UNIVERSITY OF CALIFORNIA

COLLEGE OF AGRICULTURE

AGRICULTURAL EXPERIMENT STATION

BERKELEY, CALIFORNIA

THE SAN JOAQUIN
EXPERIMENTAL RANGE

C. B. HUTCHISON and E. I. KOTOK

CATTLE RANGE IN THE GRANITE AREA

BULLETIN 663

April, 1942

UNIVERSITY OF CALIFORNIA
BERKELEY, CALIFORNIA

4

CONTENTS

PAGE
Introduction 3

The Experimental Area, by M. W. Talbot, J. W. Nelson, and R. E. Storie 7

The Forage Crop and Its Management, by M. W. Talbot and H. H. Biswell .... 13

Experimental Herd Management, by K. A. Wagnon, H. R. Guilbert, and
G. H. Hart 50

Ranch Organization and Management in the Granite Area, by E. C. Voorhies,
L. A. Crawford, R. L. Adams, and G. A. Carpenter 83

Interrelations of Rodents and Other Wildlife of the Range, by E. E. Horn and
H. S. Fitch 96

Studies on Valley Quail, by T. I. Storer, F. P. Cronemiller, E. E. Horn, and
Ben Glading 130

Other Studies and Experiments in the Program of the San Joaquin Experimental
Range, by M. W. Talbot, H. H. Biswell, P. B. Rowe, and A. W. Sampson . . 136

Contributors and General Acknowledgments 143

Literature Cited 145

THE SAN JOAQUIN EXPERIMENTAL RANGE1

C. B. HUTCHISON2 and E. I. KOTOK3

INTRODUCTION

The total land area of California comprises close to 100,000,000 acres.
After excluding crop lands, dense forests, brush lands, desert, inac-
cessible areas and parts from which grazing is excluded, there still re-
main upwards of 60,000,000 acres of grazing lands.

The actual grazing value of individual acres is low but there are many
of them. Collectively they are the supporting base for California's second
largest agricultural industry — livestock production. This production of
meat, milk, and wool is the only method of harvesting the forage wealth
produced by these acres from year to year. Grass, using the term in a
collective sense, is the most valuable single crop produced in the state.

Under the multiple-use concept these uncultivated lands of California
have value for watershed development, timber production, livestock
grazing, and recreation and wildlife. In different areas one of these
uses will become paramount and the others fall in different places of
relative importance.

The beef cattle population of California numbers over 1,000,000 head
and there are about 3,000,000 sheep. In addition, about 750,000 feeder
and slaughter cattle and over 1,000,000 sheep are shipped into the state
annually to meet consumption requirements.

Beef production is important in all areas and of major importance
to the agriculture of at least two thirds of the counties. The foothills of
the Sierra Nevada are of great importance for grazing and five of the
ten leading counties in number of beef cattle are in this area, namely
Tulare, Kern, Merced, Fresno, and Madera.

There are great variations in forage production from one part of the
state to another, as influenced by soil types, rainfall, and elevation.
Important as these factors are, however, climatic variability for any
one given area markedly affects the quality, time of growth, and quantity
of vegetative cover from season to season. The San Joaquin Experimental

1 Received for publication June 3, 1941.

2 Dean of the College of Agriculture and Director of the Agricultural Experiment
Station.

8 Director of the California Forest and Range Experiment Station, maintained by
the Forest Service, United States Department of Agriculture, in Berkeley, California,
in cooperation with the University of California ; transferred, January 10, 1941.

[3]

4 University of California — Experiment Station

Range (for brevity hereinafter referred to as "the Range"), located in
the granitic area near the center of the state, exemplifies range lands of
rather medium value and reflects many of the unsolved problems deal-
ing with seasonal variabilities and nutritive deficiencies.

In the Sierra Nevada foothills the vegetation is largely composed of
annual forage plants. These sprout from seed in the fall following the
first rains, make a varied rate of growth according to the weather condi-
tions, and reach maturity in March or April, with shattering of their
seed. They dry up in May or June and become bleached by the sun and
leached by occasional summer rains. This dry material constitutes the
only ground cover and gives the brown appearance of the range until
the fall rains again germinate the seed and the green mantle appears.

Throughout the year these lands are grazed by varying numbers of
livestock and support many species of wildlife. The nutritional value
of the plant cover to animal life depends on the stage of growth and this
in turn on the season of the year. In the early vegetative stages the dry
matter of the green plants is uniformly high in minerals and protein with
the total weight composed largely of water. The vegetation as grazed,
therefore, is in the nature of a watered concentrate feed. It gradually
changes from this condition as the plants mature, dry, and their seeds
shatter ; late in the dry season it becomes a very poor roughage — low in
protein, phosphorus, and vitamin A, high in fiber, and with low palata-
bility.

On account of the quantity and quality of this feed, part of the live-
stock are moved to higher mountain ranges in the national forests to
the east under Forest Service grazing permits during the summer, and
to valley salt grass, stubble, or irrigated pasture lands in the late summer
and fall.

The problem which these lands present is how best to utilize their
feed resources and still maintain the vegetative cover of desirable plant
species as well as to improve its quality or quantity. From the livestock
standpoint, therefore, it has been necessary to study the grazing man-
agement of animals in order to utilize best the feed on the ground by
the addition of supplements to meet its nutritional deficiencies at certain
seasons of the year.

With the establishment of the San Joaquin Experimental Range by
the U. S. Forest Service in 1934, the University of California entered
into cooperation with the Forest Service on several projects.

The University Agricultural Experiment Station placed the experi-
mental herd of cattle on the Range, under the supervision of the Division
of Animal Husbandry, to study how best to maintain on a year-round
basis herds of commercial cattle on these lands.

Bul. 663] The San Joaquin Experimental Range 5

Studies on forage management, on the effect of the domestic animal
life on the plant cover, the botany of the range, and the possibility of
introducing new species of greater value have been carried out by the
Forest Service. Responsibility for administration of the area, and for
coordination of all studies thereon, is also vested in the Forest Service.

The impact of wild animal life on the plant cover has become recog-
nized as an important factor and one regarding which little attention
has been paid and on which data are limited. Studies of these matters
have been carried out by the Division of Wildlife Research, Fish and
Wildlife Service, United States Department of the Interior.

The economic aspects of the livestock industry, ranch organization
and management, land values, proper-sized units and general practice
in the entire granite area surrounding the experimental area will have
a bearing on conclusions to be drawn after biological data on plant and
animal life have been accumulated. Information toward this end has
been gathered by the Giannini Foundation of Agricultural Economics
of the University of California.

Game animals and birds are coming to be recognized as an asset to
open patented lands through game management and controlled hunting
programs. Valley quail are one of the most important upland species in
this regard. Studies of these game birds on the Range lands have been
made by the Division of Zoology of the University Agricultural Ex-
periment Station in cooperation with the Forest Service, the Fish and
Wildlife Service of the Department of the Interior, and the California
Division of Fish and Game.

Several other units of the University have actively participated in
the program. For example, the Department of Forestry has made exten-
sive chemical studies of the principal forage species, the Division of
Soils has covered the Range area with a detailed soil survey, and the
Division of Agronomy is cooperating in the revegetation work. The Divi-
sion of Agricultural Extension organized the important advisory com-
mittee and cooperates in many other ways. The cooperating agencies,
with their personnel, are enumerated in more detail in acknowledgments
appearing in another section of this bulletin.

It can thus be seen that on the San Joaquin Experimental Range,
conceived and developed by the California Forest and Range Experi-
ment Station, a comprehensive group consideration and attack on the
problems of the range lands of California have been developed. The
University has encouraged the cooperation of its specialists wherever
their specialized knowledge was needed to direct and assist in accumula-
tion of pertinent data. The data presented in the seven sections of this
report by a total of eighteen authors constitute the information obtained

6 University of California — Experiment Station

by the group attack on the range problems encountered from the begin-
ning of the studies in 1934 up to 1940. The purpose of this publication is
to picture the San Joaquin Experimental Range with the kinds of going
work and to report on progress and the findings to date. Such aspects
of the work as have already been reported upon appear in "Literature
Cited," at the end of the bulletin.

THE EXPERIMENTAL AREA
M. W. Talbot, J. W. Nelson, and E. E. Storie

The San Joaquin Experimental Range is located near the center of the
state in the heart of the granite-soil section of the Sierra Nevada foot-
hills in Madera County adjacent to the Sierra National Forest, 23 miles
northeast of Madera and 28 miles north of Fresno.

The area is rolling. Elevations range between 700 and 1,700 feet ; most
of the area lies between 1,000 and 1,500 feet. Exposures in general are
southwesterly. The drainage basin empties into a small tributary of the
San Joaquin River. Numerous springs occur but no permanent streams
flow through the experimental area although the many small drainage
lines carry surface flows during the rainy winter months.

Experimental Facilities. — To serve the purposes of the studies and for
convenience in handling livestock, the tract of approximately 4,600
acres is subdivided into numbered pastures of various sizes as indicated
in figure 1. Numbers 1 to 6 are grazing-capacity pastures devoted pri-
marily to studies of the effect of different intensities of grazing on both
range and cattle ; numbers 7W and 7E are for weaning calves, as bull
pastures, and for various other handling purposes; numbers 8 to 10
are for calf pastures; and 11 to 14 are for holding the experimental
herds during the periods that they are not in pastures 1 to 6. The natural
area around headquarters is not grazed ; a portion serves as a check of
vegetative growth under total exclusion of livestock. Saddle horses used
at the station are held in the horse pasture.

For the range studies the pastures are equipped with various fenced
plots to exclude rodents or livestock or both, with many permanently
marked observation spots and photography stations. A forage-plant
nursery is located at the headquarters ; reseeding test plots are in pas-
ture 8 ; and plots for measuring runoff and erosion are in number 7.
Rodent enclosures are located in pasture 1. A more detailed discussion
of these pastures, enclosures, and plots will be found elsewhere in this
report.

In all the pastures, springs and seeps have been developed and
equipped with troughs to furnish water for livestock. Additional cattle
facilities include two sets of corrals, holding pens, scales, dehorning
chutes, and a shed for storage of supplemental feeds.

Soils. — About 87 per cent of the land on this tract consists of residual
soils formed in place through the decomposition and disintegration of
the underlying coarse-grained granitic bedrock. As a general rule the
bedrock occurs at a depth varying from about 6 inches to 4% feet from

[7]

University of California — Experiment Station

= Pasture numbers

( ) = Pasture acreages -
approximate for pastures
7 to 14 inclusive

G- = Corrals and scales

Fig. 1. — Pasture diagram of the San Joaquin Experimental Range.
Some of the original pastures, designated by numbers, were later divided
into north and south, or east and west, sections. For instance, 9 N is the
north part of the original pasture 9.

the surface. Over much of the tract the soils are less than 2 feet in depth.
Characteristically they are of sandy-loam or stony-sandy-loam texture,
of neutral reaction, and of brown or grayish-brown color with low
organic-matter content. These soils occupy a rolling to steep surface.
Nearly all the residual soils are classed with the Vista series, there being
a small area of the related Fallbrook series — soils of redder color. Al-
though these residual soils are permeable to the penetration of rain, their
water-holding capacity is low owing to their coarse texture and the small
volume of soil overlying the bedrock.

Bul. 663] The San Joaquin Experimental Range

Fig. 2. — Soils of the San Joaquin Experimental Range. References to numbers
and areas are given in figure 1.

About 12 per cent of the tract has darker-colored alluvial soils oc-
curring in the swales and small valleys. These have been classed with the
Visalia series, and are deep and productive. The period when forage is
green and nutritious is longest on these soils ; owing to their position,
they retain moisture much longer than the surrounding hill soils.

Miscellaneous material classed as riverwash and placer diggings make
up about 1 per cent. These are of no agricultural importance.

10

University of California — Experiment Station

The occurrence of these various soils on the Range is shown in figure 2.

Over most of the area, with the exception of the swales, by early sum-
mer the moisture content of the upper soil is reduced below the minimum
amount from which plants can extract water for their needs. Below a
depth of 2 feet, however, some moisture is usually available throughout
the season.

Climate. — The climate is characterized in general by mild winter tem-
peratures and high summer temperatures and by a distinct winter rainy
season followed by a comparatively rainless summer period. Snow is
rare. From these broad features of climate and weather arise some of the

Season

1934-35
1935-36
1936-37
1937-36
1938-39

Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug.

0.00 2.22 3.39 3.64 5.76 3.21 4.85 6.33 0.00 0.00 0.00 0.00

0.00 1.63 0.52 1.56 3.67 11.20 2.47 1.35 0.15 0.04 0.06. 0.00

0.00 2.17 0.10 5.16 3.22 7.08 4.58 0.65 0.00 0.00 0.00 0.00

0.00 0.03 0.50 6.23 5.99 7.75 8.69 2.69 0.02 0.00 0.19 0.00

0.14 1.37 0.75 1.28 2.87 2.50 2.89 0.26 0.03 0.14 0.00 0.02

Total

29.40
22.65
22.96
32.09
12.25

Fig. 3. — Average monthly and annual rainfall at the San Joaquin
Experimental Eange since the station was established.

most difficult problems of range and its management. Seasonal precipi-
tation as treated here begins in September; and has averaged 23.83
inches (fig. 3) for the five-year period since the experimental tract was
established in 1934. This short-period average is considerably in excess
of a probable average of approximately 17 or 18 inches, as derived from
U. S. Weather Bureau records from older stations in the locality for a
period of twenty-five years. During the first four seasons, 1934-35 to
1937-38, precipitation averaged 55 per cent higher than the probable
average for the preceding twenty years. In contrast, precipitation dur-
ing the fifth season fell 30 per cent below the twenty-year average. Such
fluctuations in rainfall from season to season are not unusual. *

Distribution of precipitation during the five years, by seasons and by
months (fig. 3) shows that nearly 80 per cent of the rain fell during
the 4 months of December through March. Virtually none fell during the
5 summer months from May through September. Occasional storms

Bul. 663]

The San Joaquin Experimental Range

11

bring a rainfall of 1 to 3 inches in 24 hours, but a greater intensity is
seldom recorded. Most of the precipitation comes from comparatively
gentle rains.

The maximum, mean, and minimum air temperatures, averaged by
5-day periods for four seasons, 1935-36 to 1938-39 inclusive, are shown
in figure 4. Maximum temperatures during the dry summer season often
exceed 105° F for several days in succession and have reached 115°.
Minimum temperatures during this period are usually between 60° and
70°, and very seldom remain above 75°. Winter temperatures are mild
but frequently fail slightly below freezing. Temperatures have occasion-

Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug.

Fig. 4. — Air temperatures, averaged by five-day periods for four
seasons, 1935-36 to 1938-39 inclusive.

ally averaged around 25° F for several days in succession. The lowest
temperature recorded was 17° in January, 1937, but this was during one
of the coldest periods on record for the surrounding region.

Maximum temperatures at the soil surface are high during the sum-
mer months, occasionally reaching 150° F, with one record of 160°. The
highest average reading for a month at three stations was 147°.

Relative humidity is high during the winter, but low during the sum-
mer period, with readings on some days as low as 5 to 8 per cent.

In contrast to numerous other localities with semiarid forage types,
average air movement is comparatively slight and fairly constant
throughout the year. Strong and continued winds seldom occur.

During the early stage of forage development in winter, cold, drying
northerly winds frequently slow down growth; and in the spring the
curing of the forage is often hastened by drying winds. The risk of
losing forage from fire in the summer is greater during periods of in-
creased wind movement.

12 University of California — Experiment Station

Sun intensity is high during the summer ; cloudy days seldom occur
during May, June, July, August, or September. Fog is frequent during
the winter rainy period, when several foggy days may appear in suc-
cession.

Vegetation and Life Zones. — The experimental area lies in the lower
portion of the woodland zone between the treeless plains below and the
higher brushy and timbered belts. This section of the foothills is charac-
teristically open and everywhere accessible to livestock. Much of the
area has an orchardlike appearance, with scattered trees and bushes
and occasional dense clumps of shrubs. The low ground cover, consisting
chiefly of annual grasses and herbs instead of perennials that dominate
most western range areas, is made up of a large number of different
species. In the language of biologists, the area lies near the lower boun-
dary of the Upper Sonoran life zone, with its characteristic plant and
animal life as described in later sections of this bulletin.

THE FORAGE CROP AND ITS MANAGEMENT

M. W. Talbot and H. H. Biswell

Many of the range-management problems of the foothill section are
directly related to the character and behavior of the forage crop, a crop
consisting almost entirely of annual plants and often referred to as the
"annual type." This peculiar forage is made up of a large number of
plant species, both native and introduced. The various kinds have dif-
ferent growth habits and they fluctuate widely in volume and abundance
from year to year. They make some growth during the winter, reach
maturity during a short active growing period in the spring, and then
quickly dry. Tenacious seeding habits, nutritional deficiencies in sum-
mer, and different reactions to grazing and protection are additional
properties of the forage crop of this area, in which range problems are
wholly different from those on "perennial-type" ranges. A thorough
understanding of these problems can be obtained only through knowl-
edge of the character and behavior of this unique forage type around
which management practices must largely be based, for perennial herbs
are so sparse that they play only a negligible part in management prac-
tices in this area.

This section contains a description of the forage, a discussion of its
characteristic growth and behavior, and a report on experiments dealing
with the practical question of how it can be grazed to give maximum
yields and at the same time be maintained and, if possible, be improved.
It should be considered a progress report since the studies are in their
early stages and some of the trends or conclusions may be modified by
study over a longer period. This discussion, which stresses forage, is
closely related to the following sections which emphasize the animal.
Both are important in a clear understanding of the range problems of

this region.

PLANTS THAT MAKE UP THE FEED

Although three species of plants account for more than half of the
forage on the experimental area (table 1), the forage cover here and on
the adjoining foothills is composed of a large number of different kinds
of plants, many of which occur in small amounts and are seldom noticed.
Some 225 different species (other than trees and shrubs) have been
recorded on the area during the past five years and the list is still in-
complete. Of these, 190 are broad-leaved plants, 30 are grasses, and 5
are grasslike plants (rushes). During the period of study the forage
crop has been composed almost equally of broad-leaved herbs and grasses,
except in 1935 when the broad-leaved herbs made up about three fourths

[13]

14

University of California — Experiment Station

of the cover. Of the grass species present, only a few are important range
plants. The sparse grasslike plants have varied from less than 1 to almost
4 per cent. Perennial herbs have totaled less than 1 per cent during four
of the five years. The herbaceous plants that have represented 1 per cent

TABLE 1

Percentages of All Grasses and Herbs That Comprised as Much as One
Per Cent of the Cover at the Height of the Growing Season*

Species

Broad -leaf filaree

Soft chess

Foxtail fescue

Red brome

Tomcat and white-tip clovers

Yellow tarweed

Maiden clover

Slender oat

Spanish clover

Popcorn flower

Toad rush

Ground lupine

Ripgut grass

Birdseye gilia

Goldfields

Red-stem filaree

Lay ia

Linanthus

Rattlesnake weed

Tree clover

Fine-leaf lotus

Godetia

Poverty fescue

Others and traces

1935

1936

41.7
19.7
17.2
2.1
0.3
0.6
0.5
2.4
1.1
0.3
0.4
0.1
1.0
0.1
0.5
0.7
0.4
0.3
0.3
0 1
0.1
0.2
0.6
9.3

1937

33.5
31.7
9.3
3.0
0.5
0.2
0.4
4.0
0.5
2.4
0.5
0.2
3.5
0.1
0.1
0.1
0.1
Trace
Trace
0.1
0.1
0.1
Trace
9.6

1938

15.5
17.0
10.1
2.0
5.3
3.2
5.1
0.5
4.9
1.3
3.1
2.9
0.8
0 5
0.8
0.2
0.2
0.3
1.0
0.5
1.3
1.1
Trace
22.4

1939

28.1
26.7
12.9
2.7
0.7
1.2
1.8
0.8
1.8
2.0
0 0
2.6
0.7
0.6
0
0.
0
0.
0.
1.6
0.2
0.1
Trace
13.0

Average

30.6
19.3
12.6
2.2
2.2
2.1
1.9
1.8
1.8
1.7
1.4
1.3
1.2
1.0
1.0
0 9
0.7
0.6
0.5
0.5
0 4
0.4
0.4
13.5

* Based on inventories covering over 1,440 acres. The least number of plots used in any inventory was
1 for each 2 acres in 1938 and 1939. The data for the 1937 crop were based on clippings, and for the other
years, on estimates.

or more of the cover at the peak of growth are listed in table 1 and
illustrated in figures 5 to 7 inclusive.

Introduced, or alien, species now comprise a large percentage of the
cover, averaging over 60 per cent for the five years (fig. 9) . Several are
among our most valuable range plants and have added to the value of
the forage. Others are inferior and even detrimental. The kind of plant
cover existing throughout this region before the white man came can
never be determined precisely since there is no early literature that
adequately describes it (i)4 and virtually no remaining samples of it.

A few of the most important plants, those on which management
practices must largely be based, are discussed in more detail in this

4 Italic numbers in parentheses refer to "Literature Cited" at the end of this
bulletin.

Bul. 663]

The San Joaquin Experimental Range

15

section. These plants are important for various reasons : some are abun-
dant ; others are relished by cattle ; and still others reveal something of
the range condition.

The common and botanical names of the various species mentioned
in this bulletin are given in the following list :

Grasses and Grasslike Plants

Australian chess, Bromus arenarius La-
bill.

Bermuda grass, Cynodon Dactylon (L.)
Pers.

Common foxtail, Hordeum murinum L.

Common spike rush, Eleocharis palustris
R.&S.

Foxtail fescue, Festuca megalura Nutt.

Meadow barley, Hordeum nodosum L.

Melic grass, Melica calif ornica Scribn.
and M. imperfecta Trin.

Pine bluegrass, Poa scabrella (Thurb.)
Benth.

Poverty fescue, Festuca confusa Piper
Purple needlegrass, Stipa pulchra Hitchc.
Red brome, Bromus rub ens L.
Ripgut grass, Bromus rigidus Roth.
Slender-leaf rush, Juncus oxymeris En-

gelm.
Slender oat, Avena barbata Brot.
Smilo, Oryzopsis miliacea (L.) Benth &

Hook.
Soft chess, Bromus mollis L.
Toad rush, Juncus bufonius L.
Wimmera rye grass, Lolium rigidum var.

strictum (Presl.) Jansen

Broad-leaved Herbs

Alfalfa, Medicago spp.

Baby blue-eyes, Nemophila Menziesii H.

&A.
Birdseye gilia, Gilia tricolor Benth.
Blue dicks, Brodiaea capitata Benth.
Broad-leaf filaree, Erodium Botrys Ber-

tol.
Buckthorn weed, Amsinckia Douglasiana

DC.
Bur-clover, Medicago hispida Gaertn.
California buttercup, Ranunculus cali-

fornicus Benth.
California everlasting, Gnaphalium cali-

fornicum DC.
California poppy, Eschscholtzia calif or-
nica Cham.
Catchfly, Silene gallica L.
Common chickweed, Stellaria media (L.)

Cyr.
Creeping-rooted alfalfa, Medicago sp.
Dock, Bumex spp.
Eriogonum, Eriogonum spp.
Filaree, Erodium spp.
Fine-leaf lotus, Lotus subpinnatus Lag.
Garden burnet, Sanguisorba minor Scop.
Godetia, Godetia spp.

Goldfields, Baeria chrysostoma F. & M.
Ground lupine, Lupinus bicolor Lindl.
Hairy lotus, Lotus strigosus (Nutt.)

Greene
Hansen larkspur, Delphinium Hansenii

Greene
Layia, Layia pentachaeta Gray
Linanthus, Linanthus filipes (Benth.)

Greene
Locoweed, Astragalus didymocarpus H.

&A.
Loosestrife, Lythrum hyssopifolia L.
Maiden clover, Trifolium microcephalum

Pursh
Napa thistle, Centaurea melitensis L.
Navarretia, Navarretia pubescens

(Benth.) H. & A.
Phacelia, Phacelia spp.
Popcorn flower, Plagiobothrys nothoful-

vus Gray
Rattlesnake weed, Daucus pusillus Michx.
Red maids, Calandrinia Menziesii T.

&G.
Red-stem filaree, Erodium cicutarium

L'Her
Sage, Salvia spp.

16 University of California — Experiment Station

Broad-leaved Herbs — (Continued)
Spanish clover, Lotus americanus ( Nutt.) Turkey-mullein, Eremocarpus setigerus

Bisch. Benth.

Spurge, Euphorbia ocellata D. & H. White-stem filaree, Erodium moschatum

Subterranean clover, Trifolium subter- L'Her

raneum L. White-tip clover, Trifolium variegatum

Tomcat clover, Trifolium tridentatum Nutt.

Lindl. Wright's tarweed, Hemitonia Wrightii

Tree clover, Trifolium ciliolatum Benth. Gray

Yellow tarweed, Hemitonia virgata Gray

Shrubs and Trees

Blue elderberry, Sambucus glauca Nutt. Interior live oak, Quercus Wislizenii

Blue oak, Quercus Douglasii H. & A. A. DC.

California buckeye, Aesculus calif ornica Mariposa manzanita, Arctostaphylos

Nutt. mariposa Dudley

Coffeeberry, Ehamnus calif ornica Esch. Poison-oak, Rhus diversiloba T. & G.

subsp. cuspidata (Greene) Wolf Saltbushes, Atriplex canescens (Pursh)

Deerweed, Lotus scoparius (Nutt.) Ottley Nutt. and A. nummularia Lindl.

Digger pine, Pinus sabiniana Dougl. Wedgeleaf ceanothus, Ceanothus cunea-

Hollyleaf buckthorn, Ehamnus crocea tus (Hook.) Nutt.

