of the eggs had
dried up. At Pomona, Ontario, and Riverside almost all the young
insects had been destroyed, and fully 90 per cent of the eggs beneath
the old scales. In Orange County Dear the coast a very small per-
centage of eggs was affected by this hot period, while recently hatched
young scales were much in evidence. The black scale occurs on a
wide range of hosts, including trees, shrubs, and herbaceous plants.

The red scale (fig. 5) thrives exceedingly well in the drier interior
regions of southern California. Jt can be found within a few miles

Fig.

■Leaves and branch of orange infested with red scale (Chrijsompfialusauranti/). (Original.)

of the ocean or as far inland as Redlands. The limits of its distri-
bution are much the same as for the black scale. This species can
be found on several host plants other than citrus species.

The yellow scale is even more of a heat-withstanding form than
the red scale. Infestation by this insect appears to be most marked
in the foothills region of the San Gabriel Valley, and along the Sierra
Madre Range through Upland and ( ucamonga, It is also broadly dis-
tributed at Redlands, where it has become a more serious menace than
elsewhere in southern California. That it is capable of withstanding
excessive heat is demonstrated by its prevalence in citrus orchards
in the San Joaquin Valley, at Marysville, Oroville, and other parts of
the hot interior valleys of northern California, where the purple scale
and to a large extent the black scale appear unable to survive.

14

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Mealy bugs occur in various parts of the southern end of the State.
Their appearance is usually spasmodic and in restricted areas. These
insects are at present more serious than elsewhere in Ventura County,
where they occur in great numbers.

The citrus red spider is general in distribution, whereas the silver
mite is restricted to the region about the city of San Diego.

INJURY RESULTING TO SCALE-INFESTED TREES.

In scale insects the mouth is an elongated beak or tube. This
tube is inserted into the bark or covering of the host plant when the

insect is feeding,

and is used to draw
up the plant juices,
which are the scale-
insect's only food.
When great num-
bers of these insects
draw sap from the
tree, even though
they are very mi-
nute, the tree's
vitality is greatly
reduced. This ef-
fect is very marked
in the attack- of
the red and purple
scales. Both of
these species cause
much destruction,
yet the writer is of
the opinion that the
red scale will de-
stroy a citrus tree
in less time than
will the purple

Fig. 6.— Orange tree almost destroyed by red scale. (Original.) qpqIp oil nthpr far-

tors being equal. Trees have been noticed from two to three years
after planting which had been killed by the red scale. Large
orchard trees are frequently destroyed by the pest (fig. 6), while it
is a very common sight in regions of severe infestation to see large
branches killed back to the trunk. Although no trees have ever
come to the writer's attention which were completely killed by the
purple scale, severe infestations result in the destruction of many
branches (fig. 7), and cause such a drain on the tree that the produc-
tion of fruit is greatly decreased. Moreover, the purple scale spread-

INJURY RESULTING TO SCALE [NFESTED TREES.

15

to the fruit, as docs also the red scale, resulting in expense for the
cleaning of fruit in- rendering it of a lower grade and, in extreme
cases, entirely valueless.

The black scale, although a much larger insect than either the red
scale or purple scale, appears to have, generally, little effect on the
vitality of the tree. Trees severely infested with the black scale may
appear as healthy as neighboring trees which are clean. Branches
are seldom if ever destroyed by its attacks alone.

The commercial importance of the Mack scale arises largely from
its habit of secreting honeydew, which spreads over the leaves, fruit,

Fig. 7.— Orange tree show;

branches at center partly destroyed and stripped of leaves by purple
scale (Lepidosaphes beckii). (Original.)

and branches, furnishing a growing medium for a black or sooty-mold
fungus, resulting in a black coating throughout the tree. This
coating is removed from the fruit by washing, or in light attacks by
brushing. In the investigation by Mr. (1. Harold Powell0 of the
causes of decay of oranges while in transit from California, it was
shown that the decay was greater in washed than in unwashed fruit.
To avoid the washing of fruit ii is necessary to destroy the scale in
the orchard-.

a Bui. Ii1:;. Bur. Planl Industry, U. S. Dept. of Agriculture, L0O8.

16 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

The black scale confines its attack mainly to the branches, vet it is
commonly found on the leaves during its earlier stages of develop-
ment and sometimes matures in this situation. Seldom does it
mature on the fruit. The red and purple scales infest the branches,
leaves, and fruit. The yellow scale occurs on the leaves and fruit.
Occasionally it is found to a very slight extent on the branches.

The more directly injurious effect to the tree resulting from the
attacks of the red, purple, and yellow scales appears to the writer to
be due to their ability to produce some toxic effect in the host plant
in addition to the injury caused by the removal of sap. These scales
cause a discoloration of the plant cells at the place where the sap is
extracted, whereas the larger black scale causes no discoloration
whatever.

METHOD OF PROPAGATION OF THE MORE INJURIOUS SCALE PESTS.

The young purple and black scale-insects hatch from eggs deposited
by the adult,- while the red and yellow scales produce their young
alive. The red and yellow scales are thus susceptible to the applica-
tion of remedial measures at any time throughout the year. The
eggs of the purple scale are much more difficult to destroy than the
insects, for the latter can be killed readily in any stage of develop-
ment although more easily in the early stages. The black scale is
capable, after it has reached its mature and leathery condition, of
resisting extreme insecticidal applications. Its eggs, also, are quite
as resistant as the mature insect, if not more so. In its early stages,
however, it can be readily destroyed by the proper insecticides.

In all species the different broods on citrus trees are seldom, if ever,
distinct, but overlap one another to varying degrees. At certain
periods the breeding is more marked than at others for each of these
insects; yet it is possible to find adult red, yellow, purple, or black
scales in the egg-laying stage at any time throughout the year in
any extensive citrus locality in southern California containing thrifty
trees and in which these scales are known to thrive. This over-
lapping of broods is due largely to the forcing and artificial conditions
of citrus culture.

METHODS USED IN THE CONTROL OF SCALE PESTS OF CITRUS TREES.

The methods generally resorted to in the control of citrus insect
pests are (1) fumigation, (2) spraying, and (3) the use of beneficial
insects.

The question of beneficial insects is too large for discussion in this
limited report; suffice it to say that their work is of the highest
importance in many respects.

Sulphur sprays are employed against the red spider and the silver
mite of the lemon.

SHEET TEN I S. 17

Distillate* sprays have been employed by southern California
horticulturists For many years, and at one time very extensively in
the control of citrus scales. The accumulated experience with these
sprays appears to have demonstrated that the results secured arc
not entirely satisfactory. To-day distillate sprays are used only on
a small acreage of citrus groves, having been supplanted by the more
satisfactory fumigation with hydrocyanic-acid gas. Nothing illus-
trates more distinctly the superiority of fumigation over spraying
with distillate oils than the readoption of fumigation by (he more
successful citrus growers, and the attitude of the officials of (he
county horticultural commissions of this region who, almost to a
man, now recommend fumigation for the control of scale-insects.

A kerosene-water spray has found a Limited use4 during the past
year in Riverside and Ventura counties.

FUMIGATION.

Fumigation with hydrocyanic-acid gas originated and was first
practiced in California by Mr: 1). \V. Coquillett, of the Bureau of En-
tomology, in 1SS6, in combating citrus insect pests. Since that time
it has gradually risen in favor as a means of destroying scale enemies
of citrus plants until to-day it is in use in almost all the important
citrus-producing countries of the world. The apparatus first used
in fumigation was somewhat complicated and cumbersome, making
the operation very expensive." As the use of this gas became more
widespread a gradual improvement in equipment as well as methods
has taken place, so that to-day the process is comparatively simple.

SHEET TENTS.

Sheet tents exclusively are now used in southern California. The
manipulation of sheet tents and the general procedure in fumigation
have been so clearly explained in Bulletin No. 76 of this Bureau
that it will not be necessary to devote space to them here. The tents
are octagonal in shape, the standard sizes being 17, 24, 30, 36, 41, 43,
45, 48, 52, 55, and 64 feet, but larger ones up to 72 or 84 feet have been
employed. The size of this style tent is property based on the dis-
tance between the parallel sides, not on the distance between opposite
corners.

The materials especially recommended, and now generally used for
fumigation tents in southern California, are 6^-ounce special drill
and S-ounce special army duck, although 10-ounce special army duck
is sometimes used in very large tents. The 62-ounce special drill is
made of single threads twisted hard and closely woven. It is light,
strong, and flexible. The special army duck is made of double

--Aim. Rept. 1' S. Dept. A.gr. for L887, p. L23, 1888.
77488— Bui. 70—09 2

18

FUMIGATION INVESTIGATIONS IX CALIFORNIA.

threads twisted hard and woven fairly close. This double-twisted
material is heavier and much stronger than the special drill, but not
so closely woven; consequently it is somewhat more porous. In
field work the special drill will adapt itself more closely to the
irregularities of the ground than the army duck, and particularly if
the tents become damp. The special 6^-ounce drill is generally con-
sidered the best
obtainable for use
in all fumigation
tents up to 45
feet standard size.
Special 8- ounce
army duck is rec-
ommended for
use in tents of
larger size. Prob-
ably the most sat-
isfactory method
of making large
tents is to have
the center of spe-
cial duck and the
sides of special
drill. This dis-
tributes the heavy material at the points of greatest wear, while the
drill makes the tent much lighter and more flexible.

Fig. S.— Tray commonly used for carrying the chemicals of fumigation
from tree to tree. Cans above contain cyanid; pitchers below contain
acid. (Original.)

POIXTS OX PROCEDURE.

The number of men making up an outfit varies from three to six.
In San Bernardino County most of the outfits consist of six men; else-
where they more commonly consist of four.

In estimating the dosage, the usual method is to make the estimate
before the trees are covered with a tent. Sometimes this scheduling
is done in the daytime, sometimes by night. The schedule for a row
of trees to be fumigated having been given, either one of two methods
of procedure is followed. In the first and more common method the
dosage of cyanid and acid for each tree of the row is measured out
into small cans and pitchers, which are placed in a tray after the man-
ner shown in figure 8. When ready for use this tray is carried from
one tree to the next down the row (fig. 9). Frequently two trays are
necessary to carry the material required for the entire row or set of
trees. The water is carried in a pail and measured at each tree.
The receptacles in which the gas is generated consist of earthenware
jars holding 1J to 2 gallons, having the handle on the side (fig. 10).
If dosages in excess of 16 ounces are used in a U-gallon generator or

DOSAGE SCHEDULES OF MORE IMPORTANT WRITERS

L9

in excess of 20 ounces in a 2-gallon
quently boil over, especially if the

generator, the contents \\ ill Fre-
cyanid is in small lumps or is

])ow dered.

DOSAGE SCHEDULES OP THE MORE CMPORT ANT WRITERS ON FUMIGATION.

Since the publication by Morse, in L887, of the first dosage schedule
for use in fumigating citrus trees with hydrocyanic-acid gas, a great
many tables of
dosage have been
rec ommen d e d
through publica-
tions in this coun-
try and abroad.
Among the more
mi t hori t a t ive
contributions on
this subject are
those of Coquil-
lett . Morse, Craw,
Marlatt, Johnson,
Havens, Wood-
worth, Pease, and
Morrill, of this
country; C. P.
Lounsbury, of
Sou tli Africa, and
W. J. Allen, of
New South Wales.
A careful study
has been made of
the dosage sched-
ules proposed by
these different in-
vestigators with a result most surprising. In the first place, we
must consider that uniform dosage will not be given to trees
unless based directly on their cubic contents when covered with
a tent. Secondly, dosage t aides prepared for trees merely with
regard to their cubic contents and without regard to the varying pro-
portions of leakage surface present in trees of different sizes are
faulty to a large degree. Of all the dosage tables which have come
to the writer's attention only those by Lounsbury, in South Africa,
by Morrill, in Florida, and a recent one by Wood worth in California,
have been based on the proper assumptions. The other tables were
either based directly on the cubic contents without regard to leakage
surface, or were4 prepared without any knowledge whatever of the
cubic contents represented by trees of given dimensions. Several

FIG. '.i.— Man carrying tray and water bucket. (Original.)

20

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

belong to the latter class. The following statement and comparative
table have been prepared which indicate the wide range of variation
in these schedules:

Dosage schedules. — For the information of those who may be inclined to doubt
the writer's contentions in this bulletin with relation to the generally chaotic con-
dition of fumigation schedules published in the interests of California citrus growers,
a table has been prepared which includes nearly all of the more important sched-
ules, together with a comparative analysis of the same. The dosages for trees of
given dimensions were duly computed. Having this data at hand and utilizing
the dosage allotted to each individual tree, it was possible to work out the rate of

Fig. 10.— A typical California lemon orchard with row of fumigation generators placed ready for use
the following night. (Original.)

dosage per 100 cubic feet of inclosed space at which that particular tree was being
fumigated. This has been done for all trees in the schedules proposed by several
writers, and the results have been arranged in the latter half of the table.

A glance at this table will show that the schedules of Morse, Coquillett, and Wood-
worth were all based on the cubic contents of the trees, which were dosed at a uni-
form rate, but without regard to the leakage of gas. Large trees are dosed at the
same rate as small ones, thus giving a lack of uniformity in results. All of the other
schedules detailed in this table were apparently prepared with little or no regard
to the cubic contents represented by trees of different dimensions. Although it
would appear from the table that leakage was taken into account, inasmuch as the
smaller trees receive a greater rate than the larger ones, proper allowance could not
be made for this factor without definite consideration of the cubic contents. Con-
sequently the decrease of rate recommended is in all cases irregular and widely
removed from a rate proportionate to the actual leakage. The fact that trees in-

DOSACK SClIKlUl.r.s of MORE [MPORTANT WRITERS.

21

crease in dimensions much more rapidly than in cubic contents la seldom taken
into consideration. The result is that the larger trees receive a relatively Bmaller
dosage than they Bhould.

Morse's schedule was prepared especially for the cottony cushion Bcaleand probably
bar the red scale. The schedules of Coquilletl and Pease, and doubtless thai of i raw,
were prepared for the red scale. Those of Johnson, Woodworth, i In • Riverside Com-
mission, and the Rural Californian were intended especially for use against the black
scale. The red scale was generally known to l>e harder to destro) than the black scale.

In Morse's schedule all trees receive practically three-fourths ounce per LOO cubic
feel of inclosed tent space; in Coquillett's, practically one-half ounce to LOO cubic
feet; in Woodworth's schedule they receive one-third ounce for the same space.
In 'raw's table, the smallest tree receives approximately 9 times as great .1 do age
rate as tin' largest; in Johnson's table, the smallest receives about \\ times the rate
iA the largest; in that of the Riverside Commission, the smallest ie allowed about
L3 times that of the largest: in that of the Rural Californian, the smallest receives
about 8 times thai of the Largest ; while in thai of Pease, I he smallest receives a dosage
rate about li' times as great as the Largesl tree.

This short analysis seems sufficient to call attention to the irregularities of these
schedules. A study of the following table will reveal many other interesting points.

Doaag< schedules recommended by several recognized authorities, with computed dosage
rates per 100 cubic feet of spaa inclosed by tent.

AMOUNT OF CYANID (OUNCES) PER TREE RECOMMENDED.

Height

of tree.

Width
of tree.

Cubic

con-
tents
of tree.

Morse .a

Coquil-

lett>

Craw.c

T. B.
Johnson. d

Wood-
worth.*

River-
side Hor-
ticultural
Commis-
sion./

Rural

( aii-
fornian.0

Pease.A

Feet.

4

Feet.

4
5
5
4
5
6

Cubic ft.

40

60

80

70

100

140

225

Ounces.

0.3

Ounces.

Ounces.

Ounces.

Ouncis.

Ounces.

Ounces.

On lies.

4

4

.6

6

1

1

l

i

6

4

6

1
1.0

;

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

5

s

5J

I

8

6
8
9

200
335

47.')

