CALIFORNIA
AGRICULTURAL EXTENSION SERVICE

CIRCULAR 79

NOVEMBER, 1933

Control of Injurious Rodents
in California

TRACY I. STORER

Cooperative Extension work in Agriculture and Home Economics, College of Agriculture,

University of California, and United States Department of Agriculture cooperating.

Distributed in furtherance of the Acts of Congress of May 8, and June 30, 1914.

B. H. Crocheron, Director, California Agricultural Extension Service.

THE COLLEGE OF AGRICULTURE

UNIVERSITY OF CALIFORNIA

BERKELEY, CALIFORNIA

CONTENTS

PAGE

Introduction ... 3

Ground squirrels 8

Burrows 9

Feeding habits .... 10

Estivation and hi})ernation 11

Breeding 11

Methods of control 13

Costs of various control methods .• .19

Tree squirrels 19

Pocket gophers 20

Burrows 20

Breeding 22

Damage due to pocket gophers 23

Methods of control 23

Moles 32

Nature of damage 32

Methods of control 33

Introduced rats and mice 34

Breeding 35

Nature of damage 35

Methods of control 36

Meadow mice 45

Wood rats 46

Kangaroo rats 47

Muskrats 47

Rabbits 48

Black-tailed jack rabbits 48

Snowshoe rabbits 49

Cotton-tail and brush rabbits 49

Methods of control 50

Formulas for preparing poisoned baits for rodents 52

General precautions 52

CONTROL OF INJURIOUS RODENTS
IN CALIFORNIA'

TRACY I. ST0EER2

INTRODUCTION

California has a large and varied stock of native rodents, some 270
species and subspecies, to which have been added a few species intro-
duced from other places. Among all these are a few widespread,
aggressive, and adaptable rodents of serious economic concern to man
because of their ability to thrive in cultivated areas, because of the
damage which they do to crops, gardens, levees, or buildings, and because
of the diseases which some of them carry and may transmit to man.
Several others are of occasional or local concern, while many do no harm
whatever, and some are beneficial. The principal injurious rodents are
the ground squirrels, the pocket gophers, the introduced Norway, black,
and roof rats, the house mouse, and the jack rabbits and cottontails.
The financial loss inflicted by rodents in California is difficult to
determine. Gross estimates of more than 20 millon dollars annually have
been cited. Some specific estimates of earlier years are : ground squirrels,
5 to 12 millon dollars; pocket gophers, 8 million dollars; introduced
rats, 2 million dollars. Such estimates are necessarily crude and do not
consider the probable effect of the additional crop (if saved) upon price
depression. Nevertheless individual producers suffer material losses
from rodents, and protection by moderate control is thoroughly justi-
fied. Losses to cereals by ground squirrels, where unchecked, may
amount to 10 or 15 per cent of the crop. Pocket gophers damage field
crops in ways not easily measured in crop units ; they may injure or
kill individual orchard trees. Ground squirrels, pocket gophers, or musk-
rats may do important damage to ditch banks or levees. Ground squirrels
if unchecked may materially reduce the available forage in a pasture.
Jack rabbits and — less often — cottontails, may gnaw bark on individual
trees or vines necessitating replacements which will need several years
to reach productive age. Jack rabbits may seriously reduce the pro-

1 Particular indebtedness is acknowledged to previous publications by Mr.
Joseph Dixon in this field (Bulletin 340, Circular 296, Extension Circular 29),
material from which is used herein.

2 Professor of Zoology and Zoologist in the Experiment Station.

[3]

4 California Agricultural Extension Service [Cir. 79

duction in a truck patch by eating the young plants. On the other hand,
certain rodents, in places, benefit the soil by "cultivation," and in other
ways not well understood.

Efforts at suppressing injurious rodents date from the earliest days
of farming in California; attempts at ground squirrel destruction were
made by the mission fathers. Governmental aid early took the form of
bounties paid by counties for squirrels and gophers. The introduction
and spread of bubonic plague among ground squirrels in the coastal
region of California during early years of the twentieth century stimu-
lated federal and state organizations to reduce the numbers of these
animals. In 1917, by legislative enactment, the control of rodents was
made a duty of the county horticultural commissioners (now agricul-
tural commissioners), and this led to establishment of cooperative
relations between federal, state, and county agencies in the work.

Coordinated and persistent official efforts, together with stimulated
activity among landowners, have reduced the ground-squirrel popula-
tion in many places well below" earlier levels.

Cooperative drives, especially during the '90's in the San Joaquin
Valley and the coastal area of southern California, and less often in
recent years, measurably reduced the jack-rabbit population.

Pocket-gopher control has always been mainly a problem for the
individual farmer. Great variation exists in the numbers of these
animals today ; some large farms, especially orchards, are relatively
free of them, while other cultivated areas have as many gophers as under
primitive wild conditions, or more.

The introduced rats and mice are a perennial problem in stored
products and about buildings where any kind of food is available ;
reduction of losses involves exclusion rather than destruction. While
spasmodic "cooperative" campaigns of trapping or poisoning rats and
mice kill limited numbers, there is no lasting effect unless these are
followed up at frequent intervals.

The control of injurious rodents is now a function of several govern-
mental agencies. In California, the Bureau of Biological Survey of the
United States Department of Agriculture supervises rodent control on
federally owned lands and, together with the State Department of Agri-
culture, advises and helps to direct operations in the counties. The
agricultural commissioners in the several counties prepare and dis-
tribute poisoned grain for use against ground squirrels; each of the
larger counties has one or more special rodent inspectors to study
populations and to direct or actually apply control measures. The
United States Public Health Service studies rodent-borne diseases and

1933] Control of Injurious Kodents 5

conducts campaigns against rodents in and about the seaports, while the
California State Board of Public Health through its Bureau of Sanita-
tion does similar work elsewhere in the state.

Control on federal lands is financed by Congressional appropriations ;
supervisional work of state and county officials by appropriations from
the state legislature and the county boards of supervisors. Some coun-
ties, in the past, have paid for materials and even labor in poisoning
rodents by grain baits and gas, but in most counties the individual
landowner or tenant pays for materials and either provides or pays for
the labor needed to apply such materials to the land. During the decade
1921-1931 the annual expenditures for rodent control were approxi-
mately as follows : federal, $10,000 to $25,000 (average $15,000) ; state,
$10,500 to $24,000 (average about $12,000) ; counties (including both
supervision and materials, the latter often refunded by landholders),
about $290,000 to $554,000 ; and the landowners have contributed labor
estimated to be worth $100,000 to $130,000. The total therefore has been
from half to three-quarters of a million dollars annually. In 1929-1931
the state expended a special fund of $100,000 on rodent control work in
the plague-infected central coast counties. Most of the funds indicated
have been used against ground squirrels, especially the species known
as the California ground squirrel. If private expenditures are con-
sidered, it is probable that all kinds of rodent control operations in
California have involved as much as one million dollars annually in
certain recent years. The area treated, in the decade indicated, has
varied from 1.9 to 9.2 million acres.

From time to time governmental and private agencies have offered
bounties for the destruction of rodents, requiring a scalp or tail as
evidence. This system has everywhere proved costly, ineffectual, and
open to corrupt practice — in short, unjustifiable. If tax-raised monies
are to be used for rodent control, greater benefit will accrue from using
trained personnel to direct or execute the operations.

Numerous commercial preparations are sold for rodent control.
Besides the elementary poisons such as strychnine, carbon bisulfide,
barium carbonate, red squill, and cyanide, various compounded poisons
are sold under special proprietary or trade-marked names. Some are
satisfactory, a few are dangerous, and others relatively useless. Several
commercial houses make up the "government" ground-squirrel poison
formula in proper manner ; for small applications, such a preparation is
economical. But certain proprietary poisons, especially for rat control,
are too expensive. If the individual farmer or homeowner does not care
to make up the preparations mentioned in this circular he will do well to

6 California Agricultural Extension Service [Cir. 79

consult liis local county agricultural commissioner as to the best
materials.

It is recommended that methods other than poison be used wherever
practicable, thereby reducing the danger to human beings, to domestic
stock, and to harmless or beneficial wild animals. Such methods are,
therefore, especially emphasized in this circular. Poisons other than
strychnine alkaloid against ground squirrels, strychnine against pocket
gophers, barium carbonate and red squill against rats, and strychnine
against jack rabbits, are not recommended. Various other poisons, such
as phosphorus and thallium sulfate, though exploited and used in rodent
control operations, either give too uncertain results or endanger other
animals or human beings too greatly to be used by individuals in routine
rodent control.

Especially to be censured are the so-called "virus" or bacterial cul-
tures intended to be placed upon food baits and disseminated among a
rodent population. Not only do they often fail to reduce the rodents, but
they sometimes spread the infection to man. The bacteria used belong
to the paratyphoid group (Salmonella). If disseminated among a popu-
lation of rats some die, but a certain percentage always survive and
become "carriers" of the infection, and sometimes by their droppings
contaminate supplies of food intended for human beings. The use of
such "virus" in efforts to control rats about institutions has resulted in
several outbreaks of food poisoning among the human inmates.

There are certain diseases among rodents living in the wild which at
times decimate the populations of these animals ; but too little is known
about such diseases to employ them profitably, or else they are trans-
missible to man, which fact alone makes their employment for rodent
control undesirable.

There has been much criticism recently of certain control work under
official agencies. Scarcely any control operation (save selective shoot-
ing) is without some possible danger to some other forms of wild life, but
proper application of the more conservative methods reduces this
danger to a minimum.

There is no one easy means for controlling all kinds of rodents. Per-
sistent application of tested methods will keep most species in check;
knowledge of the habits and especially of the food preferences, seasonal
or otherwise, of the animals is essential to successful control. Brief indi-
cations of such habits are included below, but the individual who
desires to control rodents must study the animals continually and,
especially, must check on the results and worth of each effort at control.

Wherever practicable, exclusion is the best method. Though more

1933 J Control of Injurious Rodents 7

expensive at the outset it produces lasting results. Unfortunately, it
cannot be applied to all species under all conditions. Traps of specific
kinds will control pocket gophers, house mice, woodrats, and sometimes
other species; poisonous gases can be used against species inhabiting
burrows in the ground where a deadly concentration of gas will reach
the animal before being diffused. Shooting is applicable for small
numbers of ground squirrels and jack rabbits. Poison baits may be used
with effectiveness against most rodents but not with universal success
and there is always danger of poisoning other animals, including
domestic livestock, and even human beings.

All the native species of rodents had originally natural enemies which
constituted one of the several checks on their increase. Many of these
natural enemies have been reduced and some even eliminated, although,
as detailed beyond, they constitute one of the assets of agriculture. As
some of these also prey on domestic animals their economic status as
checks becomes a matter of relative values. But certain natural preda-
tors such as the red-tailed hawk, the barn owl, and the gopher snake
are so preponderantly beneficial that only a very short-sighted person
would destroy them. Yet many farmers habitually kill every hawk and
snake they see without regard to its economic status.

Bubonic plague has been found to be present in the California ground
squirrel of the coastal region from Alameda, Contra Costa, and San
Francisco counties south to Los Angeles County. This disease originally
came in on ship rats and spread from them, probably through rats in
the cities, to the squirrels; it may be retransf erred from ground squirrels
to rats. Although usually transmitted by fleas, in coming from ground
squirrels to man it may take on the much more deadly pneumonic form
in which it is transmitted directly from one human being to another by
droplet (coughing) infection. No one, therefore, save a trained public
health official, should handle ground squirrels in the area indicated, and
under no circumstances should squirrels from this region be skinned or
used for food.

A native disease, tularemia, first discovered in the California ground
squirrel, is also present in rabbits and, less often, in other rodents; it
varies in virulence according to time and place. Man may contract the
disease by skinning infected animals; one case is recorded where the
disease was contracted by eating rabbit flesh improperly cooked. Rocky
Mountain spotted fever, a tick-borne disease, is present in the north-
eastern plateau region of California embracing Lassen, Modoc, and
eastern Siskiyou counties. At times it is transmitted to man by the bite
of infected ticks inhabiting ground squirrels and rabbits. The mortality

8 California AGRicuj/ruRAL Extension Service [Cir. 79

in man is considerable. Another disease, relapsing fever, lias recently
come to attention in the Tahoe region and about Big Bear Lake in the
San Bernardino Mountains. Carried by small chipmunks, it is trans-
missible to man, several cases having occurred in the past decade. The
handling of rodents or their capture as pets in any of the areas where
these diseases occur is therefore not desirahle.

GROUND SQUIRRELSs

Ground squirrels are among the most widely known and most destruc-
tive mammals in California. The large, long-tailed, grayish-brown
California or "digger" squirrels {Citellus heecheyi and subspecies'*) are
the most important in relation to man : first, as destroyers of crops ;
second, as carriers of disease ; and third, because of the damage they do
to irrigation structures. Three varieties occur in California : the dark-
colored Douglas ground squirrel, with much black between the shoul-
ders, ranges from San Francisco Bay northward throughout the region
west and north of the Sacramento and Feather rivers; the brownish
Beechey ground squirrel occupies central California and the coast
district from the Golden Gate, Carquinez Strait, and the Feather River,
to the Mexican line ; and the gray-toned Fisher ground squirrel lives in
the southern San Joaquin and Owens valleys and south along the western
border of the Mohave Desert, The differences between these three are
not of significance in relation to the work of eradication (save that the
Douglas ground squirrel seems to be somewhat easier to control) so they
are here treated as one. Knowledge of their common habits of feeding,
breeding, and hibernation is important in control work.

In the northeastern plateau region, embracing most of Lassen, Modoc,
and northeastern Siskiyou counties, there is the smaller Oregon ground
squirrel {Citellus or eg onus), of stocky build, with a short tail and plain
brownish-gray coloration ; it is essentially an inliabitant of grasslands.
Other species of ground squirrels inhabit the higher mountains of Cali-
fornia and portions of the Mohave and Colorado deserts, but are of little
or no economic importance.