Nutt. var. ilicifolia (Kell.) Greene

Species of Filaree. — Of the three species of filaree (known also as
alfileria, alfilaria, and alfilerilla) found on the Range, the broad-leaf
filaree is by far the most abundant (fig. 5, A). This species alone com-
prised from 15 to 42 per cent of the forage cover during the study, being
the most common of any forage plant (table 1) . Red-stem filaree (fig. 5,
B) is exceedingly sparse on the area, having made up only 0.1 to 3.0
per cent of the plant cover, while white-stem filaree (fig. 5, C) occurs
only in small quantities around headquarters.

The filarees, all annuals introduced in early days into California from
the Mediterranean region, have become widely spread and now consti-
tute an important part of the range feed. They are well distributed over
the range, but generally occur in greatest abundance on south exposures.
The seeds are among the first to germinate after the fall rains begin,
usually in October. The young plants may make appreciable growth
during November if the rainy season is early, but during 8 or 10 winter
weeks the plants often turn reddish in color and make practically no
growth except leafy rosettes, which may lie nearly flat on the ground.
They are shortest on areas grazed heaviest during the previous season.
The rosette tendency appears to be slightly less developed in the red-
stem and white-stem forms ; consequently these two species may furnish
more feed than that of the broad-leaf during the winter period.

With the advent of warmer weather, usually during the latter part of
February or early in March, a central stalk begins to develop and soon

Bul. 663] The San Joaquin Experimental Range

17

Fig. 5. — Eight important broad-leaved herbs occurring on the San Joaquin Ex-
perimental Eange: A, broad-leaf filaree; B, red-stem filaree; C, white-stem filaree;
D, Spanish clover; E, popcorn flower; F, maiden clover; G, tomcat clover; H, tree
clover.

attains an average height of 5 to 8 inches ; but in favorable sites it may
reach 18 or 24 inches. In 1935, some plants in favorable spots grew to a
height of 36 inches. Flowering and fruiting continue from the time
spring growth starts until the plants dry. On south exposures drying
usually begins around the middle of April, but on north exposures and

18 University of California — Experiment Station

in moist places a few plants may remain green until late in June. Because
of this long fruiting period a fairly good seed crop is likely to be pro-
duced if any growth takes place at all. In years when growth is scant
and mature plants are small, the fruits or "seeds" may account for about
a third or half of the height growth of the plants. Although adverse
growing conditions decrease the number of seeds produced, the size of
plant appears to be decreased even more. Seed development reaches its
peak during a period of 4 or 5 days of very rapid growth around the
first of April, when the seed heads increase in length by one-fourth inch
to almost an inch per day. This is the period when stockmen often re-
mark that one can "almost see the feed grow." This spurt in growth is
a kind of seed-crop insurance, for grazing is not likely to be close enough
during this short period to keep pace with seed production.

After the plants have dried, they are very brittle. Stands of filaree,
even if protected from grazing, actually decrease in forage density
through the summer period, from crumbling alone. Such plots, in 1939,
showed a decrease in total forage density from 59 to 48 per cent from
April through July. The dominant grass species on these plots showed
no decrease in density. On many pastures much more filaree herbage is
lost during the summer from trampling than by grazing.

Filaree plants are grazed readily throughout the growing season. They
provide considerable fall forage during favorable seasons, but during
the average winter, while in rosette form, they furnish very little feed.
Fruits in the dough stage are sought by cattle, and when plentiful con-
tribute at that time an important part of their diet. Some intensive
studies made by Klugh6 during the green-feed period in 1936 showed
that broad-leaf filaree was grazed most heavily at this time and com-
prised over 35 per cent of the feed.

During summer, when the plants are dry, the crumbled remnants of
broad-leaf filaree plants are seldom taken by cattle unless the forage
becomes exceedingly scarce. Red-stem filaree plants, on the other hand,
are not so coarse and stemmy, and portions that remain after crumbling
and trampling are taken more readily. Also, Gordon and Sampson (7)
found that protein content of the dry material of broad-leaf filaree is
not so high as it is in the red-stem species. Analyses showed a consider-
ably lower crude-protein content for awns than for the dry leafage and
stem material.

Percentage utilization of broad-leaf filaree was estimated at the end
of the grazing season in 1939 on 60 permanent plots distributed over 6
grazing-intensity pastures, to provide key areas to aid in comparing the

5R. H. Klugh, formerly Junior Eange Examiner, California Forest and Eange
Experiment Station.

Bul. 663] The San Joaquin Experimental Range 19

degree of grazing, year by year. The plant occurred on 57 of the 60 plots
and the measurements showed 86 per cent of this species to be either
grazed or trampled down. On only 3 of the plots was any stubble left
standing after grazing, and this averaged 2 inches in height. Eighteen
of the 57 plots were classified as grazed closely, 27 as moderately, and
12 as lightly grazed. The filaree on the closely used plots was 99 per cent
utilized or trampled flat on the ground; that on the moderately used
plots, 89 per cent ; and on the lightly used plots, 58 per cent. Much of the
trampling apparently was done by ground squirrels and other rodents.

Because broad-leaf filaree is an aggressive plant, occurs in abundance,
and can be depended on to provide feed under a wide range of conditions,
it can well be rated one of the two most important forage species on the
experimental area.

Clovers. — True clovers common on the area are maiden clover, tree
clover, white-tip clover, and tomcat clover. Six other species of clover
have been recorded but none of these is sufficiently abundant at present
to be of importance as a forage plant. During the five years of study
the combined percentage of the four most common clovers in the forage
stand has varied from less than 1 to almost 11.

The clovers are all native annuals. Their distribution over the range
is influenced by moisture content and soil conditions. Maiden clover
(also called pinhead clover) is found widely over the slopes and ridges,
while tree clover (fig. 5, H) is most common around borders of swales
and creeks. Tomcat clover (fig. 5, G) and white-tip clover grow in wet
swales and occur most abundantly in wet years. These two clovers are
the ones most frequently noticed on the range, since they grow in almost
pure stands. Maiden clover (fig. 5}F) is the smallest of the four species
and does not provide as much forage per plant. It grows to a height of
5 to 8 inches ; the others often attain a height of 12 to 14 inches.

Clovers are classed as forage species of high nutritive value, and are
taken readily whenever available ; stockmen usually expect cattle to do
well when clovers are plentiful. Swales of clover are usually grazed so
closely that by the end of the season the soil is almost completely bare.
Swales remaining wet until late in the season may furnish some especially
prized green feed after most of the forage on adjacent slopes and ridges
is dry.

Spanish Clover. — Spanish clover (fig. 5, D), sometimes called pea
vine, occurred in only minor abundance during any of the five years.
It varied in amount from 0.5 per cent, in 1937, to 4.9 per cent, in 1938
(table 1). This native annual, well-distributed over the range, differs
from most of the other forage plants in its ability to make some summer
growth.

20 University of California — Experiment Station

The seeds do not germinate so readily as do those of broad-leaf filaree,
and the young plants make a slow growth during the winter period.
Growth, however, is more rapid for a time around the middle of April
after the bulk of the other forage has matured, and continues at a slow
rate during the summer. In shallow soils the plant seldom exceeds a
foot in height, but in deep soils in swales where moisture is plentiful
late in the season, it may grow to 3 feet. The plant produces a slender
central stem somewhat resembling a pea vine.

Duration of green summer feed varies with the season. During the
four wet seasons of 1935 to 1938, much of the Spanish clover remained
green all summer. During the dry spring of 1939, in contrast, some
plants on shallow soils of south exposures dried by the middle of May ;
and even on the best sites only a few plants remained green all summer.
Seed pods are developed continuously throughout the summer and seeds
are cast over a long period.

Spanish clover is not especially sought by cattle in the early stages of
growth, but late in the season when the bulk of the other forage has dried
it becomes a choice forage plant and is readily taken even when dry.
Its importance on the range lies in the fact that it provides some green
feed during the dry summer when it is badly needed. On pastures saved
for late-season use, Spanish clover is among the first forage plants to be
grazed.

Popcorn Flower. — Popcorn flower (fig. 5, E) , a native annual, ranged
in abundance from 0.3 per cent, in 1936, to 2.6 per cent in 1935 (table 1) .
It is widely distributed over the Range but usually is most common on
south exposures. Like filaree, it germinates readily after the first rains
and, provided the season is early, it may make considerable fall growth.
A rosette, formed almost at the beginning of growth, during midwinter
remains almost flat to the ground on closely utilized areas. With the
coming of warm weather in early spring some of the rosette leaves soon
point upward and a flower stalk develops early. This is one of the first
plants to flower in the spring and the many blooms produced in favorable
years spot the landscape like a scattered skiff of snow. Flower stalks
average from 8 to 14 inches high ; on good sites frequently 24 inches. The
plant begins to dry early and by the first of May is almost completely
dry. As the leaves dry they shrivel; most of them soon fall, and leave
only a dry stem standing.

Since popcorn flower usually makes considerable early spring growth,
it provides feed at a time when it is badly needed. In this habit of growth
lies its chief importance. Its green flower stalks are readily eaten by
cattle but dry stalks are not grazed.

Yellow Tarweed. — Yellow tarweed (fig. 6, A), also a native annual,

Bul. 663] The San Joaquin Experimental Range

21

Fig. 6. — Broad-leaved herbs of secondary importance on the San Joaquin Experi-
mental Eange: A, yellow tarweed; B, goldfields; C, layia; J), rattlesnake weed; E,
godetia; F, fine-leaf lotus; G, ground lupine; H, birdseye gilia; and I, linanthus.

varied in amount from 0.2 per cent to 5.1 per cent during the five years
of study. Each year when the forage inventory is made in April at the
peak of growth of the forage crop as a whole, tarweed plants are still
comparatively small and immature; therefore their amount as repre-
sented by the inventory is not fully indicative of the abundance of the
plant at maturity. The plants usually average around 10 inches in height
and have a spread of about 6 inches.

22 University of California — Experiment Station

Although widely distributed over the range, yellow tarweed is usually
most common in swales where the soil is deep. The seeds germinate a
few days later than those of filaree and popcorn flower. Growth of the
young plants is very slow through winter and the plant is little noticed.
Growth speeds up, however, as the bulk of the forage begins to dry
around the middle of April, and thereafter the plants rapidly become
conspicuous. The plants begin to bloom during the latter part of July
and continue to grow throughout the summer. Full bloom occurs during
August and September, and drying usually does not take place until
after the fall rains start in October or November. During the dry spring
of 1939 some plants dried as early as June on south exposures where the
soil was shallow, and premature dying continued throughout the sum-
mer. A heavy infestation of grasshoppers also killed many tarweed
plants in June of 1939.

This tarweed is grazed by cattle with but little discrimination until
it is about half grown. Thereafter it is mostly avoided. After it has
matured it is rarely taken. It produces a gummy tarlike substance which
adheres to the head and legs of grazing animals and to clothing of anyone
who brushes against it. Moreover, the plants may grow locally in stands
so dense that they interfere with normal grazing.

Yellow tarweed has been occasionally regarded as a possible indicator
of range use and condition. As an illustration, an apparent scarcity of
this plant on the range may indicate heavy use since it is grazed in the
early stages of development. On areas where utilization is close during
the green-feed period, any remaining plants are usually spindly, whereas
the plants are more robust under moderate and light grazing. Even
under close use enough seed is usually produced for another year's crop.
Under light use and complete protection the plant gradually decreases
in abundance. Trampled areas are favorable to yellow tarweed, as in-
dicated by its increased abundance in old abandoned roads, cattle trails,
and other similar places. For this reason a range used heavily one year
and protected the next may support a denser stand of tarweed the
second year, owing to the trampling of the previous season. In view of
these observations, tarweed as an indicator plant must be used with the
greatest of caution, especially on ranges that are grazed during the
current season.

Turkey -Mullein. — Turkey-mullein, another native annual, was not
sufficiently abundant in 1935 to be represented in the forage inventory,
and during the following four years it composed less than 0.5 per cent
of the total forage cover. However, this summer annual, like yellow tar-
weed, is still small when the forage inventory is made in April, and is
far more conspicuous later in the season.

Bul. 663] The San Joaquin Experimental Range 23

On the Range turkey-mullein is most apt to be abundant in places
that are grazed off early in the spring. The seeds do not begin germinat-
ing until late in December and continue to germinate throughout
February, and even into March. The young plants average about three-
fourths inch both in height and diameter when the bulk of the forage
over the range dries in April. At this time many of the seedlings die
on lightly grazed areas, probably because the upper soil is so completely
dried out by the maturing forage crop. Where much of the forage crop
has been removed through heavy grazing the soil is not so completely
desiccated, and more of the turkey-mullein plants survive; this appar-
ently accounts in large part for the heavy stands in places that are
grazed close early in the season. The average height of the mature plants
is nearly 7 inches, and the spread of tops 5 inches — with extremes of
14 to 16 inches.

No grazing of turkey-mullein by cattle has been observed on the Range
in any stage of development. Possible indicator value is the chief reason
for including a discussion of this species. Abundance on one range and
comparative scarcity on an adjoining one may indicate heavy use of the
first and light use of the second. However, an abundance of the plant
does not always signify that the range has been used heavily. When a
heavy rain falls at about the time the main forage crop is at least half
dry, the depleted soil moisture may be replenished and the turkey-
mullein plants survive and become widely conspicuous. This plant, there-
fore, may also be used to a limited extent as an indicator of range use
and condition. Still, like yellow tarweed, it must be so used with caution.

Minor Broad-leaved Species. — Many broad-leaved species other than
those already discussed are found in small quantities. Most of these are
native annuals. Some of the most common are birdseye gilia, goldfields,
layia, linanthus, rattlesnake weed, godetia, ground lupine, and fine-leaf
lotus. Some of these are shown in figure 6. As with other species, the
relative abundance of these fluctuates up and down from year to year.
They are taken along with other forage when young and succulent, but
after they mature and dry some are grazed but slightly. The California
bur-clover, another introduced annual and one of the most valuable
range plants in the annual type, is found only as a trace on the Range,
and occurs very sparingly throughout the granite-soil section.

A few native perennials such as the California buttercup, Hansen
larkspur, and California poppy make up only about 0.33 per cent of the
total herbaceous forage cover. Owing to their scarcity they are com-
paratively insignificant on the range. Only the California buttercup is
worthy of mention as a forage plant. It makes a very early growth and
provides some of the earliest feed in several of the swales and other wet

24 University of California — Experiment Station

places. It was iy2 inches high on the first of February in 1939 when the
other feed was only one-fourth inch tall.

Soft Chess. — The most abundant grass on the Range during four of
the five years of study has been soft chess (fig. 7, A). During the years
1936 to 1939, inclusive, it made up from 17 to 32 per cent of the cover,
although in 1935 it accounted for only a little over 1 per cent. Of the
grasses alone, it composed 7 per cent in 1935, 61 per cent in 1937. This
grass is widely distributed over the Range, but usually is most abundant
on north exposures and in moist places.

Soft chess is an annual introduced from Europe. Its seeds germinate
about two days later than those of filaree, provided conditions for germi-
nation remain favorable. Fall and winter growth is only moderate. Rapid
growth in the spring starts 2 or 3 weeks later than that of broad-leaf
filaree ; consequently, rains late in the season are much more beneficial
to its development. To insure a heavy yield of soft chess, a good rain is
almost necessary at about the time filaree has reached maturity. Numer-
ous height measurements of ungrazed plants in the natural area in 1937
and 1939 averaged 6.3 and 5.4 inches, respectively.

On poor sites the plants usually are shorter, but on good sites height
growth of 20 to 25 inches is not uncommon. In the spring of 1935, follow-
ing a very favorable growing season, some plants over limited areas
were found to be 52 inches high. When the plants are small and stunted
the seed heads also develop poorly.

Soft chess is considered about the best of the annual grasses in this
plant type. In areas where broad-leaf filaree makes up a high percentage
of the plant cover it seems that the crop is better for yearlong grazing
where soft chess makes up at least 20 or 25 per cent of the cover. For
the season as a whole, it might well be considered one of the two most
important forage plants on the Range, being equalled or surpassed only
by broad-leaf filaree. During the summer it has thus far been rated as
the most important forage plant in the area. Soft chess also provides
considerable winter feed but is grazed most heavily at the time it is
reaching maturity, just after broad-leaf filaree has begun to dry. In
1936, for example, it composed nearly 60 per cent of the forage at this
growth stage. The seeds do not shatter readily and the plants are grazed
extensively during the summer. Because of this, the grazing animals get
more nutriment than they otherwise would from the dry stalks and
leaves alone.

The percentage utilization of this species was estimated at the end
of the grazing season in 1939 on the 60 permanent plots. Soft chess was
present on 59 of the 60 plots and the measurements showed 55 per cent
of this species to be grazed down to an average height just short of 2

Bul. 663]

The San Joaquin Experimental Range

25

Fig. 7. — Important grasses and grasslike plants on the San Joaquin Experimental
Eange: A, soft chess; B, red brome; C, ripgut grass; D, slender oat; E, foxtail
fescue; F, poverty fescue; G, toad rush; H, common spike rush; I, slender-leaf rush.

inches. This amount of utilization appears in the field to be about mod-
erate use. The 59 plots on which this species was found were classified
as to whether they showed close, moderate, or light use. According to
this classification 20 of the plots were closely used, 27 moderately used,
and the remaining 12 were used only lightly. The plots showing close

26 University of California — Experiment Station

utilization had 80 per cent of this species grazed down to about V/2
inches. Those showing moderate use had 50 per cent grazed down to
about 2 inches ; and of the 12 plots that appeared to be lightly used, only
28 per cent of the forage showed use and the stubbles that remained were
slightly over 2 inches high.

Foxtail Fescue.- — The second most common grass on the Range is fox-
tail fescue (fig. 7,E), which composed from about 9 to 17 per cent of the
total cover during the five years (table 1). It accounted for 69 per cent
of the grasses in the spring of 1935 when soft chess was scarce, although
during the following four seasons it composed only 18 to 40 per cent.

Foxtail fescue is a native annual. The plants are evenly distributed
over the range and appear to be especially well adapted to shallow soils.
Its seed and those of soft chess germinate at about the same time. The
stems are slender and the forage never becomes sufficiently coarse to
make it undesirable for cattle. Compared with soft chess, foxtail fescue
is shorter and ripens and dries a little earlier — almost as early as broad-
leaf filaree. The seeds shatter readily; very few remain on the plants
by the end of the summer.

Foxtail fescue, the third most abundant species of all plants on the
area, is fairly palatable to cattle during all seasons of the year. During
the dry summer it ranks next to soft chess. It composed a little over 30
per cent of the forage taken in February of 1936 in one study on utiliza-
tion. Its use began to diminish in March when filaree and soft chess
began to make more growth, and it continued to decrease through April
and May. By the middle of May it composed only 5 per cent of the forage
taken.

This species was found on 52 of the 60 utilization plots and the meas-
urements showed 59 per cent of the forage to be grazed down to an
average height of about 1% inches. This amount of utilization has been
rated as about moderate use. Of the 52 plots where foxtail fescue was
found, 19 showed close use, 23 moderate use, and 10 light use. Those
grazed closely had 82 per cent of the foliage of this species grazed down
to an average height of less than IV2 inches ; those estimated to be grazed
moderately had 53 per cent grazed down to about 1% inches ; and on the
10 plots that showed light use, only 29 per cent of the foxtail fescue
showed any grazing at all — with an average height of just over 2 inches.

Minor Grass Species. — Grasses other than those already discussed are
relatively sparse on the Range. Slender oat (also known as slender wild
oat) , an introduced annual, is fairly common but in none of the five years
has it composed more than 4 per cent of the cover (table 1). It is illus-
trated in figure 7, D. Over the foothill section in general this is one of
the most common range plants. It is considered good forage during the

Bul. 663] The San Joaquin Experimental. Range 27

winter period when it makes more growth than many of the other species,
especially in protected places, bnt during the summer when the forage is
dry it is not taken very well.

Ripgut grass (fig. 7, C) makes considerable winter growth during
which time it furnishes good forage, but after it has dried the plant is
not taken readily. The awns are long and rigid and often cause trouble
to grazing animals. For this reason it is usually considered an inferior
species on the range.

Red brome (fig. 7, B), also commonly known as foxtail chess, is an-
other introduced annual ; it comprised 3 per cent or less of the forage
cover. This species is fairly well distributed and, like foxtail fescue,
appears to be well adapted to shallow soils. Its growth during the winter
is about like that of soft chess and it furnishes fair feed during this time.
It matures early, at about the same time as foxtail fescue, and after it
is dry it is not well liked by cattle. The awns are rather rigid but they are
not especially injurious.

Poverty fescue (fig. 7, F), one of the least abundant of the annual
fescues, is common on the Range but makes up only a small portion of
the cover — never more than about 1 per cent. This species is commonly
associated with foxtail fescue and is similar in growth and value. Stock
graze the two species together without discrimination.

Common foxtail, also called mouse barley and wall barley, is sparse
on the Range but is abundant over many areas in the annual type. Stock-
men regard it as a good forage plant during the winter, when it makes
considerable growth. After it dries, it is a decidedly inferior forage,
because of the troublesome awns.

Pine bluegrass is a native perennial. Even though it makes up a very
small percentage of the cover, it is the most common perennial grass on
the Range. It has composed less than 0.2 per cent during each of the five
years of study. Other perennial grasses which are exceedingly sparse
are melic grass and purple needlegrass (also known as California needle-
grass and nodding needlegrass). Bunches of melic are confined chiefly
to rock outcrops. These perennial grasses have not increased on this area
during the past five years, either under grazing or total protection.

The perennial grasses now occurring on the range as a whole are
mainly dormant through the dry summer period. They dry almost as
early as do the bulk of the annual species, and then become green again
after the start of the fall rains. Pine bluegrass, in particular, dries very
early and none of the top remains green during the summer. Drying of
purple needlegrass is delayed nearly 2 or 3 weeks later and a little green-
ness may remain in the base of the clump practically all summer, al-
though the amount of such green feed is small.

28 University of California — Experiment Station

Cattle graze pine bluegrass, both green and dry, much as they do the
annual grasses. Its organic constituents are similar to those of annual
grasses (7). From these and other observations on growth and utiliza-
tion, the advantages of the native perennial grasses are chiefly two :
They make more fall and winter growth than annuals, and they can be
relied on to make grazeable fall growth with less moisture.

Grasslike Plants. — The grasslike plants, or rushes, (fig. 7) are com-
paratively sparse on the Range. The three most common species are
slender-leaf rush, common spike rush or wiregrass, and toad rush. The
first two are perennials; all three are natives. Even though the two
perennials together have composed only 0.5 per cent or less of the cover
each year, they are rather important grazing plants for a short time at
the end of the green-feed period. They are found chiefly in the wettest
swales and usually remain green for a time after most of the annuals
have dried. It is during this time that rushes are grazed most heavily by
cattle and are most important on the range ; for example, in 1936 they
made up well over 15 per cent of the feed for a short time near the middle
of May. In the dry spring of 1939 they dried early and therefore fur-
nished little feed at this season.

Toad rush (fig. 7, G) varied in amount from 2.8 per cent of the forage
cover, in 1935, to nothing in 1939. Like the perennial rushes, this species
is found mainly in wet places, and the season of 1939 was a dry one.
When this species is plentiful it is most common on heavily grazed and
trampled areas, everything else being equal, and therefore it may be
considered as an indicator of heavy grazing, or of other disturbance. The
plant is grazed readily when it is green but only moderately after it has
dried.

Browse Plants. — Shrubs commonly browsed on the Range (fig. 8),
listed in order of amount of feed they provide on this particular area,
are : wedgeleaf ceanothus, also variously known as buckbrush, chamise,
chaparral, and greasebush; hollyleaf buckthorn, locally often called
redberry ; coffeeberry, also called California buckthorn; blue elderberry,
known also as blueberry elder ; poison-oak ; and Mariposa manzanita. In
order of their palatability they are rated thus : blue elderberry, coffee-
berry, hollyleaf buckthorn, poison-oak, Mariposa manzanita, and wedge-
leaf ceanothus. Three tree species commonly browsed, both in order of
amount taken and their palatability, are the California blue oak, interior
live oak (also known as highland live oak), and California buckeye. The
first of these trees probably furnishes more browse than any of the shrubs
on the Range. It is estimated that the shrubs and trees together roughly
cover, on the average, from 10 to 15 per cent of the Range area; at higher
elevations the proportion of browse plants is greater.

Bul. 663]

The San Joaquin Experimental Range

29

Fig. 8. — Seven shrubs and trees commonly browsed on the Experimental Eange:
A, wedgeleaf ceanothus; B, coffeeberry ; C, hollyleaf buckthorn; D, Mariposa man-
zanita ; E, blue elderberry ; F, blue oak ; G, interior live oak.

It is difficult to estimate the value of these browse species on the range.
Even though they do not furnish a large quantity of feed, probably not
exceeding 1 or 2 per cent year after year at the experimental area, they
do provide some green material during the summer when the herbaceous
forage is mainly dry ; thus they help out on protein, and possibly vitamin
A (11, 12). This, no doubt, is their chief value on ranges where they
compose only a small part of the diet. A few browse species on any range
are highly desirable plants.

30 University of California — Experiment Station

GROWTH AND DEVELOPMENT OF THE ANNUAL FORAGE CROP

In planning his ranch enterprise the skillful operator considers care-
fully the times when forage grows and is available for grazing. The
forage crop on this area passes through three distinct phases each year :
(1) Seeds germinate and growth starts shortly following the first effec-
tive rains of the season (expected in October) but is usually very slow
during 3 or 4 months of the mild winter, being retarded mostly by low
temperatures but occasionally by insufficient rainfall. (2) Rapid growth
starts during the latter part of February or early in March and termi-
nates near the end of April. The forage is of most value to livestock dur-
ing this period of rapid growth. (3) The mature crop dries rapidly
during May and June and remains dry throughout the summer — a period
of nutritive deficiencies which must be considered in management. The
dry-feed period extends through the summer and until such time as the
first autumn rains of the next rainy season start growth of a new forage
crop. These three phases, and their bearing on practical handling of the
range, are discussed under their separate headings.