2

2

n

1J

5

8

2.4
3.4

1

7

9

10

6§ sin

1

10

8
10
11

8
10
12
14
13
10
12
14
10
15
14
16
17
14
16
18
19

435

645

37

535

800

1,130

1,490

1,440

960

1,355

1,790

1,036

2,210

2,105

2,680

3,215

2,400

3,080

3,815

4.470

"Te"

6.2

21

3

4*

21

2

7

10

2

8

11

12

2

12

.........

'"io.2
"""12.4"

H

5

5

^

8

12

4

10

12

83

7

7

0

4J

13

14

14

7
8

10

14

s

5

12

15

4
8

15

15. 7
........

23

16

12

9
10i

12

16

9

8

5J

14

17

....

18

15

18

10

6

14

18

27.1
28.3

19

16
8

20

13j 2. 175

20

16
18
20

::
18

3, (85
5,235
4,835

11

L3
15

10

14
10
11

6j

20

20

20

22

12

aRul. 71, Univ. ofCal. Agr. Exp. Sta. (1887).

6 Insect Life (1889).

e Destructive Insects (1891 .

H A. S1 '■••■ Bd. of Horticulture

i Bui. 115, Univ. ..Mai. A.CT. Exp. Sta. I I89i

■ ' Bui. 127, U. 3. Dept. Agric.
.'/ From ■• Fumigation Methods," by W. <;. Johnson.
f> California Cultivator | I9i

22

FUMIGATION INVESTIGATIONS IX CALIFORNIA.

Dosage schedules recommended by several recognized authorities, with computed dosage
rates per 100 cubic feet of space inclosed by tent — Continued.

AMOUNT OF CYANID (OUNCES) PER TREE RECOMMENDED— Continued.

Height
of tree.

Width
of tree.

- Ho.se.
of tree.

Coquil-
lett.

i raw.

T. B.
Johnson.

Wood-
worth.

River-
side Hor- '
ticultural
Commis-
sion. |

Rural
fornian.

Feet.
24

Feet.
IS
20
22
28
20
20
25
28
30
25
30

Cubicft. Ounces.

5,340

6,490

7,735

11,900

Ounces.

Ounces.
13*

Ouna s.
18

20

Ouna ».

14
16

16

24
16

24

.'4
24

Ounces.

Ounces.

24

i

24

24

7,120

1..'
14"

_
30

8,375

30

30

30

16,675

36

21,915

COMPUTED DOSAGE KATE (OUNCES OF CYANID) PER 100 CUBIC FEET OF INCLOSED

SPACE.

Height
of tree.

Width
of tree.

Cubic

SSL *-■

of tree.

Coquil-
lett.

Craw.

T. B. Wood-
Johnson. ! worth.

River-
side Hor-
ticultural
Commis-
sion.

Rural
Cali-
fornia]!.

Pease.

Feet.
4

Feet.

4
5
5
4
5
6

Cubicft. Ounces.
40 0. 75

Ounces. Ounces.

Ounces. Ounces.

Ounces. Ounces. Ounces.

4

60
80

(> 6

5

6

70

1.4

1.4

1.4 0.77

6

100
140

4

6

72

0.36

3.0

225

.71

8

5| HiO

.31

8

6

8
9
61

s

10

11

8
10
12
14
13
10
12
14
10
15
14
lb
17
14

200

335

475

310

435

645

S70

535

800

1,130

1,490

1,440

960

1,355

1,790

1,036

2,210

1

2.5

8

. 72
72

.3

2. 1

9

10

.32

10

11 M) . 7fi

.8

. 7

.57 .46

1.6

10

72

.31

1. 25

11

12

.37

12

.56 . 62 . C)2

.56

1

12

.71
------

.35

.88

12

.57 .47

.33

3

13

14

14

.51

. 45 . 45

.74

14

.28
..........

.67

15

5
.36

15

71

16

2,105

2,680 -71

. 57

43

.34 .38

.57

16

.3

a

52

17

3.215

O JOO

.71

IS

IS

16 3.0S0
18 3 MS

.32

2

.45

IS

. 71

19

19
13

16
IS
20
22

4,470
2, 175
3,485

5,235

R oto

.63

.36
.32

20

20

. 31 . 37

.29
.32
.19

.19

20

.35

20

. 7>

20

22

18 4,835
18 5,340

20 6,490
22 7,735
28 11,900
oq r ion

. 25

.15

■~>A

.34
.31

2
.25
.18

.13

°-i

20

.12

°4

24

96

.19

.17

.12
.1

30

20
25
28
30

25
30

12,680
14,365
16,675
15,630
21,915

.19
.19
.11
.14
.15
.11

30

30

30

36

36

PRESENT SYSTEM OF SCHEDULING DOSAGE. 23

THE PRESENT SYSTEM OF SCHEDULING DOSAGE.

When we understand thai up to the present time only one approx-
imately accurate4 dosage schedule has been proposed by the fumiga-
tion experts of California, and, what is more confusing, thai no two
tables agree in all respects, we can do1 wonder thai the practical
fumigator has turned from them in perplexity. Finding the tables
o( Little assistance, the fumigator has had to determine his own
dosage from practical experience and the results secured. It' be
failed to destroy the scale on a 6-foo1 tree in using l ounce of cyanid,
he increased his dosage for the next 6-fool tree, and so on. lie has
also Learned that the dosage required to destroy some scales is
greater than that for other species, ruder the system at present
in vogue the dosage is usually estimated in the daytime. The
estimator, who ordinarily is the man in charge <>!' the outfit, starts
out in an orchard equipped with cross-section paper or a schedule
sheet. lie walks between two rows of trees, jotting down in the
corresponding squares of the schedule sheet the dosage which he
believes the trees should receive. If he is a careful scheduler he
will look at the trees from different side^ before indicating the
dosage, as trees are sometimes more compact on one side than
on another. Less careful men set down the dosage for the two
rows of trees while moving along as fast as they can walk. The
writer has seen some schedulers walk through the field at a rapid
pace, taking four rows at a time.

The estimation of dosage in this manner is mainly guesswork.
Measurements of the trees are made by the eye: consequently, suc-
cessful results depend very largely upon the accuracy of the estimator's
eye-measurement, supported by his experience in fumigation. The
most careful of estimators are very irregular in their scheduling.
This point has already been mentioned by Professor Woodworth.0
From measurements taken after many fumigators, we have found
none who did not at times vary more than 50 per cent in dosage esti-
mates for t pees containing exactly the same cubic contents after being
covered with a tent. Frequently the variation is as high as 100 per
cent. The results secured by a few of the more careful and expert
schedulers have been good as a whole. These men, however, can
cover but a small portion of the citrus groves of southern California
in one season.

The writer has been shown orchards in which it was stated that all
the scale had been destroyed by the use of heavy dosages. Even if
this were the case it would show that the smallest percentage or
strength of dosage used on any tree in those orchards was sufficiently
large to desl roy the scale. Since, as we have found, expert fumigators

a Bui. L52, I'niv. of Cal. A.gr. Exp. Sta., L903.

24

FUMIGATION INVESTIGATIONS IX CALIFORNIA.

vary considerably in their estimates, many trees in the above-men-
tioned orchards must have received a much greater dosage than was
necessary for scale eradication, thus resulting in a waste of cyanid

and acid.

In Table I have been arranged the dosage estimates which were
scheduled in different orchards by three different rumigators. After
the trees had been covered with tents the exact contents were com-
puted by the writer from actual measurements. The dosages given
in these tables are not for scattered individual trees selected because
of their irregularity in size, but each table embraces a continuous
number in a single row taken at random, regardless of the size or
regularity of the trees. As great a lack of uniformity as that shown
in each table might be looked for throughout the orchard. These
schedules of dosage were used against the red and purple scales,
species considered by most fumigators to be about equally resistant
to the gas. The reader will note the wide difference in the dosage in
the estimates of the different fumigators.

Table I. — Variation in the dosages estimated for several consecutii t tret s by thret different

fumigators.

Work of first fumigator.

Work of second f amigator.

Work of third fumigator.

Dosage
recom-
mended.

Actual
volume

of

treated

tree.

Ounces.
11
12
10
10
10
12
12
12
12

Cubic feet.
1,090
2,050
1,369
1,603
1,516
1,440
1,663
1 . 755
1,350
1,175

Volume
of space
in tent
to each
ounce of
dosage.

Cubicfeet.
150
170
135
165
150
120
140
145
110
130

Dosage
recom-
mended.

Actual
volume

of

treated

tree.

Volume
of space
in tent
to each
ounce of
dosage.

Dosage
recom-
mended.

Ounces.

Cubicfeet.

11

1,000

7

400

15

1,800

10

600

16

2,200

15

1,800

16

2,400

1

2,000

IS

2,200

17

2,200

12

1,250

16

2,500

Cubicfeet.
90
CO
120
60
140
120
150
125
120
130
105
156

Actual
volume

of

treated

tree.

Volume
of space
in tent
to each
ounce of
dosage.

Ounces.

6
4

4
5
4

4

Cubic feet

1.500

1.100

803

1,200

1.100

900

1,300

1,150

950

950

900

1.550

Cubicfeet.
250
275
200
240

260

230
190

258

THE INITIAL PROBLEM CONFRONTING THIS INVESTIGATION

After becoming acquainted with the existing methods of fumiga-
tion, it was realized that one of the first problems to be solved was
to devise some accurate system of determining dosage which would
obviate the errors due to guesswork. It at once became apparent
that the only way in which this result could be attained was by
determining accurately the cubic contents of the space inclosed by
the tent and giving the tree a dose proportionate to the content-.
It was also apparent that before such a system could be put into
operation, after having been worked out in practice, it would be

COMPUTING VOLUME and DOSAGE FOB TENTED CBEES. 25

necessary to determine by an extensive series of experiments the
dosage required for different-sized tree- for the various scale pots
infesting the citrus orchards.

METHOD OF COMPUTING VOLUME \\i> DOSAGE FOB TENTED TREES.

Although most citrus trees possess a certain genera] similarity id
shape, they are nevertheless somewhat irregular, no two ever being
identical in all respects. This renders it impracticable to determine
the exact contents o( any giveD tree. For field work, however, this
i> unnecessary, and all that is Deeded is to approximate it with a
fair degree of accuracy. In order to calculate the cubic contents of
an object, it must be considered as shaped like .some regular geomet-
rical figure or figures. The figure which most closely approximates

in shape an orange or lemon tree before it lias been pruned 1^ a cylin-
der surmounted by a hemisphere, and in computing the volume we
have considered them of this shape.

If we know the height and width of a tree covered with a tent, it
is a comparatively simple matter to calculate its contents.

In the past in California work the dosage has been based upon
these two measurements. After a tree is covered with a tent it is a
matter of some dilhcult y to determine the height and the width. By
using as factors the distance around the bottom of the tent and the
longest distance over the top of the tent we arrive at a more prac-
ticable method by which to compute the cubic contents of a given
tree. Using these measurements as a basis the writer has invented
a formula a by means of which the cubic contents of a tree may be
computed. To avoid computation work in the field as far as possi-
ble, the writer has formulated a table approximating the cubic con-
tents of trees of different dimensions, which is, he believes, suffi-
ciently extensive to include any eitrus tree in southern California.
During tins investigation no tree has been found whose dimensions
did not fall within the limits given in this table. The distance

a Professor Wood worth (Bui. 152, Univ. of Cal. Agr. Exp. Sta., p. 5, 1903) was the
first to propose a formula for obtaining the contents of tented trees by computing the
distance around the bottom and over the top. An analysis of this formula during
the early part of the writer's field work proved that it was inaccurate, thus necessi-
tating the determination of a new formula. The writer has worked out a formula
based od the two measurements above mentioned, it is as follows:

r-( <> C(3k-4) x
-1-^2 L2;r J

In this formula C=the cireumforenec of the tree.

0 = the distance over the top of the tree.

C° C(3r I)

If a person work- oul and notes down in a charl the values of r- and -^7) —

for different valuesoi ( of which beisapl to make common use, ii is possible by its
use in connection with the formula to determine the contents of trees with fair
rapidity.

26 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

around and over a given tree being known, the table will show the
approximate cubic contents of the tented tree. The dosage can then
be applied in proportion to the contents and at any strength desired.
A lemon tree, after being pruned, is ilat on the top. Therefore
we can not consider the geometrical figure which is applicable to an
orange or unpruned lemon tree as also applicable to a pruned or flat-
topped lemon tree. The figure which approximates the latter i> a
cylinder. Xow it so happens that the contents of a cylinder having
certain dimensions over its top and around its bottom are almost the
same as for a figure of the same dimensions composed of a cylinder
surmounted by a hemisphere. This is a great advantage inasmuch
as the schedule of dosage proposed for orange trees may also be used
for all lemon trees, thus obviating the necessity of preparing two
different schedules.

METHODS FOR OBTAINING THE MEASUREMENTS AND DOSAGE OF TREES.

WITH APPARATUS.

Of the various methods suggested for obtaining the measurements
of tented trees, the first was naturally by the use of a tapeline. It
was an easy matter to ascertain the distance around the tent with a
tape, but to measure the distance over the top was much more diffi-
cult. This required the services of two men and repeated efforts.
For field work on a commercial scale this was impracticable.

Woodworth0 explains a method of securing measurements, which
consists in the use of a fishing rod and a wire line, the latter marked
off by knots into 1-meter lengths. His description of this method
is as follows:

Having first attached the line at about its middle to the end of the rod, one end
of the former is made fast to the tent. The most convenient way to accomplish
this was found to be by means of a hook, like a fishhook from which the barb had
been removed. The most convenient place of attachment was at a point 1 meter
from the ground.

After attaching one end of the line to the tent the rest of that half is caused to lie
up to and over the center and top of the tent by means of the rod. The one making
the measurement then walks around to the opposite side of the tent, rod in hand,
holding the line constantly in position over the top. The other end of the line is
carried around the tent at the same time and is then drawn taut, measuring the lasl
fraction of a meter by means of the graduation on the lower joint of the rod. Adding
now 1 meter, the distance the first end is from the ground, we have the measurement
of the distance over the top of the tenl from the ground on cue side to the ground
on the other.

A second measurement was then taken by throwing the line off the top of the tenl
by means of the rod and holding it so that as the measurer proceeds around the tent
to the point where the line is attached, it will encircle the tent at a point about 1
meter from the ground. The end of the rod is again brought into requisition and the
last fraction of meter read in centimeters.

Both measurements are thus made by one person in a single trip around the tent.

"Bui. 152, Univ. of Cal. Agr. Exp. Sta., 1903.

METHODS FOB OBTAINING MEASUREMENTS \.\l» DOSAGE. 27

This method might be practicable with a medium-sized tree,
but for trees of Large size, especially seedlings, which arc sometimes
more than 30 feet in height, its use would doubtless prove difficult,
and for field operations multiplication of apparatus should be avoided

as far as possible.

WITIIOI T Ml' \i: \ II B.

The Woodworth system. — The first scheme, so far as the writer's
knowledge goes, for obtaining the measurements and dosage of

trees without the use o\' apparatus was suggested by Professor
Woodworth.0 This method consists of marking od the tent, on two
opposite sides and parallel with the edge, a series of lines winch are

placed at such distances from the center of the tent that they will
correspond with differences of i ounce in the dosage of trees of the
average shape. Upon each i^ these lines are marked three num-
bers; the first indicating the dose (in ounces), the second the cir-
cumference on which the dose is based, and the third the amount
the dose must be varied when the actual measured circumference
IS greater or less than that marked on the lent. For trees having a
circumference greater than the average between the second figure
on the line that is nearest the ground on one side of the tent and
the second figure on the corresponding line on the opposite side,
the average dose is increased for each additional yard of circum-
ference by the amount (in ounces) given by the third figure on the
line; for trees Inning smaller circumferences the figures are corre-
spondingly decreased.

Although the system is fairly accurate, its adaptability for use
under the present condition of fumigation in southern California is
somewhat questionable. The amount of calculation required to
ascertain the dosage for each tree gives large chance of error and is
wasteful of time. The possibility of error is still further increased
through the necessity of varying the dosage for different species of
scale-insect>.