Ground squirrels are essentially terrestrial animals finding most of
their food on the surface of the ground and digging their burrows in the
soil of open flat country, hillsides, or artificial embankments. The
digger squirrels, however, climb readily and in places ascend nut and

3 For a detailed account see : Grinnell, J., and J. Dixon. Natural history of the
ground squirrels of California. California State Commission of Horticulture
Monthly Bulletin 7(11-12) :597-708. 1918 [1919].

4 These have all recently been referred to as subspecies of OtospermopJiilus
grammurus.

1933]

Control op Injurious Rodents

fruit trees such as almonds, apricots, and peaches. The Oregon ground
squirrel practically never climbs.

Ground squirrels are ^'fair-weather animals," active by day in the
warmer season of the year and on the occasional warm days in winter.
The young are brought forth in the season of plentiful food. Part of the
population in the lowlands spends a period from late summer to mid-

6 6 /O /;i H /6 m ;?0 Zl Z4 /?^ ^g 30 32 3^ 36 dd -40 4Z 44 46 46 ^0

rig. 1. — Plan and section of "colonial" burrow system of California ground
squirrel : In, entrance ; a, food store ; nests as indicated. Each square is 2 feet. Exca-
vated near Bakersfield, May 3, 1918, by Joseph Dixon and H. G. White. (From
Grinnell and Dixon, 1919.)

winter in "summer hibernation" or estivation (as detailed beyond), in
contrast to those digger squirrels living in the mountains and all the
Oregon ground squirrel population which regularly hibernate in the
colder season of the year.

Burrows. — All species of ground squirrels dig burrows, which are
used for safety retreats, for shelter during inclement weather, and
during estivation or hibernation, for the storage of food, and for the
rearing of young. Burrows are made in flat land, in hillsides, and in
ditch, road, and railroad embankments. These burrows vary in length,
depth, and complexity. The diameter depends upon the species of
squirrel. Dixon found that the burrows of the California ground squirrel
averaged 4.3 inches in diameter, and that there were three kinds of bur-
rows, "male," "female," and "colonial," the last being longest and most
complex with numerous openings (fig. 1). The individual burrows were
from 5 to 34 feet in length (including all side branches), while one

10 California Agricultural Extension Service [Cir. 79

"colonial" system had a total of 138 feet with 8 entrance holes. Squirrel
burrows vary in extreme depth; the deepest found by Dixon was 66
inches. Of six burrows of Douglas ground squirrel excavated at Davis,
the deepest reached to 80 inches below the ground surface. In one
instance reported from Fresno County California ground squirrel
burrows were encountered in a chalk pit down to 28 feet below the
surface. The estimated volume (which has a direct bearing on the
amount of carbon bisulfide used in the gas method) varied from 1.03 to
17.8 cubic feet in the burrows excavated by Dixon ; one Oregon ground
squirrel burrow excavated by Grinnell and Jacobsen averaged 2.5
inches in diameter, was 66 feet in length, and had an estimated volume
of 3.5 cubic feet.

Feeding Habits. — Although ground squirrels are often called "sper-
mophiles" because the preferred food for many species consists of
seeds, most of them will take other kinds of food, including grasses and
herbs. Every ground squirrel has two thin, membranous internal cheek
pouches, opening just inside the lips, one on each side of the mouth. The
cheek pouches thus differ from the external, fur-lined pouches of pocket
gophers and kangaroo rats, but are used for the same purpose in all
these animals — namely, to carry food and (in some cases) nest material
also.

The feeding habits of ground squirrels vary with the season and the
locality. During the rainy months, from November to March or April,
the digger squirrels eat some of the grain, seeds, or nuts stored below
ground the previous summer; and they also glean scattered seeds on
the surface of the ground, eating these wherever found. At this season,
however, their principal food is grass and green herbage. As seeds begin
to ripen the squirrels gather these in their cheek pouches to carry them
below ground. Seeds of both wild plants and the cultivated grains are
taken. The digger squirrels are fond of the soft, growing pits of peaches
and apricots and in getting these they may destroy much fruit. They
also eat almonds and walnuts.

The damage by ground squirrels, by reason of their feeding habits, is
chiefly to grain and pasturage. In fields they dig up sprouting grain
and later pull down the ripening heads. After harvest they concentrate
about shocks and stacks, continuing to eat and carry off quantities of
grain. On pasture lands they may take appreciable amounts of grasses
and herbage that could otherwise be utilized by domestic stock. Accord-
ing to Dixon 200 California ground squirrels on open range might con-
sume as much green forage as one steer or ten sheep. On isolated ranches

1933] Control op Injurious Rodents 11

bordered by wild lands on one or more sides, squirrels may seriously
deplete grain, fruit, and nut crops.

Estivation and Hibernation. — A period of torpidity obtains among
some of the ground squirrels in the valleys of California. It may extend
continuously from late summer (estivation) through midwinter (hiber-
nation). A female ground squirrel regularly and continually observed
under normal conditions in a dooryard in southern California did not
estivate until its second year ; thereafter, each succeeding year of its life,
it estivated regularly, becoming very fat and retiring to its burrow
during the last week in August. If removed from the burrow during this
period, it proved to be torpid, without perceptible respiration. About
February 22 of each year following, with marked regularity, it emerged
in an emaciated condition. This habit of estivation affords an explana-
tion for instances where all the squirrels active in a field during the
autumn are killed, and yet old breeding squirrels suddenly appear the
following spring, as sometimes happens when there is seemingly no
chance for reinfestation from the surrounding fields. The extent of this
habit of estivation among our ground squirrels is unknown because it
is exceedingly difficult to follow an individual squirrel through its
activities for any length of time. However, this suggests the advisability
of control in the late spring and early summer rather than in the fall
and early winter when some of the breeding stock may be underground
out of the reach of poisoned grain. There is also some question as to
whether dormant animals, in which respiration is extremely slow, are
fatally injured by fumigation in their burrows.

The Oregon ground squirrel is reported to go into its quarters from
the first week in August to mid-September, and may emerge as early as
mid-March when there is still much snow on the ground.

Breeding. — The California ground squirrel, in the central part of the
state, breeds during the first half of the year according to data secured
on more than 20,000 females collected by the United States Public
Health Service in 1910. Few females were pregnant in January, but the
proportion rose steadily until at the end of February 42 per cent of the
females contained embryos, in mid- April, 16 per cent, and in mid-May,
5 per cent ; no females with embryos were found after June 4. However,
examination of smaller numbers of squirrels in Los Angeles County
during 1925 revealed a few pregnancies each week between July 4 and
October 13. Regular breeding activity began there by December (71
pregnant females among 1,698 squirrels) ; and in March, 1926, no less
than 70.3 per cent of 1,976 female squirrels contained embryos. Thus a
longer season and a higher percentage of pregnancies is indicated for

12 California Agricultural Extension Service [Cir. 79

the southern region.^ Scattered data indicate that the breeding season
is slightly earlier in the warmer interior valleys and somewhat later
along the coast and in the mountains.

The exact period of gestation in the California ground squirrel is
unknown ; for the Douglas ground squirrel in western Oregon a period
of 25 to 30 days is indicated, followed by about 7 or 8 weeks of growth
in the nest and about the parent burrow. If the records for the year 1910
cited above represented, on the average, embryos half-way through the
period of gestation, then embryos of any given date would be repre-
sented by young running about aboveground about 9 weeks later ; this
calculation indicates early May as the time when the greatest number
of young squirrels should appear, and, indeed, coincides with field
observations.

Ground squirrels in California usually produce but one litter an-
nually. The number of embryos was found by the United States Public
Health Service to average 7.2 per female. In southern California the
averages were slightly higher ; the actual number of embryos in indi-
vidual females varied from 4 to 15 in the latter region.^ In Tulare
County, on an area where squirrel control had been carried on for two
years or longer and where the available food supply for the squirrels
remaining was therefore presumably greater, larger litters were found,
an average of nearly 10 embryos per female being noted by W. C.
Jacobsen.^

Less is known about the breeding of the other ground squirrels. The
Oregon ground squirrel has about 8 in a brood ; the season of appearance
is from mid-May at the lower altitudes (3,000 feet), until early June at
higher levels.

The important thing to remember regarding the breeding of squirrels
is that one female killed before April 1 is equivalent to the destruction of
about 8 squirrels later in the year — hence the great desirability of con-
trol in late winter and early spring.

The rate of reproduction in ground squirrels is such that any cam-
paign which does not ordinarily succeed in eliminating 85 per cent in
a given year will not effect any general reduction in numbers. Theoreti-
cally, if 90 per cent are removed each year it would require 8 to 9 years
to eliminate completely the squirrels on a given parcel of land, provided
no reinvasion occurred by migration of squirrels from adjacent areas.
Squirrels have been known to migrate one or more miles into cleared

5 Storer, T. I. Summer and autumnal breeding of the California ground squirrel.
Journal of Mammalogy 11:235-237. 1930.

6 Jacobsen, W. C. Eate of reproduction in Citellus hrecheyi. Journal of Mam-
malogy 4:58. 1923.

i^'^3] Control of Injurious Rodents 13

areas. These facts indicate the importance and indeed the necessity for
persistent and intensive work if a land owner wishes to keep down the
squirrel popidation on his property. Furthermore, cooperative effort
on the part of all ranchers in a region is necessary, else cleared areas
will be invaded by squirrels from lands where control is not practiced or
is ineffective.

Methods of Control. — The five most effective methods of destroying
ground squirrels are: (1) poisoning with strychnine; (2) fumigation
with carbon bisulfide ; (3) trapping; (4) shooting; (5) encouragement
of natural enemies.

1. Poisoning: Strychnine-coated barley (formula 1) ' is to be used in
that part of the dry season when squirrels are gathering seeds and grain
in their cheek pouches for storage underground. The grain is coated
with the less-soluble strychnine alkaloid, together with substances to
disguise the taste. The kernels are picked up by the squirrels and quickly
placed in the pouches, the strychnine coating is absorbed by the delicate
membranous lining of the pouches, and causes the death of the squirrels.
Before the storage season, squirrels eat scattered seeds as found and
then will often reject strychnine-coated grain. After grain crops begin
to mature the animals will be attracted to the more abundant supply
and will be less interested in poisoned bait. The exact period, in any
locality, when exposure of poisoned grain will produce results must
be learned by the person doing control work ; it will vary according to
locality and may vary from year to year in any one locality.

The poisoned grain should be scattered from the hand, a tablespoonf ul
at a time, on hard bare ground, as on squirrel runways radiating from
burrows or along fences. If dropped into tall grass or in the soft earth
about burrows, it is less likely to be found by the squirrels. One quart of
poisoned whole barley will provide enough for 30 to 35 baits. Around
farm yards or in pastures where livestock are present the poison, for
the sake of safety, must be put well down the burrows. Placing of
poisoned grain in the burrows has been used in some localities, especially
where no control work with this material has been carried on in the
immediate past. It is reported that the best results with grain placed
in burrows are obtained during the summer and autumn months.

Failure to reduce or eliminate squirrels may follow use of poisoned
barley when ram washes off the poison coating, when squirrels refuse,
for one reason or another, to accept the poison, or when squirrels are
feeding on filaree. F. E. Garlough reports that in areas where green

T Eefers to section, "Formulas for Preparing Poisoned Baits for Eodents,"
appearing at the end of this circular.

14 California Agricultural Extension Service [Cm. 79

filaree is being eaten abundantly, stryclinine-coated barley is ineffective.
His explanation, that the high tannin content of filaree counteracts the
effect of the strychnine, has lately been questioned by Burnett.^

Animals other than ground squirrels, including domestic stock and
various wild rodents and birds, are readily killed by strychnine. Quail
and pheasants, and presumably also domestic poultry, can stand higher
doses of strychnine, but this fact should not be taken as an excuse for
careless placement of poison. Strychnine-coated barley is also poisonous
to man ; hence, supplies of the poisoned grain must he locked up, so as to
be inaccessible to children and irresponsible adults.

Burnett in studies on the Wyoming ground squirrel, a species about
half the weight of the California ground squirrel, has found considerable
variation in the susceptibility of individual squirrels when fed strych-
nine-poisoned oats in captivity. He infers that difficulty in poisoning
with strychnine under field conditions, where certain individuals sur-
vive, may result from this variation and also from a tolerance built up
by repeated sub-lethal doses. Much individual variation was shown by
the captive squirrels; some adults were killed by as few as 10 or 12
kernels ; others survived relatively heavy doses ; one squirrel ate over
1,000 kernels of poisoned hulled oats in 13 days before being killed. No
corresponding studies have been published on the California ground
squirrel. It has long been believed that 20 kernels are sufficient to kill
an average squirrel of the latter species.

Oats (formula 2) and dandelion greens (formula 3) are recommended
for the Oregon ground squirrel in place of barley. The oats are to be
distributed in the same manner as barley ; two or three of the dandelion
greens are to be put in each burrow. Birds are not endangered by the
poisoned greens.

2. Carbon bisulfide : This chemical, suggested for use against ground
squirrels by E. W. Hilgard in 1876, has found extensive use ever since.
From a half million to a million pounds (10.76 pounds per gallon) were
used annually in California during the decade 1921-1931.

Carbon bisulfide is a liquid that volatilizes easily. The following facts
should be kept in mind in using this material: (1) The fluid is highly
inflammable and the gaseous form is highly explosive ; it should be kept
away from all fires and exposed lights ; (2) being heavier than air, the
gas settles in the lowest places in the underground burrow and hence
will not go over an elevation higher than the entrance unless a pump is
used to force it in ; (3) when the ground is dry and full of cracks the gas

8 Burnett, W. L. The action of strychnine on the Wyoming ground squirrel,
Citellus elegans elegans. Colorado Agr. Exp. Sta. Bui. 384:1-19. 1932.