Fall and Winter Period of Slow Uncertain Growth. — Annual plants
in this type of range make very little growth during the months of
December, January, and February. Consequently, the supply of new
forage during this period is generally not sufficient to maintain animals
without additional feeds. When the first good rains are fairly early, for
example in mid-October, before the temperatures become low, there may
be considerable growth in the fall ; but the amount produced is usually
not sufficient to carry the animals through the ensuing winter weeks.
Winter growth is usually a little greater if the fall rains are early,
everything else being equal, since the plants are better established and
the increased top growth offers protection against cold.

Height measurements from both protected and grazed plots, as shown
in table 2, afford an illustration of the slow rate of growth of range
forage during the three winter months : December, January, and Feb-
ruary.

During the fall and early winter, perennial grasses and rushes make
more growth than annuals. This observation is illustrated by some meas-
urements taken in January, 1939. Clumps of pine bluegrass were 2.5
inches high and California melic averaged 10.3 inches, whereas the
average height of soft chess on 60 permanent plots at this time was 1.4
inches, foxtail fescue was 1.4 inches, and broad-leaf filaree only 0.5 inch.
Perennial grasses thus provide, in proportion to their abundance, more
fall and early winter feed than the annuals.

Early Spring Period of Rapid Growth. — Most growth of annuals takes

Bul. 663]

The San Joaquin Experimental Kange

31

place in about 6 weeks following the first of March. Usually there is
plenty of moisture at this time and rising temperatures further stimu-
late growth. An illustration of the accelerated growth during this period
appears in table 2 ; the measurements were made on the same species and
plots as those given for the winter period. All three species had reached
maturity by April 20. Therefore, during the period from March 1 to
April 20 broad-leaf filaree, on the two sets of plots combined, made 83

TABLE 2

Winter and Spring Growth of Three Main Forage Species, 1938-39

Range use

Date

Height in inches by species

Broad-leaf
filaree

Soft
chess

Foxtail
fescue

Winter growth

Protected against livestock.

Grazed by livestock .

December 1

0.4

1.5

March 1

0.7

2.0

December 1

0.4

1.2

March 1

0.8

1.5

1.3
1.6

1.1
1.3

Growth during six weeks of spring

Protected against livestock.

Grazed by livestock.

March 1
April 20

March 1
April 20

0.7

2.0

4.4

6.4

0.8

1.5

4.3

4.7

1.3

4.9

per cent of its height growth, soft chess made 68 per cent, and foxtail
fescue 74 per cent. Although there is considerable variation from year
to year, it appears that beginning of rapid growth cannot be expected
much before the end of February or first week in March.

On ranges that are stocked on a yearlong basis there is usually no
scarcity of feed at this time of year. The forage grows far faster than
the animals are able to utilize it. "Where feeder cattle are shipped in, in
order to utilize the forage during the green-feed period, the situation
may of course be different. As the feed becomes more abundant during
this period of rapid growth, cattle begin to do some selective grazing.
Livestock gain rapidly during this period since the forage crop as a
whole is lush, nutritious, and ample.

Summer Period of Greatest Feed Deficiencies. — The forage crop is
exceedingly poor in nutritional qualities after it has dried and weath-
ered. Hart, Guilbert, and Goss (13) and Wagnon, Guilbert, and Hart, in
the section "Experimental Herd Management," have shown crude pro-

32 University of California — Experiment Station

tein to be the chief deficiency as far as cattle are concerned. The crude-
protein content of young plants is high but it decreases rapidly as the
plant grows, according to Gordon and Sampson (7), and Guilbert and
Rochford (10), until it reaches the full-bloom stage. Afterwards the
percentage of crude protein decreases slowly until it is very low when
the forage has dried and weathered. Hart, Guilbert, and Goss (13) state
that phosphorus also may become deficient in some years under certain
conditions. A detailed discussion of these deficiencies, and of their sig-
nificance to the grazing animal, will be found in the section mentioned
above.

There are some exceptions to these general statements regarding the
decreased value of the range forage plants in summer. A few species such
as Spanish clover and perennial rushes, being late growers, furnish some
green feed until late in the season or, in some years, practically all sum-
mer ; they thus help out on the crude-protein-deficiency problem. Never-
theless, as explained in another section, the amount of these plants is not
sufficient to balance the ration and provide for normal animal nutrition.

VARIATIONS IN THE FORAGE CROP FROM YEAR TO YEAR

In practical ranch operation the stockman is ever troubled by the
uncertainty of range forage. In the annual-plant types, big differences
in the crop from year to year are the rule — not the exception. Owing to
the size of these differences, to their uncertainty, and to the fact that
they cannot be predicted far in advance, proper stocking of ranges is
extremely difficult. Consequently, fluctuations in yield, mixture of
species, and time of growth add up to one of the biggest problems facing
livestock operators in these foothills.

Changes in Yield of Forage. — The total yield of forage varies notice-
ably from year to year, and occasionally extreme fluctuations take place.
It is a common observation that these variations are generally greater
in the annual-forage type than in perennial-grass types.

It is estimated that variations amounting to nearly twice as much feed
one year as compared with another have occurred on portions of the
Range during the period covered by these preliminary studies. Differ-
ences of fully 50 per cent have been recorded in the fenced pasture plots,
weights of forage clipped from small areas in these ungrazed plots being
equivalent to 1,460, 1,010, 1,490, and 1,570 pounds per acre for the years
of 1935, 1936, 1937, and 1939 respectively. Forage production during
these years probably has been above that of a long-time average, since
conditions for growth (particularly precipitation) have been excep-
tionally good; and 1935 was considered by old residents as the most
productive in many years.

Bul. 663] The San Joaquin Experimental Range 33

These fluctuations are not unusual in the grazing lands in California
that are covered with annual plants. Far greater differences in yield
between consecutive years were reported by Talbot, Biswell, and Hormay
(14) in the southwestern part of the San Joaquin Valley for the years
1934 and 1935. The season of 1934 was very dry, and growth conditions
were so adverse that practically no forage was produced over large areas.
The following season, with an abundance of well-distributed rainfall
and with favorable temperatures, livestock grazed over the same areas
in feed literally "knee deep."

Changes in Kinds of Range Plants. — In this annual type, not only the
amount but also the kind of forage varies. As already pointed out and
shown in table 1, the percentage of most species on the Range during the
five years of study varied widely at the peak of growth ; however, a few
have held fairly constant. As a result the range feed as a whole is better
some years than others, a fact that reflects differences in palatability and
nutritional qualities of the various plants, and also supports a common
observation of local stockmen that "this is a poor bur-clover year" or "a
good bur-clover year" — according to the current abundance of this supe-
rior forage plant. Also, table 3, based on annual inventories, shows how
several groups of the range vegetation vary from year to year. Further
evidence is diagrammed in figure 9, which also emphasizes the surprising
degree to which alien plants have invaded range lands of this region.

It is an interesting observation that several kinds of plants appear to
make parallel fluctuations. The behavior of the legumes to date illus-
trates this tendency. "When proportions of the clovers have been high in
the plant cover, the percentages of Spanish clover, fine-leaf lotus and
lupines have also been relatively high ; and when one of these groups has
been low, the others have been also. A similar type of fluctuation was
noticed with several species of grasses, as well as with the grasslike
plants. Apparently the growth factors favoring one species or group is
also favorable to the others and, as a result, a similar response is made
by all. Also, fluctuations by groups tend to cause corresponding shifts
in the average forage density over the range. It has been noticed particu-
larly that a shift up or down in the legume group will cause a similar
shift in total density, since most of the legumes represented in the plant
cover have a dense spreading habit of growth.

The number of species making up an appreciable part of the flora also
varies considerably from year to year. For example, in 1935 the number
of species each making up 1 per cent or more of the forage crop was 16 ;
in 1938, 14 ; in 1939, 10 ; and in 1936 and 1937 the number was only 7.

But little information has been obtained thus far concerning fluctua-
tions that extend over a long period of time. However, indications are

34

University of California — Experiment Station

that the percentage of some species varies with long-time wet and dry
periods.

Variations in Time of Growth and Drying. — Growth dates vary widely
from year to year — a general observation confirmed by numerous meas-
urements. In 1934, growth was abundant in the autumn and continued
well through the winter; in 1937, fall and early winter growth was
practically nil. During the fall and winter of the years 1935, 1936, and

TABLE 3

Fluctuations Between Plant Groups from Year to Year

Plant group

Percentages by years

1935

1936

1937

1938

1939

Annual broad-leaved plants

74.8

20.6

2.9

53.2
45.5
0.5

43.5
55.5
0.5

61.7

34.1

3.6

52 6

47 1

Trace*

Total

98.3

99.2

99.5

99.4

99 7

* A trace is less than one-tenth of one per cent.

90
80

70

fr-. 60

Z

u

O 50
K
S 40

30

20

10

iM

1935 1936 1937 1938 1939

Fig. 9. — Variable percentage of total forage represented by alien plants.

1938, plant growth was moderate. The period of rapid growth during
the spring varied by about one full month. In 1935 and 1936 rapid
growth began during the last week of February, while in 1938 it was
exceptionally late, not beginning until the last week of March. In the
spring of the years 1937 and 1939, rapid growth began about the first
of March.

The length of time that the forage crop is dry and low in crude protein
also varies from year to year, which complicates supplemental feeding.
The forage may dry earlier some years than in others ; and new feed
comes later some years than others, according to the occurrence of fall

Bul. 663] The San Joaquin Experimental Range 35

rains. Drying at the end of the season took place at about the same time
during the four years 1935 to 1938 ; but in the spring of 1939 drying on
south exposures was nearly 3 weeks earlier than for the four preceding
years, and in swales the difference in time was still greater. Data col-
lected in 1938 and 1939 from permanent plots located on south exposures
illustrate the wide differences in time of drying. Soft chess, for example,
in 1938 was only 33 per cent dry on May 15, whereas in 1939 it was
58 per cent dry by April 28. Foxtail fescue on the same plots was 90 per
cent dry by May 15 in 1938, and was 84 per cent dry on April 28, 1939.
Drying of forage in the swales in 1939 was only a little later than it was
on most slopes, but during the wet seasons of 1935 to 1938, inclusive, the
swales were wet until late in the season and the forage was correspond-
ingly late in drying. During the years of heaviest precipitation there was
still some green feed in the wettest swales in mid- June.

These wide fluctuations in forage produced from year to year necessi-
tate careful and skillful operation on the part of the stockman, in pro-
viding supplemental feeds or additional acreage during adverse seasons,
and in other ways keeping his business as flexible as possible.

Conditions That Bring About Forage Variations. — Climatic factors
no doubt have most to do with yearly fluctuations in forage yield, com-
position, and season of growth. Precipitation and temperature are the
most important, and a seasonal variation in either may cause a marked
fluctuation in the feed. The close tie between climatic factors and forage
growth has encouraged the thought that certain characteristics of the
forage may be predicted some time in advance of their development.
Evaluation of the more complex relations awaits further study, but
certain promising trends are already beginning to take shape.

It appears that about 0.8 inch of rain is sufficient for the germination
of a normal stand of annuals, the required amount depending somewhat
on such factors as humidity, temperature, cloudiness, and wind. Under
the same conditions of weather, less rain will start growth of perennial
grasses and rushes. These began to grow following a rain of 0.25 inch
in the fall of 1937 ; and other light showers up until November 23 — all
totaling 0.5 inch — were enough to maintain their growth.

With favorable conditions, annuals usually germinate very quickly.
One year, new seedlings were coming up within 41 hours after the start
of fall rains and repeated observations have shown that warm slopes
will begin to take on a green cast within only 3 or 4 days following the
start of rains. Cattle will begin to graze the new growth within a few
days, although for some time the volume of new green feed may be in-
sufficient to maintain the animals.

The percentage of each species in the mature forage cover is influenced

36 University of California — Experiment Station

somewhat by the time the first rains occur. When the rains are early,
conditions usually favor the germination of certain species ; while late
rains bring about the germination of other kinds of plants. This was
first observed through greenhouse and plant-nursery experiments and
later verified in the field. Some broad-leaf filaree plantings made in the
unheated greenhouse in December of 1936, after temperatures had be-
come fairly low, yielded very low germination percentages, but in the
early warm fall of 1937 a good stand was obtained from a similar plant-
ing. Broad-leaf filaree plantings made in the forage nursery late in the
season also yielded a poor stand the first season. Similar behavior was
observed in the field in the fall of 1937 when the first effective rains did
not occur until December 9. The percentage of broad-leaf filaree, and also
of soft chess, was low in the forage cover the following season. Several
other species, mainly natives, were more abundant. If this relation con-
tinues to hold true, broad-leaf filaree and soft chess are likely to be
abundant when the first effective rains are early, provided of course
there is a good seed supply of each in the soil and a drought period does
not follow the first rains. When the first rains are late (December 10 or
later) these species are likely to be comparatively sparse.

When the rains are late a greater number of species are likely to be
found in the forage cover, common examples being some of the clovers,
lotus, and lupines. One acquainted with seedlings can usually make a
fairly close guess early in the season as to what the abundance of certain
species in the forage cover in the spring will be. Obviously, this cannot
be done with any high degree of accuracy since the plants respond dif-
ferently to varying climatic growth factors.

Good plant growth usually follows the first effective early rains. Dry
periods may follow, however, and cause the feed to cease growth almost
completely, or in some cases even cause the death of many plants. In the
fall of 1934, the first effective rains were not especially early but growth
throughout the fall was excellent since the rains kept up and moisture
was sufficient. Growth in the fall of the years 1935 and 1936 started
early but dry periods which followed and lasted until the middle of
December brought it almost to a standstill. Growth was excellent for a
short time following the first rains in the fall of 1939, but the ensuing
period of clear dry weather stopped growth ; and most of the plants on
south exposures died.

After the first of April, rainfall no doubt becomes the most critical
factor in growth. The maturing vegetation requires considerable mois-
ture as the days lengthen and temperatures increase. Rainfall during the
latter part of March and first of April may determine largely the amount
of forage produced. In the spring of 1939, plant growth was slowing

Bul. 663] rpHE gAN Joaquin Experimental Range 37

rapidly prior to March 26 and 27 when a total of 1.28 inches of rain fell.
This was sufficient to cause continuation of growth and a good forage
crop developed. Had it not been for this rain, soft chess probably would
have made little growth and developed little seed since it matures late
in the season.

During wet years some forage in the swales may stay green nearly 2
months after the vegetation has dried elsewhere and thereby the length
of the green-feed season is increased. This was the case during the four
wet seasons of 1935 to 1938, when there was heavy seepage into the
swales. In many places the perennial rushes provided green feed until
the middle of July, and in 1935 some plants were still green even at the
middle of August. In contrast to this the feed in the swales in 1939,
following the dry winter season, dried only little later than it did on the
ridges, being mostly dry by the middle of May and almost completely by
the first of June. Cattle graze heavily on the green feed in the swales
late in the season, and as long as they get sufficient quantities of this,
protein supplements are not needed. Therefore, spring rains sufficiently
heavy to cause seepage until late in the season are of value and may be
expected to extend the green-feed period and thus delay for a time the
critical period of dry feed.

Heavy rains after the bulk of the palatable vegetation has matured
and dried, and is not using soil moisture, are a boon to the summer-grow-
ing species such as yellow tarweed, turkey-mullein, and Spanish clover.
In 1935, a rain of 1.76 inches coming on April 29 partly explains why
yellow tarweed in some places attained a height of 6 feet and the Spanish
clover remained green all sumer. Again, yellow tarweed and turkey-
mullein plants were very abundant and moderately large during the
summer of 1938 as a result of heavy rains late in April. During the
seasons of 1936 and 1937 when the rainfall was light through the latter
part of April the summer-growing plants were sparse and only of mod-
erate size. In 1939, when practically no rain came during all of April
and the soil was left dry as a result of the forage crop maturing, most
all summer-growing species died early in the season. Only a few turkey-
mullein plants remained along creeks, where the soil was deep, and in
other places where the forage crop was grazed off before the soil moisture
was depleted.

Rains — even light showers — coming late in the season after most of
the forage has dried are usually harmful to the dry feed through leach-
ing (13). For example, it was the general consensus of opinion that a
rain of only 0.15 inch in May 1936 had this effect.

In summarizing the several climatic factors that leave their mark on
forage, two stand out — and they operate together. Rain in itself, im-

38 University of California — Experiment Station

portant as it may be, is not the determining factor, for it operates in re-
lation to temperature. It is possible to have, for example, a better forage
crop with 10 or 12 inches of rain well distributed and combined with
favorable temperatures, than with 25 or 30 inches of rain with unfavor-
able distribution and temperatures. The study period has not yet in-
cluded any bad drought years. So this "annual type," which to the casual
observer appears to be a simple type, is actually a very complex one.

As in other branches of agriculture, any forecasts of conditions would
be helpful in arranging the complicated range livestock business. To be
able to know, in advance, of impending shortages in range forage would
give the operator a greater chance to provide supplemental feed, adjust
herds, or arrange credit facilities to carry him over the unfavorable
periods. As more is known about the effects of weather on feed (for even
a short period in advance) the operator may be better equipped to deal
with the changes and uncertainties of feed that prevail in this type of
range ; for the relation between climate and feed bears directly on the
everyday questions of proper grazing season and intensity of use.

EFFECTS OF DIFFERENT INTENSITIES OF GRAZING

The many practical problems of forage management in the annual
type in need of study include the effects of different intensities and
seasons of grazing, the possibilities of improvement by artificial reseed-
ing or through fertilizing or other soil treatment, and the effects of
different uses and treatments on rainfall runoff. Most of the forage-
management studies reported upon in this bulletin concern chiefly the
effects on the range of different intensities of use — what happens, for
example, when this type is grazed heavily or lightly. The question upper-
most in many stockmen's minds is : "How can I graze my range to get
the most out of it and still not hurt it? And if possible, improve it V

Since the measured effects of different degrees of use of annual-type
ranges were little known when the San Joaquin Experimental Range
was started, this was considered one of the major studies. Its aim is to
find out just what the relation may be between the closeness and period
of grazing a range and the kind, amount, and duration of forage which
it produces. Concrete facts on such problems bear directly on one of the
fundamentals of successful ranch operation.

A set of pastures was established for beginning the study. It consists
of two small, two medium-sized, and two large pastures (fig. 1), all
stocked with the same number of cattle for approximately 7 months of
the year. Each pasture is grazed by 15 cows with their calves to weaning
time, together with a bull for nearly 4 months of the period. The average
number of calves produced per pasture is about 9, so figuring each of

Bul. 663] The San Joaquin Experimental Range 39

these as one-half animal unit, the pastures were stocked during the three
years of 1937-39, inclusive, at the approximate rates of 9, 14, and 18
acres per animal unit for 7 months. This stocking has been followed
during the seasons of 1937, 1938, and 1939, so the pastures have been
grazed at these rates for only three years. The livestock are usually
turned on in January, as soon after the first of the year as the feed is
sufficiently far along to carry them without much weight loss, at least
in the middle-sized pastures. They are usually removed early in August,
after the feed has become short in the small pastures and the animals
have begun to lose considerable weight. During the remaining 5 months
the cows are in pastures 13 and 14 where one group receives supplements
when the feed is deficient in nutritional qualities.

As a preliminary indication of carrying capacity during the five-year
period it was estimated that about 20 to 25 acres per cow, with calf to
weaning time, was needed per year, and this rate of stocking resulted in
fairly close use of the forage. For moderate utilization 8 or 10 acres
more per head was required. Roughly, almost 50 per cent more acreage
has been required from August to February than for the remainder of
the year. The difference in acreage requirement during these periods is
due to various factors, including poor summer feed, and to loss of forage
during the summer through trampling.

The six pastures vary considerably in forage productivity as judged
by degree of utilization at the end of the grazing period. In one instance
this difference in total available forage per animal unit has thus far
consistently offset the difference in size of pastures. Pasture 2, which is
230 acres in size, has shown lighter use every season than pasture 4 with
309 acres (table 4). The differences between pastures in such things as
amount of rock outcrops, swales, inaccessible areas, brush, trees, and
soil depth, greatly complicate any attempt to compare the pastures on
an acreage basis. In the present stage of the experiments the degree of
forage utilization at the close of the annual grazing period affords a more
usable basis of comparison, and is the basis here discussed.

The pastures, as well as spots in each, are not all grazed equally close
every year, because the total forage crop is not the same from year to
year. At the end of the 1937 and 1939 seasons, all the pastures showed
closer use than they did in 1938, when only pasture 3 showed compara-
tively close use — but not so close as during the other two years. At the
end of each season the pastures were rated as to whether they showed
close, moderate, or light utilization. The three main degrees of utilization
pictured in plate 1 were arbitrarily chosen for convenience of the study.
In defining and applying them, however, the pastures were critically
looked over, permanent plots were staked for comparative and repeated

40

University of California — Experiment Station

examinations, many measurements of stubble height were made, and the
visibility or concealment of small ground objects was compared. For
example, on closely utilized areas (plate 1, A) the remaining portions of
filaree were mainly flat on the ground and less than IV2 inches of soft
chess and foxtail fescue stubbles remain. A very small percentage of soft
chess and foxtail fescue plants show no use — approximately 10 per cent
where soft chess makes up 35 per cent of the original cover and foxtail
fescue 10 per cent. Small objects on the ground the size of golf balls
usually show plainly. On areas moderately utilized (plate 1, B) a small

TABLE 4 ^

Estimates of Degree of Utilization Observed at the End of Each
Grazing Period in the Six Pastures

Year

Pasture

1
(small)

Pasture

2
(medium)

Pasture
3

(small)

Pasture

4
(large)

Pasture

5
(medium)

Pasture
6

(large)

1937

1938

Close

Moderate
to close

Close to
moderate

Light to
moderate

Light
Light

Very close
Close
Very close

Moderate

Moderate
to light

Moderate
to light

Close

Moderate
to close

Moderate
to close

Moderate
to light

Light

1939

Light

amount of filaree, approximately 10 per cent, does not show utilization
or trampling. Approximately 50 per cent of soft chess and foxtail fescue
are grazed down to nearly 1% inches. On areas lightly utilized (plate 1,
C), approximately 40 per cent of filaree is not utilized or trampled and
only about 30 per cent of soft chess and foxtail fescue is grazed at all,
and that down to only about 2^ inches. Here, objects the size of a cro-
quet ball are often obscured by the remaining vegetation. It will be noted
that the height of stubbles on differently grazed areas did not vary so
much as the percentage of plants grazed.

Based on such examinations, the pastures were rated as indicated in
table 4. On the basis of average utilization for the three-year period, the
pastures may be arranged in three groups : the closest, pasture 3 ; inter-
mediate, pastures 1, 5, and 4; and lightest, pastures 2 and 6. If the pas-
tures were classified on the basis of four degrees of average utilization,
instead of three, the arrangement would be : pasture 3 ; pasture 1 ; pas-
tures 5 and 4 ; and pastures 2 and 6.

The reader is again reminded that the results were obtained on range
areas grazed for about a 7-months' period, areas where there was defi-
cient feed for only a short time early in the season and again near the
end of the season. Results would probably be different on ranges grazed

Bul. 663]

The San Joaquin Experimental Range

41

for shorter periods, or on a yearlong basis. Further study will be needed
to determine the full effects of varying degrees of use on these ranges
stocked for different seasons — and including bad drought years, not yet
encountered.

Effect of Different Degrees of Grazing on Early Winter Growth. — The
presence or lack of some leftover feed has a marked influence on growth
of new forage during the fall and winter. On moderately and lightly used
areas, growth is usually 2 or 3 weeks in advance of that on areas closely
utilized. This contrast has been consistent during the last four seasons

TABLE 5
Effect on Height Growth of Previous Year's Grazing

Species

Height of plants (inches) on plots utilized at
different rates in preceding season

Closely
utilized

Moderately
utilized

Lightly
utilized

0.30
0.92
0.79
0.67

0 57
1.46
1.19
1.07

1 18

2.14

1.69

Average

1.67

reported on the Range, both between differently grazed pastures, and
between differently grazed spots in any one pasture.

Measurements of height — an important factor in determining avail-
ability of forage early in the season — were made on 120 sample plots
on December 30, 1938, and on 121 small additional areas on January 27,
1939. These observations are indicative of the effects of old feed on new
growth. The December examinations were made near the fence line
separating moderately and lightly used areas from closely used ones. In
the moderately and lightly used areas height measurements of the broad-
leaved herbs (all species, but chiefly broad-leaf filaree) averaged 1.0
inch ; in the closely used areas they averaged only 0.6 inch. Correspond-
ing measurements for the annual grasses (all species combined) were
2.4 inches and 1.8 inches. The January figures were limited to the three
dominant species in closely utilized areas, moderately utilized areas, and
lightly utilized areas. The results, summarized in table 5, show that the
height of broad-leaf filaree in the lightly used areas was over 3% times
higher than it was in those that had been closely used. The leaves were
held in the upright position by the old vegetation ; in the closely used
areas they were almost prostrate on the ground. At this time, broad-leaf
filaree in the lightly and the moderately used areas provided consider-
able feed for cattle, while that on the closely used areas was almost too
short to be grazed. Soft chess and foxtail fescue were fully twice as high

42

University of California — Experiment Station

on the lightly utilized areas as on the closely utilized and about one half
higher on the moderately used plots.

A moderate amount of feed left on the ground at the end of the
season not only offers protection to growth of new feed the next fall and
winter but for a time early in the new season it makes up a large part
of the animals' diet since it is taken along with the young forage.

When the weather warms in the spring, growth of new feed for a short
time becomes very rapid on closely used areas, even faster than where
considerable old feed is left. This partly results from the straightening
up of broad-leaf filaree. Later in the season, after the weather has become

TABLE 6

Estimated Forage Density from 1935 to 1939 in Pastures of
Different Degrees of Utilization*

Year

Very close
(pasture 3)

Close
(pasture 1)

Intermediate

(pastures 4

and 5)

Light

(pastures 2

and 6)

1935

per cent
35
30
31
36

per cent
49
36
44
45

per cent
44
28
33
40

per cent
44

1936

36

1938

40

1939

47

* Exclusive of 1937 when densities were not recorded. Density percentage refers
to proportion of the ground area covered by vegetation.

still warmer, height growth in the moderately to lightly used areas again
surpasses that in the closely grazed ones.