The Morrill system.1 — Dr. A. W. Morrill, in the course of his
work against the white fly (Aleyrodes citri R. & II.) in Florida, has
devised a method of marking tents which is easily the most practi-
cable yet proposed for obtaining the distance over the top of a tented
tree. Although apparently a. modification of the idea presented in
the Woodworth method, it is really quite different. In the Wood-
worth system the actual dosage is calculated from the figures on
the tent. The Morrill system is merely a rapid and simple way of
obtaining the distance over the top <>f a tented tree.

a Bui. 152, Univ. of Cal. Agr. Exp. Sta.. L903.
&Bul. 7G, Bur. Ent. U. S. Dept. Agr., 1908.

28

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

In figure 11 is shown an outline of a regulation fumigating tent
marked after the Morrill system. Three parallel lines and one line
at right angles to them are indicated on the tent. The middle one
of the three parallel lines passes through the central point in the
tent canvas, running lengthwise of the central section or strip of
which the tent is made and passing over the top of the tent from
the edge on one side to the edge on the opposite side; these lines

1/

9/

9/

*/

£/

Z/

//

0/

6

8

Z-

9

S

I i 1 1 I i i 1 1 i 1 1 1

//

9/

9/

P/

€/

d/

//

0/

6

8

I

9-

$

//

9/

9/

PI

€/

2/

//

0/

6

8

Z-

9

S

VOVON^O^^^^^^^^^

: ; i i i i I i i I

Fig. 11. — Outline of a fumigation tent marked according to the Morrill system.

also run in the direction in which the tent should be pulled on or ofT
a tree. The single line at right angles to the parallel lines passi -
through the central point, as does the middle one of the three parallel
lines, and extends also from the edge on one side to the edge on the
opposite side. Beginning at the center these lines are graduated in
feet toward either edge of the tent, after the manner shown in the
diagram. For tents above 36 feet (average size) it is unnecessary to
commence the graduation nearer than 5 feet from the center of the
canvas. When one of these lines is over the middle of the tree

METHODS FOB OBTAINING MEASUBEMENTS \NI» DOSAGE.

29

(fig. 12), the distance over can be calculated by merely adding
together the two numbers on t ho opposite sides of the lent where

the e«le;e touches the ground. For instance, supple that on the line

over the center of the tree 12 Is nearest the ground <>n one side and
1.") on the other. The distance over the center of this tree would

be the sum of these numbers, which i- 27 feet. With the lines

graduated after this manner it makes little difference in determining
the distance over the top o\^ the tree whether or not the geometrical
center of the tent is at the center of the tree, the single requirement
being that some part of one of the graduated lines approximates
the center of the tree.

Fig. l

A fumigation tent marked after the Morrill system. (Original.)

The two lines running parallel to this central line should be about
4 feet distant from it in the larger fumigating tents. The reason
for using these auxiliary lines is, that in practice the center of the
tent is very often pulled considerably to one side, especially in
covering small trees. If the middle line does not fall immediately
over the center of the tree, one of the other two lines is quite likely
to do so, and that one should be used in obtaining the distance over.

The cross line running at right angles to the three parallel lines
also passes through the center of the tent and is marked like the
other-. In case of an irregularly shaped tree, by the use of this line
the distance over can be taken in two different directions and the
average taken for use in determining the cubic content-. In field
work, however, this cross line is unnecessary, as measurement over
the top in one direction i- sufficient.

30 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

The measurement around the bottom of the tent can be obtained
by the use of a tapeline or by pacing. Under this system the work
is facilitated by having a chart or table of figures showing the cubic
contents corresponding to different dimensions.

THE CHEMICALS REQUIRED IN FUMIGATION.

For the generation of hydrocyanic-acid gas in fumigating, potas-
sium cyanid, sulphuric acid, and water are necessary. The hydro-
cyanic-acid gas is produced by the action of the sulphuric acid on the
cyanid of potassium. Under the early methods of generating hydro-
cyanic-acid gas the cyanid was dissolved in water before being used.
At the present time cyanid is used in the crystal form entirely. The
water is first measured out and poured into the generating vessel.
The required amount of acid is then added to the water, producing a
great increase in the temperature of the mixture. T\ Tiile the mixture
is hot it should be placed beneath the tree and the cyanid added. If
permitted to cool before the cyanid is added, the generation of gas
will not only be slower than with the heated mixture, but the amount
of available gas will be decreased, thus making the operation more
expensive, and necessarily less efficient.

POTASSIUM CYANID.

An imported cyanid designated as 98 to 99 per cent pure is used
almost exclusively for fumigation purposes in southern California,
under the popular belief that it is superior to American cyanids for
this purpose. There seems to be no real basis for this common belief,
and, in fact, experiments conducted by Prof. Wilmon Newell while
State entomologist of Georgia demonstrated that certain brands of
American cyanid met all the requirements necessary for fumigating
nursery stock, and it seems reasonable to believe that these will also
be equally available for citrus-orchard fumigation. A series of
laboratory and field tests has been planned to demonstrate the use-
fulness of all the available brands of potassium cyanid.

In the field investigation reported in this bulletin the 98 to 99 per
cent imported cyanid commonly used in southern California has been
emplo}Ted throughout and, although no chemical analysis was made,
the results proved entirely satisfactory.

SULPHURIC ACID.

Too much stress can not be placed upon the quality of sulphuric
acid used in fumigation. Operators have repeatedly informed the
writer of much burning of fruit and foliage which occurred during
the season of 1905, owing to the use of a grade of acid differing from
that ordinarily employed. An analysis of the acid used that season

THE CHEMICALS REQUIRED. 31

showed that it contained traces of nitric acid, the presence <>l' which
might explain the burning. Nitric acid is one of the most active of
chemicals and is unstable as well. When heated it readily volatilizes.
By adding sulphuric acid to water a great amount of heat results. If
nitric acid be present in the sulphuric acid as an Impurit \ it would be
far more volatile than under ordinary circumstances The addition
of the cyanid increases the heat, at the same time causing hydro-
cyanic-acid gas to be violently thrown off. This gas assists in carry-
ing off the volatilized nitric acid, which, condensing on the cool,
moist surfaces presented by the Fruit and leaves of the citrus trees,
might result in hums or pits.

In procuring sulphuric acid for Fumigating purposes, only that
should be purchased which is entirely \'vco of nitric acid, and which
is guaranteed 66° ( Baum6), or 93 per cent pure.

Some commercial sulphuric acid on the market meets all the
requirements of fumigation, while much can be found which does
not. To enter fully into the reason for this would he out of place in
this bulletin. All that is necessary is to mention briefly the char-
acter of the material and processes used by various manufacturers,
some of whom strive to place a better grade of acid on the market
than do many others.

In the manufacture of sulphuric acid, sulphur may be considered
the basic element. This is obtained from one of two source-, viz,
from free sulphur, known commercially as brimstone, or from sulphur
in combination with a metal, as iron or copper pyrites. Brimstone
is comparatively pure sulphur, containing little or nothing which
would reduce the grade of the acid manufactured from it. It some-
times contains a very small quantity of ash. Pure iron pyrites con-
tains about 53 per cent of sulphur and about 47 per cent of iron.
Copper pyrites contains much less sulphur. Ordinarily the pyrites
used in making acid contains small quantities of other elements, as
arsenic, zinc, lead, etc. To manufacture sulphuric acid, it is neces-
sary to convert the sulphur into a gas, sulphur dioxid, which is
brought about by burning the crude product in a retort. The
sulphur dioxid thus formed is conducted into certain chambers
where it is mixed with fumes of nitric acid, air, and steam, the
resulting product being dilute sulphuric acid. Where brimstone is
used comparatively pure sulphuric acid is formed. When, however,
pyrites are burned, other (dements present in the ore (as arsenic, etc.)
are volatilized, pass along with the sulphur dioxid, and are present in
the crude acid.

That which concerns us most vitally in fumigating is the presence
of nitric acid. A much greater proportion of nitric acid becomes
mixed with the products of combustion from pyrites than from brim-
stone, resulting in the presence of a larger amount of this undesirable

32 FUMIGATION INVESTIGATIONS IX CALIFORNIA.

acid in the sulphuric acid. The impurities, including nitric acid,
may be eliminated by refining. This, however, requires extra
expense, and, as these impurities are of little or no importance in
some of the lower uses to which sulphuric acid is put, the acid is not
usually refined. Such acid is unsuitable for use in fumigation.

Taking all things into consideration it is safer, in purchasing
ordinary commercial sulphuric acid on the market, to order that
made from brimstone rather than that made from pyrites ore. It is
possible, however, to secure quite as good a product from pyrites as
from brimstone, if the former be sufficiently refined. If the fumi-
gator demands that it be free from nitric acid, arsenic, etc., and refuses
to accept it unless the product is of the grade required, there is no
reason why he should not be able to secure satisfactory material.

PROPORTION OF MATERIALS USED BY FUMIGATORS.

With each dry ounce of potassium cyanid most fumigators usQ,
1 fluid ounce of sulphuric acid, although some use 1J ounces. The
proportion of water used varies all the way from 2 to 8 times the
amount (by bulk) of acid, the majority using between 3 and 4 parts
of water.

THE AMOUNT OF SULPHURIC ACID NECESSARY.

Chemical combinations take place with definiteness; that is, when
one chemical acts on another in the production of a third substance,
the proportion between the first two chemicals is always the same.
Such is the case when sulphuric acid acts upon potassium cyanid in
producing Irydrocyanic-acid gas. A given amount of cyanid requires
a certain amount of sulphuric acid of a fixed degree of purity to
carry the reaction to completion. A quotation from a letter received
from J. K. Haywood, of the Bureau of Chemistry of this Depart-
ment, illustrates this point:

In the action of sulphuric acid on potassium cyanid approximately four-fifths of an
ounce (avoirdupois) of 93 per cent acid is used up for every ounce of 98 per cent
cyanid.0

Expressed in fluid ounces four-fifths of an ounce avoirdupois equals about 0.42 of
a fluid ounce. We may say that theoretically 1 ounce avoirdupois of 9S per cent
potassium cyanid needs 0.42 of a fluid ounce of ordinary commercial sulphuric acid
(93 per cent) to convert it entirely to hydrocyanic acid. Since it is always best to
have some excess of the acid to carry the reaction to completion, it is probable that
three-fourths of a fluid ounce of commercial sulphuric acid is ample in practice to
convert 1 ounce avoirdupois of 98 per cent potassium cyanid to hydrocyanic acid.
If 1 fluid ounce of the commercial sulphuric acid is used it will certainly leave a con-

«The reaction is as follows:

2KCN+H2S04=K2S04+2HCN.

PROPORTION OF CHEMICALS.

aiderable exceee of sulphuric acid present, h La perfectly possible, however, thai
this excess of Bulphuric acid is of value in heating up the mixture bo thai more of the

hydrocyanic acid is Liberated and uot absorbed bj the liquid.

The results of some tests serve as a further illustration of this
point. It was desired to determine by experiment if l fluid ounce of
acid to each ounce (avoirdupois) of eyanid would be sufficient to
carry the reaction to completion in the liberation of hydrocyanic-
acid gas. It is to be understood throughout that the eyanid ounce

is avoirdupois and the acid and water is the fluid ounce. For this

test two series of ordinary lj-gallon Fumigating vessels were placed
in line. In one series equal parts of acid and eyanid were used.
Three parts of water were used in all cases. The amounts of eyanid
used ranged from 1 to 10 ounces, that is, in one generator were

placed 1 ounce of eyanid, 1 ounce4 of sulphuric acid, and 3 ounces of
water; in the next of the same series, 2 ounces of eyanid. 2 ounces
of sulphuric acid, and (> ounces of water, and so on in the same pro-
portion up to 10 ounces. The second series was identical with the
first except for the use of one-fourth more acid than eyanid. After
generation had taken place for about out1 and one-half hours an
examination was made of the residue. In the first series, in which
equal parts of acid and eyanid were used, the residue was in the
form of a liquid. In the second series, in which 1£ ounces of acid to
1 of eyanid were used, the residue in several pots had collected in a
mushlike mass. Being puzzled at first over this phenomenon, in
order to ascertain if eyanid still remained unchanged in the residue
the writer added more sulphuric acid, but there was no further evolu-
tion of gas. This at once demonstrated that all the available eyanid
had been dissolved. Analyses of this residue by J. K. Haywood of
the Bureau of Chemistry showed that the reaction was complete both
when 1 ounce of acid and when 1\ ounces of acid to 1 of eyanid were
used. In submitting the result of these analyses, Dr. H. W. Wiley,
Chief of the Bureau of Chemistry, wrote:

The amount of eyanid present in these samples is so small that it does not indicate
to us incompleteness of reaction, but rather indicates the amount of hydrocyanic acid
dissolved in the residue. This view of the case is strengthened by the fact that
increasing the amount of sulphuric acid in the cases above did not decrease the amount
of cyanogen present in the residue. From our work, therefore, we are of the opinion
that the same amount of sulphuric acid as of potassium eyanid is sufficient to carry
the reaction to completion.

" In an address printed in the Proceedings of the Thirty-fourth Annual Fruit Grow-
ers' <onvention of California, p. 10:5, the proportion of chemicals spoken of appears
somewhat different from that mentioned in this publication. This is due to the fact
that the parts mentioned in that address were based on parts by weight of acid and
eyanid, both of which are chemically pure — not the commercial product as given in
this bulletin.

77488— Bui. 79 09 3

34 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Summing up, it may be said that 1 fluid ounce of commercial sul-
phuric acid (93 per cent) to 1 ounce (avoirdupois) of 98 per cent
potassium cyanid is certainly enough to carry the reaction to com-
pletion in the liberation of hydrocyanic-acid gas and is perhaps an
unnecessarily large amount. In practical field work where dosages
of varying sizes are constantly being used, it is very convenient to
reckon the acid in the same number of parts as the cyanid. The
use of 1 part (fluid measure) of acid to each part of cyanid is there-
fore recommended.

The commercial potassium cyanid sold on the market is usually 96
to 100 per cent pure. The commercial sulphuric acid on the market
is sold as 66° Baume and should contain 93.5 per cent sulphuric acid.
In California fumigation work, these grades are used and are to be
understood wherever cyanid or acid is mentioned in this bulletin.
In the dosage allotments cyanid is always measured in ounces or
parts dry weight, while the acid is measured in fluid ounces or parts.

THE EFFECT OF TOO GREAT AN EXCESS OF ACID.

In the experiment mentioned, in which two series of hydrocyanic-
acid gas generations were completed, the question immediately arose,
why the residue in some generators, in which 1^ parts of acid were
used, congealed, while in the case of those in which equal parts of acid
and cyanid were used no such result was noted. The explanation is
simple: When sulphuric acid acts on potassium cyanid, hydrocyanic
acid, a gas, and potassium sulphate, a solid, are formed. If sufficient
water is present, the potassium sulphate dissolves and there is no solid
residue. This was the result when equal parts of acid and cyanid
were used. When one-fourth more acid than cyanid is employed,
there is a large excess of acid. The potassium sulphate is not as
soluble in water containing excess acid as it is in water alone ; hence
it undergoes partial crystallization, resulting in a mushlike residue or
congealing into a solid mass.

WATER AS A FACTOR IX FUMIGATION.

There are several reasons why water should always be employed
in fumigation: It is very useful in dissolving the potassium cyanid
and hastening and completing the chemical reaction with the acid.
A piece of cyanid thrown into a mixture of acid and water imme-
diately gives up a portion of its mass in solution. Scarcely has the
cyanid dissolved when it is partially converted into gas. The heat
liberated during this process assists in forcing the solution of more
cyanid, which is also partially converted into gas. This continues
until the chemicals are exhausted and the reaction stops.