1933]

Control op Injurious Rodents

15

escapes and is less effective than when the ground is wet and the gas thus
confined to the burrow ; (4) it should be used only in holes known to be
occupied by squirrels at the time of the treatment ; (5) stocks of carbon
bisulfide should be kept tightly corked for the fluid evaporates rapidly
on exposure to the air.

r

1

1^^^^

■m

Fig. 2. — Karbo-Killer, for pumping carbon
bisulfide into rodent burrows.

Two methods of application are in use, either waste balls saturated
with the liquid are thrown into the burrows, or the material is pumped
into the burrow by means of a special machine. In earlier years waste,
corn cobs or even horse droppings were used to carry the fluid.

In the waste-ball method, the balls are soaked in the fluid and one or
two are thrown into each burrow. Since each waste ball requires about 2
fluid ounces, 50 to 60 waste balls may be dosed with a gallon of bisulfide,
allowing for evaporation and spillage losses. As each burrow is treated,
the entrance should be closed with a shovelful of earth and quickly
tramped down.

When using the waste-ball method some operators ignite the gas by
means of a flame. Matches are dangerous for this purpose. To explode
the gas, a six-foot piece of %-inch pipe with one end closed and pointed

]6 California Agricultural Extension Service [Cm. 79

and the other end tightly wrapped with a rag soaked in kerosene adds
safety to the operation. After the bisulfide has been confined in the
burrow for a few seconds a hole is made with the pointed end of the pipe
through the dirt plug at the burrow entrance. The gas is then "touched
off" with the torch on the other end of the rod which has been previously
lighted. The operator should stand well to one side of the burrow when
igniting the gas. Exploding the gas may reveal open burrows in grass-
land which otherwise would not be noticed. The gas should be ignited
only when the ground and surface vegetation are damp, never in dry
grain fields, about buildings, or other places where there is risk of fire.
The explosion and burning of the gas produces a mixture of carbon
dioxide, carbon monoxide, and sulfur dioxide. The latter two are
poisonous, especially the carbon monoxide. Stewart and Burd^ con-
cluded that there is probably a slight advantage in exploding the gas.
The use of an adequate pump is much the better method.

The first practical "pump" was developed by Surgeon J. D. Long
of the U. S. Public Health Service. In recent years the Karbo-Killer
(fig. 2), a heavy metal cylinder with contained pump and reservoir
for the liquid carbon bisulfide, has replaced practically all other ma-
chines. Application by pump tends to drive the gas into all portions of
a burrow.

In using a pump, place the flexible hose in a burrow, throw a shovelful
of earth over the hose to prevent loss of gas, and deliver 20 strokes with
the pump. After the hose has been withdrawn tramp down the earth in
the burrow entrance.

In Santa Barbara County, where as many as 50 Karbo-Killers have
been used, workers go over an area and close all open squirrel burrows
with mattocks, then apply the gas several days later. This practice
obviates the need for treating uninhabited burrows. The gas should be
pumped slowly. On retreatment of burrows found open after the first
coverage, the number of pump strokes is doubled.

Surgeon Long found that his "destructor" would treat 200 to 250
holes with one gallon. A test of the Karbo-Killer in the laboratory (with
no back pressure on the hose, as in a burrow) indicated that from % to
% ounce of carbon bisulfide was delivered in 20 strokes of the pump, or
192 to 384 dosages per gallon. The slower rate of pumping delivered
smaller dosages. Field tests of the Karbo-Killer by various rodent-
control officials indicate that from 125 to 250 holes can be treated with
one gallon of carbon l)isulfide, using 10 to 20 strokes of the pump.

9 Stewart, G. R., and J, S. Burd. Control of ground squirrels by the fumigation
method. California Agr. Exp. Sta. Bui. 302:205-224. 1918.

1933] Control of Injurious Rodents 17

In flat country the exhaust gas from an automobile may be employed
to kill ground squirrels, as described beyond for pocket gophers.

Although many other materials have been tried for gassing squirrel
burrows, both experimentally and otherwise, none has proved so satis-
factory as carbon bisulfide. From 1925 to 1980, calcium cyanide flakes
and dust were tried by official agencies but the material did not find
favor.

3 and 4. Trapping and Shooting : Where ground squirrels are digging
into ditch banks, and in other cases where they must be disposed of
promptly at any cost, special means must be adopted. Trapping and
shooting are valuable for such local application. These two methods can
be used at any season, but the time required to keep traps properly set,
which is essential to success, and the high initial cost of traps as well as
of ammunition make them too expensive for general use on large acre-
ages. Trapping and shooting are also useful in cleaning up the few
'Vise" squirrels that escape the poison and gas, and in reducing the
breeding stock in the early spring before the annual increase.

The No. 91 Oneida jump trap is the best, being lighter, easier to set,
and having larger catching surface (pan or treadle) than the ordinary
steel trap with an outside spring. The jump trap lies flat on the ground ;
the jaws have wide contact surfaces that reduce the chance of breaking
the animal's leg ; and the inside spring causes the trap to jump up and
take a high grip on the leg instead of on the foot only. Such a trap may
be set without bait in the entrance to a burrow, or may be baited with
rolled barley, and set in a squirrel path, or near a feeding place. In any
case it is well to scrape out a slight depression so that the upper surface
of the trap when set will be flush with the surrounding ground. The trap
should be well secured. A long stake sharpened at one end, passed
through the ring in the end of the trap chain, and then driven well into
the ground will serve to hold the trap and also as a marker, enabling one
readily to find the trap again.

Shooting with a long-range .22 caliber rifle may be used to clean up
small numbers of squirrels.

5. Natural enemies : The natural enemies of the ground squirrel con-
stitute one of Nature's checks on these rodents. Some of these are
coyotes, badgers, weasels, wildcats, red-tailed hawks, golden eagles,
rattlesnakes, and gopher snakes. Badgers, weasels, and snakes capture
the ground squirrels in their burrows. Wildcats and coyotes lie in wait
near the burrows until the squirrels venture forth in search of food.
Dixon examined 186 stomachs of wildcats from forty different localities
in California; 26 contained ground squirrels and these with other

18

California Agricultural Extension Service [Cir. 79

rodents were found to constitute more than half of the food. Hawks and
eagles swoop down on the squirrels from the air. The importance of pre-
serving as many as possible of these native enemies of the ground
squirrel is evident. The aid of coyotes and rattlesnakes may be dearly
bought; but badgers, gopher snakes, hawks, and eagles cost little or
nothing, and each catches ground squirrels almost throughout its life-
time. The accompanying table shows what certain hawks and eagles
actually accomplish. The dead rodents counted in the nests listed repre-
sent merely the surplus which the old birds had carried to their young,
in addition to food eaten on the days of observation. Hawks and eagles
it thus appears capture, when prey is abundant, far more than they and
their young can actually consume.

TABLE 1

Destructive Rodents Found in Nests of Hawks and Golden Eagles
IN San Diego County by James B. Dixon

Species of bird

Date

Young and eggs

Locality

Evidence

Red-bellied hawk

Western red-tailed
hawk

April 3 (1916)....

March 28 (1906)

March 26 (1909;
April 4 (1907)..

3 young, one week old

1 day-old young, 2 pipped

Pala

Vista

Rincon ..
Lilac

1 ground squirrel and 2
gophers

Remains of 2 ground

Golden eagle

1 day-old young

squirrels
9 jackrabbits

2 young, one week old

11 ground squirrels in
and about nest

6. Other methods : Small orchards of deciduous fruits or nuts in iso-
lated regions may be protected from ground squirrels by fastening
smooth cylinders of tin about the tree trunks. This tinning, if started
about 2 feet above the ground and continued upwards for 2 or 3 feet
will usually keep the ground squirrels out of the trees unless there are
drooping branches on which they can climb.

Flat collars or disk-like shields of sheet metal about 2 feet in diameter
are often used to protect young roadside trees against defoliation by
ground squirrels. A hole in the center of each shield admits the trunk
of the tree, and a radial cut, from circumference to center, enables the
shield to be placed in position, below the first branches.

Seed corn can be protected from ground squirrels during germination
by being treated with coal tar.^° For this purpose add one large spoonful
of coal tar to a gallon of boiling water. When the mixture has cooled

10 Lantz, D. E. Destroying rodent pests on the farm. U. S. Dept. Agr. Yearbook
for 1916. p. 381-398. 1917.

1933] Control op Injurious Eodents 19

somewhat the corn may l)e stirred in and allowed to remain several
minutes ; germinating qualities will not be impaired.

Costs of Various Control Methods. — The cost of ground-squirrel con-
trol varies with the degree of infestation, the method used, and the cur-
rent cost of control materials and labor. Material costs have been as
follows: Strychnine alkaloid per ounce, 1919, $1.95; 1925, $1.00; 1928,
$0.75 to $0.60; 1932, $0.45 to $0.36; 1933, $0.23. Carbon bisulfide, per
gallon, 1912, $0.90 ; later, $1.25 ; 1932, carload lots, $0.55, or 1 to 5-gallon
lots, $0.75 per gallon. Waste balls, per thousand, $3.00. Prepared strych-
nine-coated barley was sold by county agricultural commissioners in
1931 to ranchers at 6 to 7 cents per pound. Labor costs in the decade
1921-1930 were about $4.00 per man per day or $5.00 for man and horse.

It is rather difficult to give exact current figures on costs per acre. In
1910, according to C. H. Merriam of the U. S. Biological Survey, single
treatments with grain cost 3 cents per acre; with waste-ball gas, 1.5
cents per burrow. In 1912, on the basis of 10 holes per acre, Surgeon J. D.
Long gave costs per acre as follows : poisoned grain (4 to 5 treatments) ,
35 cents ; waste-ball gas (2 treatments) , 68 cents ; ''destructor" (2 treat-
ments), 20 cents. In 1922 F. E. Garlough of the Biological Survey cited
costs as: poisoned grain, 6.5 cents per acre; waste-ball gas, 8.5 cents.
The skill, speed, and accuracy of laborers, the degree of infestation, and
the nature of the ground to be covered all tend to vary the costs.

TREE SQUIRRELS

In the foothill and mountainous parts of California there are two
kinds of tree-inhabiting squirrels, the gray squirrel {Sciurus griseus)
and the red squirrel {Sciurus douglasii), also called chickaree or pine
squirrel. The former of these often lives adjacent to nut and fruit trees
and, at times, takes the products of these trees, to the dismay of the
grower. One grower in Santa Cruz County reported that in years of
short crops of acorns the gray squirrels took many English walnuts and
also gnawed into boxes of apples stored in the open. By setting box traps
he captured several squirrels, transported them less than a mile from his
ranch, and released them ; no further damage was noted. Tin guards, as
detailed for ground squirrels, would exclude tree squirrels as well, pro-
vided there were no low-hanging branches on the trees and provided
further that the trees were sufficiently isolated so that the gray squirrels
could not travel overhead from tree to tree and so gain access to the
nuts. Tree squirrels are protected under the California Game Code.

20 California Agricultural Extension Service [Cm. 79

POCKET GOPHERS

Pocket goi)hers are small, stout-bodied, short-legged rodents, with
conspicuous front (incisor) teeth, a pair of external fur-lined cheek
pouches or pockets — one on either side of the mouth — used for carrying
food materials (and not earth), small eyes and ears, and a short, nearly,
naked tail. The term "gopher" is applied in the northern Middle West
to one or more species of ground squirrels, and in Florida to a land
tortoise ; so the special designation '^pocket gopher" is used for certainty
of identification here.

The work of pocket gophers is sometimes confused with that of moles.
The two animals are quite dissimilar, the pocket gopher being a rodent
with blunt head, conspicuous incisor teeth, and a brownish-colored coat,
a somewhat longer tail, and small front feet. The mole has a slender,
conical snout; no eyes or ears evident externally; small, needle-like teeth;
very large fore-feet and claws; and silvery black fur, velvety in texture
and reversible. The mole has no cheek pouches of any sort. The accom-
panying illustrations (figs. 3, 9) show important differences between the
workings made by moles and pocket gophers.

Pocket gophers are widely distributed over most of California, being
absent only on certain portions of the southeastern deserts. They inhabit
practically all kinds of soil, save very rocky areas, although they are
more abundant on the better soils. Between 30 and 40 different species
and varieties of these animals have been described from California, each
inhabiting a special geographic area, but no two live in the same locality.
For practical purposes they may all be treated as one, their habits and
the methods for their control being essentially the same over the entire
state.

Burrows. — The pocket gopher is strictly an inhabitant of the soil, liv-
ing in burrows of its own construction, never climbing, and only seldom
coming out on the surface of the ground. It digs clean-cut round tunnels,
al)Out 2 inches in diameter, more or less parallel with the surface of the
ground, at depths of from 6 or 8 to 12 or 14 inches. The earth from these
(fig. 3) is thrown out on the surface through short lateral tunnels made
at frequent intervals. This results in a series of rounded surface mounds
which, by their position, often give a clue to the location of the main
tunnel. When i)utting earth out of a lateral tunnel the gopher pushes
the successive loads of earth into a more or less crescentic pattern and
when the lateral is closed a low central plug or depression in the mound
often indicates the location of the mouth of the lateral.

1933] Control op Injurious Rodents 21

The age of a mound may often be determined by the amount of
moisture in the earth that has been pushed out and also by the degree of
blanching (loss of chlorophyll) in any grasses or herbs in the mound.
Trapping seems most productive in fresh workings.

Besides the laterals used to push out earth the gopher makes short,
almost vertical laterals in coming to feed on surface vegetation. These
often are closed with earth that does not rise above the adjacent ground
surface.

t.

Fig. 3. — Method of pocket gopher in pushing earth out of main tunnel through
a lateral tunnel and onto the surface of the ground. The mouth of the lateral re-
mains open when excavation is in progress; successive loads of earth are pushed
in different directions from the mouth of the tunnel, resulting often in a mound
of crescentic outline.