Effect of Different Degrees of Grazing on Forage Yield, Composition,
and Density. — As a result of the different degrees of use over the 7-
month period each year from 1937 to 1939, certain changes in range-
forage conditions have shown up. Although no pronounced changes due
to grazing have been detected in either total forage density (table 6) or
species composition, the effect of very close grazing on height growth of
forage has been apparent. From measurements made in a series of fenced
plots established during the third year of this pasture experiment, the
average height of forage in the closest-grazed pasture (no. 3) was ap-
proximately 1 inch shorter than in the adjoining lightly used pasture
(no. 2).

Slight differences in forage yield were noted in 6 fenced plots placed
at random astraddle and at intervals along the fence line separating pas-
tures 2 and 3, in the fall of 1939 before any stock were turned in. Later
from each of these plots the vegetation on 40 random samples in open
areas was measured, clipped, dried, and weighed. The yield from the
plots in the lightly grazed pasture was found to be slightly greater than
in the closely grazed one, although the difference was not significant.
Comparisons in yield for the pastures as a whole are not yet available.

Bul. 663] The San Joaquin Experimental Range 43

It has also been observed that the litter on the most heavily grazed
areas is thinning out. More bare soil (less litter-covered) is becoming
evident, in comparison with the more lightly grazed areas.

Another interesting comparison of range forage conditions between
areas grazed most heavily and others grazed less closely is afforded by
the cattle weights — difficult as they are to interpret at this stage of the
experiments. Roughly, pounds gain per animal unit decreased as close-
ness of use increased. For example, the gain per animal unit was small-
est on the closest-grazed pasture (no. 3) where "the lower gains and
thinner condition of the animals show that the degree of feed utilization
was beyond optimum for the well-being of the animals and for maximum
net returns." For a more detailed discussion by the animal husbandmen,
see the section "Experimental Herd Management," which presents the
data showing the effects on the cattle.

Obviously, some undesirable conditions are developing on the heaviest-
grazed areas, but whether these trends will continue and what their
cumulative effect on the range may be will require a longer period to
determine.

Comparison Between Grazed and Protected Areas. — In contrast to the
more stable plant associations the annual forage cover, if unused, may
change readily and rapidly. The rapidity and extent seem largely de-
pendent on the rate of accumulation of litter on the soil ; hence indirectly
on soil productivity. When production is high and litter accumulates
rapidly, rapid changes in the cover have been recorded ; when production
is low the observed changes have been comparatively small in extent and
slow in occurrence.

In areas on which old litter has begun to pile up as a result of nonuse
the trend of change in composition is toward shade-enduring species of
plants. If the litter becomes exceedingly heavy, as it may in only two or
three years on highly productive soils under favorable growing condi-
tions, even the most aggressive species may be unable to penetrate it in
spots ; consequently both density and yield may be reduced. On the other
hand, where the litter accumulates slowly on poor sites, there may be
comparatively little change in plant mixtures for several years following
protection from range livestock and rodents. These observations on what
happens to forage when grazing is excluded have direct bearing on
management.

Supporting data are now available from several areas. One is a series
of small plots on comparatively shallow soils, fenced against rodents
and livestock, established on the Range in 1935. Another is a tract of
about 80 acres in the natural area (fig. 1) which has not been grazed
by livestock since the station was established in 1934. The third location

44 University of California — Experiment Station

is a set of small supplementary plots established in 1933, 20 miles east-
ward on the Sierra National Forest, near Northfork, at a higher eleva-
tion (2,300 feet) where the plant yield has been greater than on the
plots at the Range. The fourth area was a plot in productive soil about
20 miles in the opposite direction at a lower elevation, fenced against
livestock for the past ten years.

In the small completely protected plots on the Range broad-leaf filaree
increased for one year, then decreased noticeably by the fifth year. Fox-
tail fescue decreased considerably and was not recorded on the plots in
1939. A marked increase was recorded for soft chess. In general, two
species of plants accounted for the most conspicuous changes in cover in
these plots in comparison with the grazed pastures during the same
period. Slender oat changed but little in the pastures, yet increased
several-fold in the tightly fenced plots; and foxtail fescue practically
disappeared from these plots but persisted in the pastures without dis-
cernible change.

Forage density on these same plots showed no marked change during
the period of study. The density in 1935 was 41 per cent ; in 1936, 37
per cent ; in 1937, 40 per cent ; and in 1939, 43 per cent. The average was
40 per cent for the five-year period. These density estimates from year
to year are not greatly different from those in the grazed pastures, where
the five-year average was 38 per cent.

In the natural area where livestock were excluded but rodents were
undisturbed, there was a general decrease in broad-leaf filaree, a notice-
able increase in foxtail fescue, and an even larger increase in soft chess.

Changes in forage were more spectacular in the outpost plots, 20 miles
away, than on the Range. In these plots the soil was deeper, forage pro-
duction heavier, and the rodent population lower. Grazing around the
plots was heavy in 1934, 1935, and 1936 ; light in 1937 ; and practically
nil in 1938 and 1939. Systematic estimates and measurements of vegeta-
tion were recorded in 1933, 1935, 1937, 1938, and 1939.

When the protected plots were established at this outpost in 1933, 98
per cent or more of the cover consisted of broad-leaved herbs (table 7),
the chief of which were filarees. Two years later, in 1935, a general
increase in grasses was recorded for all the plots, but it was much
greater on the protected ones. After grazing was lightened on the area
grazed by livestock and rodents, grasses continued to increase and by
1939 comprised about 75 per cent of the cover.

Some of the changes indicated in table 7 are of course due to climate ;
but the replacement of broad-leaved herbs by grasses is significantly
greater in the fenced plots. The change was also surprisingly rapid. In
the completely protected plot the change from 98 per cent herbs to 96

Bul. 663]

The San Joaquin Experimental Range

45

per cent grasses occurred in two years — changes far greater and more
rapid than those observed on the Range proper, and probably unusual
for the forage type as a whole.

The response of this delicately balanced vegetation to extremes of
treatment has been further evidenced by the appearance of undisturbed
portions of protected plots compared with portions from which the
accumulated old dry growth has been carefully removed by hand at the
beginning of the growing season. Such comparisons have been made at

TABLE 7

Changes in Percentages of Broad-leaved Herbs and Grasses on Plots

Under Different Grazing Treatments at the Outpost

Treatment of plots

Plant group

Percentage in forage cover
for various years

1933

1935

1937

1938

1939

I Broad-leaved herbs

\ Grasses

J Broad-leaved herbs

98

2

99

1

98
2

4

96

15

85

79
21

1

99

4
96

67
33

20
80

20
80

43
57

1

99

1

99

f Broad-leaved herbs

\ Grasses

22

78

both the Range and at the outpost. The differences in growth on the
cleared and uncleared areas were especially marked in the winter of
1939-40. Some of the slender oat was 24 inches high and heading on
February 23 on the protected area, while on the cleared area, in contrast,
slender oat was lacking and broad-leaf filaree was only slightly over 1
inch in height. This striking contrast was conspicuous for weeks as
indicated in plate 2 (upper) , which shows conditions in early April.

The plot fenced against livestock for ten years prior to examination
in 1938 was in a deep-soiled area where bur-clover, red-stem filaree, and
common foxtail are common. Among the observed changes here were the
following : Comparatively inferior grasses had increased at the expense
of more valuable forage plants — chiefly herbs. Ripgut grass composed
75 per cent of the cover and common foxtail 5 per cent. Soft chess and
slender oat together made up 5 per cent. In the outside grazed pasture
the grasses, mostly soft chess and foxtail fescue, composed only 50 per
cent of the cover ; bur-clover, red-stem and broad-leaf filarees made up
most of the remainder. These findings indicate that on low-yielding sites
only slight changes in forage composition and density may take place
for several years following complete protection from all grazing. On
high-yielding areas, however, the litter accumulates rapidly and com-

46 University of California — Experiment Station

paratively tall and inferior grasses may soon crowd out some of the
shorter and better herbs. The indications are that on such productive
areas the degree of grazing can be used, within broad limits, to keep
collective groups of plants in approximately the desired proportions.
Thus, not only can annual-type ranges be grazed too closely, but some
areas can also be grazed too lightly — at least from the standpoint of
maintaining a high percentage of herbs.

The Need for Two Broad Classes of Range Indexes. — In the coopera-
tive studies reported in this bulletin two classes of indexes of range stock-
ing and utilization are employed and are being further developed to aid
in understanding: (1) what happens to the grazing animal, and (2)
what happens to the range itself — under any given system of manage-
ment. The first class is dealt with in detail in the following section.
Obviously the stockman should utilize fully this simpler and more widely
recognized class — livestock, the end product of his business. The second
class of index supplies additional information on whether a range "is
going uphill, downhill, or holding its own" — basic information for a
permanently sustained grazing industry.

Simple plant indicators which reveal general trends of range condi-
tions have been fairly well developed in other western range types, but in
the California annual type it is doubtful whether equally simple plant
indicators will be discerned. More dependence will likely have to be
placed on groups of plants.

A major purpose of the forage-management studies is to build up
better range indexes or indicators, more particularly to aid the stockman
during bad years when he needs the information most. In good years the
need may not be so important. Unfortunately (from an experimental
standpoint) the period covered by these studies has not yet included bad
drought years ; consequently we do not yet know, for example, whether
or to what extent a short recuperative period of one or two good years
(following drought) may overshadow the effects of several years of
grazing at different degrees of closeness.

Even a well-treated range may present vexing problems to the range-
livestock operator in a critical drought year. A range that is running
down may seriously intensify these problems. An underlying objective
of the studies is to work out techniques both in range handling and live-
stock handling so that the shock of these critical years may be reduced to
a minimum.

Thus the two classes of range indexes (animal and plant) complement
each other. Both are important in the appraisal of general range condi-
tions and especially so in this type of range forage because of its big and
sudden changes.

Bul. 663] The San Joaquin Experimental Range 47

SUMMARY OF THE PRINCIPAL RESULTS OF THE FORAGE STUDIES

Studies at the San Joaquin Experimental Range show how many of
the range-management problems of the foothill section are directly re-
lated to the character and behavior of the forage crop.

Range-management practices are built almost entirely around annual
plants, which compose about 99 per cent of the herbaceous forage cover
on the Range. The sparse perennial grasses have shown no increase dur-
ing the period of study, even in completely protected places.

Introduced species, mostly from the Old World, account for a sur-
prising portion of the forage cover ; they average over 60 per cent.

The herbaceous plant cover consists of some 225 species (190 broad-
leaved plants, 30 grasses, and 5 grasslike plants), but 10 of these com-
prise three fourths of the cover, and 2 have made up one half of the
livestock forage.

The ratio of broad-leaved plants to grasses was 3 : 1 in 1935 and 1 : 1
for each of the following four years.

Of the broad-leaved plants on the Range, broad-leaf filaree is the most
common and important species, and averages 31 per cent of the cover.
Since this introduced plant is aggressive, occurs in abundance, and can
be depended upon to furnish feed under a wide range of conditions, it
is rated as one of the two most important forage species on the area. True
clovers, Spanish clover, and popcorn flower are rather sparse but are
important range plants at certain seasons of the year. Tarweeds and
turkey-mullein are undesirable forage plants but have some value as
plant indicators.

Soft chess, another introduced annual, is the second of the two most
important forage plants of the area ; it averages 19 per cent of the cover.
This plant is the most abundant grass and is especially important during
the summer. Foxtail fescue, a native annual, ranks second in abundance
(13 per cent) among the grasses and third among all of the range plants.

Pine bluegrass, a native plant, is the most common perennial grass
on the area, although it makes up less than 0.2 per cent of the cover.
Other perennials, still more sparse, are melic and purple needlegrass.

Several native rushes, perennial grasslike plants, are sparse on the
Range, yet they composed a considerable part of the diet of the experi-
mental herd of cattle near the end of the green-feed period when most
of the other forage had dried.

Several browse species occur on the area but none makes up a very
large percentage of the animals' diet. Their greatest importance prob-
ably lies in the fact that they provide some green feed during the dry
period when the forage crop is deficient in protein and vitamin A.

48 University of California — Experiment Station

Three distinct forage periods underlie the whole problem of foothill
range management. The first extends through 4 or 5 months of the
summer when feed is mainly dry and low in nutritive values ; the second
is a period of 3 or 4 months during the fall and winter when green feed
starts at the beginning of the rainy season but makes little growth and
is generally uncertain or insufficient ; and the third period covers 3 to
4 spring months during which rapid growth occurs and the principal
forage plants ripen. During the first two periods the forage is usually
so poor in either quality or quantity that it does not provide a suitable
and dependable ration for cattle.

Most forage plant growth takes place in 4 to 6 weeks during the
spring.

Annual-type ranges are exceedingly difficult to stock because the
period and rate of growth and the yield of forage vary greatly from
year to year. The proportions of the different kinds of plants making
up the forage cover also vary considerably, and affect nutritive qualities
of the feed and complicate supplemental feeding.

The wide and irregular changes in forage produced from year to year
in this annual type necessitate skillful operation on the part of the
stockman, in providing supplemental feeds or additional acreage during
adverse seasons, and in other ways keeping his business as flexible as
possible.

Climatic factors have most to do with yearly fluctuations in forage
yield, composition, and season of growth. A seasonal variation in either
precipitation or temperature may cause a marked fluctuation in the
feed. Rainfall and temperature operate together. The effects of weather
on feed appear to be most pronounced early in the growing season and
again late in the season ; 10 to 12 inches of rain with favorable distribu-
tion and temperatures may make a better forage crop than 25 to 30
inches with unfavorable conditions. When more is known about the rela-
tion between weather and forage, the operator will be better equipped
to deal with the changes and uncertainties of feed that prevail on this
range type.

Successful, sustained ranch operation depends upon definite knowl-
edge of what happens when a range is grazed to different degrees and
for different periods. The effects of approximately 7 months of heavy,
moderate, and light use on this type of foothill range have been studied
in a set of 6 pastures — two small ones, two medium-sized, and two large,
averaging approximately 155, 240, and 312 acres in size. All have been
grazed with approximately the same number of animals, equivalent to
about 9, 14, and 18 acres per animal unit in the various-sized pastures.
Considerable variation exists between the pastures and they are not

Bul. 663] The San Joaquin Experimental Range 49

grazed uniformly ; consequently, degree of utilization at the end of the
grazing period offers the best basis of comparison.

Three degrees of utilization (close, moderate, and light) have been
recognized and described by camera and word pictures. One index of
utilization is stubble height ; another is the extent to which small ground
objects of different sizes are screened or masked by remaining ungrazed
vegetation.

These experiments must run for several years before sound conclu-
sions can be reached ; nevertheless certain definite trends are apparent
from results to date.

Some leftover dry feed means earlier green feed. New green growth
has been available from 2 to 3 weeks earlier on areas where some old
feed has been left to serve as protection than on areas closely used the
previous season.

Under the conditions of this study, forage composition and density
were not appreciably affected in three years by the different rates of use
for the grazing period of approximately 7 months each year.

Surface soil litter on the most closely used areas has decreased, but
the full effect on future composition, density, and yield will require more
time to determine.

This annual forage cover under certain conditions may change very
readily if completely protected from grazing. On low-yielding sites only
slight changes in forage composition and density may take place for
several years following complete protection. On high-yielding areas,
however, the litter accumulates rapidly and comparatively tall and
inferior grasses may soon crowd out some of the shorter and better herbs.
The indications are that on such productive areas grazing can be used,
within broad limits, to keep collective groups of plants in approximately
the desired proportions. Thus, not only can annual-type ranges be grazed
too closely, but some areas can also be grazed too lightly — at least from
the standpoint of maintaining a high percentage of herbs.

A major purpose of the forage-management studies is to develop better
indexes or indicators of range condition, more particularly to aid the
stockman during bad years when the information is most needed. Indexes
which reveal what is happening to both the animal and the range, under
any given method of management are, therefore, important in range
appraisal in this annual type.

EXPERIMENTAL HERD MANAGEMENT
K. A. Wagnon, H. R. Guilbert, and G. H. Hart

The College of Agriculture has for several years studied problems in
beef -cattle and sheep production on the range lands that comprise 60
per cent or more of the total area of California. This work has shown
the common occurrence of low-percentage calf crops (11) and deficien-
cies in the range feed with respect to several dietary essentials — notably
protein, phosphorus, and vitamin A (7, 12, 13). It has also been demon-
strated that the low value of the dry feed is further reduced by leaching
when summer rains occur (9) .

There are many important practical problems involved in raising
livestock in the area where the Range is located. The green-feed season
is relatively short, and a great variation exists in the total dry matter
produced per acre from year to year. Some browse is available, although
the native species are such that only limited quantities are utilized by
the grazing animals. Some ranchers depend upon summer ranges in the
mountains under a U. S. Forest Service permit; others obtain fall or
winter feed on patented lands in the San Joaquin Valley, and some
obtain feed from both types of outside ranges. Still others maintain their
herds on home foothill patented land on a year-round basis.

The majority of these cattlemen maintain breeding herds. In favor-
able seasons veal calves and some grass-fat cows, heifers, and steers are
marketed from the area. The difficulty of producing satisfactory slaugh-
ter animals for increasingly discriminating markets has necessitated the
sale of greater numbers of cattle as feeders.

Many cattlemen in the state are faced with problems of yearlong
operation on lower ranges ; and since so large a part of their investment
is in foothill lands, the objectives in the experimental work reported here
placed major emphasis upon long-time investigations bearing on the
general problems of yearlong utilization. It was desired to ascertain the
kinds and amounts of supplements necessary to correct nutritional de-
ficiencies of the range, and thereby increase efficiency of utilization and
promote productive efficiency of the breeding herd and normal eco-
nomical development of the young animals. Also desirable was a long-
time study of the effect of natural environment on the development of
animals and on the possible improvement through breeding and selection,
comparisons to be made with similar animals receiving supplemental
feeds. A further problem was the demonstration by actual trials whether
or not it would be economically preferable to turn off finished cattle or to
sell offspring as calves, yearlings, or two-year-old feeder steers and

[50]

Bul. 663] The San Joaquin Experimental Range 51

heifers. One major objective was to ascertain what degree of forage
utilization is consistent with the animals' welfare and with maintenance
or improvement of the range.

Under agreement with the U. S. Forest Service, the University pur-
chased a foundation herd of cattle to be supervised by a member of the
Animal Husbandry staff in residence at the Range:

The foundation females were medium to good, grade Hereford heifers
bred to purebred bulls. Originating in New Mexico, they were shipped to
Madera County as yearlings. They were free from tuberculosis and bo-
vine infectious abortion and since the Range lands had not been grazed
since about July 1, 1934, complications from these two diseases were
definitely eliminated. To study the possibilities and limitations of a
breeding-up program on an average group of females in such an environ-
ment, purebred Hereford bulls were supplied from the University reg-
istered herd.

METHODS OF PROCEDURE

On June 14, 1935, the 70 head of two-year-old, bred heifers were
brought to the Range. They were run together and received supple-
mental feed during their first calving, which occurred during the ensuing
fall and winter.

On January 10, 1936, they were divided into two comparable groups,
A and B, of 33 head each, together with their calves. The remaining three
animals (one having died) were kept for possible replacement purposes.
Group A was to receive supplemental feeds during the late summer and
until good green forage was available, while B was to have no supplement
except in emergency to prevent death loss. Calf crops from the respective
groups would receive the same treatment as their dams. These groups
were to remain intact over several years, and replacement was to be made
with heifers within the respective groups. The effect of using different
sires was minimized by annually alternating the bulls used early in the
season when the two groups of cows were separated or by using closely
related bulls of comparable quality. Later in the breeding season groups
A and B were together in the six grazing-intensity pastures described
below and were bred to the same bulls.

As female get from the cows reached breeding age, those of satisfac-
tory quality were used for enlarging the herd or for making replace-
ments. Cows were culled for poor type and for barrenness, poor lactation,
extreme wildness, and other manifest disabilities.

Concurrently, a series of investigations was designed to obtain data
on the effect of three different intensities of grazing upon the cattle and,
as described in the preceding section, upon the range vegetation.

Pastures 1 to 6 inclusive (fig. 1) were used in the* grazing-intensity

52 University of California — Experiment Station

studies. Pastures 1 and 3 were stocked at the approximate rate of one
animal for each 9 acres ; pastures 2 and 5, one animal for each 14 acres ;
pastures 4 and 6, one animal for each 18 acres. Variations in this general
procedure will be discussed as the data are presented. Each pasture con-
tained an equal number of cattle of groups A and B. Unless culled from
the herd, the same animals were returned to the same pasture each year.
The entering date was the time when the new green feed was considered
abundant enough to carry cows suckling calves satisfactorily without
supplement. For the years reported in this publication, this time came
in January or February. The animals remained in these pastures until
August or September. The criterion for removing the cattle consisted in
evidence that the feed in the smallest pair of pastures had been closely
utilized. Upon removal, the cattle were segregated into their respective
groups and placed in separate holding pastures until the next green-feed
season. Pastures 11 and 12, until 13 and 14 became available in 1938,
were used mainly as fall holding pastures for the cow herds. The other
enclosed fields shown in figure 1 were used to pasture weaner calves,
yearlings, and other stock.

Supplemental feeding began, for calves, at weaning time; for other
group-A animals, when the cattle were removed from the six grazing-
intensity pastures. The amounts and kinds of supplements used varied
with the amount and character of the range forage and with the condi-
tion and needs of the cattle. Cottonseed cake (43 per cent protein) and
rolled barley were the principal feeds used. In general, when dry feed
was plentiful, the amount of cottonseed cake necessary to correct protein
deficiency (1-2 pounds daily) sufficed to promote gains and efficient
utilization of the forage. After the first rains and when green feed was
scant, additional barley was used whenever it was a cheaper source of
the extra total nutrients needed for maintaining milk flow or for pre-
venting weight loss in young animals. The maximum supplement, usually
3 pounds daily, had to be increased to 5 pounds under the extreme condi-
tions of 1937-38 winter period. Unless otherwise stated, supplemental
feeding was done daily in the morning. The feed was placed in troughs,
and the cattle were trained to come to their feed by calling them.

The original females dropped their first calves mostly from October to
January, a time somewhat earlier than is the practice in the area. It was
desired, however, to investigate the practicability of early calving in
connection with the supplement-feeding program. Placing the bulls with
both groups at the same time was necessary because they were pastured
together during most of the breeding season. The bulls were placed with
the cows about December 28 and were removed early the following May.

Either coarse-erround or block salt was available for the cattle at all

Bul. 663] The San Joaquin Experimental Range 53

times. Observations were made on the relation of its accessibility to the
movements of cattle on pastures.

Water on these pastures was developed to a greater extent than on
many California ranges and, being more accessible in relation to the
feed supply, resulted in better average utilization of the feed over the
entire area than is frequently the case.

The main corrals and scales shown in figure 1 were built at the start
of the experiments. In 1939, a second set was established on the division
fence between pastures 13 and 14. The experimental animals became
gentle and were handled through the corrals and scales with care, so
that shrinkage in handling was reduced to a minimum.

Weights were taken when there were major changes in the method
of procedure such as moving into the six grazing-intensity pastures at
the beginning of the year, the end of the green-forage period, weaning
time, starting of the supplement-feeding period, and just before calving.
Weaner calves were weighed monthly during the first period that they
were fed supplements. All weights presented in this publication were
taken on a "full" basis; the animals were driven in from pasture and
allowed to drink before weighing.

The animals were individually identified numerically by tattooing the
ears, and by branding with caustic fluid for field identification. Indi-
vidual records were kept. Over the period covered by this report, one of
the authors collected forage samples by hand, following the cattle closely
to take samples of the forage as grazed. He also observed the grazing
habits and movements of the animals.

EXPERIMENTAL DATA

During the period when the data here presented were being collected,
calendar years 1935 to 1940, both seasonal rainfall and temperature pro-
duced favorable growth of feed, which consisted almost entirely of an-
nual species that would cause these years to be classed as reasonably
good. Before the experiments were started, forage samples were being
collected from this locality for part of the data published by Hart, Guil-
bert, and Goss (13). During that period there were much less favorable
feed years, and forage was considerably less abundant. Nevertheless,
during the time covered by the data in the present publication twice as
much feed was produced in some years as others. Since feed conditions
varied so greatly from year to year and since the data on different age
groups are so limited, in most cases no attempt has been made to present
average results for the entire period. Data covering additional years
of work must be secured before definite conclusions can be drawn.

In table 8 and in subsequent paragraphs, the times when the breeding

m

H

fe

W

a

a

7}

Oh

n

£

£

UQ

w

n

o

i

3

p

H

§

£

O

H

Pm

P

o

h

M

O

H

ft

i— (

«4

PQ

Oh

o

o

Eh

©

3

-f

N

Cw

A

En

ft

0

3

pq

2 & .Moo

4) -OS— ' s?

0 "O ^ ~ 5"
ft 8. £§3

'pfcn

BQ

05 CO
CO 05^.

£ "° _- 5"
N.Sm .T3
2&M Moo

°9=3

- ->>

fe o

"ONOMC

•*»< CO CM *0 r

WONM'

■» CD N GO rt OO

0500 | pq

to i— ■ ■* CO CO

1 §2.

& CD .J2 05

a a M cu <35

m

•J3 O .t-i C3

CD -05.-< <J
CQ "SO

N-SO .'

a« . coo
§-aa|2

m CO O

WCOrHlO.

f OOOi
OOOH

t^ o> oo 05 co
"cdojm

■ NNlOm

TON>ONH

~ +

lONOOOO
CM CO lO

I

* CM IC CO CO

CO •"* t— CO CO
CO-^i 00 "* rt

CO ^-*i t- I
OOOO I

CO »o ■* OS CO

CCWtDO'H

I Oi y-< CM CO
) -"*< r- CM rt
t~ 00 >-H

+

OS CO >— I CM O OS <

(DCBOtOrt CO <

§«1

33-5 S g'

OM <©CN CO

CON «©•<*« CO

■»*CN« CO

CMf~ »-h CO

■«f (M CM

OO M co
■ l>- CO N. i— i co

CM OO

NMOiO
CO CO O0 r-l

OO COO

COCM 00 .-c ^H

I 05 OcM 0O

O0 CM lO

MCOOOrtCO-^
COCM t^ CM ^H

£^«E

.-<*< o-

CO <L .