Potassium sulphate, a solid, is the by-product resulting from the
reaction by which hydrocyanic-acid gas is produced. Water dis-

PB0P0B1 ION OF CHEMICALS. 35

Bolves the potassium sulphate as it Forms and prevents it from coating
tlit* cyanid not yet in solution. In the presence of an Insufficient
amount of water, the potassium sulphate is not completely dissolved,
but forms a coating on the pieces of cyanid, preventing the Bulphuric
acid from penetrating to it, and thereby retarding, or even in part
preventing, the reaction. In such cases this undissolved potassium
sulphate usually congeals, causing the pots to "freeze." The phe-
nomenon always occurs where the formula is I 1 I, or where the
Bame amounts of water, acid, and cyanid are used. On agitating
the congealed residue by stirring, it is almost always possible to find
small pieces of undissolved cyanid enveloped in a coating of the
potassium sulphate. Ordinarily, when the residue is stirred the
particles of cyanid arc4 removed, to some extent, from this envelope
of potassium sulphate, allowing some of the unused acid to reach
them, and thus evolving a small amount of gas without the addition
of more acid. Under these conditions, however, the reaction is
never complete, and it is highly desirable therefore to add sufficient
water at the beginning to dissolve all the potassium sulphate.

From this last statement, as well as the data presented under the
heading "The effect of too great an excess of acid" (p. 34), it is seen
that the concealing or "freezing" of generating jars is due to either
or both of two conditions: (1) An insufficient amount of water to
completely dissolve the sulphate of potassium, or (2) a large excess
of sulphuric acid, whereby the water is rendered less capable of t aking
into solution the same amount of sulphate as it otherwise would.

Another very important function of the water in the reaction is
the heat produced by the union of the sulphuric acid and water.
Potassium cyanid introduced into this heated mixture gives off hydro-
cyanic-acid gas much more quickly and thoroughly than at a lower
temperature, and in field work rapid generation of gas is essential.

THE EFFECT OF DIFFERENT PROPORTIONS OF WATER ON THE TEMPERATURE OF THE CAS.

Anyone who has watched the escaping gas and steam from the
reaction of potassium cyanid and sulphuric acid wherein different
proportions of water were used could not fail to notice that the
violence with which the generation starts and the gas is given off is
apparently greatest with the smaller proportions of water. Fumi-
gators are aware of this, and commonly increase the proportion of
water when using large amounts of cyanid. Practice has demon-
crated that with a greater proportion of water the Injurious effect of
the resulting gas on the leaves and fruit is materially lessened. The
lessening of the injury has been attributed to the fact that the escap-
ing gas was Less heated when large proportions of wafer were used.
In order to clear up this point an experiment was performed, the
results of which are given in Table II.

36

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Table II. — Experiment to determine the effect of different proportions of water on the
U mperature of the resulting gas.

Amount of chemicals used.

Tempera-
ture of the

acid and
waU't mix-
ture.

Highest

tempera-
ture of the

hydro-
cyanic-acid
gas.

Tempera-
ture of the
gas one
minute
from start
of genera-
tion.

Cyanid.

Acid. Water.

Ounces.
5
5
5
5
5
5
5

Ounces. Ounces.

5

5 10

5 15

20

25

30

5 40

°F.
180
190
170
100
145
136
125

°F.
124
126

128
128
118
108
90

°F.
115
121

109
105
105
104

-7

In this experiment 5 ounces (avoirdupois) of cyanid and 5 ounces
(fluid) of acid were used for each test. The proportions of water were
varied, 5, 10, 15, 20, 25, 30, and 40 ounces, respectively, being used.
As a result the proportion of water to 1 part of acid or 1 part of cyanid
was 1, 2, 3, 4, 5, 6, and 8, respectively, for the different tests. These
generations were made in a H-gallon fumigating vessel in a room.
The temperature of the escaping gas was taken at the mouth of the
pot. The temperature of the acid-water mixture was taken one
minute after pouring the two together. The cyanid was then added.

The maximum temperature of the escaping gas is always realized
within the first minute, usually thirty to forty seconds after the gener-
ation commences. Examination of the maximum temperature of the
gas as noted in the third column of the table above indicates that the
temperature of the gas is reduced when large proportions of water
are used. When using from 1 to 4 parts of water, the temperature
is nearly uniform, but with 5 parts of water the decrease becomes
marked. Repetitions of the above experiment gave similar results.
The violence of the reaction and the temperature of the gas are affected
more or less by the size of the pieces of cyanid. A very violent reac-
tion results from the use of cyanid in powdered form.

We would expect that to increase the proportion of water woidd
decrease the temperature of the gas. One reason is shown in tin-
table under the column marked " Temperature of the acid and water
mixture." As the proportion of water to sulphuric acid becomes
larger the resulting temperature of the mixture is lessened. Hence
^vvhen the cyanid is added to the mixture as high a degree of heat to
start the reaction is not developed as when the smaller proportion of
water is used, and in consequence gas is evolved less violently.

THE TEMPERATURE OF THE GAS WHERE LARCH AND SMALL DOSAGES ARE USED.

In an experiment to determine the temperature of the gas result-
ing from large and small dosages (Table III) the chemicals were used
in the following proportion-: Cyanid 1 part, acid 1 part, and water

PROPOB HON OF CHEMK A I ,S.

37

3 parts. The reactions were accomplished in 2-gallon earthenware
fumigating vessels in a room where the air was moderately quiet.
The temperature of the gas was taken at the mouth of the vessels.

Table III. — Experiment to determim tht temperatun oftht gas resulting from largt <n,<l

sin <il I dosagi s.

Amount

of chemicals used

Time

Highesl
tempera-
ture ol
mixture
of acid
and

W .H«T.

after
mixing

w hen
tempera-
ture of

Temper-
ature ol
mixture

at <'iiil

tempera-
ture ol
ii\ dro-
.■\ anlc-

e ration
w hen

tempera-
ture "i

1 ••IH|MT-

ature of
minute

from

minutes

frnin

Cyanid.

\chl.

\\ iter.

mixture
is highest.

minute.

genera-
tion.

tlon.
F

-

Ounci t.

Ounces.

°F.

Seconds.

°/\

°F.

Seconds.

!

9

135

30

131

: ■

6

L8

5

157

104

86

V

B

24

167

25

15

30

L06

111

10

70

25

M

30

10

L2

12

L64

L57

,,ii

L13

95

11

14

12

L78

• '-

145

25

lit'.

16

16

[8

173

25

161

118

20

20

60

L72

25

168

L53

-''"'

123

106

An examination o( this table shows that the temperature of the
escaping gas increases somewhat as the dosages become larger.

Hence if heated gas is more injurious than cooler gas, we would ex-
pect more burning as a result of the increased dosages. This is
exactly what does happen to some extent in field operations. It is
interesting to note that the highest temperature of the acid-water

mixture occurs about one-half minute after the mixing takes place.
The highest temperature of the hydrocyanic-acid gas occurs about
one-half minute after the generation commences, and then the tem-
perature of the gas rapidly decreases during two to two and one-half
minutes, at the end of which time most of the gas has been evolved.
At the expiration of from three to five minutes the generation of gas
has practically ceased.

THK EFFECT OF DIFFERENT PROPORTION^ Ol WATER ON THE A.Mol NT OF AVAILABLE

HYDRO* YANK - A< CD GAS.

In the course of this investigation an experiment was made to deter-
mine the amount of hydrocyanic-acid gas available when generated
with different proportions of water. The results as determined by the
Bureau of Chemistry of this Department are given in the accompa-
nying chart (fig. 13).

In these experiments commercial sulphuric acid, G6° Baume or
92.77 per cent pure, and potassium cyanid 97.12 per cent pure wen4
used. Three ounce- (fluid) of sulphuric acid and 3 ounces (avoirdu-
pois) of potassium cyanid were employed in each experiment, and
3, 6, 0, 12, 15, 1 v 2 1 . and 2 1 ounces, respeel ively, of water were used
in the different experiments.

38

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

From the following chart it is evident that with the acid andcyanid
mentioned the largest amount of gas is available from two parts of
water. As the proportion of water is increased above two parts the
available gas is decreased until with eight parts of water we obtain
only about 43 per cent of gas, or less than one-half as much as with
two parts. In other words, 1 ounce of cyanid and 1 ounce of acid in
combination with 2 ounces of water will produce much more avail-
able gas than 2 ounces of cyanid and "2 ounces of acid with 1G ounces
of water.

The cause for the smaller amount of gas with one part of water
than with two parts has already been explained (see p. 35).

We can see from the chart that the proportion of water used is one
of the most important factors in fumigation practice; and many of

PRO
CYAN/D

PORT/L
AC/D

W50F
WAT£fi

PER CENT OF 6 A >S G/l/f/V OFF
10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

1

1
1
1
1
1
1
1

1
1
1
1
1
1
1
1

1

87.84
93.75
89.95
86.25
81.68
79.65
73.47
43.27

1

2
3
4
5
6
7
8

Fig. 13.— Chart showing total amount of gas evolved when different proportions of water are used.

(Original.)

the poor results in field work can be directly attributed to the use of
too much water. That the water should be measured as carefully as
the acid is beyond question.

Aside from variations in the amount of water used, due to lack of
precision in measuring, the proportion of water recommended by dif-
ferent authorities on fumigation has varied all the way from two to
eight parts. It is no wonder we see widely differing results from the
work of different men. It is a common practice with many funii-
gators to increase the dosage when fumigating a tree that is severely
infested with scale. It is also a common practice— in fact, so com-
mon as to be almost universal — to increase the proportion of water
when using heavy dosages. This is apparently done with a view to
preventing injury to the fruit and foliage. In following out this prac-
tice the fumigator has many times unconsciously prevented thevery
resull lie wished t<> accomplish — that of obtaining a more concen-
trated gas.

THE CORRE( I PROPORTION <>i WATER.

The chart (fig. 13 shows. that two part- o{ water to one part
cadi of cyanid and acid will produce the maximum amount of avail-

Mixix<; in i i 1 1 EM h \i.s. 39

able gas. It is impracticable, however, to use t w < > parts <>l water in
field work, for with this proportion <»f water the residue, especially
where small dosages of powdered cyanid are used, w ill Frequently con-
geal within an hour's time the usual period lor leaving the tents on
the trees. Although this proportion of water is apparently sufficient

to dissolve the sulphate at first .so that a complete reaction takes

place, it appears unable to hold the sulphate in solution long enough
afterwards to prevent inconvenience in held work. It is of course evi-
dent that a "frozen" generator does not always signify an unsatis-
factory generation. With three parts of water, however, the residue
seldom congeals, and this is the proportion we have used in all of our
field work and which we recommend. The water should he meas-
ured carefully with a glass or dipper graduated to ounces

THE Most ECONOMICAL PROPORTION <>i CHEMICALS 1*0 l-i I\ GENERATING
BYDROCYAN 1< -ACID GAS

In the preceding discussion it has been shown that for various
reasons l fluid ounce of commercial sulphuric acid and l ounce (avoir-
dupois) 96 to 100 per cent potassium cyanid in combination with '4

fluid ounces of water give a complete reaction. Thus the 1 1-3 for-
mula, hitherto recommended by the Bureau of Entomology, is fully
indorsed.

A review of the use of hydrocyanic-acid gas for fumigation, both in
California and elsewhere, shows frequent divergence from the more
economical and satisfactory proportion of chemicals indicated above.
One book recognized as an authority on fumigation methods recom-
mends the use of "one-half more acid than cyanid and one-half more
water than acid.'' Many of the entomologists and horticulturists in
the eastern United States advise in their recommendations for nur-
sery fumigation two parts of acid and four parts of water to each
part of cyanid.

MIXING THE CHEMICALS.

It is preferable to pour the water into the generator first and then
add the acid. The pouring of the water onto the acid is more likely
to cause splashing of the acid from the jar onto the himigator. When
the acid and water are in readiness for generating the gas the i'umi-
gator adds the pieces of cyanid to the mixture and hastily retreats.
A- already stated, the cyanid should be added while the mixture of
water and acid is hot. The advantage of this is shown in the fol-
lowing experiments performed by the Bureau oi Chemistry of this
Department. One ounce of potassium cyanid, 1 fluid ounce of com-
mercial sulphuric acid, and 3 fluid ounce- of water were used in each
case.

Experiment No. I. The potassium cyanid was added to a mixture
ol acid and water in which the heal was exhausted, and it was found

40 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

that 23.25 per cent of hydrocyanic acid remained in solution and was

not liberated.

Experiment X<>. ... — The potassium cyanid was added to a mixture
of acid and water when first combined, i. e., when the heat was great.
and it was found that only 10.08 per cent of hydrocyanic acid
remained in solution.

Caution. — The cyanid should never be placed in the water before
the acid is added. If the acid is added to the cyanid in solution, a
very violent reaction takes place, which will sometimes throw much
of the liquid from the vessel. In one instance about 1 pound of
cyanid was dissolved in water in a 2-gallon generator. Acid was then
added, producing a disturbance so violent as to throw some of the
liquid almost to the top of a two-story barn.

V 1 l-gallon generator will serve for a dose of about 15 ounces of
cyanid without boiling over, or a 2-gallon generator for approximately
20 ounces.

The residue from the reaction contains more or less sulphuric acid
which has not been used. This residue should never be deposited
against or at the base of a tree, as it may penetrate to the roots,
especially in light sandy soils, destroying a part if not the entire tree.

PURPLE SCALE FUMIGATION.

PRELIMINARY EXPERIMENTS FOR THE CONTROL OF THE PURPLE SCAJ.h.

During the month of November, 1907, experiments were under-
taken at Orange, Cal., to determine the dosage required for the
destruction of the purple scale (Lepidosa phes heckii Xewm.) in all it-
stages, as well as to determine the effect of exposures of different
durations. The orchard under treatment contained orange trees
varying from 7 to 14 feet in height. The infestation with the purple
scale was very severe on many of the trees.

In the first experiment the duration of exposure was thirty minutes.
In this experiment a series of tests was made to determine the effect
of different dosages. These tests were as follows: One series of trees
was dosed at the rate of three-fourths ounce of cyanid per 100 cubic
feet of inclosed space; a second series at the rate of 1 ounce, a
third at the rate of \\ ounces, and so on, increasing the dosage of
each succeeding series at the rate of one-fourth ounce per 100 cubic
feet. The largest dosage used was 2\ ounces per 100 cubic feet.

The second and third experiments were the exact counterparts of
the first in all respects except that the duration of the exposures was
respectively one hour and one and one-half hours.

From the data secured from these experiments it should be pos-
sible to determine the killing dosage tor the purple scale for that
particular length of time, provided a sufficient strength of gas was
reached. To insure that the dosage sought would fall within the

FUMIGATION AGAINST NIK PURPLE SCALE.

41

Bcope of the schedule, the Limits were made verj broad. Prom the
difference in strength of killing dosage between these three experi-
ments we would be able to determine the effect of length of exposure
on results secured.

To obviate as much a> possible the leakage of gas, \\ bich w ould vary
iu trees of different sizes, trees were chosen of as uniform a Bize as
could be obtained. The cubic contents of the trees chosen for the
first two experiments did not vary greatly and the tree- ranged
between 11 and 14 feet in height. As in the first two experiments
most of the larger trees had been used, for the third experiment we
were' compelled to utilize those remaining, which varied somewhat
in size, and were also, for the most part, noticeably -mallei' than
those represented in the4 lirst two experiments.

During the latter part oi January an examination was made of the
result- of these experiments. Fully two weeks were devoted to this,
and thousands of the purple scale were scrutinized. The method
employed was a very careful one. In each case the scales were
overturned ami examined with a powerful hand lens. In those
instances in which the entire contents of the scale were not at once
revealed, the delicate ventral scale was ruptured and the contents
scraped out. Through this method not a single egg could escape
observation.

Four trees were used in each test and an examination to deter-
mine results was made of each. This examination included many
infested leaves and branches taken as close to the ground as possible
and up to 6 or 7 feet above the ground. Infested fruit was also
examined when obtainable. The average condition existing in these
four trees was taken to indicate the result of the test.

The chemicals were used in the following proportion: Potassium
cyanid, 1 part: sulphuric acid. 1 part: water, 3 parts.

Table IV. — Fumigation for tfu purple scale, experiment X<>..
[Length of exposure, thirty minutes; height of trees, ll to 14 feet.]