Gophers construct deeper tunnels in connection with their nests, and
there are also short, steeply pitched "sumps," presumably to drain ad-
jacent tunnels. The nest is usually in a chamber about 8 inches in diame-
ter ; it is constructed of shredded fibers of grasses and other plants, giv-
ing a resemblance to fine excelsior. Food is often stored beside the nest
or in other enlarged chambers off the tunnel system.

The burrow system of a pocket gopher may be extensive, involving
up to 100 yards of tunnel length. Ordinarily each system is inhabited by
a single gopher, although young may remain in the tunnel occupied by
a female for a time after leaving the parental nest. The systems of ad-
jacent gophers may be connected, but connecting tunnels, and even
portions of the workings of a single animal, are often plugged firmly
with earth. When a gopher is trapped out of a tunnel system another
animal may later occupy that system, and moles have, at times, been
trapped in gopher burrows. Pocket gophers are active throughout the
year (even in mountain areas where they work beneath the snow and

22 California Agricultural Extension Service l^^^- ^9

put the surplus eartli in tunnels in the snow), and fresh workings may
be found in any month. Activity may be less on dry areas during the hot
summer season. The animals are also less active during and immediately
after a heavy rain.

Breeding. — On pasture lands and on uncultivated and unirrigated
areas there is evidently a limited breeding season, after the beginning
of the rains, when green forage becomes available in quantity ; on such
areas there is probably a single annual brood. But in irrigated regions,
and especially in alfalfa fields, where green forage is available for much
or all of the year, a longer if not year-round breeding season is evident.
Dixon" has found that in southern California breeding begins early in
November and is about two weeks later in central California. In an al-
falfa field at Turlock he surmised that two or even three broods were
raised in a single season.

At Davis, the writer examined gophers from November until May and
found embryos during each month of this period. Information on go-
phers at high altitudes (5,000 feet and above) indicates breeding in
June and July.

The average number of embryos is between 5 and 6 per female. Among
gophers from all parts of California, Dixon^- found limits of 3 to 12
embryos, with an average of 5.8 in 28 females; at Davis, in 1932, 53 of
224 females contained embryos, from 1 to 10 per female, the average
being 5.7. In this latter group, all from the University Farm, the ratio of
pregnant females varied in different weeks from 16.2 to 33.3 per cent,
but many others among the females collected showed signs of recently
having given birth or suckle to young so that a very high percentage of
the females breed in any one year. Gophers (at Davis) begin to breed
when about 3 ounces (90 grams) in weight, the head-and-body length
then being about 5^/2 inches (135 millimeters) .

The young remain in the nest for several weeks after birth but eventu-
ally leave the parent tunnel system for an independent existence, often
wandering some distance overland and starting their small tunnels in
new places. In cities, young gophers sometimes go out on the streets and
are unable to mount the vertical surface of concrete curbs. Gardens and
fields earlier free from gophers may become tenanted by young in the
spring months. The following dates indicate approximately when the
main crop of young becomes dispersed although young are scattered
both earlier and later : southern California, March 20; San Joaquin and

11 Dixon, Joseph. The breeding season of the pocket gopher in California.
Journal of Mammalogy 10:327-328. 1929.

12 Dixon, Joseph. Control of the pocket gopher in California. California Agr.
Exp. Sta. Bui. 340:338. 1924.

1933] Control op Injurious Rodents 23

Sacramento valleys, April 1; Owens Vallej^, April 15; foothills of the
Sierra Nevada, April 30 ; northwest coast region, May 15.

Damage Due to Pocket Gophers. — On wild lands gophers are probably
beneficial in the long run. Although pasture lands cleared of them often
show an immediate increase in the amount of forage available for live-
stock, it is unknown whether this increment would continue for a suc-
cession of years. On unplowed land gophers "cultivate" the soil, often
turning over considerable percentages of the surface in a single year.
Whether their burrowings contribute to or serve to check erosion on
slopes is undetermined. The rich sediments of valley bottom lands have
resulted from erosion at higher altitudes in past geologic time ; to this
process pocket gophers have contributed. The ''long-time" values need
to be studied carefully.

The immediate and specific damage to man's interests by these animals
results from their feeding habits and their tunneling. Production in
alfalfa fields can be seriously reduced by their destruction of root
crowns; truck plants suffer direct damage by consumption of roots;
valuable plants in flower gardens, especially those with bulbous roots,
are often destroyed. Gophers cut roots of trees and vines and gnaw the
bark of trees, at times completely girdling the latter so that they die
unless saved by bridge grafting.^ ^ Gopher burrows in home gardens
often divert the relatively expensive metered water. Burrows in the
banks of ditches and canals may lead to breaks in the earthwork whereby
water is lost and adjacent lands are flooded, necessitating expensive
repairs.

Methods of Control. — With persistence, pocket gophers may be ef-
fectively controlled and even eliminated over considerable areas. No
other common rodent can be held down or exterminated so successfully,
because of the slow rate at which gophers advance their tunnel systems
and their general disinclination to travel long distances overland.

In general, gopher control may be practiced at any season. Prompt
attention to the first evidence of gopher work in a garden will often save
valuable plants. Obviously the best time to exercise control is before the
young are abroad. Both trapping and poisoning are easiest soon after
green vegetation starts in the early winter or spring, when the gophers
are very active and the ground is soft. Wise old males, which usually
cause the trapper most trouble, seem to lose their instinctive caution dur-
ing the mating season, take poisoned bait readily, and often blunder into
traps that they would never enter at other times. Every female caught

13 Hodgson, E. W. Saving the gophered citrus tree. California Agr. Exp. Sta. Cir.
273:1-19. 1923. (Out of print.)

24

California Agricultural Extension Service

[Cm. 79

then, 1)efore the young- are ])orn, means from 4 to 12 fewer gophers for
the next season.

The methods in use against pocket gophers are: (1) trapping; (2)
poisoning; (3) gassing; (4) flooding; (5) fencing; (6) encouragement
of natural enemies; (7) miscellaneous. Each has its particular advan-
tages, and best results follow use of two or more methods.

1. Trapping : Many special kinds of traps have been designed to cap-
ture pocket gophers (fig. 4) . The usual snap and steel traps used against

Fig. 4. — Different kinds of traps for pocket gophers ;
all are shown as set, ready for use.

rats and mice and for carnivorous mammals are ineffective against go-
phers, which must be trapped in their burrows where space for operation
of a trap is limited. The special traps fall into two classes : those de-
signed to spring when a gopher pushes a load of earth against the flat
trigger pan of the trap, and those operated by a bait trigger, moved when
the gopher seizes the special bait.

Trapping is especially adapted to gardens, orchards, small fields, and
banks of irrigating canals. In California, more gopher control is prob-
ably accomplished by traps than by any other single method. Traps are
useful in following up and capturing individual gophers. The special
gopher traps are safe to handle and require only a limited amount of
skill and a little digging to place them. On one ranch near Davis a work-
man handled about 75 traps, set usually in pairs. These required about
5 hours of time to examine and reset as needed daily, and on 3 successive
days took 38, 40, and 37 gophers, respectively.

The most successful and most commonly used trap in California is the
Macabee (fig. 5a), constructed entirely of wire except for the trigger.
The next most popular is the box type with a choker effect (fig. 4, upper
left). Numerous other types have been designed and marketed (fig. 4)

1933

Control of Injurious Rodents

25

but none seems to have found any lasting demand. Certain explosive
traps or "gopher guns" have l)een exi)loited but the danger attending
their use makes them undesirable.

The most effective "set" for the Macabee trap is in the main runway
(fig. 6), and not in the lateral run leading to the surface mound. This
necessitates the use of two traps per setting, one in each direction ; but
the results are much more certain and quicker, so that the catch per trap

Fig. 5. — Regular Macabee gopher trap (a) ; modified (reconstructed)
Macabee trap for "wise" gophers and for moles (Z>) ; both traps are set, ready
for use.

per day is greater than if only one trap is set in a lateral run where it is
often filled with dirt by the gopher. An ordinary stiff-handled 12-inch
iron spoon helps greatly in finding the main run and in placing the trap
properly. It should be supplemented with a light short-handled shovel,
for digging down to the main run. The freshest mound should be se-
lected, and the probable direction of the main run determined by noting
the angle of the dirt-plugged hole. The mounds are usually situated 1 or
2 feet distant from, and the laterals nearly at right angles to, the

mam run.

The procedure is as follows : push the handle of the spoon into the
ground where the lateral is believed to be. If the handle enters an open

26

California Agricultural Extension Service

[Cm. 79

lateral it will drop through the opening. If the lateral is filled loosely
with dirt, the drop will be less noticeable but still plainly felt. If it is
plugged tightly one must dig down a little distance with a shovel before
probing again. Should this fail, try a new mound. When a lateral is
located, follow it down to the main run, which is always kept open by
the gopher. With the shovel clear a place so that a trap can be set in each
direction. Clear out the main run with the spoon, disturbing it no more
than necessary. Set the treadle, or pan, so that a slight touch will spring

c,^

.^---Q

%.

o__^-—

Fig. 6. — Section through portion of burrow system of pocket gopher, showing main
tunnel (a, a) and closed laterals {h, b) together with surface mound (c). Traps are
best placed at a, a. Poisoned baits should be inserted Avith a probe in the main tunnel.

the trap, and place the trap well into the hole. A little loose dirt should
be left in the bottom of the hole to cover the prongs and front end of the
trap when the latter is pushed into place. Press the trap down firmly
so that it will not slide backward if a gopher pushes against it. Then
cover the burrow with a clod or a handful of grass or alfalfa so that
only a little light can reach the trap. A gopher instinctively closes all
open burrows tightly to keep out natural enemies. A trap placed in an
open hole may be sprung by the dirt which the gopher pushes ahead in
plugging the hole.

Each trap should have a light wire or chain attached so that it can be
fastened to a stake or some other object to mark its location ; this plan
also prevents the trap from being dragged far back into the tunnel by a
wounded gopher or being removed by a carnivorous animal when it
contains a gopher. If conspicuous stakes 2 feet or so in height are used,
traps are less liable to be lost, should the field be cultivated while they
are set. Stakes are essential to mark trap locations in alfalfa fields or
truck patches, otherwise many will be lost. On a ranch where traps are
used regularly it is well to have some distinctive kind of stake so that

1933J

Control op Injurious Rodents

27

all the farm laborers will recognize gopher sets. In case only one trap is
available to set in a main run, a careful watch will often, but not always,
reveal one or more little flies emerging from the run when it is first
opened. These flies seem to have a direct relation to the gopher ; experi-
ence has shown that they come from the side then occupied by the
gopher — hence, from the tunnel in which to set the trap. Each setting
should be visited morning and evening, or of tener.

Individual gophers will frequently be found that refuse to enter any
sort of trap. These are generally old males, likely to be the ones that do

O

^...=1

a be

Fig. 7. — "Pan" of regular Macabee gopher trap (a,h) -, dotted Uiies show where
this should be bent to form the horizontal "pan" for modified trap (c).

most of the gnawing on fruit trees. Wlien such an old-timer repeatedly
fills the trap with dirt, a special method becomes necessary (fig. 7).
Take a Macabee trap and move the treadle forward about an inch and a
half, placing the wire which carries the treadle below, instead of above,
the two longitudinal wires. Cut off the wire trigger to meet this change.
Then bend the treadle 'backwards at right angles to its former position
so that it will lie parallel with the trap, instead of standing up at right
angles to the runway (figs. 5h, 7). Set the trap so that it will spring
easily. Put a pinch of loose cotton under the treadle to keep out dirt,
place the trap in the run, and cover the whole trap lightly with a thin
layer of loose earth. Another method is to take a recently caught female
and rub the reproductive parts on the treadle of the regular Macabee
trap. A few drops of urine, squeezed out during the process, will only
add to the efficacy of the decoy. If all else fails, catch a gopher snake and
turn him loose in the run.

After having put out traps, kick the tops off all mounds, so that on the
next round any new mounds will show where gophers remain and where
additional effort is needed.

28

California Agricultural Extension Service

[CiR. 79

2. Poisoning : Pocket gophers may be killed in numbers by use of
poison bait. Since their external cheek pouches or pockets are lined with
fur no poison can be absorbed there as with ground squirrels ; depend-
ence must be placed on stomach poisons. Strychnine alkaloid or sulfate
is effective for this purpose. The bait must be something relished by
gophers and must be placed in the main runs with as little disturbance
as possible; if placed on the surface of the ground it would not often be
found by gophers and might be a menace to other animals, wild or domes-
tic, while if put into laterals or open holes it might easily be buried or

I %"iron rod welded into pipe

it

it

>

/^ wrought iron pipe i ^/f

rod welded into pipe-

-^

Fig. 8. — Probes for locating main tunnels of pocket gopher and for inserting
poisoned baits: upper, probe of steel rod with wooden handle; lower, probe of
%-inch iron pipe with points forged into place.

])ushed out in the dirt. Cubed root vegetables dusted with strychnine
and some sweetening material to disguise the taste are the commonest
bait used (formula 4). Individual baits are placed in main runways by
means of a special probe (fig. 8), with which burrows can easily be
located and baits inserted with a minimum of disturbance.

The probe can readily l)e made by any blacksmith from a 2-foot length
of %-inch round steel. The metal should be pointed at the lower end
and bent near the upper end to form a foot-rest that may be stepped on
by the operator to force the probe into the ground. For work in hard
earth the steel point should be slender ; in loose, sandy soil a blunter
point is better. The upper portion of the probe can be made from an old
shovel handle. Tt should be reinforced at the base by a heavy iron ferrule.
An alternative method of construction is to use half -inch pipe, forging
a ])ointed %-inch rod in one end and a larger ])oint at the opposite end.
Such a probe is slightly heavier but more rigid.

To use the probe, sink the smalh'r pointed end into the ground between
the rows of gopher mounds or plugs, repeating the operation until the
main burrow is recognized by a sudden drop as the probe goes through it.