•* CM CO

■IO00 CO CM CO

OC"!(l»t^H

»3

ilSeJB

<B C3 +J

is £-a

03 £ M

m '£

icsJsll

O a OT3 UOOOS
<hih CO CU «*-" Q>«*i CD P
OOmSoOmOm^ot
S CD'S

) MM O

3J.15PS»S.S

©s

M (H

03 cu
Ma

c c

III

CD 43

S M

T3 fc

9 CD ^ CD

£t3 O M

CD £ CD CD

^ O fl >

sa

||

5 ll

3s §^

gi SB

St) 13^
-O 03 «•«

■j J Is

'P ^c3^

Lais

OS'S

Bul. 663]

The San Joaquin Experimental Range

55

herds were confined to the six grazing-intensity pastures are designated
as the "grazing periods." The times when the A and B herds were sep-
arated in the fall and winter pastures and when group A received sup-
plemental feeds are designated as "supplement periods."

The principal data obtained from the breeding herds for 1935-1940
are presented in table 8.

Weight Variations of Breeding Cows. — The average weight of the 70
foundation heifers when delivered on the Range, June 14, 1935, was 794
pounds. Although Jhere was a plentiful supply of forage, the animals

Fig. 10. — Foundation heifers as they appeared September 24, 1935, at the
beginning of their first calving period.

just maintained weight up to September 24. They had advanced in preg-
nancy during this time and had actually lost considerable flesh. They
were moved to another pasture of ungrazed dry forage, and supple-
mental feeding started. The appearance of the heifers at this time is
shown in figure 10. Ninety-four per cent of these heifers calved between
September 9, 1935 and January 10, 1936. Milk supply on dry feed with
supplements was satisfactory for first-calf heifers, green feed developed
early, and most of the weight loss at calving had been regained by the
end of the period. Table 8 shows the subsequent pasture gains for the
grazing period January 10 to August 4, 1936, for the cattle now divided
into two permanent groups.

During the feeding period beginning August 4, 1936, until second
calving, animals of group A (fed supplements) gained on dry forage
and 1 to 2 pounds of cottonseed cake, while group B lost weight. After
calving, group A received up to 3 pounds of supplements daily. They had
good udder development and the calves were thrifty. The animals of

56 University of California — Experiment Station

group B, though given twice as much acreage for grazing, were con-
spicuously lacking in udder development after calving on dry feed ; and
the calves looked rather "dry-haired." Green forage again came on early,
however, evidently before stimulus for milk production was lost, since
weaning weights of the calves from the two groups did not differ sig-
nificantly when ages and sex were considered. Though there was an 81-
pound average weight loss in group B and although the animals were
thin, they were in strong condition at the end of the period. A snowstorm
followed by cold weather occurred after the beginning of the subsequent
grazing period (January 2 to August 10, 1937). The forage regressed
and the cattle lost weight during the first 2 months. During the 1937
grazing period, animals of group B gained more than group A, because
those of the latter were in better condition at the beginning of the period.

During the early part of the supplement period August 10, 1937 to
February 25, 1938, group A animals again made some gain on dry forage
plus cottonseed cake. Dry feed became scarce in December, and cold
weather prevented significant growth of new forage until February.
During this time double quantities were fed every other day. The amount
was increased to 8 pounds every other day during part of the period to
prevent excessive losses in weight and to maintain milk flow. Despite this
liberal feeding, the animals were 158 pounds lighter at the end of the
period than before calving. The inclusion of two-year-old, unbred heifers
in the averages shown in table 8 somewhat obscures the results with cows
that calved. The loss in weight of group A cows that calved was 239
pounds; that suffered by the corresponding group B cows was 331
pounds.

Group A carried some reserve flesh at the beginning of the period and
remained strong ; the calves did well. Group B animals became extremely
thin and weak. Two calves starved because their dams failed to lactate,
and one cow died of exposure and of the infection resulting from a re-
tained placenta. To prevent serious death loss, supplemental feeding was
started. Three more cows died subsequently as a result of the impoverish-
ment. The others soon showed marked improvement. They then traveled
to outlying portions of the pasture for forage, whereas they had pre-
viously stayed close to water and salt, and thus accentuated the effect of
feed shortage by limiting their grazing area. The calves from group B
appeared unthrifty. Many of their mothers were "dry" or nearly so be-
fore good forage was available.

The gains of the cows were large during the next grazing season (Feb-
ruary 25 to September 2, 1938) after these heavy losses. Again group B
animals gained more during the grazing period.

Eight two-year-old, unbred heifers were added to each group at the

Bul. 663] The San Joaquin Experimental Range 57

beginning of the supplement period of September 2, 1938, to January
25, 1939. Pastures 13 and 14 became available and were used for the
first time for groups A and B respectively. Although the acreage per
cow was somewhat greater for group A, there was no conspicious dif-
ference in the amount of total forage available for the two groups. Be-
cause of the short time that remained before calving, supplement-feed-
ing with group A was started at the rate of 4 pounds every other day.
After rains in late October the amount was increased to 6 pounds on
alternate days and later decreased as the forage improved. Table 8 shows
the total supplements fed. The average weight losses of the groups (as
shown in the table) again obscure results on the cows that produced
calves, not only because two-year-old heifers were included but also
because there were more dry cows in group B. The weight loss of cows
calving in group A and group B averaged 155 and 268 pounds respec-
tively. In general, group B cows and heifers were thin, especially those
that had also calved the previous year. The calves from these animals
showed effects of lack of milk compared with those of the group fed
supplements.

In the grazing season, January 25 to August 1, 1939, there was, on
account of low rainfall, an absence of late-growing green feed in the
swales. Maximum weight was attained in May, when all feed dried and
the cattle lost weight until removed from the grazing-intensity pastures.
Net gains for the period, therefore, were less than in previous years.

At the beginning of the supplement period, August 1, 1939, to Feb-
ruary 8, 1940, group A received daily 1 pound of cottonseed cake. Rains
started forage growth in September. Lack of rain thereafter resulted
in regression of the forage in December. In the interim, while forage
was good, group A refused supplements. Table 8 shows the total supple-
ments fed. There was over 100 pounds' difference in the average weight
of the two groups of cows at the end of the period, though five less
animals in group B had calves.

Death Losses and Retained Placentas. — Table 8 shows a loss of two
cows in group A and six cows in group B, during the entire five-year
period ; one animal had died before the groups were originally divided.
The deaths in group A resulted from eversion of the uterus at calving.
All the losses in group B were caused by impoverishment, with or with-
out complications from retained placentas and subsequent infection.

As table 8 shows, there were 37 retained placentas in group B, and 11
in group A. Some of these cows received treatment ; others did not. Treat-
ment doubtless saved additional death losses in the thin group B ani-
mals. The effects of these complications on subsequent pregnancies was
not serious, for little or no permanent sterility resulted. Only during

58 University of California — Experiment Station

the 1939 calving season did similar numbers of cases occur in each group.
In that year green feed started early ; group A refused supplements for
a time and then forage became poor. Forage on north slopes in pasture
14 suffered less from drought than that on the south exposures of
pasture 13, and the animals not fed supplements had more favorable
feed conditions than usual.

Calf Crops and Weaning Weights. — As previously stated, 94 per cent
of the original heifers dropped calves the first year. Seven of these off-
spring were lost shortly after birth, four being killed by coyotes.

Because these cattle calved earlier than is customary in this area, there
was an accumulation of coyotes upon the range. Although actual attack
or slaughter of a calf by a coyote was not observed, K. A. Wagnon
counted twelve separate coyotes in the fields containing the cattle one
afternoon and observed a coyote chasing a calf. In all cases the calves
attacked were isolated animals hidden by their dams away from the
group. The Biological Survey, having been requested to hunt the Range,
trapped 35 coyotes in 30 days. No dogs or other marauders were seen
by the trapper or the Range personnel. Since this time coyotes have been
hunted annually during the calving period and no more calves have
been lost from this cause. The percentage calf crop weaned (table 8)
was 84.3 per cent.

The 1937 calf crop represented the second calves from the original
heifers. Both groups were being treated alike at the time of breeding,
and no significant difference in the calf crop is shown. The third crop
of calves (dropped between October 3, 1937, and January, 1938) was
the first in which variations in reproduction resulting from the difference
in feeding management of the two herds could be expected.

Twenty-seven calves were dropped in group A ; 25, in B. Four calves
(one premature, one still-born, and two whose mothers died from ever-
sion of uterus) were lost in group A. Five were lost in B from causes
directly attributable to undernutrition. The difference in percentage
calf crop weaned is shown in table 8, which also reveals nearly 100
pounds' difference in the weaning weights of the two groups.

The fourth calf crop (dropped mostly between September 25, 1938,
and January 1, 1939) included calves from first-calf heifers that had
been added to the herd as well as from the mature cows. A still birth
from a group B heifer was the only calf lost this season. The difference
in percentage calf crop between the two groups is, therefore, due mainly
to the number of dry cows. There was little difference in percentage calf
crop from the heifers of the two groups. Calf crop from group A cows
was 91 per cent in contrast to 61 per cent for group B, the cows of which
were very thin the previous winter. Besides the difference in percentage

Bul. 663] The San Joaquin Experimental Range 59

calf crop, there was also a significant difference in weaning weight of
the calves from the two groups. This difference was greater between the
calves from heifers than from the mature cows of the two groups.

The fifth calf crop was dropped between September 26, 1939, and
January 23, 1940. The percentage calf crop born was 94 per cent in
group A, 73 per cent in group B. These subdivide by age groups as fol-
lows: the percentage from group A cows was 100 per cent and from
group B, 77 per cent ; from second-calf heifers of group A, 78 per cent
and from group B, 58 per cent. For the years 1938, 1939, and 1940 the
percentage calf crop and weaning weight difference between the two
groups is largely attributable to the supplements fed group A cows. The
total pounds of calf production at weaning time per breeding cow for the
two groups was as follows :

Year Group A Group B Difference

1938 335 234 101

1939 432 294 138

1940 424 333 91

If supplement-feed cost is computed at 1.5 cents per pound and
the value of all calves at 8 cents per pound, the added income over feed
cost for group A varied from $3.03 to $6.36 per head. In addition to this
difference in pounds of production per breeding cow under the two sys-
tems of management, the margin is further widened by the fact that the
calves from group A have been appraised each year at about 50 cents
per hundredweight higher than those of group B.

During all the periods covered, the calf crop was influenced by five
abortions, four of which were in group A. Three of these cows, when
tested, proved negative to the Brucella abortus agglutination test. One
cow aborted after severe illness, the cause of which was not ascertained.
None of these cases could be attributed to B. abortus infection or to
vitamin-A deficiency.

Weaner Calves. — Each year the calves have been weaned about July
1. The two calf groups A and B, were separated and held in corrals until
weaned. Both groups were fed concentrates in addition to hay during the
few days at weaning time in order to get them started on supplements
and so that the group B animals might more readily take supplements
should emergency arise later. After weaning, the groups were placed in
separate pastures, usually a combination of pastures 7, 8, 9, and 10.
Separate weight records were kept on steers and heifers. Supplemental
feeding was continued with group A. Table 9 gives the data for the 1935,
1936, 1937, and 1938 calf crops.

The 1935 calf crop (designated as groups A5 and B5 to indicate year
of birth) was from mothers that had received similar treatment and was

< CQ
O

« P

Eh O

DQ

w
o
pa

w

O EH

tf Ph

g g

3 M

■ H

o
ft

Eh
H

H

vl

Pi

Ph

P

co

CM

CO

© 03

CT> CM CO

©

ffi

^H CO

© CO *«

03
03

<3
'5

CM

"* +

+

»o

-O

w

»rt

c^2'm

CO

2 s

CO CO CM

©

OS t-

CM m

0> . >>

CM

■rti m ^

CM

Geo S

«** o> cm

+

+

CM
CM

0>

Ph

co

CD -*l
^ CNI

CO CO o
CM t^ IC3

©

g

CM

t*h r}(

s

CP

CQ

+

+

us

CO

uo »o

C3 CO T*

©

OS t^

<1

CM

rt Tf (M

lO CD i-i
+

+

CM lO
CM

©

ffl

CO OS

00 •* to

©

CM •"*! i— i

£

CM

CO CO

CO

CO
OS
i-H 05

.co

+

+

CO
CD

CO "5

CM CM O

©

■ co co

k.t~i £°

«*J

t^ © co

. CO CO

>> . >»

CM

Tf CD i— i

+

+

ffl

Tfl CO

i-l lO

iH rf CO

©
©

o o

co

CM

Tti 1*1

3

CQ

+

+

PL,

>o

r--

CO lO

I-- -H ^f<

©

• CO CO

<l

CO

O "O w

. CD CD

•«*! CO ^H

< it*

+

+

CO

© OJ

NlON

©

CQ

i-H CM

co co co

CM CM

s

CM

i* CO

0>

1

1

CO
Oi

"-■co

'5

w

to

-co

CO

CM i*l

i-l OS CO

©

t^ ©

<!

i-H O

i-H 1*1 CO

CM f~

>>rt. >>

CM

1*1 lO -^

CO CM

^N o3

+

+

<** o3 co

CM

OhJ1

—.Sc'l

CO

CM CO

CD Tfi CM

©

if 1*1

T3 w

—

0> i*i lO

CM CM

.2 3

GO

o>
o>

CO

*"*

CO CO

1

1

©

CD

CO

^H CO

US © >o

©

t- ©

<J

CM

O rt if
^f< CO ^H

+

+

<M r^

CO CM

lO

©

IC

■* 1C

CO CO ©

©

PQ

^h CM

i-i CM i-H

CO
CO

1-1 CO

co

i*i •*

+

+

lO

"5 CO

co en i-i

CM

©

i-H ©

CO ,— , ^>

<l

i-i CM

-H if CO

CT> CM

0> - >>

<M

1*. IC i-H

C® o?

+

+

1-5 S"

©

*~ 0>0O

^ o

>o

■"*> CO

M N »

©

PQ

^H CM

co i*i

CM

_0 g
PL)

3
S

*• +

+

CO
CO

*C

i*i CM

m os i*i

©

^H ©

<3

i-H CM
CM

lO N N

•* IO H

+

+

a> cm

in co

13 "0

tI

>>

G G co"

o3

03

5 O J

0>

a a ^

y

0) jj

tf ■- ©

cu

a

"1 *

03 03 £

™%i

"8

cs

-a .s I1

CD

i

W CD

ill

«^ "^ "G

O & .S

G 03 JS

G
3

|

!K o> o>

■O « * o

11

1

® M bo

-Q c3 03

G M M M

g c3 c3 <5
a >

G tH <-

G 0) 01
3 > >

o 'o .1

O tf &■

o

2

«< <

<J <3 <

H

Bul. 663] The San Joaquin Experimental Range 61

equally divided as to weight and quality. Death loss of a light calf in
group A5 and a heavy calf in B5 subsequently unbalanced the average
initial weights of the two groups of steers. Group A5 was fed 1 pound
of pea-sized cottonseed cake for 110 days. After the first autumn rains
2 pounds of rolled barley daily were added to maintain gains till the
end of the period. Except for the month after the first fall rains, group
A5 made continuous gains. Group B5 gained during the first month on
dry feed, lost weight for 3 subsequent months, then gained on new green
forage, but at the end of the period was only slightly heavier than at
weaning time 6 months before.

As previously discussed under the breeding herd, the difference in
weaning weights of groups A6 and B6 (1936 calf crop) is not significant
when age and sex variations are considered. One to 1.5 pounds daily of
cottonseed cake was fed to group A6 for 2 months, and then 1.5 pounds
barley was added for about 2 months. Continued cold weather after the
first of the year prevented forage growth until March. Range feed be-
came very scarce, and 5 pounds daily of supplements during part of the
period were required to prevent weight losses. Severe weight loss oc-
curred in group B6 ; the animals became thin and weak towards the end
of the period and received supplemental feeding for 48 days, to prevent
death losses. Group A6 was in thrifty condition at the end of the period.
Figure 11 shows these two groups as they appeared in December, 1937.

There was a very significant difference in the weaning weights of
groups A7 and B7 of the 1937 calf crop, as was previously explained.

The possibility of self-feeding of fish meal on the range, and also of
self-feeding a mixture of fish meal and molasses, was investigated with
group A7. The calves readily ate fish meal alone at weaning time. When
they were self -fed in the pastures, the quantity taken was too great to be
economical. Fish meal was then self -fed mixed with molasses. Consump-
tion of the mixture increased to about 9.5 pounds daily ; dry range feed
utilization was poor, bloating occurred, and gains were not commensurate
with cost. The hand-feeding of restricted amounts was therefore neces-
sary. The gains and condition of this group over the entire period were
reasonably satisfactory, whereas the group not fed supplements only
maintained their weight over the 6% months' period and were in thin
condition. Because of the difference in forage and weather conditions
from year to year, direct comparison of the feeding value of fish meal and
molasses with cottonseed cake and barley is not justified.

The difference in average weight of the two groups by the end of the
period was further increased to 250 pounds.

The 1938 calf crop again showed a significant difference in weaning
weights of the calves from the breeding herds that were fed supplements

62

University of California — Experiment Station

contrasted with those that were not. The feeding procedure for group
A8 resembled that for the 1935 and 1936 calf crops. Feed and weather
conditions were favorable and, as table 9 shows, there was less difference
than usual in gains between the two groups during the period.

Tig. 11. — In A are shown the animals of group A6 (fed supplements) as they
appeared on December 6, 1937. Those of the corresponding group B6 (not fed
supplements) are shown in the lower picture, B.

Pasture Gains After the Weaner-Calf Period. — Since most cattle are
sold at the end of the forage season rather than at the end of the winter-
ing period, the net value of supplemental feeding of calves is best judged
by the weight advantage existing at the end of the next grazing period.
Table 10 gives the data for the various groups.

In all cases, group B animals that had lost or only maintained weight
from weaning through their first winter, gained somewhat more the
next pasture season than group A animals which had made continuous
growth. There was, nevertheless a variable but significant difference be-

M
U

OJ

CO
OS

"""os

PQ

CD
» ^ (O N H

fcs^

2

■^< t-i t^
CO to N

rt »fe

flt-i ca

V

c3 .tj

1-5 m£

"3

2^s

13 0

w

r-

s

0 ^J

<J

CO IC OO CO H

CM CM O

a>

tD OO N

Ph

00

22

CD

05

«-< 00

CD

»o

pq

O »0 lO O H

tCos,,^

I-H CD T# OO

. '— ' to

e

CO CD CM

t- - >>

c3<M c3

CD

SU*

'3

O 01 £-

^M

CNI

.2 5

<!

CM CT> O i-l i-H

i-H ->f IO O

CD

19 N N

p-i

00

CO

>o

OS
i-l OO

CD

"0

-CO

pq

CM Tfi 00 -f CM

"^S^-v

H ^ »0 ^H

So 5?

09

CO »Ci CM

33

5

.2 °

OQ

CD

CM

<-i O CO CO <M

rt »-i ,-1 O

o>

CD 00 CM

Ph

OO

uO

°?

pq

* 00 N O) H
^H CM CO O

73

^fi lr~ CO

O

'53

co

CO

w

co

oj'co

0

IO » to S rt

nS'S

h ■* OO M

U5 t» N

CD ■* C3

Q «£

<*- 3^2

0<JCM

"2 o~

«5

US

g+5

PQ

■*f t— CO OS 1— 1

>— 1 Tfi OO CO

4)

■<*< t^ CO

Ph

w

1

QQ

CO

»o

<1

■>*< OS O T+ ^H

T-l t^ OS T-l

•O 00 CO

or

^3 »

c -a

a §

-3
3

rt £ CD O 03

<h '3 c '3 c«

O .3 cc bfl T3

«-. CD CD CD CD

£ bO bO bl) bfl

^ S tj oj s

C (-, I-, (-, (_.

►5 CD CD CD CD

3 > > > >

2

<

<,

<

<

e
0

52

OS

c

•a

n

>

33

0

Cy

CD

OJ

3

*

e
"3

O

pq

ot

c

pq

1/1

a

:3

3

H

a

M

a

"3

C

CD

O

a)

g

d

83

-3

0)

O

.3J3

2^

* >>

"0

0

0

a;

O
c

2

3

ga

CD -2

.fa
CD^

CD CD

'a a

8

64

University of California — Experiment Station

tween the final weights of comparable groups each year. Data are given
for only 84 days for groups A6 and B6 steers because the former were
placed on a finishing ration after this time. Aside from the higher gains
of both groups for the short period, pasture gains varied from 1 to 1.5
pounds daily.

The possibility of marketing 450- to 500-pound weaner calves or 800-
to 900-pound yearling feeders through supplemental feeding of weaner
calves is indicated. These animals, by reason of greater weight and con-
dition, were suitable for finishing in a short time in the feed lot and

TABLE 11

Summary of Feed and Weight Records on Replacement Heifers

from Weaning until Start of First Calving Period*

Group
A5

Group
B5

Group

A6

15

14

10

418

418

425

549

428

573

311

597

786

737

765

723

579

790

462

125

312

1,007

950

953

773

125

909

Group
B6

Number of animals

Average weaning weight, pounds

Final weight, first supplement period, pounds

Supplements per head, pounds

Final weight, first grazing period, pounds

Final weight, second supplement period, pounds

Total supplements per head, pounds

Final weight, second grazing period, pounds

Total supplements fed to first calving period, pounds

437
370

48
651

824
48

* Group A animals received supplements during that part of the year not covered by the grazing periods ;
group B animals received none, except in emergencies. The figures placed after the group designation are
the terminal numerals of the year of birth; for example, the animals in B5 were born in 1935.

would, therefore, ordinarily bring a higher price for this purpose than
the lighter, thinner cattle that received no supplements.

Data on Replacement Heifers. — Table 11 presents data on heifers used
for additions or replacements in the breeding herd, and covers the time
from weaning until the beginning of the period during which they first
calved. As previously mentioned, there was no significant difference be-
tween groups in percentage calf crop at first calving. Group A5, how-
ever, had calves averaging 86 pounds heavier at weaning time and the
following year produced a 78 per cent calf crop as compared with 58
per cent for group B5.

Since this work was started, practically all the heifers have been used
for enlarging the breeding herd or for replacement, as shown in table 12.

During the first four years there was a total of 98 heifers exclusive
of death losses. Of these all but 25 were added to the breeding herd.
Because of difference in calf crop and chance variations in sex ratio there
was more opportunity to select the most desirable replacements from
group A than from B heifers. Seventeen group A heifers were finished
for slaughter as long-yearlings. Two group A and six group B heifers

',

■'■

Plate 1. — Areas showing close (A), moderate (B), and light (C), utilization
in August at the end of the seven months' grazing season.

[65]

Plate 2. — Upper: Differences in height of vegetation in areas cleared of old
growth at the beginning of the growing season, compared with undisturbed
areas. Both these areas were imgrazed by either rodents or livestock. A, Cleared
part of plot ; B, uncleared. Lower : Typical range inhabited by kangaroo rats,
ground squirrels, pocket gophers, rattlesnakes, and cottontail rabbits.

[66]

Bul. 663]

The San Joaquin Experimental Range

67

were sold as feeders. When this manuscript was written, the 1939 heifer
calves were still too young for a decision regarding them to have been
made.

Management and Disposal of Produce. — Because of the demand for
information on finishing cattle, both as yearlings and as two-year-olds,
the original plan of disposing of surplus as yearling feeders was tem-
porarily abandoned.

The data for group A5, table 13, include periods on pasture alone
during 1937 and 1938, a period of supplement-feeding in the fall and

TABLE 12

Heifer Groups as Used for Enlarging the Breeding Herd and

for Replacements

Year

Group A
heifers

Group B
heifers

Total
groups
A and B

Replace-
ments
in group A

Replace-
ments
in group B

Disposal

of
group A

Disposal

of
group B

1935

15
12
13
18

14
10
6
10

29
22
19
28

15
10
6

8

14
8
4
8

2
7
10

1936

1937

1938

2
2

2

winter of 1937-38, and a short finishing period. Their check group (B5)
received no supplements throughout life. Group A5 gained only 47
pounds during a 215-day period of supplement feeding. Feed shortage
during this period not only prevented good gains but also necessitated
heavy supplemental feeding during part of the time to prevent weight
loss. The corresponding group (B5) lost nearly 200 pounds during this
period. The carcass weights for group A5 were somewhat heavy for
general demand, but graded Good. The carcasses of group B5, on the
other hand, had practically no fat covering and graded Medium and
Common. These animals, though unsuitable for slaughter, would have
been desirable feeders. The appearance of the two groups one month
before marketing is shown in figure 12.

Group A6 indicates possibilities of producing satisfactory slaughter
cattle as long-yearlings through promoting continuous gains as weaners
followed by a short finishing period at the end of the green-feed season.
Ten carcasses graded Good, and one, Top Medium ; the yield was 57 per
cent. Their weights may be compared with their control group (B6) , sold
as feeders a year later. Size and condition of average steers from each
group are compared in figure 13.

Group A7 steers and heifers not only were fed supplements during
the period from weaning to the next grass season but also had the feeding
of fish meal and molasses continued during the green-forage season and

Bul. 663]

The San Joaquin Experimental Range

69

until they were marketed. Their carcass yield indicates less finish than
was obtained by group A6, marketed the previous year. Group B7 was
held over for sale as two-year-old feeders the following year.