Number <>"ani'1

.\umn.r per 100

treated. „

i in leaves and
branches.

on fruit.

[nsects
alive,
approxi-
mately.

Eggs
normal.
approxi-
mately.

[nsi cts
ili\ e,
approxi-
mately.

normal, approximately.

Ouncfi.

4

4 1

i 11

4
1
1
1 2}

I'i r n ill .

II
0

0

n
0
0

"

Per a nt.
Over 75.

MlMllI 7."..

i • than i.

ii
(i
n

Per a nt.
10

2
0

0
0
0

(1
1)

Fully 'in percent.

Many normal eggs found under ever]

scale containing eggs.
Some normal eggs found under almost

evi ry scale containing eggs.
15 per cent.
5 7 pi r cent.
i per cent.

Two instances of ippari ntly normal eggs.
None.

42

FUMIGATION INVESTIGATIONS IX CALIFORNIA.

In this experiment, when three-fourths of an ounce of cvanid per
100 cubic feet of space was used, live adult females were found on
the leaves and branches, but the insects were killed by all greater
dosages; normal eggs were found after the use of a dosage as high as
1} ounces per 100 cubic feet. Live insects were found on the fruit
after both the three-fourths-ounce and 1 -ounce tests, but were de-
stroyed by the heavier dosages; normal eggs were found on the fruit
after dosages up to and including the 2J-ounce rate: with 2$ ounces
per 100 cubic feet, all were apparently destroyed.

This experiment indicates that for normally shaped orange trees,
from 11 to 14 feet in height, situated in a region with conditions
comparable to those at Orange, and exposed to the gas for thirty
minutes, a dosage of about 2 ounces per 100 cubic feet is required
for eradication of the purple scale from the leaves and branches. If
the trees contain fruit infested with scale, it is necessary to increase
the dosage rate to 2\ ounces to accomplish the same result.

Table V. — Fumigation for the purple scale, experirm nt No.
[Length of exposure, one hour; height of trees, 11 to 14 feet.]

Number

of trees
treated.

Cvanid

per 100

cubic feet

of space.

On leaves and branches.

On fruit.

Insects
alive,
approxi-
mately.

Eggs normal,
approximately.

Insects
alive,
approxi-
mately.

Eggs normal.

4
4
4
4
4

4
4
4

Ounces.

1

U

IV
lj

2

2l

2£

ooo ooooo

1-5 percent.

1 percent or less.

2 doubtful cases.
0
0

0
0
0

0
0

0
0

0
00

Many instances.

Several instances.

(°)

One doubtful

Few instances of normal eggs on

one fruit.
No instances of normal i e
CO
(a)

n No infested fruit on these trees.

With an exposure of one hour all insects were destroyed on the
leaves and branches at a three-fourths ounce dosage rate. All eggs
were destroyed at the H-ounce dosage rate. Since very few oranges
infested with scale were found on the trees used in this experiment, it
is considered that further investigation will be necessary before the
effect of different dosages on scale infesting the fruit is definitely
known. Xo live insects Avere found infesting the small amount of
fruit available. Normal eggs were found after a dosage as high as
the 1 v-onnce rate.

This experiment would lead to the conclusion that for normally
shaped orange trees, from 11 to 1 i feet in height, exposed to the gas
for one hour, and situated in a region with conditions comparable to
those at Orange, a dosage rate of 1J ounces per 100 cubic feet will

LEAKAGE OF GAS IN SM \i.l. 1 BEES.

43

destroy the purple scale in all its stages on the leaves and wood. If
the tree contain fruit infested with this scale it will be necessary to
slightly increase the dosage. The exact amount of this increase can
not be stated with accuracy at tins time, owing to the fact that in the
single experiment performed very little infested fruit from winch
data might be secured was available.

Table VI. Fumigation for (hi purpU scale, experiment No
[Length of exposure, one and one-hall hours; size of trees mostly 7 to 10 feet; oe< aslooall; on 11 oi I

» >n ii'.i\ es

and branches

< »n fruit.

Number
of trees
treated.

( yanld

per 100

cubic feel

of space.

Insects all \ e,

i normal,

I n ects ;iin e,

approximate^ .

approximate^ .

ipproxlmatelj

i g i normal

Ounce*.

4

.

3 5 per cent.

10 50 per cent.

10 15 per cent.

Vbove 75 per cent.

l

l

l live female.

20 25 i«t cent.

i 2 per cent.

i i 75 pel cent

4

! 1

0

i 5 per cent .

4

1*

i)

l 2 percent.

0

rwoor inges examined;
many norma] i
present .

1

i;

0

3 Instances ol doi •
in ;i l

"

A I'm normal

4

2

0

o

(")

(")

4

• >i

0

0

(")

4

u

0

0

(«)

'">

a \o material for examination.

In this experiment live insects were found on the branches and
leaves in the cases where three-fourths ounce and 1-ounce dosages were
employed. Normal eggs were found up to and including the 1 |-ounce
rate, but were destroyed by dosages exceeding this. As in the case of
experiment Xo. 2, so little scaly fruit was available at that time that
we are inclined to consider the results in this part of the test as yet
incomplete.

THE LEAKAGE OF GAS IN FUMIGATING SMALL TREES

When the results of experiment No. 3 are compared with those of
experiment No. 2 we are at first led to believe that an error has been
made. Jn experiment No. 2 it was found that the 1 \-ounce dosage
rate destroyed all insects and eggs on the leaves and branches, whereas
in this experiment it required one-half ounce more cyanid per 100
cubic feet, or a 2-ounce dosage rate, to accomplish the same result.
Since the period of exposure was thirty minutes longer than that of
experiment No. 2, we would naturally expect that tin' results accom-
plished would be as good or better, all other conditions being the
same. The apparatus and chemicals employed were identical in both
cases; and the conditions under which the fumigation w as conducted
were practically the same. There was, however, one difference: The
trees involved in the one and one-half hour fumigation were much
smaller than those of the one-hour test. This fact accounts for the

44 FUMIGATION INVESTIGATIONS IX CALIFORNIA.

less satisfactory results in eradicating the scale, in this experiment.
We know that a leakage of gas takes place through the tent and that
more gas will escape through 2 square feet of cloth than through 1
square foot in a given time. It will be shown in one of the following
discussions that the leakage surface of tented trees is proportionately
much greater for smaller trees than for larger ones. This would lead
us to expect a greater escape of gas and consequently the requirement
of a heavier dosage rate with the smaller than with the larger tree-.
The last experiment demonstrated the correctness of this deduction.

THE LENGTH OF EXPOSURE.

All considerations were the same in experiments Nos. 1 and 2 ex-
cept the length of exposure. In using a 2-ounee dosage rate, wo were
able to destroy the purple scale in all of its stages on the leaves and
branches with a thirty-minute exposure, whereas with a one-hour
exposure we were able to accomplish the same results by using a H-
ounce dosage rate. This demonstrates that decidedly better results
can be secured by leaving the tents on the trees one hour than is
possible with thirty minutes gassing. Whether more favorable
results can be accomplished in one and a half hours than in one hour
can not be determined from these experiments, since the trees in
experiment Xo. 3 were of a smaller size than those in experiments
Nbs. 1 and 2. This matter of the large or small size of the trees is a
vital factor in affecting the results obtainable.

Judging solely from the data at hand, we are forced to the conclu-
sion that one hour is the more satisfactory length of exposure. Fur-
ther experiments may show that a longer exposure will produce better
results, or even that a forty-five or fifty-minute exposure will produce
results as satisfactory as are obtainable in one hour. ^Ye hope in the
near future to be able to fully settle this question. Until this is done,
however, it would appear advisable to adhere to the one-hour length
of exposure which is now generally employed in southern California.
The considerations upon which this conclusion is based are as follows :

Experiments have demonstrated conclusively that with an expo-
sure of one hour we can obtain decidedly better results than with an
exposure of thirty minutes. If we give the tree an exposure of
thirty minutes, it will require a considerably larger amount of cyanid
to accomplish the same result. It requires approximately one hour
for an outfit to go through the complete operation of preparing the
chemicals and shifting 30 to 33 tents — the number usually employed.
The tent pullers, by the time the end of the row is reached, are usually
as much as five minutes, sometimes more, ahead of the one who
handles the chemicals. A- a result, the last trees of a row are ex-
posed to the gas about fifty-five minutes, or a little less, under the

DURATION OF EXPOSURE. If)

present system, whereas an hour is supposed to be the length of
exposure throughout. Thus it' fifty minutes is found to give as
satisfactory results as an hour, it would be poor policy to reduce the
general exposure to this basis, inasmuch as with a general exposure
of one hour some trees are already receiving but little more than fifty
minutes.

As a rule, very little gas remains under the tent at the expiration of
one hour. The amount is usually so small that the mortality among
the scale-insects could he hut slightly increased by greatly lengthen-
ing the exposure. \ arious authorities have recommended two hours
or more as the duration of exposure, and it i> possible that these long
exposures would produce slightly better results than an exposure <>!'
one hour.

From the standpoint of the fruit grower, who requires the best
results at the least possible expense, the item of time is highly impor-
tant. The question which must be considered is whether it i> more
advantageous to sacrifice time or cyanid. No doubt it is cheaper to
sacrifice time up to a certain point, hut beyond this it is cheaper to
sacrifice cyanid. As previously stated, the mortality among scale-
insects, when a two-hour exposure i> employed, might he slightly
greater than at one hour. Before advising a two-hour exposure,
however, we must determine whether or not it would he more eco-
nomical to employ an exposure of one hour and use sufficient cyanid to
accomplish the same results secured by the longer time. Fumigators
are usually paid by the hour. Where tents are left on the trees two
hours, with the same number of tents the cost for labor is exactly
twice that for one hour. From 4 to 6 men, at an average wage of :;.",
cents per hour, are used on an outfit (infrequently 3), making the
hourly cost for labor from SI. 40 to $2.10. This would purchase from
5 to 7 pounds of cyanid/' Under these circumstances, if we can
obtain as good results in an hour by using 5 to 7 pounds more of
cyanid — or a smaller amount, according to the number of men in the
outfit — it would be more economical in the (Mid to use the additional
cyanid and expose for the shorter time. The writer's own held
experience Leads him to believe that as good results can be accom-
plished in one hour as in two hours by using an amount of cyanid
costing far less than would the extra hour's labor.

It will be seen that the question before the fumigator is not simply
one of using that length of exposure which will produce the best
results, but that which will at the same time be most economical.
From field experience and other considerations the writer is led to
believe that this will be between (il'ty minutes and one and one-half
hours.

a Cyanid i- here considered as including acid, both costing aboul 28 cents per pound.

46 FUMIGATION [INVESTIGATIONS IN CALIFORNIA.

ERADICATION OF THE PURPLE SCALE.

The foregoing experiments have shown that the purple scale can
be eradicated from citrus trees, provided a dosage of sufficient
strength be used with a sufficient exposure. This dosage strength
is much greater than that at present used in fumigation.

If the purple scale can be everywhere eradicated by using a dosage
of definite strength 'which we hope to determine in due time), the
question will immediately arise in the orchardist's mind whether it
will be profitable to use this heavier dosage provided it can be em-
ployed without injury to the tree and fruit. In deciding this ques-
tion several practical considerations must be taken into account.
The trees, as will be shown later, are in a condition to stand this
heavy dosage without injury during but a limited portion of the year.
It would he impossible for the number of outfits at present in exist-
ence to fumigate the infested area within this limit of time. More-
over, unless compelled to do so the orchardists in any locality would
not all use this dosage. Whether it would be advisable for a grower
to incur the additional expense for this heavier dosage in his orchard
when the infested orchards on all sides of him are fumigated with
lighter dosages, if at all, must be determined by large-scale tests.
The foregoing are some of the difficulties in respect to the use of this
heavy dosage.

DIFFICULTY OF DESTROYING THE SCALE OX THE FRUIT.

There is one more important point which must be considered in
connection with fumigation for the purple scale. It will be seen in
an examination of the data from the foregoing experiments that an
orchardist, fumigating trees containing purple scale in its different
stages on the fruit as well as on the leaves and branches, would,
except with the heaviest dosages, leave on the fruit healthy eggs soon
to hatch and infest other parts of the trees. It would be impractical
under most circumstances to use a dosage heavy enough to destroy
the eggs on the fruit. The cost of the extra cyanid required, above
that necessary for the destruction of the eggs on the leaves and
branches, would be more than the scaly fruit is worth. Therefore
in fumigating for eradication it is advisable to remove the infested
fruit, and it is advisable to remove the old scaly fruit in any fumi-
gation. At picking, fruit badly infested with scale is usually left
on the tree, and frequently from one to a half dozen or more old,
scale-infested oranges per tree remain throughout an orchard. Even
after a good fumigation one of these old fruits might carry more
healthy purple-scale eggs than all the rest of the tree, and on the
hatching of these eggs the insect- will spread to other parts of the
tree. The danger from old scaly fruit is evident and all such should
be removed from the tree- before fumigating an orchard.

LEA K \>,i « >i GAS im RING I 'I'l l; \ I [0N6. 17

GENERAL CONSIDERATIONS.
LEAKAGE OF GAS Dl RING OPERATIONS.

One of the most important questions relating to the proper dosage
in fumigation is that of the leakage of gas through the tent; in fact,
the dosage depends directly upon the leakage. To measure with
accuracy the amount of gas which escapes through tenting fabrics
of various grades during a given length of time, or the rapidity with
which the gas within the tent is diluted under different condition-,
is a difficult problem. In this work, as far as we have progressed, no
attempt has been made to measure directly with instruments the
rapidity with which the gas is diluted, but rather to measure it indi-
rectly and roughly through determining the effect on insects by using
different durations of exposure. The easiest and most practical
method of determining the influence of leakage is by fumigating
trees of the same size, in which all factors affecting the results arc
identical with the exception of t lie length of exposure.

There is, however, one consideration of value relative to the leak-
age of gas, which it is quite necessary to understand in successfully
fumigating an orchard containing trees of a wide range of size. In
geometrical figures which approximate in shape a citrus tree, the
volume decreases at a more rapid rate than does the surface area.
In order to bring out the relation of this fact to orchard fumigation,
the following table has been prepared:

Tablk VI I.— Leakage of gas from t> nts covering trees of different dimt nsions.

Dimensions of tree.

( ontents
oi volume
of tented

lice.

Exposed

surface of

tent.

Leakage
surface as
per cent of

volume.**

Around.

Over.

Feel.
12

id
28
36
44
54

Feet.

20
30
40
50
60
70

( ubic U t.

MM

364

1,040
2.147

.,.(,1!.-,

S quart in I.
85
205
420
675
995
1 . 1 15

/ '< r rent.
86
56
40
31
26
22

a The comparison here and in the discussion which follows is between square feci of surface and cubic
feet of volume.

Taking the first tree, 20 feet around by 12 feet over, representing a
volume of 1)9 cubic feet and an exposed surface area of 85 square
feet, the ratio of leakage surface to volume is Mi: loo. For each cubic
foot of volume within that 20 by 12 tree there i^ 0.86 square loot
of leakage surf ace in the tent. The tree lo by 28 feel has 0.4 square
foot of leakage surface for each cubic fool in the tent, while a tree
70 by 54 has but 0.22 square foot of leakage surface to each cubic
foot within. Suppose that these tented trees were charged with

gas and that all the gas were to ocape through the tent. In the

48 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

first tree, 20 by 12 feet, there would be 0.86 of a square foot of tent
surface for each cubic foot of gas to escape through; whereas in the
last tree, 70 by 54, there would be only 0.22 of a square foot of tent
surface for each cubic foot to escape through. This would mean
that there would be about four times as great an opportunity for
leakage, or that the leakage would be approximately four times as
rapid in the smaller tent as in the larger one.

There can be little doubt that the leakage of gas in tents covering
different-sized trees is nearly in accordance with these figures. Hence
it can be readily seen that, in order to secure uniformity of results,
this leakage must be taken into consideration, and small trees must
receive more cyanid per 100 cubic feet than do the larger trees.