1933] Control of Injurious Rodents 29

The opening should then be enlarged, by rotating the probe or by using
the opposite larger end, so that a poisoned bait may be dropped easily
into the burrow. After putting in the poisoned bait, the hole made by the
probe should be closed by the operator's heel. As the work proceeds, the
tops of all mounds should be trampled down or kicked off ; it is then easy
to locate new mounds indicative of gophers that escaped the previous
treatment and to place new baits only where needed.

Experiment has shown that 40 per cent more baits are taken when
inserted by a probe than when burrows are dug open with a spoon or
shovel, besides which the probe method is much faster. One man using
this implement, can treat, in a single day, several hundred holes, cover-
ing as much as 40 acres. Use of the probe is easiest when the soil is damp
and soft down to the level of the main tunnels, less easy when the ground
is hard, and quite unsatisfactory in adobe or other soil that cracks when
dry since the probe then drops as easily into a crack as into a burrow.
In finely cultivated fields the dry surface soil should be scraped back
before closing the hole, which may be done with a clod of earth.

The best time to use the probe is in the late autumn during the first
cool weather or just after the first good winter rains. Burrows are easier
to locate when the mounds are conspicuous, before green vegetation be-
comes abundant. The land should be gone over thoroughly at this season.
Alfalfa fields, because of the abundant and continuous food supply, are
sometimes more difficult to treat effectively than orchards or open fields.

Gophers are most apt to gnaw or girdle orchard trees during late
summer, after the surface of the ground has become dry and green vege-
tation is scarce. Thus at the very season when the gopher is doing the
most serious damage, and speedy destruction is most to be desired, the
probing method is less easy and the operator may have to dig down to
the main runways to place poisoned baits. In a garden, nursery, or lawn
where it seems desirable to use poisoned baits rather than traps, formulas
4 or 5 may be employed.

In dry ground, where use of a probe is not practicable, the main runs
of the gopher can be opened up in the same manner as for setting a trap.
Then, using a slender pointed stick, place one of the poisoned baits a foot
back in each runway, which should then be tightly closed. The hole may
be opened two days later ; if the bait is gone and the hole remains open,
the gopher is probably dead.

3. Gassing : Poisonous gases are not often effective against pocket
gophers. The great extent of the burrow system of gophers, the great
chance for leakage of gas through the softer earth of laterals, the close-
ness of the main runs to the surface of the ground, and the fact that

30 California Agricultural Extension Servicp: L^ir- ^^

gophers may at times quickly plug off their burrows when a poisonous
gas is sensed and so evade destruction, are all factors which make the
gas method less efficient. Some workers have reported success with gases,
but in general other methods are believed to be more efficient.

Materials which have been used against pocket gophers include carbon
bisulfide, calcium cyanide (flakes or dust), and the exhaust from an au-
tomobile. The waste-ball method (as applied to ground squirrels) of
saturating a wad of cotton waste, corncob, or other material with the
liquid, putting this into a gopher burrow, and then covering the opening
with earth has been practiced. A pressure pump, such as the Karbo-
Killer, tends to force the gas farther into the numerous branches of a
tunnel system and is to be preferred. If gopher holes are accessible to
an automobile, a hose may be attached to the exhaust pipe behind the
muffler and the other end inserted in a burrow. A slight increase in en-
gine speed, and enriching the exhaust gas by use of the choke on the
carburetor, will force an abundance of carbon monoxide into the burrow.
Calcium cyanide flakes placed in the dampish earth of a gopher burrow
give off hydrocyanic acid gas, but are reported to be unsatisfactory.
Whether use of cyanide dust forced in by a small hand duster is any
more efficacious is unknown.

From time to time, various gopher "bombs" have appeared on the
market. These, when lighted and placed in the burrows, are supposed to
generate a poisonous gas that overcomes the gophers. On the whole, these
preparations have not given much success, and their use is not wide-
spread.

4. Flooding: In most sections of California where irrigation is prac-
ticed, crop lands and orchards are periodically flooded. On alfalfa, this
is done regularly. At such times, the gophers are either drowned or
forced out by the incoming water. If their tunnel systems include runs
in the levees, they may avoid the water by entering the latter. Some,
driven out into tlie open, seek the higher borders of the field. At this
time they may be dispatched readily by a good dog or by a stroke of
the irrigator's shovel.

5. Fencing: Small areas, such as flower or vegetable gardens, adjacent
to wild lands over which the gardener has no control, sometimes merit
especial protection against the entrance of gophers either by burrows
or by overland migration. A fence of 1-inch mesh buried not less than
12 inches below the ground level and extending an equal distance above,
will protect against gophers. If it is carried 24 to 36 inches above ground
it will also be adequate protection against rabbits.

1933] Control op Injurious Rodents 31

6. Encouragement of natural enemies : Two important aids in the con-
trol of pocket gophers are the barn owl and the gopher snake. The
former, which commonly nests and roosts in barns, steeples, and holes in
earth banks or cliffs, is one wild predatory bird whose activities are
wholly beneficial to man. Its diet is always preponderantly of rodents —
sometimes almost exclusively of pocket gophers. One pair of barn owls
studied by Dixon caught 3 to 6 pocket gophers daily for their young.
Among all the records of its food, instances of birds being eaten are ex-
ceptional. It has never been known to molest poultry. The gopher snake
is likewise an efficient ally in gopher control; though it occasionally raids
poultry houses and the nests of wild birds, it eats many gophers and in
open fields and orchards merits full protection. Every gopher eaten by
either of these animals means just one less for the farmer to catch.

7. Miscellaneous methods : Trees may be accorded some measure of
protection against gnawing by gophers if a cylinder of 1-inch wire net-
ting about 12 inches in diameter and 18 inches tall, is sunk in the hole
around the tree when it is planted ; the top of the wire should not quite
reach to the surface of the ground, else difficulty will later be experi-
enced in cultivating around the tree.

Trenching is successful for small-scale operations. A steep- or ver-
tical-walled ditch 18 inches wide by 24 inches deep is dug around the
plot that is to be protected against gophers. Open-topped 5-gallon cans,
spaced at intervals of 25 feet, are sunk so their tops are level with the
bottom of the ditch. Gophers getting into the ditch will be likely to fall
into the cans whence they cannot escape.

Cementing ditches is effective where gophers are active in burrowing
through the banks. A power company that had much trouble with pocket
gophers in a large ditch dug a trench 4 inches wide and 6 feet deep
straight down through the middle of the bank. The dirt was loosened
with an iron bar and removed with a narrow shovel of the type used in
digging telephone-pole holes. The trench was then filled with a "lean"
mixture of cement and sand, carried by a barge that floated on the water
in the ditch. The cement was conveyed to the trench by a galvanized
iron chute built in sections. A small irrigation ditch having a 7-foot sur-
face was protected from gophers, weeds, and leakage by applying, to the
sides and bottom, first a %-inch coat of 1 to 7 cement and then a surface
layer i/i inch thick of 1 to 3 cement. Such costly preventive measures are
advisable only where the usual control methods are ineffective.

"Gopher repellent" plants are sometimes mentioned as helpful in
gardens but there is no satisfactory evidence that any particular plant
will repel gophers.

32 California Agricultural Extension Service [^i^- 79

MOLES

As already pointed out, the work of moles is often confused with that
of pocket gophers. Moles are entirely different anatomically and in gross
appearance, their workings differ considerably, and they subsist, in the
main, on other kinds of food. Moles characteristically feed upon in-
sects and worms in the soil (and are known, therefore, technically as
Insectivora) althougli recently it has lieen found that they do, at times,
take some plant materials.

g^<J^. ^.-.

■■'^^^$^^^^^^..

Fig. 9. — Deep tunnel and lateral of mole showing how successive loads of earth are
forced up from below to form an irregular surface mound wathout exposing the
mouth of the lateral.

Moles are common in the northwestern humid coast belt of California
south to IMonterey Bay ; small numbers are found in the river bottoms
of the lowlands and in other places affording damp soils. From much
of the agricultural area of the state they are entirely absent. Two en-
tirely different kinds of workings are made by moles: surface (or sub-
surface) ''runs" and deep burrows, the first being more extensive. In
searching for food a mole goes along very close under the ground surface
and pushes up a low rounded ''ridge" of earth, leaving a tunnel or "run"
below, through which the animal may travel repeatedly at irregular
intervals. The deep burrows, like the main tunnels of pocket gophers, are
farther below the ground surface ; the earth from such excavations, in-
stead of being pushed out of an open lateral, is forced up from below,
there being always a central core of loose earth, so that the mound is
"volcanic" in surface appearance (fig. 9).

Nature of Damage. — The runs and surface mounds disfigure lawns,
golf greens, and flower beds, and the making of runs may loosen or up-
root small plants. In addition, moles sometimes cut the roots of plants

1933

Control op Injurious Rodents

33

and are now known to feed on sprouting seeds and on l)ull)s to a limited
extent.

Methods of Control. — On the average, moles are more difficidt to con-
trol than pocket gophers. Their extensive runs and tunnels generally
preclude the use of gas, they do not always take poisoned baits placed
for them, and trapping is slower.

1. Trapping: For individuals or small numbers of moles, the per-
sistent use of traps is recommended. Even the ordinary Macabee trap
will sometimes catch a mole if set in the lateral tunnel ; the modified

Fig. 10. — Mole traps and wooden traps for rats and mice.

Macabee (fig. 56) is even more useful. Several special types are also
available. "Choker loop" traps are used successfully against moles in the
Pacific Northwest, in the Middle West, and in Europe, but these seem
not to have gotten into the California trade to any extent. These traps
have two loops of wire or metal which are pressed down to encircle a
run and set off by a pan type of trigger over the top of the run. Two
traps commonly offered by the hardware trade in California are the
Out-o'-sight and Reddick traps (fig. 10). These are constructed so as to
be set astride of a surface runway ; each is equipped with a trigger "pan"
that is to be pressed down on the top of the run. When the mole comes
along, the upward pressure of its body releases the trap. The first-named
trap is a "scissor-jaw" type, while the second has a series of downward
directed spears. The spring of the first causes the two pairs of jaws to
clamp the animal firmly and fatally ; the spears of the second are driven
downw^ard through the earth and into the mole's body. The choker loop
is generally reported to be the most effective ; the spear type, least so.
The latter has the disadvantage of puncturing the mole's skin, thus re-
ducing its value if it is to be saved as fur.

34 California Agricultural Extension Service [Cir. 79

To deteriiiiiie, before setting a trap, wliicli rims are in nse press down
the soil here and there on several rnns ; if a run is in nse the mole, in
passing, will raise it again. Thereupon, press the soil down lightly once
more and set a trap, pushing it down enough so that the trigger pan
rests firmly against the earth over the run. Upon the next round the
mole's body will force the trigger upward and release the trap.

2. Repellents and gassing: Scheffer^'* states that lye, paradichloro-
benzene ('TDB") , or naphthalene, introduced into mole runways, a tea-
spoonful every 10 or 15 feet, is sometimes helpful in repelling moles.
Dixon reports that 2 ounces of carbon bisulfide poured into a deep run-
way will sometimes kill the mole tenant. The latter chemical is, however,
also likely to injure nearby plants. The Karbo-Killer may be used for
deep tunnels. Exhaust gas from an automobile engine has also been
tried in the Middle West, where moles are more abundant and cause
greater economic concern.

3. Poison baits : In England, poison baits of red squill mixed with
cereal and either meat or earthworms have been used experimentally on
areas w^here moles were particularly numerous and troublesome. The
results were satisfactory.

INTRODUCED RATS AND MICE

Three kinds of rats and one mouse introduced into California have
considerable economic importance : the Norway rat, the black rat, the
roof rat, and the house mouse. The Norway rat {Rattus norvegicus), is
a large species, measures about 16 inches in total length, the tail being
about 7% inches ; adult individuals weigh from one pound to as much as
1^2 pounds. The nose is blunt, the ears are moderately small and slightly
haired, and the tail is blunt and shorter than the head and body, so that
when bent forward it does not reach to the nose. This rat lives chiefly
at the ground level, seldom going above the first floor of buildings. It is
generally distributed along the sea coast and in the interior lowland
areas of the state, about residences, farm buildings, warehouses, and
slaughter houses, on garbage dumps, and in sewers. The black rat
(Rattiis rattus rattus) and the roof rat (Rattus rattus alexandrinus)
are smaller and more slender, about 15 inches in head and body length,
the tail being 8^/2 to 10 inches long. The weight is seldom over % pound.
The nose is sharply slender, the ears are relatively large and naked, and
the slender tail when bent forward reaches to the tip of the nose or be-
yond. Both of these rats climb readily and may occupy the upper stories

14 Scheffer, T. II. American moles as agricultural pests and as fur producers. U. S.
Dept. Agr. Farmers' Bui. 1247:1-20. 1930.

1933 J Control of Injurious Rodents 35

of buildings, and they can travel overhead on heavy electric wires. The
black rat occurs in sea ports and adjacent towns the entire length of
California, while the roof rat lives in all these places and in the interior
of the state, not only in the Sacramento and San Joaquin valleys but
also up into the Sierra Nevada to as high as 5,000 feet altitude. The Nor-
way and roof rats are grayish brown in color ; the black rat is almost
solidly black. The house mouse (Mus musculus) is small, 3 to 4 inches
in head and body length with a tail 3 to 3^/2 inches long. On the upper
surface it is nearly a uniform brown. The ears and eyes are small, the
tail is scantily haired. The house mouse occurs about human habitations,
barns, warehouses, and office buildings, and also lives entirely apart
from man in some of the interior lowland districts. It is probably present
in all counties of the state.