Grazing -Intensity Pastures. — Data on the animals grazing pastures
1 to 6 inclusive were collected in relation to the study of the effect of

Fig. 12. — Two-year-old steers on July 17, 1938, one month before they were
marketed. Those of group A5 are shown in A ; those of group B5, in B.

stocking on the forage in these pastures. Several factors complicate in-
terpretation of the record. The period covered is short, pastures com-
parable in size were not comparable in forage production, and the
animals were not on the area the year-round. The equal division of group
A and B animals in each pasture made small groups of each. This diffi-
culty, together with variation in the number of replacement heifers and
possibly other factors, thus far obscures any difference in calf crop that

70 University of California — Experiment Station

might be attributed to feed supply during the period reported. For ex-
ample, culling of wild cows and those that aborted contributed to heavier
replacements in pasture 3, though the thinner condition of these animals
may in some cases have inadvertently also influenced culling.

Table 14 presents average data for the four-year period 1936 to 1939
inclusive.

For this 6- to 7-months' period the difference in net gain per animal
unit between pastures 1 and 3 is greater than between pasture 1 and any

Fig. 13. — A steer of group AG, on the left, and one of group B6, on the right,
at the time of marketing group A6, August 7, 1938. The weight of each of these
steers was the same as the average for their respective groups.

of the larger-sized pastures. The net gain was the same in pasture 1 as
in 4, a field containing twice the acreage. Pasture 2 produced larger calf
gains and gains per animal unit than either of the largest pair of pas-
tures (4 and 6).

The highest net gain per acre was produced by pasture 1, in which the
forage was well utilized ; yet the gain during the green-forage period and
the loss in the dry period were not significantly different from those of
the larger pastures. In pasture 3, on the other hand, feed shortage early
in the season and greater losses during the dry-forage period resulted
in lower net gains and returns per acre. The lower gains, the thinner
condition of the animals, and the poorer development of replacement
heifers show that the degree of feed utilization was beyond optimum for
the well-being of the animals and for maximum net returns. The pre-
dominance of heifer calves in pasture 3 may partially account for the
difference in calf gains on this as compared with other pastures.

~H <M ^ lO

lO O

r— oo

S 55

1C Ol h rt N

■* t— t~ o

En

O
w

>

OS OS O O —i CC

lfl> 1-1

o r- co >o

(M CO OS

if «

. c3

a o g

§ a

CO _ <C

03 o a

^ « a a » ea

g"S.gS.§

« 5 S S ? fl
bO . bO bfl fe 03
03 fe _ _ £ —

a o o co ;>

H «! tf

► aj o o

•< W H H

c +3

0) o

2 H

o3 -a

a co os

+2 co 43
O > o

H <! H

H H

*3

"-a

3*2

s*

72 University of California — Experiment Station

From the standpoint of degree of forage utilization it was agreed that
on the average during 1937, 1938, and 1939, pasture no. 2 (medium
sized) and no. 6 (large) were lightly grazed; pasture no. 1 (small), no.
4 (large), and no. 5 (medium) were moderately grazed, and pasture
3 (small) was closely or very closely grazed.

The average net gain of cows and heifers and of calves in relation to
degree of forage utilization for this period is shown in the following

Lightly grazed Moderately grazed Closely grazed
(pasture nos. (pasture nos. 1, (pasture

Average net gain per animal unit, cows 2 and 6> 4« and 5> no. 3)

and heifers, pounds 229 207 155

Average calf gain, pounds 271 281 247

It is doubtful* whether the difference in gains between light and mod-
erate grazing conditions is significant but the difference between these
and the consistently lower gains made under close grazing are obviously
significant. From the livestock standpoint it is definitely indicated that
the maximum utilization compatible with maximum gains during the
green-forage season and minimum losses in the dry season gave greatest
returns during the period under consideration. This optimum degree of
utilization from the standpoint of the livestock corresponded approxi-
mately with moderate use as defined in the chapter "The Forage Crop
and Its Management." The data again show that in the absence of le-
gumes the dry forage, regardless of acreage, will not maintain the ani-
mals.

Grazing Habits of Cattle. — Observations of cattle show a close rela-
tion between their grazing habits and seasonal variations in forage com-
position, sequence in appearance or maturing of various species, and
climatic conditions as presented in the preceding general section.

Generally, broad-leaf filaree appears about 2 days after the first sub-
stantial fall rain and is readily taken by the cattle in the first leaf stage.
To get this new growth, the stock must eat considerable old litter. The
bromes, wild oats, and fescues shortly become another item in the ani-
mals' diet. If the weather turns cold the broad-leaf filaree and other
rosette-forming plants such as popcorn flower tend to lie close to the
ground, and the stock must depend mainly upon the grasses and weeds
such as buckthorn weed (locally known as fiddleneck) for the new forage
growth.

With the advent of favorable weather for plant growth, filaree becomes
the major item in the forage taken until it starts to produce seed and
matures. The cattle then begin to take larger amounts of grasses, toad
rush and true clovers if these are available. With the drying of the
filaree, fescue, and other plants growing on the higher elevations and

Bul. 663] The San Joaquin Experimental Range 73

shallow-soil areas the cattle begin grazing the maturing soft chess heads
if the season has favored their development.

After that source of feed has been exhausted or has matured, the cattle
turn to the swale areas containing slender-leaf rush, the major remain-
ing source of green forage. These areas are then heavily grazed until
bare. If the season is favorable for late-growing plants, these species are
also taken until exhausted or until they have matured. After complete
maturing of all forage, soft chess (which does not shatter its seeds so
readily as other plants) and fescues are the principal source of forage
until new growth appears. When available, Spanish clover is also sought
out and licked off the ground. Yellow tarweed, one of the two major
late-growing weeds, is taken in substantial quantities in its early stages
of growth.

Browse represented only a small part of the forage consumed. Blue
oak was the principal species taken, followed in importance by coffee-
berry, hollyleaf buckthorn, wedgeleaf ceanothus, and interior live oak.

During all seasons the cattle had two general grazing periods during
daylight hours. The first, starting at daylight, continued until various
times in the morning. The second began in the afternoon and continued
until dark except during the periods of bright moonlight, when cattle
grazed until midnight or later. Generally water was taken at midday
(when some grazing may be done) or just before the afternoon grazing
period. The cattle spent considerably more time grazing during the cool
fall and winter months, when the feed was short, than in late spring and
summer after the feed became abundant and the temperature was high.
During the cool months the cattle generally bedded down about mid-
morning and did not graze again until early afternoon. After the daily
temperature became high, the cattle would bed down at 8 :00 a.m., or
earlier and would do little grazing again until 4 : 00 p.m., or later. If they
watered at noon they often grazed a little on their way to the water
troughs.

This range land at the time of purchase was heavily infested with
rattlesnakes ; and despite a more or less continuous campaign against
them, a study discussed in a later section has shown a population of ap-
proximately one snake per acre. In two instances grazing animals were
observed to walk directly over rattlesnakes while grazing without either's
paying any attention to the other ; the snakes did not rattle. A def anged
snake was released near a group of cattle. Its movement and rattling
attracted the attention of the animals, which approached and smelled
it. The snake then actually struck the animals without their manifesting
any fear.

A total of 21 cattle were known to have been bitten, of which 3 year-

74

University of California — Experiment Station

lings died; 5 developed abscesses requiring treatment; and 2 were so
disabled that they were culled from the herd. Of the 21 animals bitten 18
were struck about the head, 1 on a front leg, 1 on a hind leg, and 1 on
the udder.

In one instance loud breathing called attention to a six-months-old
calf partially hidden in a clump of brush. The animal had evidently been
bitten by a rattlesnake a short time before, and the enormous swelling
about its head and throat made breathing difficult. Its eyes were swollen

TABLE 15

Total Salt Consumption in Grazing-Intensity Pastures

Pasture
no.

Location of lick

Total

consumption

in pounds

Consumption

per
animal unit

193S

Inconvenient

Inconvenient

Convenient

Convenient

Convenient

Convenient

1939

4.86
3.22
7 03
6.64
6 01
5.58

1

163
189
203

56
91

87

9.64

2

10.86

3

11 80

4

3.20

5

5 11

6

Inconvenient

4.97

shut, and blood trickled from the inner canthus of both eyes. The tongue
protruded from the mouth. Although death then appeared imminent,
iy2 hours later the calf had improved; and 8 hours later it was out of
danger and recovered completely without treatment.

Observations on Salt Consumption. — Until August, 1937, coarsely
ground stock salt had been kept available for the various groups, but
large amounts were lost by leaching in the rainy season. During the re-
maining time covered in this publication, block salt was used in the
rainy season, coarse-ground salt the rest of the year. A record was main-
tained of salt consumption in the various pastures. Table 15 gives data
for the six grazing-intensity pastures.

Although variations were made in the location of salt in the six pas-
tures, only one salt lick per pasture was supplied. In the beginning the
salting place was chosen mainly with regard to convenience in servicing.
The size of the pasture proved to be a minor point compared with con-

Bul. 663] The San Joaquin Experimental Range 75

venience of the lick as a factor influencing the average salt consumption.
Table 15 shows the difference in salt consumption when the licks were
alternated between locations convenient and inconvenient from the
standpoint of normal movements of the cattle.

The placing of salt in least utilized portions of the pasture had little
influence on frequency of movement of cattle into these areas. The cattle
at times became noticeably salt hungry before they went to inconven-
iently located licks. The daily requirements for water and the con-
venience of securing a "fill" of feed apparently were major factors
influencing grazing habits and movements. No indices used in this work,
including appearance of the cattle, weight, grazing methods, or others,
indicated that high salt consumption was physiologically desirable. Such
consumption may, indeed, be evidence of poor feed conditions and of
depraved appetite caused by deficiencies other than common salt. For
example, the supplement-fed groups consumed 25 to 50 per cent less
salt in the fall and winter of 1937-1938 than the others.

Analyses of Range Forage Samples. — Throughout the period a careful
attempt was made to collect samples representing the actual forage
grazed. To accomplish this, one of the authors did practically all the
collecting. Following various quiet animals while they were grazing, he
noticed the plants and parts of the plants grazed. Then he attempted
to duplicate this grazing by hand from the edges of the same spots to
get representative material for analysis. This procedure was followed
almost daily the first two years and at weekly intervals thereafter. The
collections were composited over a two-week period to consitute the lab-
oratory samples. After the specimens were collected, dried, and com-
posited, a broad ocular estimate of the general botanical composition,
the quantity of old feed from the previous year, and any other pertinent
data were recorded. About 200 such samples have been analyzed. Data
on crude fiber, crude protein, calcium, and phosphorus are presented
graphically in figure 14.

The data for 1936 and 1937 are averages of duplicate series of samples
collected from pastures 2 and 5 during the grazing periods and from 11
and 12 during the supplement periods. The 1938 data are averages of
duplicate series from pastures 2 and 3 and from 13 and 14 during the
grazing and supplement periods respectively. The 1939 samples are a
single series collected from all the grazing-intensity pastures and from
pastures 13 and 14 during the latter part of the year.

The seasonal variation in the chemical composition of the forage was
not significantly different during 1936-1938. The protein early in the
season, and after old feed was no longer taken with the new, varied from
20 to 25 per cent of the dry matter. The amount declined similarly in all

76

University of California — Experiment Station

three years, reached 10 per cent early in June and about 5 per cent by
July 1. The fiber increased as the protein decreased; and the nitrogen-
free extract, which is not shown in figure 14, remained fairly constant.
The increases in protein and the decreases in crude fiber during August
and September coincide with the time the cattle were changed from the

Jan. Feb. Mar Apr. May Jane Ja// Aag. Sept Oct. A/ok Oec.

Fig. 14. — Seasonal variation in chemical composition of forage samples collected
to represent forage grazed by the cattle during 1936 to 1939 inclusive.

grazing-intensity pastures to the fall holding pastures. During the first
few weeks in these ungrazed pastures, enough Spanish clover and other
legumes were present to improve the value of the forage. The change
from a diet largely of grasses to a mixture containing the legumes is also
reflected by the rise in calcium content of the sample.

As figure 14 shows, the forage in 1939 was conspiciously different in
seasonal trend of chemical composition from that in previous years.
Filaree made up much of the forage during the growing season — a fact

Bul. 663] The San Joaquin Experimental Range 77

that explains the unusually high calcium content of the samples. The
protein, varying from 10 to 15 per cent during the growing season, was
much lower than in other years. Because of lack of rain, the feed dried
early ; the protein reached a level of 5 per cent by the first of June and
remained below this level until rains fell in October. The crude fiber was
correspondingly high. After the forage dried, grasses made up most of
the feed taken. The low calcium, characteristic of these grasses, is shown
in the graph. The absence of the sharp rise in calcium in May that oc-
curred in other years is a result of lack of late-developing species such
as Spanish clover.

The phosphorus picture for the various years was similar. The longest
period when phosphorus may have been a limiting factor was in 1938,
when the amount dropped to 0.10 per cent or lower from the latter part
of September until late in November.

Calcium content is more erratic than the other constituents studied.
The changes reflect alternative grazing on grasses and grasslike plants,
low in calcium, and on high-calcium plants such as filaree, legumes, and
other broad-leaved species, as discussed under grazing habits of the
cattle. The amount of calcium supplied was adequate at all times.

In general, protein deficiency coupled with the low-energy value in-
dicated by the high fiber content is doubtless the chief limiting factor
for producing gains. The weight records of the cattle show that gains
cease and losses begin when the protein content falls below 7 or 8 per
cent.

During 1938, a series of forage samples was collected from each of
pastures 2 (lightly grazed) and 3 (closely grazed). The seasonal trend
from early stages to maturity was the same in both series ; there was,
however, a consistently higher fiber content and lower protein content in
the samples from closely grazed pasture 3 throughout the season. Thus
there was no evidence that close and frequent grazing maintained the
forage at an earlier stage of development such as usually occurs with
close grazing of perennial forage. Apparently these annual plants tend
to mature and produce seed regardless of the amount of vegetative
growth permitted.

Acorns as Range Supplements. — Livestock men in the area attribute
the following conditions to heavy feeding on acorns in years of heavy
mast: deformed calves, loss in weight, gain in weight, abortion, and
death from impaction.

In 1936 there was a heavy acorn mast, and the three animals kept from
the original group for replacement purposes had not been used. They
were in different stages of pregnancy — 4, 5, and 7 months, respectively —
when an experiment was started to feed them acorns with the natural

78 University of California — Experiment Station

dry feed. Beginning August 1, 1936, the acorns, harvested green from
the trees, were fed with the cups intact. Later, as the nuts ripened and
fell, they were collected from the ground ; and at that time the cups were
not attached. Feeding started at the rate of 3 or 4 pounds per day, and,
as the acorns ripened in September, consumption rose to 14 or 18 pounds
per head daily. There was marked difference in the palatability of the
acorns from different trees, although all were of the same blue oak species
(Quercus Douglasii) . After green feed became available, consumption
increased to as much as 37 pounds in one day for one cow.

During a 60-day period on dry feed, the acorn supplement resulted
in an average loss of 150 pounds per head in contrast to an average of
20 pounds per head by the group B animals not fed the supplement.
Normal calves were born to two of the animals. The one animal which
was 5 months pregnant when the acorn-feeding started, gave birth, at
the end of 117 days, to a bull calf weighing only 31 pounds. This calf
had short legs and was humpbacked in the dorsolumbar region of the
spinal column. It lived until three years of age, and meanwhile the de-
formity became more marked.

The last animal to calve did so after green feed became available.
Although she continued to eat acorns in large quantities, she regained
from the third to the fifth month all she had lost in the first 2 months on
the dry-feed and acorn ration.

These results agree with other evidence that acorns ingested with ad-
ditional protein supplements will not cause weight losses. Since we also
have evidence of deformed calves from cows not having contact with
acorns during gestation, evidently acorns cannot be considered the
primary factor in causing the deformity. The anatomical anomalies of
"acorn calves" appear to be similar to those produced in chicks and rats
by manganese deficiency. No evidence, however, of deficiency of this
element in forage produced in areas where the deformity occurs has
thus far been obtained.

DISCUSSION OF RESULTS WITH THE EXPERIMENTAL HERD

Because of the seasonal variation in quantity and quality of forage,
the length of the dry period, and the amount of winter growth, no defi-
nite recommendation regarding time or amount of supplemental feeding
can be given. For the breeding herd, the amount fed varied from 270
to 476 pounds yearly per cow.

Without supplements, losses in weight or in condition occurred in
cattle of all ages on dry forage except when first turned on such forage
that had not been grazed during the growing season. With the rates of
stocking employed, maintenance or slight gains were made for 2 to 4

Bul. 663] The San Joaquin Experimental Range 79

weeks, while the better forage was being consumed. Thereafter, regard-
less of abundance, the protein of the dry forage was low, the fiber con-
tent was high, and weight losses ensued unless supplements were fed.

The practicability of maintaining the breeding herd in thrifty condi-
tion yearlong on this range by supplemental feeding has been demon-
strated. Such feeding resulted in higher average percentage calf crops
and permitted early calving of uniform-aged calves that attained a
weight of 430 to 501 pounds when weaned at 7 to 8 months of age. These
early calves received ample milk and had full advantage of the entire
season when forage was good and when, in their development, the im-
pulse to grow is greatest, as is shown by normal growth studies (8) . They
attained sufficient size and development to utilize the poor dry forage
after weaning, and cows were spared the drain of lactation for a long
time on dry forage. The disadvantage of weaning in July lies in the
relatively long period requiring supplements, needed to continue normal
development until the next green-forage season. The optimum time for
calving on this and similar ranges should be further investigated.

Even on dry forage, little trouble was encountered with difficult calv-
ing and with retained placentas in cows and heifers receiving supple-
ments. Serious difficulty in this respect occurred with the herd not
receiving supplements, and death loss was higher. Breeding the females
which did not receive supplements, somewhat later, would probably be
advantageous ; this postponement was expected to occur naturally as the
experiment progressed. There was some tendency for cows in this group
that calved regularly to breed later in the season. Many, however, failed
to breed during the four months' period allowed ; and they, being dry,
developed good condition and again bred early the following year. The
tendency to shift the breeding period later in the season was thus offset.

The labor involved in feeding supplements to the cows was little, if at
all, greater than the labor involved in a careful supervision of the other
group, because of frequency of retained placentas, death losses, and the
like. Evidence from these data and other observations would tend to
support the probability that low protein intake is associated with in-
creased incidence of retained placenta.

Feeding supplements on alternate days reduced labor without notice-
ably reducing the effectiveness of the supplements as compared with
daily feeding.

After 2 or 3 months on dry forage the cattle not fed supplements
tended to become light red or yellowish red, in contrast to the dark-red
color and smoother condition of hair maintained by the supplement-fed
groups. An animal in a supplement-fed group that refused to eat supple-
ments could almost invariably be distinguished at a distance by this color

80 University of California — Experiment Station

characteristic. Cattle not fed supplements were also usually slower in
shedding.

Because of lower calf crops and the chance occurrence of a smaller
proportion of heifer calves from the groups not receiving supplements,
there has not been, thus far, so much opportunity for selective culling
and replacement in these groups as in the others. Moreover, animals in
the groups not receiving supplements are usually thinner and less well
developed, a fact which renders judgment of them difficult and uncertain.

A certain environmental standard, mainly determined by feed condi-
tions and plane of nutrition, is necessary for satisfactory progress in
the improving of quality through breeding. Breeding for the higher
market grades is an important factor in the ability of feeders to utilize
profitably the feed in feed lots. Its continued absence in any given area
reduces the value of the feeder cattle.

Insufficient data are available to show definitely the relative economy
of the supplemental feeding employed with group A replacement heifers
as compared with no supplement or more limited supplemental feeding.
There was no significant difference in the percentage of first calves
dropped, but there was a wide difference between the groups in weights
of calves at weaning and in second-calf -crop percentage. These advan-
tages, combined with the additional weight of the supplement-fed heif-
ers, must be considered in comparing the two methods of management.

Weaner calves were fed amounts of supplements varying in the dif-
ferent years from 286 to 597 pounds over periods of 6% to 8 months.
The average daily gain varied from 0.55 to 0.75 pound for those receiv-
ing supplements as compared with a little more than maintenance on the
average in the others. The final weights of the former at the beginning
of the green-pasture season varied from 550 to 650 pounds ; the latter
from 345 to 495 pounds. This difference was decreased somewhat during
the subsequent pasture season, and a limited amount of data indicated
this to a greater extent in the heifers than in the steers. Not only weights
and feed costs, but also variation in the value of the animals per pound,
should be considered in evaluating the data.

The possibility of producing good-grade slaughter steers as long-year-
lings and as two-year-olds through supplemental feeding has been dem-
onstrated. The practice, though economically feasible thus far, is not
necessarily the best use of such land. It involves a certain amount of
chance determined by weather, price of feed, and market price of the
finished animals. The data on finishing are significant as indicating
means of maximum utilization of grass in the production of slaughter
cattle desirable in weight and finish on ranges better adapted than the
experimental area is for fattening cattle.

Bul. 663] The San Joaquin Experimental Range 81

Since heifers fatten more readily than steers at the yearling stage and
sell to better advantage at this age than later, supplemental feeding to
finish surplus heifers on this and similar ranges appears practicable.

Experience indicates that without supplemental feeding, two-year-old
steers cannot be depended upon to reach slaughter condition. The evi-
dence also supports the view that this and similar areas will sell most
animals of the various ages as feeders even when supplemental feeding-
is practiced.

Attention is called to the possibility of producing and marketing
heavy feeder calves at weaning time through liberal supplemental feed-
ing of the cow herd and by early calving. Fluharty (3) shows this to be
economically feasible only when percentage calf crop is high and heavy
weights are obtained. Under such a system supplement-feeding is limited
to the breeding herd and to a shorter period than that required by calves
carried over from weaning to the next green-grass season. Such calves,
however, make continuous gains when fed supplements and offer another
means of marketing the range feed through sale of 800- to 900-pound
fleshy long-yearling feeders at the end of the green forage season.

The data from these experiments, in common with general experience,
show that greater gains are made during the yearling than during the
two-year-old stage.

Except for one group of cattle fed fish meal and molasses, the locally
produced supplements, cottonseed cake and barley were used. Because
of the protein deficiency of the range, a protein supplement is a first con-
sideration. Protein supplements also furnish phosphorus and added
energy. Because these feeds are usually higher priced than barley or
other grains, the latter can be used for additional energy after protein
requirements have been met.

Abundant information is available on replacement value of various
protein concentrates that may be substituted for cottonseed meal and
other grains, and of by-products that may be substituted for barley.
This information, summarized by Guilbert and Rochford (10), may be
utilized in selecting alternative feeds in case cottonseed cake or barley
prices become discouragingly high.

According to evidence from the study on the six grazing-intensity
pastures, reduction of individual animal production by excessive stock-
ing in relation to feed supply may be such that returns per acre are
likewise reduced. The results also show that grazing over a 6- to 7-month
period to an extent designated as "moderate to close" produced gains per
animal unit comparable with those resulting from light grazing; and
the return per acre obviously was higher. The data suggest that the con-
cept of maximum forage utilization compatible with maximum produc-

82 University of California — Experiment Station

tion per animal unit has possibilities of wide application for defining
proper rates of stocking from both economic and ecological standpoints.
A policy of maintaining a hay reserve as an insurance against extreme
drought and feed shortage has been adopted, on the principle that such
feed under cover could be kept for many years without much deprecia-
tion, whereas feed on the range is spoiled from year to year by rain and
exposure. Hay reserve thus permits as complete utilization of the range
in average years as is compatible with its maintenance and the well-being
of the animals without the hazard of feed shortage and high feed costs
under extreme conditions. Apparently about 50 per cent more acreage
is required to carry the cattle from August to February than from this
time until August 1.

RANCH ORGANIZATION AND MANAGEMENT
IN THE GRANITE AREA

E. C. Voorhies, L. A. Crawford, R. L. Adams,
and G. A. Carpenter

The discussion under this heading describes the general ranch manage-
ment practices in the region surrounding the San Joaquin Experimental
Range (fig. 15). The area covered by the survey totals approximately
500,000 acres situated in Madera, Mariposa, and Fresno counties. It is
fairly typical of the foothill belt of the Sierra Nevada in which some of
the most pressing problems of livestock production are met.

The "granite area" derives its name from the frequent and often sub-
stantial outcroppings of granite found throughout the section. Land-
surface features of this type are found to extend in a southeasterly
direction from the town of Mariposa for approximately 50 miles, and
from the open range or "bald hills" section bordering the San Joaquin
Valley floor in a general easterly direction to the present boundary
(1940) of the Sierra National Forest. The width of the area varies from
about 6 to 20 miles.

While the granite outcroppings give the region a fairly uniform ap-
pearance the topography varies greatly. Low hills, broken slopes, ra-
vines, and small valleys are found. The elevation rises from an altitude
of approximately 700 feet above sea level on the western border ("bald
hills" section) to about 3,000 feet on the eastern border. The valleys are
relatively small, tortuous in contour, and widely different in soil. Some
of the land is arable, but by far the greater part of the acreage is suitable
only for livestock grazing.

Three principal rivers drain the area — Chowchilla, Fresno, and San
Joaquin. Only the last-named has an appreciable flow during the entire
year. In all of the streams the largest flow occurs in the late winter and
early spring months. With the exception noted above, most of the streams
are dry at some time in the late summer or fall.

Development of the Cattle Industry. — The beginnings of the cattle
industry here date back to the period shortly after the discovery of gold
in 1848. During the next two decades this and other foothill areas in
central California attracted a large influx of miners. The population of
the region was probably greater at that time than at any period since.
There were numerous operators using the public range at the time, sup-
plying meat and dairy products to the then important mining communi-
ties. Such enactments as the Homestead Act (1862) were utilized to
acquire meadows and key tracts which might be used for the control of

[83]

84

University of California — Experiment Station

the public range. As numbers of cattle were increased the use of range
gradually expanded both in the foothills and in the higher mountains
to the eastward, the latter area being utilized in the summer months.