The correctness of the foregoing deduction has been frequently
demonstrated in the field. In using on a smaller tree a certain dosage
strength with which on large trees we were able to secure splendid
results against the purple scale, we were always much less successful.
In other words, if we used 1 ounce of cyanid per 100 cubic feet on
the 70 by 54 foot tree, we would get far better results than had we
used the same dosage rate on the 20 by 12 foot tree. A very forcible
exemplification of this condition has been given in experiment
No. 3, in fumigating for the purple scale. In this particular experi-
ment much less satisfactory results were secured on the small trees
when using a one and one-half hour exposure than on the large
trees of experiment No. 2, with a one-hour exposure.

TIME OF THE YEAR FOR FUMIGATION.

Although fumigation is carried on in California at all times of the
year, there are certain periods in which the operations are more
general. There are two main factors to be taken into consideration
in fumigating, i. e., the species of scale-insect and the condition of
the tree. As to the latter, it may be said that at certain periods of
the year trees are in such a tender condition that they can not
withstand a heavy dosage without injury, especially to the fruit.

The bulk of fumigation in California at the present time is carried
on between the latter part of August and December. Probably the
principal reason for fumigating during this period is that at this
time the black scale is most successfully reached. The eggs of the
black scale, and the insects themselves when full grown or nearly so
(commonly spoken of as in the u rubber'' stage), require very heavy
dosages. On the other hand, the young of the black scale, or those
which have not reached the so-called "rubber" stage, can be
destroyed with a moderate dosage. Although the life history of
the black scale has never been thoroughly worked out for the region
with which we have to do, it is generally understood that the majority

TIME OF Vl.vi; FOB FUMIGATION. 49

of the insects of the large and more regular brood are batched and
in their least resistant stage during September and October. In
some favorable seasons the eggs are almost all batched in August.
Moderately light fumigation dosage may be used against the black
scale during this period with success.

The black scale occurs in practically every citrus-growing locality
of southern California, while the purple, red, and yellow scales, the
other principal citrus pests, are more localized. A heavier dosage
is u>cd for the latter insects than for the black scale. Where the
other species occur in orchards infested with the black scale, it is
a common practice to fumigate during the regular black-scale
period, using the heavier dosage. The majority of these scale insects
can thus l>e caught at one time. When fumigating for the purple
scale alone, operations may be commenced as early in the season as
tin4 trees are in a condition to withstand the heavy dosage without
injury, although probably it would be preferable to fumigate a little
later in the fall. The purple scale is to he found in the egg stage
throughout the year. There is a period in the fall and one in the
early spring, however, during which the smallest proportion of eggs
is to 1)0 found. With dosages lower than those of eradication, the
best work can be accomplished at these times.

The rod and yellow scales are viviparous and can be successfully
destroyed throughout the year.

In fumigating for any of the scale-insects there is one point
worthy of consideration. Aside from trying to save the tree from
destruction or from having its vitality impaired by the attack of
scale pests, the orchardist fumigates principally in order to have
his fruit come into the packing house as clean as possible. It would
be well, therefore, to fumigate as nearly as possible to the time
which would insure him the cleanest fruit. Although lemons are
gathered throughout the entire year, the bulk of the orange crop is
taken during the first six months. Thus fumigation during the fall
and early winter would be sure to place the cleanest fruit in the
packing house. If carried on in the late spring or early summer,
such insects as remain undestroyed would have the opportunity to
breed through a period of several months and infest much fruit.

FUMIGATION DURING TIM: BLOSSOMING PERIOD.

The statements by experts on fumigation as to tin4 amount of
injury resulting from work while the trees are in blossom are very
conflicting. Some fumigators hold that a very light dosage will
destroy the tender blossoms, while others believe that the blossoms
will stand a heavy dosage. In order to decide this point much experi-
mentation was carried on and many observations made throughout
77488— Bui. 7«i— <»!» 1

50

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

the blossoming period of 1908. Some of the results secured are
given in the following paragraphs.

Experiment No. 1. — On February 28 and 29 about one-third of an
acre of mixed Valencia and Navel orange trees was fumigated at
Upland, Cal., using dosage rates of 1 ounce and 11 ounces per 100
cubic feet. The trees were about 12 feet in height. At this time
the blossoms were just appearing on the trees, none of them being
far enough advanced to open. The general conditions of the blos-
soming may be understood by an examination of figure 14. This

Fig. 14.— Orange blossoms at an early stage of development. (Original.)

may be considered the tenderest stage of blossoming. An examina-
tion of these trees two weeks later showed that no apparent injury
had resulted and that the trees at this tune contained as heavy a
set of blossoms as the surrounding unfumigated trees.

Experiment No. 2. — On March 30 fully 1 acre of Navel and Valencia
orange trees about 10 feet high were fumigated at Orange, Cal.. using
dosage rates of 1 , 1 ], and 2 ounces per 100 cubic feet. The condii ion
of blossoming at the time of fumigation ranged from no open blos-
soms on some trees to full blossoms on others. An examination of
these trees at a later date showed that with the 1 and \\ dosage
laics no apparent injury had been done. The 2-ounce rate had caused

1 T.\| l(i \ I [ON N\ I I II. I I I : I 1 I l> >.M \i.l.. 5 1

b considerable percentage of the blossoms to drop, ye\ not enough
to lessen the coming crop of fruit to any great extent . if at all.

Experiment No. 8. During the months of April and May, 25 acres
of Valencia and Navel oranges at Glendale, Cal., were fumigated by
an expert under the direction of the Los Angeles horticultural com-
mission. While this fumigation was in progress, trees could be
found in all stages of blossoming, from those with blossoms just
appearing to those in full bloom. The dosage rate used was esti-
mated to be from three-fourths to l ounce per loo cubic feet. Of
course this rate varied with different tree-, since the dosage was
estimated after the usual guesswork method. Several examinations
of the orchard were made. Although blossoms were injured on some
of the (roes, the number was so small as in do u ay to lessen i he ful ure
crop of fruit.

Other instances might be mentioned, but the results correspond
practically with those in the three experiments already described.
Trees in which there were blossom-shoots and tender leaf-shoots
side by side would have the leaf-shoots burned back while the blos-
soms remained uninjured. Also numbers of cases could be found
where the tender leaves on the blossom-shoots were burned while
the blossoms themselves remained uninjured. This, as well as the
heavy dosage which the blossoms will stand without injury, would
lead us to conclude that the blossoms will stand a heavier dosage
than the tender leaves and leaf-shoots. These experiments also
show that fumigation can be safely conducted during the blossoming
season, using such dosages as are at present generally employed by
fumigators, or are advised in dosage schedule 1 (p. 65).

FUMIGATION WHILE THE FRl IT is OF SMALL SIZE.

Experiments and observations to determine the effect of fumiga-
tion on fruits of various sizes, and more especially on small fruits,
Were made during the season of 1908. Conflicting opinions on this
subject are prevalent.

Juperiment No. 1. — On June 16 two Valencia orange trees about
S feet in height, in a healthy condition, and containing young fruit
from three-eighths to one-half inch in diameter, were fumigated at the
2-ounce dosage rate. Fully 25 per cent of the fruits on these trees
were pitted or burned.

Krperiment No. 2. — On June1 24 a somewhat unhealthy Navel
orange tree about L2 feel in height, 'with the fruits about one-half
inch in diameter, was dosed at the rale of U ounces. Fully 50 per
cent of the fruits were pitted. Two healthy Valencia orange trees
about io feel in height, with fruits practically the same size as in
the case of the Navel tree, received a dosage at The rate of 2 ounces.
About 40 per cent of the fruits were burned

52 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Experiment No. 3. — On July 11 and 13, four Valencia orange trees
were fumigated, using a lj-ounce dosage rate, 4 trees receiving a
H -ounce dosage, 8 trees a l§-ounce dosage, and 4 trees a 2-ounce
dosage. These trees were in a perfectly normal condition, about
7 to 8 feet high, and contained young fruits fully three-fourths of an
inch in diameter. With the lj-ounce dosage rate no fruit was
burned; with the H-ounce rate an occasional orange was slight ly
burned; with the 1 |-ounce rate1 a very small percentage was burned,
while with the 2-ounce rate a considerable percentage was injured.
This demonstrates that a 2-ounce dosage rate could not be safely
used on trees of this size.

Experiment No. 4- — During the middle of July a large number of
orange trees of all sizes were fumigated at Santa Fe Springs, CaL,
using various dosage rates. The trees fumigated were of several
varieties, in a healthy condition, and all well filled with fruits about
the size of an English walnut and slightly larger. It was found from
this experiment that a dosage rate of 1 ounce to 100 cubic feet could
at this time be used without injury on orange trees 15 to 16 feet high.
Only an occasional orange was burned by l\ ounces. Smaller trees
proved able to stand a heavier dosage than larger ones without
appreciable injury.

On the basis of information obtained from experiment No. 4,
dosage schedule 1 (p. 65) was prepared. This schedule was put into
use during the latter part of July and has been in use, up to the
time of writing, by two outfits, at Whittier, Cal. Although no
noticeable injury to the fruit has resulted from the use of this dosage,
the general effect on the tree has indicated that a heavier dosage
could not have been used with safety.

A further example of the tender nature of small fruits was shown
in some work done by an excellent fumigator at Downey, Cal., during
the latter part of May. The fruits were for the most part three-
eighths of an inch or less in size, while the trees were thoroughly
infested with scale and in a generally unhealthy condition. So far
as could be determined, a dosage rate of approximately three-fourths
to 1 ounce was used. The larger percentage of the fruits on these
trees was burned. Other instances of like fumigation, where the
fruits were one-fourth inch or less in diameter, have been seen. The
fruit at this period is very tender. Doubtless it is the most critical
period of any during which fumigation is conducted.

From the foregoing, it is evident that heavy dosage can not be
used while the fruits are small without more or less injury, and that
the most critical period during which fumigation may he conducted
is between the time when the fruits are set and the time when they
attain the size of a walnut.

HANDLING THE ACID. 53

SIMPLE METHOD OF REMOVING ACID FBOM DRUMS AND CARBOYS.

The writer has ai times been obliged to employ rather awkward
methods in drawing acid from drums and carboys, and other fumi-
gators have doubtless met with the same trouble under like circum-
stances. Brief mention will he made of some of the best methods
which have been brought to notice to obviate this difficulty.

From drum*. The best method of taking acid from drums known

to the writer is that at present in use in San Bernardino County and
is .shown in figure 15. The apparatus consists of a, lead-lined tank

large enough to hold a. drum of acid and having an outlet through

FlO. 15.— Lead-lined tank used in San Bernardino County for removing sulphuric acid from drums and

for filling jugs. (Original.)

which the acid may be drawn into carboys, jugs, or whatever vessels
are preferred for field use. A drum of acid is rolled from the wagon
upon two parallel beams and along these beams onto a small turn-
table at the tank. This turntable is thenTevolved through a quarter
circle, permitting the drum to be rolled out over the lead-lined tank,
into which the acid is then allowed to How. The acid may be drawn
as previously mentioned. The outlet is made of lead tubing, fitted
at the tank end with a lead valve by which the (low is regulated.

Another w\y satisfactory way of drawing acid from drums came
to the writer's attention in examining some operations at Glendale,
Cal. It consists in the use of a short iron pipe threaded at one end

54

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

so as to fit the opening in the drum. The one difficulty with this
device is that the flow of acid is uneven and spouting. To offset
this, Mr. William Wood, of Whittier, Cal., has contrived a small

copper tube for
attachment to the
pipe, one end of
the tube being ex-
posed to the open
air, the other end
extending up above
the level of the acid
within the drum,
thus allowing an
uninterrupted How
of air into the latter. This apparatus is illustrated in figure 16.

A third method in use is to transfer the drums from the wagon to a
platform 2 or 3 feet high. The acid may then be removed very easily
by means of a piece of rubber hose employed as a siphon (fig. 17).

Fig. It

mproved pipe for removing acid from drums. (Original.)

Fig. 17.— Siphoning acid from drums by means of a rubber hose. (Original.)

From carboys. Two common methods used for removing acid
from carboys in the held are shown in figures 18 and 1!). In the first
method a small amount of dirt is placed againsl one side of the car-

'I'll i: MARKING 01 I IN IS.

boy, Furnishing a son of rest when the latter is tipped to remove the
acid. It is well to scoop out a small pit below this ridge of dirt, into
which the vessel receiving the acid may be lowered when the acid is
bo largely removed that it is necessary to turn the carboy far on its
side in order that all may be withdrawn.

In figure 19 the handles on the carboy are substitutes for the heap
of dirt and the pit. They are also of service in carrying the carboy.

Tin: PROTECTION OF

(YAM I).

Many fumigators do
not attempt to cover
their cases of cyanid,
hut leave them open

during the day. Tins
not only constitutes a
source of danger to

various animals, but
also during the wet
season allows water to
reach the cyanid. Fig-
ure 20 shows a simple
lid covered with zinc
which is suitable for
placing on a cyanid
case to protect its con-
tents.

HYDROCYANIC-ACID

GAS IX DRUMS.

Some discussion has
arisen during the past

Year relative to the Fig. 18.— Carboy resting against a heap of dirt to facilitate pouring

possibility of introduc- • "kL (0rigInal-

ing hydrocyanic-acid gas into drums under pressure, and using it
directly from tin1 drums, thus doing away with all generation in the
field. The use of this gas under pressure from drums is impossible
at the present time for two reasons: (1) No drums are made which
will hold hydrocyanic-acid gas without corroding; (2) we know of no
instrument which will measure gas accurately under varying degrees
of pressure, such as would exist in removing a gas underpressure from

drums.

THE MARKING <>l TENTS.

Before new tents are marked they should have been m use for a
-hort time, so that they will be thoroughly shrunken. This shrinking

56

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

may be accomplished, in regions of heavy dews or fogs, by simply
leaving the tents exposed in the open for a few days. Dipping in

water or sprinkling by
means of a hose and then
allowing the tent to dry
in the sunshine will an-
swer the same purpose
if repeated several times.
The shrinkage of a new
45-foot tent will some-
times be as much as 3
feet. Tents marked be-
fore being shrunk will
have erroneous gradu-
ations.

The most satisfactory
material to use in mark-
ing tents is diluted
printer's ink. This ink
is commonly used in ( Jal-
ifornia in marking walnut
bags. If the ink is too
thick to mark freely, it
maybe diluted with kero-
sene. Printer's ink does
not cause the cloth to de-
teriorate. A mixture of
lampblack and turpentine may also be used with entire safety. The
latter, however, will sometimes rub off to a slight extent.

Fig. 19.— Carboywith handles attached to facilitate pourin
the acid and carrying the carboy. (Original.)

A DEVICE FOR COVERING FUMIGATION GENERATORS.

During the course of this investigation much effort has been
directed toward perfecting a device for attachment to the top of the
commonly used open-style fumigation generator that will serve to
interrupt the direct rise of the hydrocyanic-acid gas. The result of
these efforts, in which the writer was greatly aided by Mr. Frederick
Maskew, is shown in figure 21. The device itself consists of a copper
cover of such size as to make it available for use with any of the
regular-pattern generators now employed by the lumigators of
southern California. It is stamped in a concave form from a sheet
of copper, with corrugations to permit the escape of gas. The shape
is such as to conform to the size of the opening of generators of dif-
ferent capacities and also to direct the course of the escaping gas
downward and distribute it uniformly through the lower part of the

A COVER FOB n M l«. \ l l»».\ ,,i..\ ERA CORS.

i) <

tent. It is attache*! to the generator by iiinges of stout copper wire
Becured by a key bolt passing through the handle. The cover is
raised by a slight pressure of the thumb od a projecting piece which
ie curved in such a manner that the cover will remain in an upright
position when so required. When the generator is emptied of its
contents, t ho cover swings clear by its own weight. A glance at the

illustration will satisfy the practical I'ninigator that it is adapted to

all the requirements of rapid work in the dark, while its use has

demonstrated that it is simple, strong, and durable. It is \<t\
possible that if the copper cover were lined with a thin covering

o[ Lead its durability
would be increased.