All these alien rodents have been in California at least since the 1850's.
They have been distributed by commerce, the house mouse especially
having often been carried in shipments of household goods, in bales of
hay, and in packages of food materials. The aggregate population of
these alien intruders is enormous. They cause serious financial losses to
mankind, not only by the destruction of foodstuffs, but by damage to
buildings, packages of food, furniture, paper records, and even elec-
trical wiring. The alien rats were responsible for the introduction of
bubonic plague and have been concerned in at least three epidemics of
this disease in California cities. All are unpleasantly abundant, the mild
climate of California favoring their survival both indoors and outside.
In 1926-1927, house mice in the Buena Vista Lake region increased so
in numbers as to constitute a "mouse plague."

Breeding. — The Norway rat has an average of between 8 and 9 young
per litter and may rear 4 or more broods in a single year. The young are
able to shift for themselves when only about three weeks of age and
may breed when less than four months old. The black and roof rats
average over 6 young per litter and are otherwise about as rapid in
maturing as the Norway rat. House mice average 5 to 6 young per
litter, the young can run about within three weeks after l)irth, and can
breed when only 42 days old. Captive females have been known to pro-
duce 100 young in a year. Small numbers of any of these rodents may,
therefore, increase to very large populations in a relatively short time.
In an experiment with captive Norway rats, more than 1,500 were pro-
duced in one year from a single pair.

Nature of Damage. — Rats and mice feed upon practically anything
used for food by human beings and domestic animals. They may, there-
fore, damage or destroy almost any kind of foodstuffs. Even more im-

36 California Agricultural Extension Service l^^^- 79

port ant than the actual amounts of food eaten is the damage to sacks,
boxes, and other containers. In most warehouses constant campaigns
against rats must ])e carried on, and despite these, much re-sacking of
grain is necessary, involving expenditures for new sacks and for labor.
Cardboard i)ackages of cereals and numerous other food supplies in
grocery stores are torn open, fouled, and otherwise rendered unsalable
by the work of rats and mice. On poultry farms there is, in addition to the
loss of foodstuffs and the cutting of sacks, some destruction of eggs and
killing of young birds. Both rats and mice gnaw wood, plaster, and any
material except strong metal in their efforts to get in or out of buildings
and to gain access to food supplies, causing unsightly damage and facili-
tating the entrance of others of their kind. It is generally believed that
rodents sometimes cause fires by gnawing at the insulation on electric
wires and causing short circuits. Telejihone service has been interrupted
by the work of house mice. In addition to spreading bubonic plague, the
alien rats carry other diseases, some transmissible to man, and these are
becoming increasingly important in the United States.

Methods of Control. — Rats and mice have caused damage and distress
throughout the recorded history of mankind. Every means that human
ingenuity could devise has been used to combat these adaptable, hardy,
and prolific pests, and from all of these a few methods have come to be
recognized as of major importance: (1) exclusion, (2) trapping, (3)
gassing, and (4) poisoning.

1. Exclusion: This is the only proper solution to the rat and mouse
problem. All other methods give only temporary relief. Exclusion in-
volves construction and maintenance of buildings so that none of these
rodents can enter (figs. 11-13). It is practicable and not excessively ex-
pensive, but must be followed by proper maintenance and proper use of
the structures. It involves not only the closing of all openings to a mini-
mum aperture of % inch (since house mice can pass through an open-
ing only 1/2 inch wide), but also protecting all food supplies against
access by these rodents. In brief, exclusion calls for concrete floors,
foundations carried well down into the ground, screening of all ven-
tilators with hardware cloth or metal grilles with small apertures, ce-
menting in or sheathing with metal around all pipes or electrical con-
duits which enter buildings, and provision for tight fitting of doors and
windows. Once such rodent-proof construction is provided, continual
care must be exercised that doors and windows are closed when not
actually in use. Household supplies of food should be kept in covered
containers of metal, crockery, or glass. Small stocks of animal feed
should be put into clean galvanized-iron garbage cans; larger stocks

1933]

Control of Injurious Rodents

37

38 California Agricultural Extension Service [Cir. 79

should be housed iu rooms with tightly sheathed walls, or, best of all, iu
bulk containers such as metal "elevators."

Concrete floors aid in preventing entry of rats and mice into build-
ings ; unless a foundation wall extends down into the ground all around
the margin of the floor, rats will often burrow underneath and establish
nests there. Such a wall should reach about 18 inches below the ground
surface but may be thin — 3 or 4 inches in thickness — and of relatively
low cement content ; it can often be provided for by using a trenching
spade. For aboveground construction solid concrete or masonry is
the most permanent rat-proofing known; although initially expensive,
it requires practically no upkeep. For all out-buildings and farm
structures of frame construction on wooden studs, single walls are
preferable. Double walls, especially those sheathed on both sides with
wood, often provide ideal retreats for rats and mice and also passage-
ways by which these rodents may gain access to upper floors (fig. 11).
Old buildings with double wooden walls may often be made rodent-
proof by tacking pieces of galvanized sheet iron over the base of the
outer wall, removing the lowermost boards on the inside, and putting in
one or another of the stops illustrated in figure 11 (I-M) . Windows and
ventilators close to the ground should be screened with "hardware
cloth" of % or %-inch mesh securely nailed on all sides (fig. 11, E).
If such openings must be screened against insects, it is safer to use both
hardware cloth and fly screen, placing the heavier mesh outside. Above-
ground windows and doors are adequately protected by fly screen if the
frames fit tightly in the openings and there are no breaks. Where pipes
or conduits pass through walls of buildings, mechanics are prone to
leave larger openings than necessary (fig. 11, B, F) . Such spaces consti-
tute the commonest paths of entry for rats and mice. Three methods of
treatment may be used to close such openings : (1) in brick, tile, or con-
crete walls the space around each pipe is filled solidly with concrete
(fig. 11, D) ; (2) on wooden walls a sheet-metal collar is fitted around
each pipe, extending well beyond the edges of the opening and tacked
or nailed to the wall (fig. 11, G) ; (3) entrances to steam tunnels, and
other places where ventilation is necessary and workmen must enter
from time to time, should be fitted with close-fitting but removable
frames covered with hardware cloth of y^ or ^-inch mesh and bolted or
screwed into place.

If food materials are stored in such ways as to prevent mice and rats
from getting at them, a major step in control will have been accom-
plished. Not all materials can be thus protected. Hay, either in bulk or in

1933]

Control of Injurious Rodents

39

STOP
SHEET METAL

i <; FLO<>RlNG ^

Fig. 12. — Sectional view through a door
showing use of sheet-metal around edges of
door (sides as well as top and bottom), and on
the stop and sill to prevent entrance of rats
and mice. Maximum clearance, % inch. A sill
abutting against the door is preferable to a sill
beneath the door.

bales, is almost impossible to protect against mice, but practically all
other food, materials can be protected if sufficient pains are taken.

About the home, large covered tin cans and glass bottles, crocks of
2 to 10-gallon capacity, and even small clean garbage cans for nuts, flour,
cereals, dried fruit, etc., will protect against rodents besides affording
clean containers from which the contents can be removed without incon-

40

California Agricultural Extension Service

[Cm. 79

venience or spilling. Incidentally, glass jars and cans with close-fitting
tops will prevent moths and other insects from attacking foods so stored.
Small stocks of poultry feed can be kept in large garbage cans, whence
they may be easily removed ; if the covers are kept in place at other times
there will be little attraction for rodents. The larger commercially

Fig. 13. — Construction for corn crib to exclude rats. The
ends of studs and joists are nailed together to resist lateral
pressure when the crib is filled. Spaces between studs are
closed by wooden stops, at floor level. Hardware cloth is
placed between joists and flooring and between studs and
outside slats. The whole crib is elevated on concrete, field
stone, or wooden sills so that the floor is at least 12 inches
aboveground.

available cans will hold about two sacks of feed. If the amount of feed
given to poultry at each feeding is limited so that the birds will clean
it up rapidly, little or none will remain to attract rats and mice.

If the necessary supplies of animal feeds are large, economy of time
as well as good practice demands that they be stored in bulk containers
on the principle of the grain elevator. On the market are fairly cheap
cylindrical tanks of galvanized sheet iron with metal covers and with
metal spouts at the bottom ; these are made in several sizes up to about
two tons' capacity. Storage in such containers, besides excluding
rodents, lielps to reduce attacks by insects on the grain. Still another
alternative is to construct l)ins of tongue-and-groove boards, with tight-
fitting lids. As a further protection against rodents these may be
sheathed on the inside with galvanized sheet iron or on the outside with
hardware cloth. They should be framed on the outside so that the

^^33 J Control of Injurious Kodents 41

interior is of smooth wood. The supporting legs should be sheathed with
sheet-metal at some distance above the floor so that rodents will not be
able to climb up the supports.

Doors to granaries, feed rooms, grocery stores, and feed stores should
either be covered entirely with metal or provided with metal sheathing
around the edges, and should fit against a metal-sheathed stop (fig. 12) .

Corn cribs constitute a special attraction to rodents. The ordinary
wooden crib being placed on the ground provides harbor for rats and
mice beneath the floor while the open construction aboveground gives
ample opportunity for them to obtain all the food they want. Metal corn
cribs now available in manufactured form, find considerable use in
eastern states. A rancher who cannot provide an all-metal structure
may markedly reduce his losses by raising the crib 12 to 18 inches above
the ground (using wooden sills, concrete blocks, or field stone for sup-
ports), and covering the aboveground part of the structure (including
the door) with hardware cloth in addition to the usual wooden slats
(fig. 13) . New cribs should be constructed with the wire cloth.

Corrugated sheet iron as commonly applied to buildings does not
provide complete protection, for mice and young roof rats can pass
between the corrugations and the wooden frame. Save for failure to
close such small apertures, however, this material is ideal for many
kinds of structures. If the construction is altered slightly so that these
small openings are stopped up, complete exclusion may be had.

Garbage and slaughter-house offal serve as food for rats in many
places. All such refuse should, therefore, be put into metal containers
until removed for final disposal. Although most communities now require
their citizens to use metal cans for garbage, the carelessness of many
housewives in not keeping the cans covered often nullifies the efforts
of city officials to reduce rodent infestation. Even on the farm a garbage
can will help keep down both rats and house flies. Throwing all kitchen
scrap into a chicken or hog pen gets rid of much of the material, but too
often such pens are never cleaned of accumulated refuse and so
attract rats.

Irregular piles of empty boxes and of lum])er afford rat harbor (pro-
vided there is a food supply adjacent), but the amount of such harbor
can be markedly reduced if the material be piled in an orderly manner
and the whole pile elevated a foot or more aboveground on trestles or
other suitable supports.

Since piles of trash, accumulations of brush, old prunings from fruit
trees, bundles of cornstalks, and all similar materials afford rat harbor,
they should be promptly burned. Often they may be completely cleaned

42 California Agricultural Extension Service [Cir- 79

up if some old crankcase oil or diesel oil is used to start the pile to
burning well.

In brief, then, exclusion is the most important and practical method
of dealing with the rat and mouse situation. It involves the actual struc-
ture of buildings, keeping them in repair, seeing that all doors, screens,
and the like are properly closed, and storing all kinds of food in such
a way that the animals cannot get at it. The most effective means to
control rats and mice is to build them out and starve them out.

The farmer or houseowner who is bothered with rats and mice should
look over his property in the light of these suggestions and determine
wherein his buildings and food supplies favor the presence of rodents.
Often by using materials already on hand he can do much to starve out
and exclude these animals.

The remaining methods described here have to do with destruction
of rats and mice already present on a property. As whole books have
been written on detailed means for control of these rodents, only the
most important methods can be mentioned here.

2. Trapping : Many different kinds of rat and mouse traps have been
devised and manufactured, from the simple wooden spring traps (fig.
10) and guillotine traps, through cage traps, to elaborate maze-like de-
vices of automatic character. Long experience has shown that the ordi-
nary wooden, spring trap is as efficient as any. Almost any trap will
catch some rodents, but even the most elaborate devices will fail to cap-
ture certain "wise" rats. The efficiency of cage traps is far below that of
spring traps.

Few precautions are needed in setting traps for mice. They should be
placed in locations where droppings, tracks, or gnawings on food indi-
cate the presence of mice. Traps for either rats or mice should be placed
with the trigger end against the wall of a room or building. They may
also be placed on beams, on pipes, or in other places where the animals
are accustomed to travel. Contrary to popular belief, cheese is not a uni-
versal bait. Rolled oats, corn meal, bread scraps, hamburger steak, bacon,
fish, apples, raisins — in fact, almost any human food may be used suc-
cessfully. The more solid materials should be pressed or even tied to the
trigger ; when using rolled oats or meal a few particles scattered under
the trigger will make for results. If traps are placed where cats are
likely to prowl they should be fastened to some nearby object with a
fine wire.

Rats are often much more difficult to trap. Greater care should be ex-
ercised in setting and baiting. The trigger should be set so as to spring
with as light a touch as possible. In difficult cases the operator should

1933] Control of Injurious Rodents 43

wear clean cotton gloves when handling the trap. Success will sometimes
follow when a trap is placed and baited several days but left unset. Then,
after the rats have become accustomed to its presence, it should be baited
anew and set. Sometimes covering the trap lightly with sawdust or
similar material will aid in trapping a particularly wary rat.

Rats that ignore traps placed in the open will often be caught if a
board or box is placed so as to shield the trap, leaving space for the rats
to enter and for the trap spring to operate.

Daily renewal of baits, especially those which decompose, should be
practiced and needless to say trapped rats should be removed as soon
as possible. Traps should be examined at least once daily and preferably
more often.

Trapping is preferable to use of gas or poison baits for destruction of
mice and rats in dwellings or office buildings, since all rodents so killed
may be removed at once from the premises. With gas or poison the ani-
mals are apt to die in inaccessible locations, behind walls or beneath
floors, creating a long-persistent stench. This practically never happens
when traps are used.