OMAfcfiA

Survey area

— fioaaaary of area
**» Coaatj/ o'aaaa'ar/'es
- To wr? $/?//> foaaa'ar/es
/?a// raaa's

\mm& Sa/7 Jaaaa//7 Exper/meafo/ fiaaye

Fig. 15. — Part of the granite area surveyed in the study on ranch
organization and management.

Prior to the gold discovery, settlement had not taken place here. An
important change occurred in land tenure, the early range being "free."
This gradually changed so that a considerable part of the range was
transferred to private individuals. Numerous enactments by the Con-
gress after the passage of the Homestead Act, for example, the Timber
and Stone Act (1878), the Enlarged Homestead Act (1909), and the

Bul. 663] The San Joaquin Experimental Range 85

Stock-Raising Homestead Act (1916), enabled settlers to obtain private
holdings in many cases considerably larger in acreage than was allowed
by the Homestead Act of 1862. Considerable land was acquired through
mining claims. By the very manner of acquisition, resulting holdings
were small. There is no doubt that many of those who engaged in cattle
production did so only after the exhaustion of their gold deposits. While
there have been consolidations of "ranches" or farms in the area by either
lease or purchase, there are still many small units — too small for efficient
operation.

About the same time the small operators were occupied in founding
small herds for the supplying of the mining communities, larger oper-
ators came into the San Joaquin Valley to the west. The herds of the
latter were kept in the San Joaquin Valley for most of the year, grazing
in the lowlands after the rains started the grass. These men moved to the
foothills and mountains with their herds after the feed of the valley was
gone, and remained at the higher elevations while the feed lasted. Both
the smaller and larger beef cattle operators are found here until the
present day. There is, however, one great change which has come about :
the small operator once had his consumer at his door — today he does not
have the same market since the miners have in most cases departed.

The type of farming carried on in an area is usually the resultant of
two sets of determinants — natural and economic. The natural conditions
may be placed in three main categories — soil, topography, and climate.
In the granite area the lands are thinly forested. The soils vary from fair
to poor and are nonirrigable and somewhat droughty. Even with a cli-
mate which might be suitable for certain crops the soil conditions and,
to a large extent, the topography favor livestock production. Some hay
is produced and on some of the farmsteads suitable areas are utilized for
the growing of fruits and vegetables for local use. Where the foothills
shade off into the Great Valley some grain is grown — but these localities
are really outside the granite area.

The chief economic forces which unquestionably have influenced the
development of the beef-cattle industry in the region have been the
varying demand for beef, and labor conditions. In addition, such factors
as the size of the enterprise, calf crop, death losses, and indebtedness in
the area itself have had a marked influence. Since the beginnings of the
beef-cattle industry here, no doubt the experience, training, capabilities,
and desires of the operators have been influencing factors in the develop-
ment of the enterprise in the area.

Size of Business. — Both the acreage controlled and the number of
cattle held vary widely. Variations in holdings range from 100 to close
to 20,000 acres while the number of cows kept by individuals vary from
4 to over 400. The term "average holding" for this area is meaningless.

86 University of California — Experiment Station

By far the greater number of operators control less than 3,000 acres, yet
the distribution by acreage up to this amount does not cluster about any
mean. A similar wide variation exists in the number of breeding cows
kept. On 88 ranches out of a total of 185 from which data were obtained
less than 35 breeding cows are kept while on 41 ranches over 100 are kept.

Composition of Cattle Holdings. — Data on the composition of cattle
holdings on 37 of the ranches surveyed were obtained in 1936-37. The
total number of all classes of cattle was 8,658 grouped by percentages as
follows : yearling heifers, 9.8 ; yearling steers, 9.4 ; two-year-old heifers,
2.1; two-year-old steers, 8.5; cows three to seven years old, 38.9; older
cows, 5.1 ; calves, 14.5 ; weaners, 8.5 ; bulls, 2.1 ; aged steers, 0.9. These
percentages varied on the different ranches. A statistical separation into
groups of different sizes did not reveal a definite trend in these classifica-
tions.

Size of Breeding Herd. — The total number of cattle handled by stock-
men fluctuated considerably as the result of available range, winter feed
supply, cattle prices, and other factors. The number in the breeding herd
seemed to remain fairly constant, but the latter varied so greatly from
ranch to ranch that the average of 91 is of little significance. Small herds
predominated ; 144 out of the 185 from which data were obtained had
less than 100 breeding cows, and of these, 88 herds had less than 35 cows
and 56 from 36 to 100 cows. Of the 41 herds with over 100 breeding cows,
26 had from 101 to 200 cows, and only 15 ran over 200. The larger part of
the operators with small herds were "part-time" farmers. Their liveli-
hood was obtained from road work, woodcutting, government aid, and
working for neighbors, in addition to the keeping of the herd of beef
animals. Most of the operators having more than 35 or 40 cows relied
chiefly upon the income from the cattle as the source of livelihood. While
some of the latter operators may have had one or two additional side
lines, their main business was raising beef.

From observations and questioning, it appeared that from 100 to 125
cows constituted a herd which required the services of an owner to be
fully utilized — perhaps with the hiring of additional labor during cer-
tain seasons of the year — or in which the labor of the family was utilized.

The outfits with over 200 head of breeding cows usually controlled
large units of range land, and as a rule employed from one to three men
in addition to the operator during a part of the year. In the main, this
latter group was composed of the older settlers, whose holdings have
been acquired over a considerable period of time.

Land Tenure. — Of the 185 ranches from which data were collected,
over half (96) were handling a combination of both owned and leased
land ; 81 owned land which they operated ; while 8 were tenants. Gen-

Bul. 663] The San Joaquin Experimental Range 87

erally speaking, the smaller operators leased a smaller percentage of
their holdings than the larger ranchers. The leased land was acquired
from private individuals and from federal, state, and county govern-
mental agencies. Approximately a third of all the operators used na-
tional forests for pasture from three to five months. As a rule, the
operators with the larger number of animals more frequently utilized
the national forest.

The acreage of the ranches from which data were procured included
350,202 acres, and of this area approximately a third was leased. Less
than 10,000 acres was recorded as crop land. The crop land is devoted
largely to growing oat hay, which is used for horse feed and during the
winter months for some of the poorer range cattle.

Somewhat over half of the leases were for a one-year period. Fre-
quently leases were renewed by the same parties. The majority of the
lessors were nonresident owners.

Rental price varied, the amount depending largely on the physical
features, forage, and location. In the year of study, prices varied from
$0.10 to $1.00 per acre, with an average of $0.31. Highest rentals were
paid for land in the slate formation bordering on the granite while the
rough land with granite outcroppings rented for the lowest price. Sev-
eral operators have had access to certain lands in return for the payment
of taxes and the upkeep of fences. The average assessed valuation of
range lands in the area ranged from approximately $2.50 to $3.00 per
acre. The tax rate per $100 of assessed valuation was $1.17 and $1.18 in
Madera and Fresno counties, respectively, for lands classified as being
outside of special assessment districts. The tax would therefore range
from about 2.9 to 3.5 cents per acre.

Feeding Practices. — Information on feeding practices obtained from
a cross section of over a hundred (105) ranchers in the area indicates a
variation from ranch to ranch according to the physical setup of the
range controlled, the resources of the operator, and other conditions. Of
the 105 ranchers, 15 stated that year-round grazing is the only provision
for feed. The remaining 90 use some supplementary feeds, consisting of
grain hay, cottonseed cake, and barley — the latter two being fed sep-
arately or in combination. Among those using supplements on the range,
the prevailing practice is to feed breeding animals and other stock in poor
condition in order that the winter might be survived ; as a result supple-
ment-feeding is confined to the one or two months' period from the time
the rains damage forage until green grass becomes available. Only 8 of
the 105 operators interviewed do any finishing of cattle for market.
Hay, cottonseed cake, and barley are used for such feeding.

Almost half of the operators (49 out of 105) use a system of seasonal

88 University of California — Experiment Station

grazing. Cattle are held on the lower foothill ranges during the winter
months, and frequently supplement-feeding is practiced. When the
green feed gains strength, the cattle increase in weight and condition ;
some steers and older cows are salable as grass-fat cattle. With the ad-
vancing season and the subsequent drying of the grass on the lower
ranges, the stock are moved to range at higher elevations. If the operator
holds a grazing permit, he moves the allotted number into the national
forests. Normally cows and calves are sent but in some instances other
stock are included.

Operators owning or controlling range at both lower and higher
elevations have access to green feed for longer periods, since the growing
season of range grasses is about two weeks later on the average for each
1,000 feet increase in elevation in this area.

The majority of those practicing year-round grazing on owned and
rented land, or both, are largely operators of small herds. A number
attempt to rotate pastures for their stock. Others, limited largely to a
few part-time operators, graze the entire holding throughout the year.

About a third of the operators graze stock in a national forest for
from three to five months. Most of these persons possess herds of over 36
cows and many of them are dependent upon the continuance of forest
permits to remain in business. In addition to lands in the national forest,
approximately 15,000 acres of Indian lands are grazed by Indians or by
ranchers having access to such lands.

In the investigation made for 28,074 cattle on ranches from which
adequate data were available, the private range furnishes slightly less
than 92 per cent of the total grazing, while somewhat over 8 per cent is
supplied by the national forest.

Stocking the Range. — In an area of varied topographical soil condi-
tions and conditions of management, there are differences in the number
of acres of range per animal within the area itself. Some units are over-
stocked, while there are parts on which livestock is not grazed. The av-
erage rate of stocking, revealed from information obtained from 185
operators, is approximately 23 acres per animal unit per year. On a few
of these ranges, some supplement-feeding is done during dry periods.
No account is kept of the amount fed. Ranchers in the lower "bald hill"
section on the slate soil report about 12 acres for each mature animal,
while ranchers using some of the poor granite range require as much as
40 acres. Few of the latter require, in addition, supplements for winter
feeding. With such great variation, any generalization would be difficult
to make.

While the amount of land not in use is small, there are such areas
resulting from the nontaking of lands of high rents, unsettled estates,

Bul. 663] The San Joaquin Experimental Range 89

certain Indian allotments, and certain lands occupied by individuals
who are not interested or concerned with the cattle business. In certain
areas, isolation and topography prevent the use of land.

When cattle prices are relatively high, it is difficult to rent range in
the area. Many livestock ranchers require sufficient forage for yearlong
operations. Such operators attempt to rent pastures by the acre or at a
stipulated price per animal per month.

The demand for national forest range is such that all available range
is being occupied. Normally, those seeking national forest ranges keep
from about 36 up to 200 cows. The use of such ranges by these persons
is so closely connected with the uses of private range that any serious
disturbance in the permittee system would cause stockmen to make con-
siderable adjustments in their operations. Such adjustments might be
the obtaining of summer range or the feeding of supplements.

In the area, there is considerable discussion among the operators as
to the difference between operations confined to the Great Valley and
those in the area. The general trend of opinion is that the former opera-
tions are perhaps more expensive as far as operating costs are concerned,
but it is generally admitted that they result in a higher-grade and better-
finished animal product. Only a few of the Valley operators are found
among the permittees of a national forest. Some have grazing pref-
erences and attempt to control private land within the boundaries of a
national forest in this manner.

Many of the national forest ranges are poor and can be used only to
aid in the production of feeder stock. Long stock drives are often neces-
sary en route to and from these ranges. This sometimes necessitates the
use of additional supplementary feeds on returning to the lower ranges.
Ranchers in the Valley agree that death losses are high in the national
forests. Another problem which confronts the stockman using the na-
tional forest is that in recent years there has been a great increase in
roads, making considerable disturbances as a result of traffic. Some op-
erators have avoided this trouble by trucking the cattle, but the smaller
operator cannot always follow this procedure.

Breeding Practices. — In general, three different breeding practices
are carried on : (1) yearlong, in which the bulls are allowed to run with
the cows at all seasons; (2) semicontrolled, in which the bulls are re-
moved from the breeding herd for from 3 to 8 months; (3) controlled,
in which the bulls are segregated from the breeding herd under a
definite plan of operation and in which a definite program of herd im-
provement is in force. By far the larger number (approximately three
fourths) of the ranchers practice yearlong breeding. Almost another
fourth segregate the bulls from the herd except in spring and summer.

90 University of California — Experiment Station

A very small number of ranchers select and segregate the heifers from
the breeding herd and breed them as two-year-olds. Most ranchers agree
that the latter is a desirable practice, but they are not equipped with the
necessary pastures and ranges to separate the young heifers from the
herds. The size of some herds is such that the additional labor necessary
for controlled breeding is far too costly.

Types of Bulls and Breeding. — Approximately one half of the op-
erators are using purebred bulls. The remaining half use grades or cross-
breds and animals with miscellaneous blood. Some of the operators using
bulls of miscellaneous blood think they cannot afford purebred sires.
They often rely upon bulls of their own breeding or else pick up a cast-off
or grade bull from their neighbors. Bulls turned on the national forests
must at least meet the state law requirements.

Considerable crossbreeding is practiced in the area, one of the most
common crosses being the Hereford bull on Shorthorn cows. The claim
is that this cross results in a larger calf and a larger milk flow. Some
operators buy yearling bulls and condition them before use, in order
to get the bulls accustomed to the range and in good breeding condition
before turning them in with the herd. This has an important influence on
the calf crop and the nature of the calves, but it is considered by some
ranchers to be too expensive a process.

A number of the purebred bulls are used for a longer period than is
thought to be desirable. Trading of bulls in order to avoid this practice
is not uncommon. Most operators plan to purchase a young bull, use him
two or three years and then trade with a neighbor, who is following a
similar practice. When bulls get too old to trade readily, they are sold to
the butcher and replaced by younger animals. In many cases where the
finances of the operator are limited, he will continue to save bull calves
from his own herd which show favorable or outstanding characteristics.

Calf Crop. — The number of calves dropped averaged 79 per 100 cows
maintained in breeding herds. The most startling situation, however, is
revealed in examining the range in numbers of calves dropped, namely,
24 to 100. Out of the 97 herds from which data were obtained, only 2
reported the number of calves dropped equal to the number of breeding
cows. One of these herds contained 6 breeding cows, the second 65. In an
examination of the averages by size of herd, no particular difference was
found between the larger and smaller herds. The figures, however, as
reported, were significant in that all size groups indicated extremes from
uneconomically low to substantially high percentages. The largest per-
centages of calves dropped per 100 cows maintained in breeding herds
were observed on those ranches that either had the best range conditions
or made provisions for the care of breeding cows.

Bul. 663] The San Joaquin Experimental Range 91

Calving Season. — The largest number of operators reported that the
majority of their calves, however, were dropped during February,
March, and April, although several reported calves coming in other
months. There is some variation as the result of ranch location and
marketing practice. Some stockmen, principally those in the lower eleva-
tions, • prefer fall-dropped calves. They can ordinarily count on green
feed in December and January and the resulting heavier milk flow. On
the other hand, some ranchers firmly believe that calves which come early
in the spring usually winter better than those which are born late. If
a rancher follows the practice of selling weaner calves in the fall, he
must produce early spring calves in order to gain the needed amount
of weight by fall. The latter is particularly the case for stockmen who
graze cattle on the national forests, in that the calves must come suffi-
ciently early to make the trip to the mountains and also to make best
use of the mountain range. Moreover, fall calves, if taken to the national
forest, have sufficient age to be counted as mature animals, a disadvan-
tage to the stockmen.

Death Losses. — Data on death losses of cows and calves were obtained
from approximtaely 65 herds for the year 1936-37. The smallest per-
centage of losses occurred in the groups which numbered from 100 to
200 cows. The largest losses, on the other hand, were in the smaller
herds of under 35 cows. The reason for the latter losses was probably
because of the fact that the smaller operators were running cattle as a
side line. It is only natural that operators who divide their time between
some outside activities and the running of cattle may tend to let the
cattle shift for themselves. For the herds from which data were obtained,
the percentage of calf mortality was 6.1 per cent, while the percentage
of cow mortality was 4.3 per cent, for the year 1936-37.

The heaviest losses occurred among calves and in calving two-year-old
heifers. Causes for death losses were varied. A considerable loss in small
calves occurred during the calving season. Losses resulting from preda-
tory animals, dehorning, and blackleg were relatively small. Fully 90
per cent of the operators reported that they vaccinated their calves for
blackleg. As a result, losses from this cause were practically nil for the
period studied. Some operators reported losses from accidents. A goodly
number of losses were reported as "cause unknown" or "animals miss-
ing." Many of the latter die on the range, and their carcasses are not
found. Several cattlemen believe that cattle rustlers steal animals now
and then, principally young unbranded calves. Two operators reported
losses from eating poisonous plants, such as larkspur and black fern.

Another cause of loss which many ranchers maintain is rather serious
is the eating of too many acorns by cows during gestation. It is said that

92 University op California — Experiment Station

this results in the typical "acorn calf," born with very short legs, over-
sized head, and undershot jaw. Some "acorn calves" also show a bloated
condition. Some stockmen are sure that the situation is more serious
following large acorn crops throughout the oak belt of the foothills.

The entire question concerning "acorn calves" is of a controversial
nature. Research work on "acorn calves" does not confirm the ideas which
many possess on the subject ; investigators state that acorns cannot be
considered the primary factor in the causation of the deformity. (See
the section "Experimental Herd Management.")

Dehorning and Branding. — The practice of dehorning is increasing.
Most operators in the area dehorn, brand, castrate, and vaccinate at the
same time, most commonly in April and May. In herds where calves are
dropped throughout the year, these operations are sometimes performed
two or three times a year. The common practice is to brand calves before
they are weaned, in order to identify ownership and thus discourage
thieving.

Weaning. — Most of the calves are weaned when from six to seven
months of age. Stockmen using the national forests wean their calves in
September or October, or as soon as they come to the lower ranges. Being
in good condition when they come out of the mountains, they can be
separated from their mothers and fed a little hay or concentrate, or else
placed on pasture reserved for this purpose.

Some of the smaller operators, nonforest permittees, and stockmen
who do not get calves of fairly uniform age oftentimes allow the cows to
wean the calves, although this practice is not generally followed. Wean-
ing practices are usually determined by the calving period and the
nature of the selling practices.

Culling Practices. — On some of the better-managed ranches, cows are
culled on the basis- of performance, as well as on age, type, and con-
formation. Cows that are shy breeders, poor keepers, or those that pro-
duce poor calves are culled. The majority of stockmen follow what might
be termed a "mechanical" culling process. The process is a simple one
of retaining even somewhat undesirable cows as long as they winter
fairly well and continue to produce fairly good calves. Smaller operators
apparently cull less consistently than others. A few of the smaller oper-
ators purchase dairy calves in the lower valleys and graze them in the
foothills, attempting to grow them and eventually sell them as milk or
beef stock, according to the sex of the calf. Dairy stock as a rule do not
make as satisfactory gains as those made by beef -type animals on the
range of the area, and as a result there are many poor animals which do
not actually pay for their keep.

The average age of the cows in the majority of herds ranges from

Bul. 663] The San Joaquin Experimental Range 93

three to seven years. When culled according to age, cows are sold when
they are nine to ten years old ; cows older than ten years are uncommon.

Marketing Problems and Practices. — Available feed supplies and the
condition of the range influence the marketing of cattle from this area
in a material way. Some cattlemen, however, operating under a load of
heavy indebtedness, must sell in accordance with stipulations which
govern the granting of a loan. Normally speaking, plentiful feed and
good range tend to hold cattle for a longer period of time. Conversely,
when the feed is short and when the range is poor, stock cattle, as well
as calves, may be sold. The market conditions also play an important
part.

There are apparently two times during the year when cattle are largely
marketed from the granite area. In the spring, grass-fat animals are
disposed of in May, June, and July ; and range-fat animals in October,
after they have been on the mountain ranges. A goodly number of the
grass-fat steers sold from the lower ranges in the spring go to the central
markets at San Francisco and Los Angeles, where they are sold by
weight. Some local sales are made to local buyers and speculators, who
in turn gather their purchases at a common loading point for further
shipment to the central markets. Fresno and other Valley cities take
considerable numbers of cattle, and these are usually purchased direct
from producers located in the area.

The majority of steers sold out of the area go as feeders, unless they
have been "warmed up" by being fed concentrates for a period. There
are, however, many grass-fat steers which are sold to the beef trade in
the spring following the green-grass period.

Ranches can be classified on the basis of the type of production pre-
dominantly practiced, as follows: (1) Operating on a steer basis and
selling steers at two years of age or over; (2) operating mainly on a
cow-and-calf basis and selling weaners; (3) having no definite yearly
plan but marketing animals mainly on finish and market conditions as
determining factors.

Approximately one half of the operators interviewed could be classed
in group 1, while groups 2 and 3 comprised the balance. A number of
small operators sold their calves as veal to local Valley butchers. Since
the advent of the truck and trailer, stockmen can load one or two animals
when going to town, and they then can visit the various slaughtering
houses or butcher shops to see where the best prices can be obtained.
Many of them supply some small beef plants regularly.

While weights and prices vary from year to year, nevertheless some
of the data which were obtained for 1936-37 present some interesting
studies. The average weight of two-year-old steers which had been fed

94 University of California — Experiment Station

supplements in some degree at least, was 968 pounds ; the range in weight
was from 800 to 1,100 pounds. The average price received for these
animals was $7.20 per hundred pounds, the range being from $6.00 to
$8.50.

During the same year the average weight of two-year-old steers from
a group of twelve ranches which depended entirely on range for mainte-
nance growth was 825 pounds. The average selling price of these animals
was $6.70 per hundred pounds. While the comparison is not such that
one can rely on the results, nevertheless it can be pointed out that the
animals fed supplemental feeds weighed on the average 143 pounds more
per head and they sold for an average of 57 cents more per hundred-
weight, because of their better condition.

During this same year the seven- to eight-months-old weaner calves
ranged in weight from 300 to 450 pounds. The sale price varied from 7
to 7% cents per pound.

Other Livestock Enterprises. — Approximately a fifth of the operators
interviewed kept two or three sows. These farrow twice a year, pork for
the family is obtained, and some hogs are sold. On most ranches feed for
hogs is hauled from the San Joaquin Valley ; therefore, hog raising is
not an enterprise of any considerable importance. A very few sheep are
kept but the number is so small that they can be ignored. Now and then
an operator is found who milks a few cows and sells the cream. The
majority maintain one or two cows for the family milk supply. Most
operators keep a few chickens, in order to provide eggs for the family's
use, and four flocks of over 1,000 turkeys were found.

Flexibility in Ranch Management in the Granite Area. — Cattle prices
are affected by influences over which the rancher within the area or else-
where has little or no control. Improvements in management practices
aid many cattlemen in taking advantage of some marketing and price
situations. Any blanket recommendations for all operators should be
viewed with caution — if not with suspicion.

Operators can be in a somewhat more advantageous position if their
ranch organization is such that some choice may be exercised as to class
of animals to be sold ; for example, whether calves, yearlings, feeders,
or fat cattle. An ample supply of range for winter feeding will often
make for a far more flexible program than is carried on at present;
market conditions and prices of different classes of cattle can then have
some determining influence on sales. For example : Are calves to be sold
to the feeder trade or carried on to the yearling stage ? Are steers sold
to better advantage as two-year-olds or three-year-olds? In a flexible
market program, operators normally would sell the type of cattle for
which prices are favorable and hold those for which the demand is slight.

Bul. 663] The San Joaquin Experimental Range 95

Normally, efforts to keep up a breeding herd would be justified — not
only because such a procedure leads to flexibility of operation, but be-
cause of the time and expense involved in bringing a new herd together.
In operations in which an improved breeding herd is not important, such
recommendations would be invalid.

Seasonal cattle movements into California for both slaughter and
feeder purposes are fairly regular ; the larger movements usually start
in October and, in general, last for approximately six months. This sit-
uation results largely from the demand 'of the California markets and
from the cattle management plans of the producers in the localities from
which cattle are shipped into California. Some adjustment in local
operations might be made to take advantage of this situation.

Feed prices would loom large in any increased flexible arrangement
made by producers in the area. Not only must kinds of feeds and quan-
tity be considered, but the transportation problem particularly exerts
an influence on these prices in the granite area.

One of the principal problems facing the operator in the area is that
of acquiring control of range that is adequate and suitably located so
as to provide range units of suitable size for economic operation. The
somewhat unstable short-time leasing system and high rental prices in
many cases prevent not only a flexible program but in many ways present
difficulties to present programs.

INTERRELATIONS OF RODENTS AND OTHER
WILDLIFE OF THE RANGE

E. E. Horn and H. S. Fitch

Many factors, including the climate, topography, amount and distribu-
tion of food, water, and shelter, determine the kinds and abundance of
wildlife occurring on any area. The animal life thus reflects all the con-
ditions of the region where it occurs. Some kinds, present the year-round,
are adjusted in their annual schedules to seasonal changes ; others are
migrants which may stay only for brief periods in a given locality, or
may merely pass through it. The entire wildlife population of an area
is dependent for food either upon vegetation or upon other animals
which directly or indirectly obtain their sustenance from plants. The
economic bearing of each species is determined largely by its feeding
habits.

In our foothill range lands the rodents are of most general importance
because of their enormous numbers and their feeding, which largely is
levied upon plants valuable as forage for domestic stock. Of the preda-
tors, some are beneficial because they prey upon rodents, although in
some cases they may be destructive by preying upon cattle, sheep, or
domestic animals, as do the mountain lion, coyote, and bobcat on oc-
casion. Any study of the wildlife in an area must therefore consider
many different creatures, all significantly interrelated in their influ-
ences ; each, by its presence and activities, affects all the others.

An economic problem in land utilization was created when the pioneers
brought domestic livestock onto ranges formerly used only by wild
grazing animals. The resultant increased use of the forage effected
changes in vegetation, and the newly created habitat conditions were
less favorable to some kinds of wildlife but favored the increase of others.
Destruction of forage by wildlife came to assume increased importance
as free or cheap range lands were used up and competition in the graz-
ing industry became keener, necessitating the introduction of more eco-
nomical practices in land use.

The stockman's interest in wildlife centers in the question of what
these animals do to the range. Ranchers continually ask, regarding these
different animals : How much forage and what kind of forage do they
take ? What good they may do ? Can they be removed, and how much
would removal cost ? Would the increased returns make control an eco-
nomical practice ?