A common result of
the use of heavy do-
sages o( line fragments
of cyanid is the burn-
ing and ultimate drop-
ping of many of the
Leaves directly above1
the generator in the
pathway of the rapidly
rising gas. This result
is usually spoken of as
the "chimney " effect.
The generator cover
eliminates this " chim-
ney' ' burning.

A second and highly
important point is the
effect of open gener- fig. 20.-7,1
a tors on the tent.
The outer part, or skirt, as it is sometimes called, of fumigating
tents is constantly being perforated with small holes, even when
used by the most careful of workers. We have noticed this effect
to some extent in our own outfit, which we believe to be as carefully
handled as any fumigation outfit could be. These holes are known
to be acid burns. A few simple tests have demonstrated conclu-
sively that many of these acid holes are due to acid carried along
with the escaping gas and reaching that part of the tent nearest
the generator. By placing large pieces of canvas in the path of gas
escaping from open generators in which dosages similar to those
often used in field work are employed, it was found that drops of
acid reached the canvas a- high as 5 feet from the ground. The
writer has frequently seen generating vessels placed not more than
2 feet inside the tent. At such a distance one can readily see that

d top for protecting cyanid in the field.
(Original.)

58

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

drops of acid might reach the tent. The generator cover described
above so deflects the gas, and Incidentally such acid as is carried with
it. that the drops are thrown to the ground, thus saving the tents.

The decreased cost in mending of tents will doubtless pay for the
cost of such a cover device several times over in a fumigating season.
A third advantage, which we have not as vet demonstrated but
which we have reason to believe will develop, i-> a better distribu-
tion of gas through the tent. Heretofore the most difficult part of
the tree in which to destroy insects is the lower part. Thi> is also
the part of the tree in which the purple scale is largely to be found.
With the open generator the gas rises straight up in a narrow column
for several feet (fig. 22. at left), being broken up and distributed

through the top of the tree
first. As the gas is lighter
than air, it is not to be ex-
pected that it will quickly
become uniformly distrib-
uted throughout the bot-
tom of the tent, even if at
any time it becomes as con-
centrated here a- at the top.
The greater burning effect
and better killing effect in
the top of the tree would
tend to substantiate this as-
sumption. Field observa-
tions in fumigating huge
trees show that the gas is
of no great strength at the
lower part of the tent for
several minutes after the
charge is set off. With this
new cover the gas is broken

Fig. 21.— A cover device attached to a fumigation generator-
corrugations in cover allow gas to escape. (Original.)

up and distributed through the bottom of the tent first dig. 22, at
right). By the time it reaches the top it is pretty generally distiib-
uted throughout the tent. As the bottom of the tree i> the first to
receive the full benefit of the gas. a more complete killing of scale at
the bottom of the tent may be expected than with an open generator.

AN IMPROVED SYSTEM OF FUMIGATION.

During the month of July. 1908, a system of fumigation which has
decided advantages over the old method was introduced into Cali-
fornia field practice. In this system the tents are marked after the
Morrill method, described on pages 27-30 (figs. 11 and 12). Only
the three parallel lines are used, the cross line being unnecessary

AN I.M PROVED >\ - II \l.

i9

and to some extent a disadvantage in practical work. The marking
on these lines gives us an easy means of determining the distance
over the top of the tree. Our experience has show n that the distance
around the tented tree can he measured very accurately by pacing.
The one whose work, in n regular out lit. i-. to <>l>tain the dimensions
of the trees, should make several pracl ice trials in advance of Fumiga-
tion, so as to determine the exact length of his pace, and i<> regulate
it. if necessary.

In pacing the dis-
tance around a
tree it is well to
keep far enough
from the edge of
t lie tent — say
from h inches to 1
foot distant — to
prevent the body
from coming into
contact with it.
The length of the
pace should be
regulated to 2\ or
8 feet when ap-
proximating the
actual distance
around the tented
tree, preferably 3
feel . if the pacer
can step that dis-
tance without

much exertion.

t ijt .i t y _ Fig. 22.— Difference in the direction taken by gas escaping from an open

in Hani} tne ais- generator and from one covered with the corrugated lid. (Original.)

tance paced will

be slightly greater than the actual circumference of the bent. From
these two measurements (the distance around and the distance over),
it is possible to approximate the cubic contents of the t rce.

SUPPLY CART.

With this system some change is necessary in the character of the
vehicle for carrying materials, inasmuch as the measuring of chem-
icals is conducted at the tree. A two-wheeled handcart of the same
genera] description as that in use l>y the San Bernardino County

outfits lias been adopted. The handle of the carl and the arrange-
ment of the lights have been improved upon: while the use of faucets
in drawing off the acid and water is also an improvement. One of

60

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

the carts equipped for use is shown in figure 23. As purchased the
cart bed consists of a plain box fitted with a two-shaft handle. This
handle is removed, and is replaced by a tongue having an enlarged

link-shaped iron about a foot long, firmly attached at the end. This
link-shaped handle is very convenient in field work. The scales for
weighing the chemicals are placed on a platform above the center
of the box. The cyanid is contained in a tin-lined box in the rear
half of the cart, while the acid and water are placed in the front end.
A 10-gallon keg firmly attached in a horizontal position to the bed
of the cart i- a very convenient receptacle for the water. A galva-

Fig. 23.— Cart used with the improved system of fumigation. (Original.)

nized-iron basin like that shown above the keg in figure 23, having an
opening at the bottom fitting into the bung of the keg. make- a very
satisfactory funnel for filling the keg. The acid may be held in an
earthenware jar or a lead-lined tank, with cover firmly attached to
prevent slopping.

By way of a cover for the earthenware jar we have used a lead-
lined lid. which fits tightly within the top (fig. 24). At the center of
this lid is an opening about 6 inches in diameter, around the circum-
ference of which is attached a leaden tube which extend- downward
several inches and prevents the slopping of acid through the hole.
A lead-lined cover lit- into the top of this tube. This opening in the
er is for use in filling the jar.

AN [MPROVED SYSTEM.

61

Very few metals will withstand sulphuric acid without corroding
For this reason all the common types of faucets are practically
worthless for drawing acid. There is
no faucet on the market thai is alto-
gether satisfactory for this purpose,
although at the present time a manu-
facturing firm on the Pacific coast is
experimenting in the hope of perfect-
ing the necessary article. We have
mei this difficulty in an entirely prac-
tical manner by attaching a three-
quarter inch iron pipe to the lower
side of the jar and regulating the flow
of acid by means of a large pinchcock
placed on a short piece of rubber tub-
ing at the end of the pipe (fig. 2 1. 1,-1.
and 5). The How of acid is rapid and
easy to control. Pure rubber is most
satisfactory and a fresh piece should
be substituted about every other night .

The water is drawn from a faucet.
In order that this may he drawn on
the same side of the cart as the acid,
a pipe of the character shown in figure
23 is required. The faucet should
have an opening of about three-fourths
inch to allow a heavy flow and should be of such a type that a half
turn will give it a full opening.

As fumigation is usually conducted at night a torch is placed on
the front of the cart to furnish a light by which to measure the acid
and water; one on the elevated platform is convenient for the man
measuring the cyanid.

This style of cart is entirely practicable for almost all fumigation
work. The chemicals can be measured quickly and accurately with-
out any slopping of acid or water. The work is also easier on the
men in charge than under the old system. On ground which is so
rough that a wheeled cart can not be drawn, a portable table may
be used. Such a table as is shown in figure 25 can be easily utilized
for such a purpose.

PROCEDURE.

Five men are required to operate this system to advantage. Two
men pull the tents and kick in the edges around the bottom of the
tree. One man takes the measurements of the tree and determines the
dosage from a dosage schedule which he carries with him. After de-
termining the dosage he should empty the generator to be used for
that tree and have it in readiness bv the time the cart arrives. The

Fig. 24. Earthenware ai I ! jar with attach'
ments for field use: t, Jar complete; /.in-
side view of lead-lined caver showing tube
at (enter; 8, copper top for opening In
cover; .;, pinchcock; .",, method of attach-
ing iron pipe to jar, and rubber tube on end
of pipe with pinchcock attached.

62

FUMIGATION INVESTIGATIONS IN CALIFORNIA.

generator should always bt emptied with one and the same hand and
with this hand "he should never touch the tent; otherwise acid burns

Fig. 25.— A table which can be used instead of a cart in fumigation over very rough ground. (Original.)

may result. The estimator should also be foreman of the outfit, as
tins is the most responsible position of all. Two men work at the
cart. One measures the water and acid, the other weighs the cyanid.
The latter holds up the edge of the tent while the acid man places
the charge beneath the tree.

(Original.)

Iii actual field practice the carl is first brought up to one end of
the row which is to be fumigated (iig. 26). The estimator obtains

ADV \.\ i \(.i ;s rxh! i: i in i m im;u\ in SYS1 EM

63

his measurements and calls out the dosage. The two men then
measure out the required amount of chemicals and dose the tree (fig.

). While they are thus engaged the estimator li

as moved on t<>

the next tree, determined the proper dosage, and holds the generator
in readiness when the cart is brought up. He l lien calls out the
dosage with which the tree is to be treated. This procedure contin-
ues in like manner until the entire row Is fumigated.

Outfits employing tins system have, on an average, been fumigat-
ing a complete set of 32 tents in from forty to forty-five minutes.

This would appear to demonstrate that the system is entirely practi-
cable from the standpoint of time economy, as tents an* usually
required to he left on the tree one hour.

ADVANTAGES UNDER THIS SYSTEM.

First. — The element of guesswork in estimating dosage and the
consequent waste of cyanid are eliminated, since the dosage is deter-
mined according to a uniform method in all cases. If a dosage of suf-
ficient strength to destroy 90 per cent of the purple scale is used,
practically 90 per cent of the purple scale is killed on c\^\y tree
throughout that orchard not 90 per cent on some trees and 50 per
cent, more or Less, on others, which has occurred at times under the
old method. Or if a dosage strength just sufficient to eradicate the
pest i> employed, a. like result will occur throughout the orchard and
there will he no great waste of cyanid by reason of man} trees
receiving a larger dosage than was necessary.

J

64 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Second. — An economy of cyanid results from the accurate measure-
ment of the water. Three parts of water are always used, resulting
in the maximum amountof available gas for practical work, as already
explained (pp. 38-39). Under the old system the water is usually
measured with an ungraduated dipper at the tree, and as the cyanid
and acid have been previously measured out into small cans, which
are in turn placed on a tray to he carried from tree to tree, the sched-
ule is not carried along for consultation in estimating the water, but
the required amount of water is guessed at from the amount of chem-
icals in the cans intended for that particular tree. Owing to the vari-
ation in the proportion of water which results in this way, the maxi-
mum amount of available gas is seldom produced by the reaction.

Third. — By the old method the cans on a tray sometimes become
confused, in consequence of which some trees get the dosage measured
out for others. This error is eliminated under the improved system.
as the dosage for each tree is measured out just before that particu-
lar tree is fumigate* 1.

Fourth. — The tent pullers seldom get more than one or two trees
ahead of the cart. As a result, all trees receive the same length of
exposure. Under the old system, when the tent pullers got far ahead
of the cart at the end of a row, these trees received a much shorter
exposure than the first trees.

DOSAGE SCHEDULE.

Having obtained the dimensions of the tented tree, the next step
is to determine the dosage. It has been previously stated that the
cubic contents can be calculated from these two dimensions. This
might be done in the field and the trees then dosed in proportion to
the contents. The time required for the calculation of the dosage,
however, even after determining the cubic contents of the tree, would
not only prevent rapid field work and allow an opportunity for error,
but would cause a lack of uniformity in dosage, from the consideration
of the cubic contents alone, as will be explained later. This diffi-
culty has been obviated by preparing a dosage schedule from which
the required dosage may be learned without any figuring as soon as
the measurements of the tree are known.

The orchardists in the citrus section about Whit tier, Cal., desired
to commence fumigating for the purple scale during the latter part
of July. The question immediately arose as to what dosage could be
used at that time of the year without injuring the young fruit. As
stated under experiment Xo. 4 (p. 52), while the fruit is small a dosage
of 1 ounce to 100 cubic feet could be used on trees from about 10 to 1 5
feel in height without injury to the fruit, whereas smaller trees
would stand a heavier dosage. A- tin- was the limit of dosage which

DOSAGE SCHEDULE,

65

could be used at that time of the year without injury to the young
fruit, the writer prepared a schedule based upon this data.

According to this schedule, trees n feel in circumference 1>\ i's
feet over the top, from ground to ground, receive I ounce of cyanid
for each 100 cubic feel of inclosed space. This proport ion is increased
on smaller trees, \\ bile on trees w hich are larger it is decreased to offset
the proportionately smaller leakage (pp. 43 14,47 18). [n preparing
the schedule the writer began with a tree ll Feel in circumference by
28 feel over the top. The cubic contents of this tree were determined

16 ie

20 22 24 26 28

30 32 34 36 38

4042 4446 48

50 52 54 56 58

60 62 64 66 68

70 72* 74 76 78

IO
12

14

*

i

1

!'.•

in

IO

i ,

IK

\%

2 2

."v

12

i, 2

2

2'h

254 2*

3

3

3/:

AA'/z

5

14-

16

2 2*

2*1 3 1 3 1 3 3*

3'/? 4 4', 5 v.-

<, ' 7

16

If*

3, 3 3 . 3 . -•

4VUj 5'/; ■ •• >■

/ /'.- 8

18

2022 2426

28

3032 3436 38

4-0 42 44

46 48

50 S2 5456 58

60 62 64 66 68

70 72 74 76 78

20

3 4 |4)4

5 IS/t 5>4|6 gft 7 1 7 8 '8#

9

9)4

20

??

AHA 5 |5fci 6 |6*l 7 |7*|7ft| 8 [8Vfe|8>fe| 9^

10

m

Z2

?A

5M (-,'.-

9)4 0 . io I0>2

ll

ii i

24-

26

7

8

a::- 9 9)4 10

'O'.-IO .11 II II'';

i: ■ 13 14

I4X

15

26

28

8'/;

9 'A 10 10, II

HV; II'; 12". 12'.- 13

13": 14 M

1514

II

28

30

32 3436 38

4042 444648

50 52 54 56 58

60 62

64

66

68

70

72

7A

76

78

30

10

II |IIV4|I2

12'- 1 3 I.V/> 14 14'/:

I4JH 15 U51HI6)4]I7)S

18 19

2OC014 2I14

30

32

I2|l2&

13 14 15 \\5'fi\ 16

1614 17 l/Yz 18 18/:

19 19%

120)4

2114

227:

32

34
36

38

13

14 15

16

17

17':

18

IB'l-

19

I9M

20

20)4

21

717;

22)4

23

34

14

15

It

i?

17*

18

m

20

20X

215!

?\'k

22

?2*

23

23'/;

2514

36

16

I6K

l/>.

18'.

19

:n-;

21

21'/;

22

23

m

74

2474

25

25'/?

38

40 42

46 48

50

52 54 56 58

60 62 64 66 68

70 72 74 76 78

40

I7J 18

19 20

21

22422V4' 23' 24

?4£ 25 25)4 : 26 267?

27 27/4 28

40

41

^

19

20 21

22

22'k 21%, 24 24'/-

25 25'/; 26 26'// 27

27'/? 28 12814 29

41

42

20

20)6

22

22)4

23

2414 25

2S'6_

26

7674

27

2714

28

2?*

?8* 29 29 30 30/4
29 '29)4 30 '30 /z* 31

?9/z*30*3l *3l/;"32"

42
43

44

43

21'/;

Z2'h

23

2314

25

;vv

26

M

27

2/V.

28

44

.'3

23)4

24

25

26

26'A

27

2/:-

2ft

%'k.ZS

1

50 52 54 56 58

60 62 64-66 68

70 72 74 76 78

45

24JJS ! 26 -26V4- 11

2Tk 28 29.29/; 30

304 31 31'/? 32; 33

4-5

46

WkivBtiiklVhlTk

23 282b 29 : .30 J '.