3. Gassing : Rats may often be measurably reduced by poisonous gases.
Much of what is said regarding carbon bisulfide as used against ground
squirrels will apply for rats. If there are fires or likely to be sparks from
electric switches around dwellings or other buildings this explosive ma-
terial should be used only with great caution. Much successful rat con-
trol is obtained with cyanide products. Ships are now regularly fumi-
gated with compressed, liquid hydrocyanic acid gas. A small hand duster
employing calcium cyanide dust gives excellent results about farm build-
ings but is dangerous in dwellings. Attaching a piece of rubber tubing to
the end of the duster makes it easier to reach certain holes. As the bur-
rows are dusted, the openings should be filled with earth. The moisture
in the ground and in the air of the burrows combines with the calcium
cyanide to form hydrocyanic acid gas. Corn cribs may be temporarily
rid of rodents by use of such a dust pump. There is no danger in sub-
sequently using corn from these cribs for stock feed, since the poisonous
gas is generated rather quickly and the residue consists only of lime.
Cyanide should never be used in an inhabited dwelling or in a farm
building where animals are present. Finally, the exhaust of an auto-
mobile may be led down rat burrows through a hose. With a rich mix-
ture, an abundance of poisonous carbon monoxide will be forced into the
burrow by the pressure of the exhaust.

4. Poisoning : Many different materials have been used to poison rats.
Some of these are successful, some are not readily taken by rats, and

44 California Agricultural Extension Service [Cm. 79

some are also highly poisonous to other animals and to man. (3f all these,
red squill is the safest and most satisfactory, being either unattractive or
less poisonous to other animals. Barium carbonate may also be used,
but requires greater care.

Ked squilP^ is obtained from a bulbous plant (Urgmea niaritinia)
native to the north shore of the Mediterranean Sea. As prepared for rat
poison it is in the form either of powder or of fluid extract, the former
being more commonly available. Since the strength of different squill
preparations varies considerably, a dependable product must be ob-
tained ; white squill is not toxic to rats. Experimental tests have shown
that poison baits as prepared for rats are not poisonous to cats, dogs,
chickens, or pigeons ; these animals either will not eat the baits or will
vomit promptly any baits eaten.

For rat control the powdered red squill is mixed with any common
foodstuff (formula 6) such as cereal, fish, or hamburger steak. In the
interior drier portions of California some meal should be mixed with the
meat or fish so that the baits will not "gloss over" by early drying, and
become unattractive to the rats. Material so prepared is to be placed, a
half teaspoonf ul at a spot, where rats occur. Even though such baits are
reported not dangerous to domestic animals, they should not be care-
lessly placed where animals or children might find them.

Barium carbonate, a heavy white powder, has also been used against
rats and until the introduction of red squill was extensively recom-
mended. This material is poisonous to man and to domestic animals.
For use (formula 7) it is mixed with cereal, meat, chopped vegetable,
fruit, or scrap bread soaked in milk. The poisoned material should be
placed as for red squill. Any baits not taken should be gathered up and
buried, and if further poisoning is attempted a change to some other bait
material should be made.

5. Miscellaneous methods : Several special ''rat viruses" containing
cultures of bacteria intended to disseminate disease through a popula-
tion of rats have been exploited. Their use seems to be more common in
continental Europe than in the United States. Extensive efforts have
been made to encourage their use in this country and one or more com-
mercial enterprises have used such cultures for rat control in California.
However, unbiased studies by rodent control officials have shown that
results are less than claimed by the manufacturers. Studies by bac-
teriologists in this country have failed to substantiate the claims set
forth for these viruses as to their effect on rat populations. Finally,

15 Munch, J. C, James Silver, and E. E. Horn. Eed-squill powders as raticides.
U. S. Dept. Agr. Tech. Bui. 134:1-36. 1929.

1933] Control of Injurious Rodents 45

public health officials have felt that there was danger in the use of these
bacterial cultures because of the possibility of transmitting the disease
to human beings. Their use has therefore been discouraged by various
official agencies and even prohibited by the State Board of Public Health
in California for several years past.

The use of ferrets in rat control is frequently mentioned. In principle,
a trained ferret is turned into a burrow and either destroys or drives out
the rats; but practically, this method has not found general favor in
the United States and is more spectacular than efficient.

Control of rodents and especially of rats is the regular business of
various ''rodent exterminating" and "vermin control" companies. Some
of these are bona fide ; others know little of rodent control.

Special rat poisons are extensively advertised in many magazines.
Some contain only the materials described in the present circular — at
relatively high prices ; others contain poisons too dangerous for use by
the average householder or farmer, and some are absolutely useless since
they contain no poisonous material whatever. Persons who seek to con-
trol rats and mice about their dwellings or farms, therefore, will do well
to inquire critically into the merits of any special preparation before
purchasing.

MEADOW MICE

The term "field mice" is applied by many persons to several kinds of
mice found in grain fields, about hay stacks, and in orchards where cover-
crops are grown or where weeds and grass are not kept down. At least
four different animals are involved: the introduced house mouse, the
native harvest mouse, the white-footed or "deer" mouse, and the meadow
mouse or vole. Control of the house mouse is discussed in preceding
pages. Meadow mice (genus Microtus) are often abundant on farm
lands and at times may increase enormously in numbers; the "mouse
plague" at Buena Vista Lake in 1926-1927 included both house and
meadow mice.

All meadow mice are blunt-nosed, with small ears, the body is covered
with a dense soft fur, and the tail is relatively short and slender. The
head and body measure about 4 to 6 inches, the tail from 1% to 2%
inches. Meadow mice live on and in the ground ; most species construct
surface runways in the grass, by shearing off the vegetation to form
little pathways about an inch in breadth running here and there in
grassland; these connect with the numerous small burrows which the
mice make in the ground.

Damage by meadow mice may involve destruction of standing grain,
cutting of green vegetation including alfalfa, damage to hay in loose

46 California Agricultural Extension Service [Cir. 79

cocks or stacks, and gnawing the bark and roots of trees surrounded by
grass and weeds. The damage is irregular as to amount, season, and place
of occurrence. These mice, more than most native rodents, fluctuate
markedly in numbers. In the arctic regions there are periodic increases
and decreases in the populations of the voles and their allies. In Cali-
fornia, changes in the meadow mouse population have been noted, but
it is not known whether these are periodic.

Preventive treatment consists of clean cultivation. Orchards where
covercrops are grown should be watched for signs of damage by meadow
mice and steps taken at once to check the mice if necessary. Otherwise
clean cultivation in orchards, removal of grass and weed growths along
fences and about farm buildings and piles of lumber will reduce the
harbor available for these animals.

Meadow mice may be controlled either by traps or by poison. For small
numbers the best plan is to use mouse traps, baited with oat meal, rolled
oats, or bits of apple, carrot, or other root vegetables and set with the
triggers of the traps across the runways. The traps will then be effective
on mice running in either direction along the surface paths. Sometimes
unbaited traps will suffice. Traps should be visited at frequent intervals,
since these mice are active by day as well as at night and the efficiency
of individual traps is increased by frequent attention.

If the meadow mice are very abundant poisoned bait must be used
(formula 8). Great care must be exercised in placing any poison in
surface runs to be certain that such baits are not taken by children, wild
birds, or domestic animals.

WOOD RATS

The native wood rats (genus Neotoma), known commonly as ''trade
rats" or "pack rats," are sometimes a nuisance in cabins in the foothills
and mountains. The wood rats resemble the Norway rat in a general
way ; a conspicuous difference is that the body covering is much softer
in the wood rat. The brown-footed wood rat which inhabits many
foothill localities in California discloses its presence by building large
conical nests of sticks and forest litter, on the ground or in trees. At
times these animals gain access to cabins and build their nests indoors.
They may also be attracted to food supplies left by the human tenants
and may remove small bright objects such as spoons, forks, knives, or
small mirrors, sometimes depositing sticks or other items "in trade."
The directions given for making buildings and food supplies proof
against the alien rats will apply here. All bedding, mattresses, or cloth-
ing stored in isolated cabins should be put into a closet or room sheathed

1933] Control of Injurious Rodents 47

with sheet metal or "hardware cloth" ; otherwise wood rats (and at times
white-footed mice) may cut it while in search of nesting materials.
Wood rats being much less wary than the alien rats, are effectively
controlled by the ordinary spring rat traps.

KANGAROO RATS

In a few localities, especially where dry farming is practiced adjacent
to wild land of desert or semi-desert character, trouble is sometimes ex-
perienced from kangaroo rats (genus Dipodomys) . These distinctive
rodents are easily recognized by their long hind legs and feet, short small
fore feet, long tails, and the presence of a pair of external cheek pouches
(like those in pocket gophers) . They live in short shallow burrows, usu-
ally in sandy ground. Over much of interior California they are of
negligible importance, having been exterminated from areas long under
cultivation. When necessary, they may be easily controlled by use of
strychnine-coated barley (formula 1) .

MUSKRATS

Muskrats {Ondatra zihethica) were earlier present in California
only in a few scattered localities in Alpine and Lassen counties east of
the Sierra Nevada, and along the Colorado River on the boundary be-
tween California and Arizona. Their numbers, and the area occupied
by them, have been increased by man. The canal constructed to carry
water from the Colorado River to the Imperial Valley formed a natural
route for migration which they were quick to use. These large rodents
are now fairly abundant and cause serious damage to the canals and
ditches of the Imperial Valley. Their burrows in the banks sometimes
contribute to causing breaks with consequent loss of water.

Elsewhere in California muskrats have been transplanted by man into
the Kern River and adjacent canals about Bakersfield, into Guadalupe
and Oso Flaco lakes west of Santa Maria, and into the Shasta River
drainage in Siskiyou County. They are reported to have been trans-
planted into Little Lake, Inyo County. In Surprise Valley, Modoc
County, they have escaped from ''muskrat farms" in two places ; early
in 1933 a serious condition developed along Fall, Tule, and Pit rivers
in northeastern Shasta County by escape of muskrats from a ''fur
farin" near McArthur.

In the eastern United States, where irrigation is not practiced, musk-
rats cause no particular damage to agriculture ; but wherever irrigation
or drainage canals are needed these animals are likely to cause trouble.

48 California Agricultural Extension Service [Cm- 79

Muskrat farming is yet to be proved economically profitable, since mil-
lions of muskrats can be trapped cheaply on the natural marshes in
other parts of North America.

Where muskrats must be prevented from damaging irrigation struc-
tures, No. 1 steel traps should be set in burrows or — better — on the
characteristic "feeding mounds" in water 2 to 3 inches in depth, and
should be partially sunk in the bottom material. The trap chain should
be passed over a slender pole driven into the bank but leaning over the
water; the outer end of the pole should have a crosspiece at the top to
prevent the trap chain from being slipped off. A trapped muskrat will
usually dive into deep water, so that the chain slides along the pole and
the animal is quickly drowned. Some trappers use a bait of parsnip,
sweet potato, carrot, or other root vegetable, suspended on a small stick
so as to be about a foot over the trap site ; but many trappers use no bait
whatever. Other methods of capture have been described.^ ^

Muskrats on canals have been reduced by fur trappers working under
permits from the ditch owners. Such helpers though often eager to catch
muskrats for their fur value, will never remain long enough to reduce
these (or any other) animals to a low level, preferring to move to new
localities when the catch declines. Control for economic purposes is more
likely with paid and supervised trappers.

At Bakersfield recourse was had to use of carbon bisulfide pumped
into burrows at a season when water had been withdrawn from the
canals.

RABBITS

Rabbits, while technically not rodents, operate in similar ways. This
group has two main divisions: the "hares," represented by the jack
rabbits and the snowshoe rabbits, living entirely aboveground, making
no nests, and bringing forth their young fully covered with fur, with
their eyes open, and able to move about at once ; and the true rabbits,
including the cottontail and brush rabbits, dwelling in dense cover,
under stone piles and brush, or in burrows, the young of which remain
in the nest for a period of growth.

Black-tailed Jack Rabbits. — The common jack rabbits {Lepus calif or -
nicus) occupy the lowland, foothill, and desert portions of California,
are sparsely represented in the humid coast region, and are absent from
the higher mountains above the yellow-pine belt. They depend upon
speed and their ability in dodging to escape enemies. They live chiefly

16 Lantz, D. E. The muskrat as a fur bearer with notes on its use as food. U. S.
Dept. Agr. Farmers' Bui. 869:1-20. 1923.

1933] Control of Injurious Rodents 49

in open places, seldom inhabiting thickets or wooded country to any
extent. The average litter is about four, usually produced in the spring
months ; but young are in evidence over a considerable portion of the
year. Although jack rabbits make no nests, individuals often have a
more or less regular retreat or "form" beneath a bush where the animal
is somewhat sheltered from the full heat of mid-day sun, and yet can
watch for the approach of enemies. The food comprises a wide variety
of plant materials, both wild and cultivated, including grain, alfalfa,
and various truck and field crops, and at times the bark and tender
shoots of small orchard trees. Formerly jack rabbits were enormously
plentiful, especially in the San Joaquin Valley and on the flatter areas
of southern California. Many spectacular "rabbit drives" were held;
the whole human population of a district would turn out and surround
a territory several miles in extent, driving the rabbits toward a central
corral bordered by wing fences. After being concentrated in such an
enclosure, the rabbits were clubbed to death by the hundreds and even
thousands. Drives and other measures have reduced the population far
below the earlier numbers, yet there are still enough jack rabbits in some
places to do considerable damage to the more intensive agriculture of
the present day.

Snowshoe Babbits. — On the higher portions of the Sierra Nevada and
in the northeastern plateau region of California there are two kinds of
rabbits adapted to living on or in the snow during the winter months :
the large Sierran white-tailed jack rabbit {Lepus townsendi sierrae)
and the smaller snowshoe rabbit (Lepus washingtonii) ; both are "hares."
These, like the black-tailed jack rabbits, are surface dwellers. Their
feet being densely covered with long fur, they can travel on the snow.
The tail in the snow-inhabiting rabbits is always white, and the animals
themselves are white in winter. These rabbits live where there is little
intensive agriculture; they occasionally nibble the twigs and bark of
apple trees during the winter months.