31 31'/? 32* 33 34

46

4-7

25 26 j 27 27^ 28

'8:'- c) M M M

32 32'/? 33 34 3b

47

48

251426ft 2714, 28 2814

2? 2914 30'// 31 32

327? 33 34 35 36

4-8

4-9

Z6JH 27 28 28ffl 29

2974 30. 31 31)4 3214

33 34; 35 i 36 '37

49

50 52 54 56 58

60 62 64 66 68

70 72 74 76 78

50

30 31

l\'li?l% 33

33/4 35 36 37 j 38

50

51

31

31 'A

32 33

33'/;

34
35

34 36 37 38 39
35*37*38* 39; 41
36T38 ' 39 40 41

51
52
53

52

m

32

33 m

53

32

32%

m 34

54

32'/?. 33 34, 35; 36

37 39 40 41 42

54

60 62 64 66 68

70 72 74 76 7$

55

33 34 35

^6J37

37)4.40 41 42 43

55

56

34

35

36

37

37/;

38'A

38 40)4 4? 43 44
39*41 42'4»3H45

/10 42 43 44 46

56
57

57

-:.r/.

V'/

KX

3714

58

35

3b

37

38

39

58

59

36

ib'/?

38

39

40

40>i42'/?43/4|45|47

59

Fig. 28.— Dosage schedule No. 1. (Original. )

and a dosage calculated which would £ive it 1 ounce to each 100
cubic feet. Trees of other dimensions, both larger and smaller, were
then considered and their contents determined. In working out the
dosage for these trees not only were the cubic contents taken into
consideration, bul also the rate of leakage as compared with that of
the tree 41 by 28 feet. Trees which were smaller than this first tree
would have a greater proportionate leakage rate, while the larger
ones would have less, lu securing the dosage for various tree-.
those smaller than -1 1 1>\ 28 feel were given sufficient cyanid in
excess of 1 ounce per 100 cubic feet to offset the increased leakage,
77488— Bul. 79—09 5

66 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

while the dosages for larger trees were decreased proportionately
below 1 ounce. This allowance for leakage so modified the schedule
thai tree- 2 4 by 16 feet received as high as 1 \ ounce- per 100 cubic
feet, while trees 60 by 44 feet received only about three-fourths of an
ounce to the same space. The results of the use of such a schedule
in practical fumigation should be thai the smaller and the larger
trees receive a dosage of uniform killing power against the scale.

After computing the dosages for tree- of such sizes as would include
all that could be covered with a tent 60 feet in diameter, a chart
was prepared (fig. 2s) and the dosages incorporated therein.

How to use tin chart. — The top line of number-, commencing at 16
and continuing through Is. 20, 22, etc.. up t<> 7s. represents the dis-
tance, in feet, around the bottom of the tent. The outer vertical
columns of larger numbers, on either side, commencing at 10 and
increasing regularly to 59, represent the distance, in feet, over the
top of the tent. The dosage of a tree of known dimensions i- found
in that square where the vertical column headed by the distance
around the tree intersects the horizontal line of figure- corresponding
to the distance over. For instance, we have a tree 40 feet around
by 28 feet over. Looking in the top line of numbers we find 40 next
after the third heavy vertical line. The dosages computed for trees
40 feet around are to be found in the vertical column headed by this
number, which commences with 6 and ends with 16. Then we glance
down the vertical column of large figures at either margin until we
come to 2S. All dosages computed for trees 2S feet over are found
in this horizontal line of figures, which commences with 8$ and ends
at 16. The dosage for a tree 40 by 28 feet is found at the intersec-
tion of this line with the vertical column headed with 40. that number
being 11 J, the required dosage of cyanid hi ounces. Before the num-
bers 20, 30, 40, 45, 50, and oo, in the lines at the right and left mar-
gins are to be found blank spaces, and in the horizontal line- corre-
sponding to these the numbers at the top of the chart are repeated
in that part of the chart containing dosage figures. These numbers,
repeated in this manner, make it easier for the eye to locate wit li
certainty the dosage figures sought. In the chart used by the
writer, the figures representing distances around and over are printed
hi red. The lines bounding these columns of figures are also red.
All the rest of the lines and figures are black.

This schedule has been called "dosage schedule No. 1." by reason
of the fact that 1 ounce to 100 cubic feet of inclosed space was taken
•a- a ba>i> in preparing it. though, as a matter of fact, only a -mall
number of the trees in an orchard receive exactly 1 ounce to 100
cubic feet.

It is not maintained that this table i- accurate to the minutest
part of an ounce for every dosage, but the writer believes that such

THE IMPROVED SYSTEM IN I 67

variations aa may be found to exist are so small thai in practical
work in the field the results in lulling scale-insects, from the use 01
part of the table, will be found as satisfactory as from the use "1"
any other part. Moderately heavy dosages almost invariably burn
the tender shoots of a tree to a greater or Less extent. Under this
schedule practice has demonstrated that the tender growth is uni-
formly burned Hack in all cases, whether Large trees or small ones
are fumigated.

As previously stated in this discussion, dosage schedule No. l
was prepared for use against the purple scale at Whittier, Cal., during
the Latter part of July when the fruits in some orchards were about
the size of a walnut. The dosage employed was as great as the fruit
would permit at that season without injury. This does not indicate

that a Larger dosage can not he safely used at other -c;i^<>!i> of the
year, if desired. The writer has at times employed a dosage of

double the strength without visible injury to the trees. This was
accomplished, however, under more favorable conditions, during the

fall and winter months, when the fruit was well grown. It is UOt
deemed advisable to use a dosage against the purple scale of less
strength than that of schedule No. 1. If complete eradication is
desired, a much heavier dosage must necessarily he employed.

The dosage in schedule No. 1 is equivalent to what i> known among
many fumigators as the "double dosage." It may be a little stronger
than the double dosage of some, rather than weaker. "Double
dosage" is usually intended to signify a dosage twice the strength
required to destroy the black scale in its earlier stages.

Since this schedule is one of uniformity, it readily permits of
manipulation. If a heavier dosage should be desired, such may he
obtained by increasing each individual number or dosage in thesame
ratio: if a lighter dosage, by proportionately decreasing each. The
schedule resulting from such increase or decrease will also he one ■)(
the same general uniformity as the first. The writer has prepared
schedules which are \, |, 1J, 1J, and \\ times the dosage indicated
in schedule No. 1 .

THE IMPROVED SYSTEM IN USE.

Two outfits of the Whittier (Cal.) Citrus Association commenced
the use of this improved system of fumigation, with dosage schedule
No. 1 . during the latter part of July, 1908. The apparent uniformity

of work, indicated by the evenness with which the tender growth
was burned hack on all trees, immediately attracted the at lent ion
of citrus growers who >;i\\ the fumigated orchard-. Their universal
approval of the method i> shown \)\~ the fact that not a single unfa-
vorable comment was brought to the writer's attention throughout
the entire fumigation work. The reception of an improved method

68 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

with such unanimous favor by a community of California citrus
growers demonstrates its value and economic importance.

Since the first outfits were placed in operation at Whittier, others
have adopted the improved system of dosage, and at the time of

writing this fully a dozen similar outfits in various parts of Los An-
geles and Orange counties, Cal., are using the new method in prefer-
ence to the old. That such a large number of practical citrus growers
in widely separated localities, who have been employing a system of
fumigation for many years, should accept an innovation within two
months, strongly indicates its superiority.

FUMIGATION SIMPLIFIED.

In the past many persons have been prone to look upon fumigation
as a process that is complex and more or less mysterious. In some
cases fumigators of years' experience have encouraged this widely
prevailing opinion, so that they might themselves be looked upon as
experts in a practice difficult to understand and only capable of
being successfully performed by men of long experience and special
qualifications. This is, of course, erroneous. The improved system
outlined in these pages shows how simple the practice of fumigation
may be made. Careful men who have never before heard of fumi-
gation can begin the practice of this system and are competent, after
instruction for a short time, to secure as good results as might be
expected from the most expert fumigator in California. This system
reduces fumigation to a matter of simple mechanical operation,
entirety intelligible to the average man, and one wherein the operator,
to obtain the best results, is required merely to proceed according to
the formulas and directions given.

This system makes it possible for the crchardists to possess outfits
of their own — either individually or through joint ownership on the
part of neighboring fruit growers — and to do the work with their
own employees.

NDEX

Pagi

Aphis goasypii, enemy of citrua fruits. 1 1

orange. (See Aphis gossy pit.)

Aspidiotus hedi r;i , enemy of cil ma fruits 1 1

Chemicals in fumigation 30 W

mixing 39 10

most economical proportion

Chrysomphalus aurantii. (Sei Scale, red.)
ci triit us. (Set Scale, yd leu

Citrus, apparent toxic effect produced l>y scale insects L6

extent of orchards in California L0

injury resulting to scale-infested tree- I I i<;

insect enemies in California, and their distribution 10 17

scales. (See Scales, citrus.

( Soccidse injurious to citrus fruits in California 10

( 'onus hesperidum, enemy of citrus fruits II

Cover for cyanid case

Distillate sprays againsl citrua Bcales 17

1 >osage schedule in improved ay3tem of fumigation 64

schedules of more important writers on fumigation 19-2 1

Fumigation against citrus scales 17-68

advantages under impnrs ed system 63-6 !

apparatus in improved Bystem 53-61

present system I v L9

chemicals 30 !<>

mixing 39 L0

most economical proportion 39

cover for generators 5<

cyanid, amount- recommended in schedules

of more important writers 21-22

cover 55

protection 55

dosage schedule in improved Bystem 64-67

schedules of the more important writers L9-24

during blossoming period 49 5 1

gae, amount, available aa affected by different

proper! ions of water 37-38

hydrocyanic-acid, in drum- 55

Leakage during operation- 17 is

in small trees 13 14

temperature a- affected by different pro-
portions of water 35 36

u lien; Large and small dosages

are used . . 36-37

69

70 FUMIGATION INVESTIGATIONS IN CALIFORNIA.

Pa po.

Fumigation against citrus Bcales, improved system

advantages 63-64

apparatus \ -61

dosage schedule

in use I

procedure 61-63

supply carl 59-61

method of computing dosage and volume for

tented trees 25-26

methods for obtaining measurements and dos-
age of trees with and without apparatus... . 26-30

present system, apparatus 18-19

dosage schedules 19-24

procedure 18-19

procedure in improved system •

present Bystem 18-19

proportion of materials used by rumigators 32-39

simplification by improved system 68

sulphuric acid, amount necessary .

effecl ol too great an excess. . . 34

jar and cover 60-61

removal from drums and car-
boys

supply cart ">!!-i;0

table to replace supply cart on very rough

ground 61, 62

tents, marking for estimating dosage... 27-30,55-^56

sheet 17-18

time of year 4S-52

tray for carrying chemicals in present system. . 18-19

water as a factor 34-35

correct proportion

effect of different proportions on tem-
perature and amount of available gas.

while fruit is small 5 L-52

purple scale 40-46

length of exposure 44-4-3

preliminary experiments 40-43

generators, cover

Fungus, black, or sooty-mold, on citrus trees infested with black scale 15

Honeydew, secretion by black scale

Hydrocyanic-acid gas. (See also Fumigation.)

in drums 55

Icerya purchasi, enemy of citrus fruits 11

Insects, beneficial, used against citrus Bcalee in ( California 16

Kerosene-water spray against citrus Bcales

Lemon. (See Citrus.)

Lepidosaphes beckii. v Scale, purple.)

Mealy bugs, distribution in California J4

enemies of citrus fruits 11

Mite, rust, of orange. (See PhyllocopUs oh ivor us and Mite, silver, of Lemon.
silver, of lemon. (See also Phyllocoptes < is.)

sulphur spray as remedy 16

1MH.X. 71

Merrill system <>!" measuring trees without apparatus

Olive, food plant of black scale i_-

Orange. I Set Citrus, I

Ornamentals, food plam.- of black Bcale ii!

PhyllocopU* oleivorus. (Sa also Mite, Bilver, of lemon.)

enemy of citrus fruits 1 1

Potassium cyanid, amounts recommended in schedules of more Lmportanl \\ tit-
ers on fumigatioD 21 22

for t'u in i u;i i ion, protection 55

puriix required 30

Vaeudoeoecus spp (Set Meal) bugs.)

Scnsst tia h misphserica, enemy of citrus fruits II

olci . (Sa Scale Mack.)

Scale, black, distribution in California L2-13

enemy of citrus fruits 11

method of propagation K>

nature of injury to citrus trees L5-16

cottony cushion. (See Icerya j>i<r<'ltasi.)
hemispherical. (See Saissi tia h mispha rim.)
oleander. (See Aspidiotus hedera .)

purple, difficult to destroy on fruit 46

distribution in California 11-12

enemy of citrus fruits 11

eradication 46

fumigation 40-46

length of exposure 44-45

preliminary experiments 40-43

method of propagation 10

nature of injury to citrus trees 11 I ... Hi

red, distribution in California 13

enemy of citrus fruits 31

method of propagation 16

nature of injury to citrus trees 14- L5, L6

m if t 1 in >\vn . (See ( 'oceus hesperidtnn . )

yellow, distribution in California 13

enemy of citrus fruits 11

met hod of propagation 1 6

nature of injury to citrus trees L6

Scales, citrus, fumigation 17 68

advantages under improved system 63 64

apparatus in improved system

present system L8 I"

chemicals 30 10

mixing 39 40

mosl economical proportion 39

cover for generators 56-58

cyanid. amounts recommended in schedules of more

important writerson fumigation -1 22

cover 55

protection 55

sage schedule in improved system <> I 67

schedules of more important writers 19-24

72 FUMIGATION ENVESTIGATIONS IN CALIFORNIA.

Page.

Scales, citrus, fumigation, during blossoming period 49-51

- amount available as affected by different propor-
tions of water

hydrocyanic acid, in drums 55

leakage during operations 47 4s

in small trees 43-44

temperature as affected by different proportions

of water

when- large and Bmall dosages are

used. 36-37

improved system

advantages 63 64

apparatus

dosage schedule 64 67

in use 67 68

procedure ; 61 63

supply cart 59-62

method of computing dosage aud volume for tented

trees 25-20

methods for obtaining measurements and dosage of

trees with and without apparatus 26-30

procedure in improved system 61 63

present system 18-19

proportion of materials used by fumigators 32—39

simplification by improved system 68

sulphuric acid, amount necessary 32-34

effect of too great an excess 34

jar and cover 60-61

removal from drums and carboys. . . 5!

supply cart 59-6 1

table to replace supply cart on rough ground 0 1 . <;2

tents, marking 27-30. "

sheet 17-18

time of year 4 s- 52

tray for carrying chemicals in present system L8

water as a factor 34-35

correct proportion

effect of different proportions on amount and

temperature of available gas

while fruit is small 51-52

methods of control in California 10- 1 7

manner of feeding 14

propagation 16

Spider, citrus red. distribution in California 14

enemy of citrus fruits 11

sulphur spray as remedy 16

. kerosene-water, againsl citrus scales 17

Sprays, distillate, against citrus scales 17

sulphur, against citrus scales 16

Sulphuric acid for fumigation, aim Mint necessary 32-34

effeel of too greal an excess 34

jar and cover «i(l 61

in ma. 7:;

Sulphuric acid for fumigation, purity required 30 32

removal from carboys and drums

Sulphur spra) s, against red spider and silver mite of Lemon L6

Supply cart for fumigation 59-61

Table to replace supply can iu fumigating on rough ground 61,62

Tents, marking for estimating dosage i'.

Tetranychu8 mytilaspidis. (Set Spider, citrus red.

Thrips, enemy oi citrus fruits 1 1

Toxic effect on citrus trees apparently produced by red, purple, and yellow

Bcales L6

Tray for carrying chemicals in present Bystem of fumigation 18 19

Water as a factor in fumigation :;i 35

in fumigation, correct proportion

effect of differenl proportions on amounl and tempera

uire of available gas 35 38

Wood worth systems of measuring trees with and without apparatus 26 27

0