Cottontail and Brush Babbits. — These true rabbits (genus Sylvila-
gus) occur over the lower altitudes in California. Cottontails are more
common in stream-side thickets and pastures ; brush rabbits on brushy
and chaparral-covered slopes in the hill country. Little is known about
the rate of reproduction or manner of caring for the young. The litters
are small, averaging about four, and are usually produced in the spring
months. Both of these rabbits will feed upon cultivated crops of garden
and field. They are classed as game, and, over most of California, may be
hunted only between November 15 and December 15. In the southern
part of the state (Districts 4, 4%, 19, and 21), because of their larger

50 California Agricultural Extension Service [Cir- 79

numbers and the greater possibilities of damage to agriculture, there is
no closed season. The California Game Code provides that owners, ten-
ants of land, or their agents may kill rabbits on their properties at any
time in any part of the state.

Methods of Control. — Rabbits are not difficult to control on small
areas but present greater difficulty on large ranches. The methods used
are: (1) exclusion, (2) shooting, (3) repellents, (4) trapping, (5)
poisoning, and (6) encouragement of natural enemies.

1. Exclusion : A fence of mesh not greater than 1% inches, buried
to a depth of 6 inches in the soil and carried 24 to 36 inches aboveground,
will exclude all rabbits, if it is patrolled at intervals to see that neither
rabbits nor squirrels have dug passages under the buried portion. Such
a fence, constructed of inexpensive poultry netting and combined with
barbed wire, will often serve as a stock fence for horses and cattle.
Sheep and hogs, however, are apt to damage a light wire fence, and hogs
especially may root at the base of such a fence and so destroy its effective-
ness. A barbed wire placed at the ground level will aid in the main-
tenance of a rabbit-proof fence. Neither jack rabbits nor cottontails will
ordinarily jump over a 24-inch fence ; but a jack rabbit when pursued
by a dog may do so, particularly if the fence is not surmounted by
closely spaced barbed wires. Such a jorotective fence must, of course,
be provided with tight-fitting gates and with sills or other means
for insuring that rabbits cannot dig below the bottom rails of gates. Its
effectiveness is lost if the gates are not kept closed except when vehicles
or persons are passing through. Small spring-closing gates may be used
in many places. This method of exclusion is best for small flower or
vegetable gardens, especially those adjoining large areas of pasturage,
grain, alfalfa, or wild land. Any high-priced crop such as a small field
of seed stock deserves the relatively inexpensive fencing here indicated.
If such a small plot of ground needs to be enclosed for but a few months,
light stakes may be used to support the wire, the bottom of which is
buried in a furrow ; then, after the crop has been harvested, the wire
and stakes may be removed to permit the use of cultivating machinery.
Around large acreages of farm land provision of rabbit-proof fencing
is ordinarily impractical. The ordinary rectangular mesh used as the
lower part of a stock fence, where sheep or hogs are being pastured, has
openings of a size that permits the entry of rabbits. Fencing for large
areas with wire of strength adequate to restrain livestock and of mesh
sufficiently small to exclude rabbits will ordinarily cost one dollar or
more per rod for the wire alone.

Over many years, farmers in California have used individual me-

1933] Control of Injurious Rodents 51

chanical protectors to guard the trunks of young orchard trees against
damage by rabbits. Earlier, thin slabs of yucca and other veneer mate-
rials were employed. Ordinary sacking has been tied around trees with
effective results. Poultry netting of 1-inch mesh, No. 20 gauge galvanized
wire, 18 to 24 inches in width, cut into strips 12 to 18 inches long and
formed into cylinders around young trees, is a common means of protec-
tion. Such cylinders should be braced so that rabbits cannot press them
against the trees and gnaw at the bark between the wire meshes. A fine
mesh wire or "hardware cloth" would also be efficacious against meadow
mice, especially if pressed down into the ground at the lower edge.

2. Shooting : Much effective control of rabbits can be had by shooting.
Rabbits are prone to feed in the early morning, late afternoon, and dusk
of evening, so that use of a gun is most effective at these times, but they
may feed also at night making this method not entirely successful.
Ranchers troubled by numbers of cottontails may obtain some relief by
inviting sportsmen to hunt over their lands during the open season. On
certain ranches spring hunting of jack rabbits is promoted, the hunter
thereby obtaining a measure of sport at a season when other game can-
not be shot, and the rancher some reduction of jack rabbits, although
such hunting will seldom clear a ranch of the animals. Reduction of
jack rabbits at this season is of especial benefit in limiting the number
of young which might otherwise be produced. At times jack rabbits have
been hunted for market use, and this aids in reducing their numbers.

3. Repellents : Various repellent substances have been used to prevent
rabbits from gnawing the bark and twigs of trees and vines. An ad-
hesive whitewash has some merit. Recently a foul-smelling sulfurized
oil (formula 9) has been developed which is to be painted on the trunks
of trees ; it is reported- to give excellent protection. A strychnine-
poisoned wash (formula 10) has been used in Idaho to protect young
orchards. Its use is contra-indicated where domestic animals are present.
Fresh blood, daubed on young trees, has found some use as a repellent.

4. Trapping : The jack rabbit drives of earlier years may be considered
to have constituted a type of trapping. In the Middle West cottontails^^
are sometimes captured in a special box trap known as the Wellhouse.
A permanent rabbit trap, constructed with a horizontal run of sewer
tile 6 inches in diameter and 4 feet long leading into a vertical 12x6 inch
tile T, with a heavy cover, has been used in Kansas. The entrance, sur-
rounded by stones and brush, gives a natural appearance. Such traps
may help in capturing rabbits in orchards.

17 Lantz, D. E. Cottontail rabbits in relation to trees and farm crops. U. S. Dept.
Agr. Farmers' Bui. 702:1-14. 1924.

52 California Agricultural Extension Service [Cir- 79

5. Poisoning : Use has been made of poison against jack rabbits in dis-
tricts where ranches are scattered and the human population is sparse.
Its use is not advised against cottontails and brush rabbits, since these
animals have value as game. Other means should be tried wherever prac-
ticable since poisoning rabbits is hazardous. The poisoned bait may be
any material relished by rabbits such as alfalfa leaves, grain heads, or
oats. Since rabbits, like many other animals, both wild and domestic, are
fond of salt, recourse has been had to poisoned salt. If any of these
methods are deemed necessary the person responsible for the poisoning
operations should make certain that all necessary precautions are taken
to prevent domestic animals, harmless wild animals, and human beings
from being endangered. On stock ranges where jack rabbits are so nu-
merous as to reduce the pasturage, substantial pens excluding livestock
but permitting jack rabbits to enter, have been used for the exposure of
poisoned materials. In the event that the owner or tenant of land is hav-
ing difficulty with rabbits which cannot be solved by methods indicated
above, it is recommended that he consult the agricultural commissioner
or farm advisor of his county as to the most appropriate methods to use.

6. Encouraging natural enemies : The red-tailed hawk and golden
eagle both feed upon rabbits, while the gopher snake is known to capture
small individuals. These and other natural enemies that subsist on rab-
bits and rodents are valuable aids to the farmer.

FORMULAS rOR PREPARING POISONED BAITS FOR RODENTS

General Precautions. — All of the substances used for poisoning ro-
dents are dangerous to human beings and to domestic animals. They
should, therefore, be handled with care. The- minimum lethal dosages
(of the elementary poisons) for a human being are approximately as
follows :

Strychnine 0.5 grain or 35 milligrams

Red squill 150.0 grains or 10 grams

Barium carbonate 150.0 grains or 10 grams

Carbon bisulfide 1 part in 1,000 parts of air

after 30 minutes' exposure

Thus, about 250 kernels of strychnine-coated barley, or 3 to 4 ounces of
red squill bait, or 2 ounces of barium carbonate bait would constitute a
lethal dose for a human being.

Antidotes for these poisons are uncertain ; egg white and milk should
be given and vomiting induced ; these may help if administered imme-
diately. In any instances of accidental poisoning of an adult or a child
a physician should be summoned at once. In the preparation of poisoned

1933] Control op Injurious Rodents 53

baits, as when dry strychnine is being dusted over bait material, care
should be exercised not to breathe the dust. All utensils and the opera-
tor's hands should be washed carefully after mixing baits. Utensils used
for this work should not be used for other purposes. Containers for
poison and poisoned baits should be kept locked up except when in use
and should always be out of the reach of children, irresponsible adults,
and livestock. Cans or sacks of poisoned bait should be clearly labeled
as POISON. Cans containing carbon bisulfide should be stored in a dry
cool place away from all fire, matches and sparks, as this material is
dangerously explosive.

Formula 1. — Strychnine-coated whole barley for California ground
squirrels (Government formula).

Barley (clean whole grain) 16 quarts

Strychnine (powdered alkaloid) 1 ounce

Bicarbonate of soda (baking soda) 1 ounce

Thin starch paste % pint

Heavy corn sirup ^4 pint

Glycerin : 1 tablespoonful

Saccharin i^q ounce

Mix the strychnine, baking soda, and saccharin together dry. Prepare
the starch paste by dissolving 1 heaping tablespoonful of dry gloss
starch in a little cold water, pour into % pint of hot water, boil and stir
until clear. Add the dry strychnine, soda, and saccharin, then the corn
sirup and glycerin; stir thoroughly. Pour the hot mixture over the
grain, turning and stirring until each kernel is coated. Spread the
coated grain out in a thin layer until the coating is thoroughly dried;
then store in a can or sack, properly labeled as POISON, until used.

Formula 2. — Strychnine-coated whole oats for Oregon ground
squirrels.

Use formula 1, but substitute 20 quarts of recleaned oats for the 16
quarts of barley.

Formulas. — Strychnine-coated dandelion "greens" for Oregon
ground squirrels.

Use formula 1, but substitute 5 gallons (bulk) of freshly cut green
dandelion leaves for the barley. Water cress or alfalfa leaves are less
satisfactory.

Formula 4. — Cut baits for pocket gophers.

Cut baits of sweet potatoes, parsnips, or carrots 4 quarts

Strychnine (powdered, either alkaloid or sulfate) % ounce

Saccharin .• YgQ ounce (5 grains)

54 California Agricultural Extension Service [Cir. "9

Cut the vegetables into pieces about % x 1/2 x II/2 inches, wash and
drain. Mix the strychnine and saccharin thoroughly in a pepperbox or
other sifting device. Sift this over the baits, meanwhile turning them
over and over, until all are evenly coated. Put the baits in a covered
container and use as soon as possible. Label, plainly, both the bucket
and the sifter : POISON.

Formula 5. — Poisoned root vegetables, raisins, or prunes for pocket
gophers.

Cut baits as directed in formula 4 ; with a slender knife-point insert
a few crystals of strychnine sulfate into the center of each bait. Prunes
that have previously been soaked for about two hours in water, or large
raisins, may be poisoned in the same way. The cut baits should be used
at once; the prunes if dried thoroughly after being poisoned, and the
raisins, will keep; poisoned prunes or raisins should be stored in a bottle
or can plainly marked POISON.

Formula 6. — Red squill baits for rats and mice. ^

Any cereal meal (oatmeal, corn meal, graham flour,

bran or peanut meal), ground meat, or minced fish. .10 parts
Red squill, powdered, dry 1 part (by weight)

Equivalent is one level tablespoonful of dry squill to one measuring-
cupful of meal ; mix.

Fresh fruit or vegetables cut in thin slices may be dusted with red
squill from a pepper sifter.

Liquid red squill may be used on dry meal or bread ; cut % pound of
stale bread into %-inch cubes ; pour one pint of liquid red squill over it
and mix thoroughly.

Formula 7. — Barium carbonate baits for rats and mice.

Cereal (corn meal, rolled oats), or bread; ground

meat or fish; finely chopped vegetables or fruits 4 parts

Barium carbonate, dry, powdered 1 part

For drier materials mix first with the barium carbonate, then moisten ;
"sloppy" baits may be more effective in warm weather ; bread baits may
be moistened with milk ; fresh fruits and vegetables should have the dry
barium carbonate sifted over them, and the poison well worked in. Baits
may be wrapped in small pieces of newspaper with the corners twisted
to make a closed container.

Formula 8. — Alfalfa leaves with strychnine for meadow mice.

Alfalfa leaves, dry 8 pounds

Strychnine alkaloid, powdered 14 ounce

Saccharin : 1^2 ounce (13 grains)

1933] Control of Injurious Rodents 55

Dissolve the strychnine and saccharin in 2 quarts of water ; sprinkle
over the leaves, turning the latter until all are moistened ; distribute the
poisoned leaves in burrows or in small piles in runways.

Formula 9. — Sulf urized oil to protect young trees against rabbits.

Eaw linseed oil 1 gallon

Sulfur, powdered 12 ounces

Heat the oil, out-of-doors, in a 5-gallon can, to 470° F (243° C).
Measure the temperature with a thermometer ; the oil gives off bluisli
smoke at tlie temperature indicated. When the proper temperature is
readied, remove the oil from the fire and ])egin adding the sulfur, a
teaspoonful at a time, stirring constantly. The oil will become hotter and
will foam a great deal. When all the sulfur has been added a black ill-
smelling compound will result. Allow the oil to cool completely before
it is used.

Clear grass and weeds from around the trees and paint the trunks
with the oil, taking pains to cover knots and injured spots. One gallon
will treat about 200 trees. The oil should be applied when the bark is
dry. Such a protective coating will last through most of one winter. If
applied in late winter a few early starting leaves may be stunted, but no
other adverse results have been observed.

Formula 10. — Poison wash to protect young trees against rabbits.

Strychnine sulfate 1 ounce

Laundry starch 8 ounces

Glycerin 6 ounces

Water 3% quarts

Prepare the laundry starch by mixing cold and then boiling in one
pint of water ; dissolve the strychnine in the remaining water by boiling ;
add the paste and glycerine. Cool and paint on trunks of trees. Do not
admit domestic animals to orchards where this poison wash has been
applied.

20m-ll.'3^