With the establishment of the San Joaquin Experimental Range, the
Fish and Wildlife Service (formerly Biological Survey) joined coopera-

[96]

Bul. 663] The San Joaquin Experimental Range 97

tively with the other agencies in a study, designed to carry over a period
of years, to determine the economic importance and interrelations of
wildlife on the granitic soil areas of the Sierra Nevada foothills of Cali-
fornia. This study is intended to bring out the relations between species
of wildlife and plants, and the interrelations between different wildlife
species which have bearing on the question of grazing and land utiliza-
tion in the foothill ranges.

Rodent activity affects the range in many ways. Each rodent species
is highly selective in its choice of plant food from the numerous species
available, and the degree of competition with livestock varies. Through
selective feeding, rodents may hold certain plant species in check and
keep them relatively scarce in the composition. Increased populations
of rodents may bring about the reduction, or even the elimination, of
certain plant species preferred by them.

Burrowing animals, most notably the rodents, but also toads, worms,
spiders, beetles, and ants, have important effects on the soils of the areas
they occupy. They move enormous quantities of earth, mixing the differ-
ent soil layers with each other, with plant material, and with their own
dead bodies and excrement. Changes in porosity, in susceptibility to
erosion, and in fertility of soil result ; and probably general productivity
is increased. Beneficial effects from this type of activity partly compen-
sate for the damage done in destroying forage. The economic role of each
species is thus determined by the sum total of its multiple effects.

Great changes from year to year occur in numbers of some animals,
and their effect varies correspondingly. Amounts eaten and trampled
each season may vary according to the total amount available as well as
the population of rodents present. Hence rodent damage is not a constant
factor, but changes greatly from year to year.

THE WILDLIFE AND ITS INTERRELATIONS

Life Zones and Habitats. — The fauna of the Range is fairly typical of
the oak and digger pine association of the foothill belt of the Upper
Sonoran Life Zone in which the area is located. From an ecologic stand-
point, one of the most distinctive features of the region is the presence
of granitic rock in extensive outcrops, or loose piles of large boulders
(plate 2, lower) . Always closely associated with the rock and apparently
in some way dependent on it here, is the most abundant tree species of
the region, interior live oak. Both the rocks and the live oaks have sig-
nificance in controlling the occurrence and abundance of various wildlife
species. Many species depend on the rocks either as a permanent home
or as a place of refuge. Notable among these are the canyon wren, the
rock mouse, the wood rat, the night snake, and the skink, all of which are
common on the area.

98 University of California — Experiment Station

Many of the vertebrates, especially birds, are partly or wholly de-
pendent on the oak woods as the essential part of their habitat. Among
these are the California woodpecker, the plain titmouse, the phainopepla,
the California jay, and the slender-billed nuthatch. Extensive areas of
open grassland constitute the preferred habitat of such other species
as the kangaroo rat, the white-footed mouse, the meadow lark, and the
horned toad. The Range is near the lower limit of distribution of chapar-
ral in the foothills, hence chaparral is scarce. Nevertheless, many of the
wildlife species on the area are closely restricted to brush patches. Among
these are the California thrasher, brown towhee, spotted towhee, Mer-
riam chipmunk, and California racer.

Mammals of the Range. — The mammals of the region represent six
main groups: the carnivores, represented by the coyote, gray fox (fig.
16, A), bobcat, raccoon (fig. 16, B), badger, striped skunk, and weasel;
the rodents — kangaroo rats (fig. 17, A), ground squirrel (fig. 17, B),
pocket gopher, and field mouse ; the rabbits — blacktailed jack rabbit and
cottontail ; the bats — several species ; the marsupials and insectivores
each with a single species — the opossum and the mole respectively.

Birds of the Range. — The birds of the Range include, in addition to
the resident species, some of which have been mentioned above, a greater
number of migrants which regularly cross the area or settle in it tem-
porarily in the course of their seasonal movements. Most abundant in
this group are several kinds of sparrows, juncos, and other small species,
but it also includes such waterfowl as the white pelican, great blue heron,
white-fronted goose, and mallard. Many of the most conspicuous birds
that breed on the Range, including the Arkansas kingbird, Bullock oriole,
and mourning dove, are not present during the winter. The raptorial
species include the red-tailed hawk (fig. 16, C), Cooper hawk, sparrow
hawk, horned owl, barn owl, long-eared owl, and screech owl, all of
which regularly nest on the area, and the golden eagle, bald eagle, rough-
legged hawk, marsh hawk, sharp-shinned hawk, prairie falcon, and
pigeon hawk, which occasionally occur on the area, but are not known
to nest there. The road runner is a common resident on the Range.

Several wildlife species on the Range are classified as game and have
economic importance in that connection. The California valley quail
heads this list.

The mourning dove is common in this region in summer and fall, and
is hunted locally almost to the same extent as the quail. Among the
mammals, the gray squirrel and cottontail are commonly considered
game, but neither is a satisfactory game animal, either on the Range,
or in the state at large. The gray squirrel is scarce and the cottontail
cannot be safely handled, because it is a carrier of tularemia.

Bul. 663] the gAN j0AQUIN Experimental Range

99

< Fig. 16.— ^4, Gray fox. Rodents constitute part of his diet. B, Raccoon— an
interesting carnivore. C, Young red-tailed hawks in their nest. Note the hind
quarters of a ground squirrel at the left.

100 University of California — Experiment Station

Fig. 17. — A, Kangaroo rats. These animals of nocturnal habit feed heavily
upon the seeds and leaves of forage plants. B, Digger ground squirrel eating
filaree.

Reptiles of the Range. — The reptiles, especially the lizards, are the
most abundant vertebrates of the region. The Pacific rattlesnake is prob-
ably the commonest of the snakes, and, with a population of at least one
to the acre, its numbers greatly exceed those of any other predator of
wildlife in the region. The gopher snake, California racer, Boyle's king
snake, and two species of garter snake are also moderately common, so

Bul. 663] The San Joaquin Experimental Range 101

that they play important roles as predators, and figure also as food
sources to the bird and mammal predators.

Amphibians of the Range. — Because of the long dry season, and scarc-
ity of yearlong water, amphibians are poorly represented, in number of
species, on the Range. However, the California toad, spadef oot toad, and
tree toad all are common. These species are limited in their season of
activity above ground to the spring months when temperature is mod-
erate and humidity is high. The remainder of their annual cycle is
mostly spent underground, usually in rodent burrows. These species all
are dependent on the short-lived, seasonal streams and pools for breeding
places, and in dry years their reproduction is greatly reduced. In favor-
able seasons, the shallow pools of water everywhere are swarming with
their tadpoles, which attract numerous predators, including the great
blue heron.

Ecological Relations. — From the foregoing account it is evident that
the ecology of the region is complex. Seasonal changes are marked. The
mild, humid winter season produces an abundance of green herbaceous
plants, but in late spring the vegetation suddenly dries out and dies.
Through the long, dry season, extending over nearly half of the year,
succulent vegetation is exceedingly scarce and many kinds of resident,
plant-eating animals are forced to alter their feeding habits, and to
subsist mainly on dry vegetation. The summer climate is severe, and
distribution of shade and moisture become critical factors. In response
to these conditions, populations of some species make seasonal movements
off the area to distant regions; those of others become less active or
entirely dormant for part of the year, while some change their habits,
and may be either diurnal or nocturnal at certain seasons. Of the small
vertebrates, rodents and lizards are by far the most numerous, and these
two groups include most of the key species which mainly support the
predator populations.

Lists of the Wildlife Species. — Of the wildlife species mentioned in
this account, the kinds of birds are well-known by standardized vernacu-
lar names, but common names are less well established for some of the
amphibians, reptiles, and mammals. To avoid any doubt as to the iden-
tity of the kinds referred to, the accepted scientific names of these ani-
mals are given below, listed in their classification order :

Amphibians
Spadef oot toad, Scaphiopus hammondii Tree toad, Hyla regilla
California toad, Bufo boreas

Kepttles
Brown-shouldered lizard, Ufa stansburi- Horned toad, Phrynosoma blainvillei

ana Whiptail, Cnemidophorus tesselatus

Fence lizard, Sceloporus occidentalis Yosemite skink, Ewmeces gilberti

102

University op California — Experiment Station

Keptile s — ( Continued )
California racer, Coluber lateralis Garter snakes, Thamnophis sirtalis and

Gopher snake, Pituophia catenifer ordinoides

Boyle's king snake, Lampropeltis getulus Night snake, Leptodeira torquata

Pacific rattlesnake, Crotalus viridis

Birds

White pelican, Pelecanus erythrorhyn- Barn owl, Tyto alba

chos
Great blue heron, Ardea herodias
White-fronted goose, Anser albifrons
Mallard duck, Anas platyrhynchos
Sharp-shinned hawk, Accipiter velox
Cooper hawk, Accipiter cooperi
Eed-tailed hawk, Buteo borealis
Rough-legged hawk, Buteo lagopus
Golden eagle, Aquila chrysaetos
Bald eagle, Haliaeetus leucocephalus
Marsh hawk, Circus hudsonius
Prairie falcon, Falco mexicanus
Pigeon hawk, Falco columbarius
Sparrow hawk, Falco sparverius
California quail, Lophortyx calif ornica
Mourning dove, Zenaidura macroura
Road runner, Geococcyx calif ornicus

Screech owl, Otus asio (sp.)
Horned owl, Bubo virginianus
Long-eared owl, Asio wilsonianus
California woodpecker, Balanosphyra

formicivora
Arkansas kingbird, Tyrannus verticalis
California jay, Aphelocoma calif ornica
Plain titmouse, Baeolophus inornatus
Slender-billed nuthatch, Sitta carolinen-

sis
California thrasher, Toxostoma redivi-

vum
Phainopepla, Phainopepla nitens
Meadow lark, Sturnella neglecta
Bullock oriole, Icterus bullocki
Sacramento towhee, Pipilo maculatus
Brown towhee, Pipilo fuscus (sp.)

Opossum, Didelphis virginiana
Mole, Scapanus latimanus
Weasel, Mustela xanthogenys
Striped skunk, Mephitis mephitis
Badger, Taxidea taxus
Gray fox, TJrocyon cinereoargenteus
Coyote, Canis latrans
Bobcat, Lynx rufus

Digger ground squirrel, Citellus beecheyi
Merriam chipmunk, Eutamias merriami
Gray squirrel, Sciurus griseus
Pocket gopher, Thomomys bottae
Pocket mice, Perognathus inornatus and
californicus

Mammals

Kangaroo rat, Dipodomys heermanni
Harvest mouse, Beithrodontomys mega-

lotis
White-footed mouse, Peromyscus mani-

culatus
Brush mouse, Peromyscus boylii
Rock mouse, Peromyscus truei
Wood rat, Neotoma fuscipes
Meadow mouse, Microtus californicus
Black-tailed jack rabbit, Lepus californi-
cus
Cottontail, Sylvilagus auduboni

RODENTS AND RABBITS OF THE RANGE

Rodents and their near relatives, the rabbits, as a group, are herbiv-
orous, and, therefore, on the range lands used primarily for livestock
grazing, their effect is mainly deleterious in using forage that would
otherwise be available to stock. Many species of rodents occur on the
Range, and each differs greatly in the amount and specificity of damage.
For most of the species, the economic bearing is either negligible or of

Bul. 663] The San Joaquin Experimental Range 103

minor importance, and three or four kinds are responsible for nearly all
of the rodent damage.

The extent of damage depends mainly on the size, abundance, and
food preferences of the species concerned. From the following account
it will be seen that this problem is exceedingly complex, involving a
knowledge of not only the amount and kind of food taken by an individ-
ual during a day, but also of the amount destroyed through burrowing
and trampling, or by cutting wastefully, and the extent to which the
foods taken are of kinds available or palatable to stock. To make any
appraisal of rodent damage it is necessary to know accurately the popu-
lation of a species and how it changes during its annual cycle, its seasonal
changes in food preferences, and the extent of increase or diminution
in its activity at different times of year. The amount and kind of vegeta-
tion present is an important factor determining the amount used by a
given population of rodents, and may even be a factor in governing the
numbers of rodents present.

The following discussion of the rodents and rabbits recorded on the
Range is intended to show the abundance and economic importance of
each. The four species which have the most important effect on the forage
are the ground squirrel, pocket gopher, kangaroo rat, and cottontail;
these species were, therefore, selected for intensive study. The ground
squirrel probably ranks first on the list as a forage destroyer, and also
is important as a reservoir of the diseases, tularemia and sylvatic6
plague, transmissible to humans. The pocket gopher was probably the
most abundant range rodent in 1938 and 1939; it subsists partly on
valuable forage plants, especially in winter, but it is also important as
a cultivator of the soil. The kangaroo rat is generally abundant, but
subject to violent fluctuations in numbers. The cottontail occurred in
a population of one to the acre in 1939 and is an important competitor
of stock for the forage crop.

Eleven other species occurring on the Range are relatively unim-
portant because of scarcity or specialized feeding habits which do not
compete seriously with livestock. The combined effect of all of them,
however, may be greater than it appears. The Merriam chipmunk is
scarce and usually found in brushy ravines, where it depends largely
on shrubby plants for food. The gray squirrel is likewise scarce. It
usually stays in or near digger pines and lives on the pine seeds. The
San Joaquin pocket mouse, present in a population of at least one to the
acre in 1939, is a small species which subsists mainly upon minute seeds.
The California pocket mouse is rare. The harvest mouse is even less

6 The term, "sylvatic" plague is used here to designate what has heretofore been
known as "bubonic" plague in native rodents.

104 University of California — Experiment Station

numerous, and is restricted to the larger swales which retain moisture
through the dry season. The white-footed mouse is one of the commoner
rodents, and was present in a population of at least one to the acre in
the summer of 1939 ; it occurs mainly in open grassland. The brush mouse
and the rock mouse are both common where there are rocks and brush
thickets ; populations were computed at several to the acre for both in
the summer of 1939. The wood rat is partial to situations where there are
large rock outcrops, brush thickets, or live oaks, where it may occur in a
population of several to the acre, but it tends to avoid open grassland.
Because it feeds mainly on leaves of shrubs and inner bark of twigs from
oak and chaparral, it is unimportant as a forage destroyer but may com-
pete with livestock for browse. The meadow mouse is scarce and confined
to a small portion of the Range, the larger swale bottoms with permanent
covering of thick vegetation, where the ground remains moist through
the summer. The jack rabbit is rare on the Range.

In the rodent study that follows special attention has been given to
the feeding habits of the important species, and detailed observations
on feeding have been carried out at all seasons. Live-trapping and mark-
ing of entire populations has been attempted as a means of determining
population densities and of obtaining information regarding the move-
ments, seasonal and long-time fluctuations in numbers, longevity, and
seasonal activity of the species concerned. All these phases of ecology
must of course be solved for each species to attain any accurate economic
appraisal of it. Some phases of these studies are still in their early stages,
but in others considerable progress has been made, and the results to
date are set forth briefly in the following account under the headings
of the species concerned.

Ground Squirrel. — The digger ground squirrel is the most conspicu-
ous rodent of the Range, because of its large size, abundance, and diurnal
habits (fig. 17 B) . This ground squirrel has a bushy tail which makes up
more than a third of the total length, well-developed and conspicuously
erect ears, well-developed limbs fitting it for either digging, climbing,
or running, and a brindled color pattern produced by alternate dark and
light bars on each hair. The underside is paler, and there is a whitish
patch in the shoulder region. The head is darker than the body, and the
eye is conspicuously rimmed with white. A large adult is about 18 inches
long. Adults weigh between 1 and 2 pounds, varying according to sex,
age, and condition. The squirrel has highly distensible internal cheek
pouches and often carries food by pouching it in the side of the mouth.
Seeds and acorns are often carried to favorite feeding places in these
pouches before being eaten, or to caches where they are stored for future
use.

Bul. 663] The San Joaquin Experimental Range 105

In their foraging, ground squirrels are influenced by a complex of
factors, including temperature, humidity, shade, location of preferred
food sources, predators, and presence of other squirrels.

With the start of fall rains green food becomes available and the
squirrels feed upon it extensively ; broad-leaf filaree, brome, and fescue
grasses are the staples of diet. During the early stages of growth, entire
plants are taken. Later in the season as the plants mature, the tender
leaves, and fruits are selected. During the spring, large quantities of
the immature fruits of filaree are consumed, and after the seeds ripen
these are extensively used as food.

As the season advances, the vegetation quickly dries out and dies. The
squirrels turn to seeds, to acorns, and the summer-growing tarweeds.
During this period of seed production the squirrels gather large quanti-
ties of seed. Some caching of food for future use is done. Table 16 gives
a list of seasonal foods of these squirrels.

The heaviest toll on the forage occurs in the spring when the popula-
tion is fully active and the young squirrels are out, for it is during this
period that the squirrels are at their highest numbers. Later, some of the
older squirrels become dormant, and many of the young have died or
have been caught by predators.

Marked differences were noted in the feeding of squirrels in the 1938
and 1939 seasons owing to the differing availability of the food sources
on the Range in these two years. These differences were mainly the result
of the heavier rainfall and later growing season in 1938. The summer of
1939 was much less favorable. During this season the crop of catchfly was
poor, loosestrife dried out early, and the yellow tarweed crop was only
a small percentage of that produced in 1938. The crop of live oak acorns
also was short. These foods, especially the live oak acorns and tarweed,
had mainly supported the squirrels through the summer of 1938. There
was, however, a heavy crop of blue oak acorns in 1939, and the squirrels
depended on these almost entirely for food throughout the dry season.
During this period adults lost weight and even half -grown young failed
to gain all summer, which suggested that acorns as a steady diet are
nutritionally deficient in some respect.

Permanently occupied squirrel burrow systems are of the colonial type
having from 10 to 50 open holes. Frequently such burrow systems are
each occupied by a single squirrel; more often two or three squirrels
simultaneously inhabit the same system. When the population is at its
maximum, large and favorably situated burrow systems may harbor ten
or more squirrels. On the Range, where the squirrel population is not
subjected to control by man, the number of burrows varies but little from
year to year and is relatively stable compared with the actual number

106

University of California — Experiment Station

TABLE 16

Seasonal Feeding of the Digger Ground Squirrel

Month

Principal food sources

Other staple food sources

regularly taken where

available

Food sources occasionally used

January

Broad-leaf filaree (leaves)

Leaves of popcorn flower,
brome, common foxtail,
and red maids

Leaves of fescue and buckthorn
weed; Mariposa manzanita
(flowers); acorns; puff balls; toad-
stools

February

Broad-leaf filaree (leaves)

Leaves of red maids and
popcorn flower

Leaves of fescue, common foxtail,
brome, tarweed, ground lupine,
common chickweed, hairy lotus,
fine-leaf lotus, buckthorn weed,
locoweed, and baby blue-eyes;
acorns; Mariposa manzanita
(flowers) ; toadstools

March

Filaree (fruits); ground
lupine (entire plants)

Red maids (flowers,
stems, leaves); filaree
(seeds and leaves)

Leaves of popcorn flower, baby
blue-eyes, and yellow tarweed;
acorn

April

Broad-leaf filaree (seeds)

Popcorn flower (seeds)

Ground lupine (fruits and entire
plants) ; phacelia (leaves) ; popcorn
flower (stems); clovers (leaves);
soft chess (seeds); blue dicks
(leaves); acorns; birds' eggs

May

Birdseye gilia (flowers);
yellow tarweed (flow-
ers) ; seeds of filaree and
ripgut grass

Seeds of catchfly, slender
oat, and common fox-
tail; red-stem filaree
(seeds and fruits) ; clov-
ers (flowers, stems,
leaves); ground lupine
(seeds and flowers)

Blue dick (leaves); puffballs; birds'
eggs

June

Seeds of Wright's tarweed
and catchfly ; yellow tar-
weed (leaves) ; loose-
strife (leaves, flowers,
stems)

Seeds of common foxtail,
ripgut grass, and digger
pine; flowers of godetia
and clover; wedgeleaf
ceanothus (berries)

Seeds of slender oat, Napa thistle,
buckthorn weed, navarretia, and
ripgut grass; leaves of hollyleaf
buckthorn and California ever-
lasting; Spanish clover (stems,
pods, leaves); birds' eggs

July

Yellow tarweed (leaves) ;
acorns

Spanish clover (fruits);
loosestrife (leaves and
stems)

Stems of Spanish clover, filaree, and
popcorn flower ; navarretia (seeds) ;
eriogonum (flowers)

August

Acorns

Turkey-mullein (stami-
nate flowers); ripgut
grass (seeds)

Loosestrife (seeds)

September

Seeds of yellow tarweed
and turkey-mullein;
acorns

Ripgut grass (seeds)

Feces of squirrels and rabbits

October

Yellow tarweed (seeds);
acorns

New grass (brome, fescue,
common foxtail)

Leaves of filaree and popcorn flower

November

New grass ; acorns

Red maids (leaves)

Leaves of filaree and popcorn flower

December

Leaves of brome, filaree,
and fescue ; entire plants
of filaree

Popcorn flower (leaves)

Leaves of slender oat, buckthorn
weed, and sage

Bul. 663] The San Joaquin Experimental Range 107

of squirrels present. Elsewhere, a reduction of 50 per cent in squirrel
population has been observed with but little disuse of holes or burrows.

The burrowing activities of these squirrels has a twofold effect upon
the production of forage that might be available for cattle. The burrows
are fairly extensive in size, being from 3 to 6 inches in diameter and from
50 to 200 feet in length. Enlarged nest and food-storage pockets bring
the displacement of earth to over 20 cubic feet for some systems. This
soil is brought to the surface where it is dumped on growing vegetation.
Digging is continuous so that a certain supply of soil is deposited on the
surface each year. Through this and other squirrel activities a certain
ground area is kept bare of vegetation in the immediate vicinity of the
burrow. It has been found in studies elsewhere that from one fifth to
one third of the square area that just contains a burrow system was
denuded of vegetation by squirrel action. Where the squirrel population
is high, and burrows numerous, this activity takes a portion of the area
out of production. Complete figures are not yet available for such de-
nuded areas on the Eange.

On the other hand, finely pulverized mineral soil, old nests, carcasses
of dead squirrels and the bodies of other animals (mice, snakes, lizards,
toads, and insects) that have died in the burrows, together with feces,
hulls, and other plant material, are all admixed in the soil of the mounds.
This results in high productivity of this soil, as indicated in the increased
plant growth there. Vegetation on these mounds on areas no longer
occupied by squirrels is more luxuriant, of greater density and height,
and remains green for from one to two weeks longer than on adjacent
undisturbed soil.

Elsewhere in California, it has been recorded that no increasing forage
production at old burrows could be found three years after the squirrels
had been removed. At burrow systems occupied during the current sea-
son, the vegetation is noticeably different from that on adjacent soil.
Red-stem filaree is a highly characteristic plant of the burrow mounds,
and may form a dense mat over several square yards within an area of
burrow mounds, although it is scarce in the composition a few yards
away in undisturbed soil. This filaree is a favorite squirrel food. Buck-
thorn weed also thrives on the edges of the burrow mounds.

In their annual feeding cycle, the squirrels dig out considerable quan-
tities of food, including certain subterranean fungi, and roots of such
plants as broad-leaf and red-stem filarees, common foxtail, and Bermuda
grass. When digging for these, particularly in sandy soil, they may
disturb the surface layer over extensive areas.

The total effect of squirrel burrowing upon soil and forage production
is the subject of further study at other locations in California.

108 University of California — Experiment Station

While ground squirrels in this region are individually restricted to
small foraging areas, there are occasional shifts of the home base, ap-
parently correlated with changing availability of food supply, shade,
and other factors. Such shifts are usually short, so that the squirrel
remains within the same general area with which it is familiar; but
occasionally much longer movements are made, taking the squirrel far
beyond the limits of its original foraging range. It may be noted that
there is little incentive for this type of movement on the Range where
the whole country is relatively uniform and seasonal changes in avail-
ability of food are mainly between adjacent knolls and swales, or trees
and open grassland. In areas providing contrasting habitats, for in-
stance, places where range land is adjacent to cultivated crops, extensive
movements from less to more favorable areas are much more general.

As stated before, evaluation of rodent pressure on the Range depends
upon the total population of rodents present. In 1934, counts of 12 to 15
squirrels per acre were secured on the intensive study area of pasture 1,
and on other favorable sites over the Range. Sick and dying squirrels
were found on several occasions during the spring of 1938 and 1939.
There has been a rather steady seasonal decline until it is estimated that
there were not more than half as many squirrels on the Range in 1940
as in 1934. Such fluctuation plays an important part in the pressure of
squirrels on a range.

In changes of populations on extensive areas, migration is probably
a minor factor — far less important than reproduction or death. The
breeding season is mainly within January and February, and only one
litter is produced annually. The litters average about seven young each
on the Range, and this litter size has been shown to be the average else-
where over the geographic range of the species. Since females constitute
about half the population in the breeding season, the squirrel population
is increased within a few weeks to more than four times the January
figure. Young usually first appear aboveground in late March or April,
when they are several weeks old. From April, 1939, to February, 1940,
the ratios of young to adults caught on an 80-acre area indicate the
abundance and ratio of young to adults to be constantly changing. In
late June and July the young outnumbered adults more than four to
one. During late summer, fall, and winter, however, the ratio of young
gradually decreased until, in February, they comprised only two fifths
of the population.

Other parts of the Range were censused by comparing number of
squirrels seen per unit time with similar counts made on the trapping
area where the population was known. The total population figure de-
rived for the 1,400 acres of the six grazing-intensity pastures was 3,465,

Bul. 663] The San Joaquin Experimental Range 109

or 2.49 squirrels to the acre. (See fig. 1.) These were distributed as
follows :

Total Squirrels

squirrels per acre

Pasture 1 (154 acres) 561 3.6

« Pasture 2 (230 acres) 871 3.8

Pasture 3 (155 acres) 629 4.1

i Pasture 4 (309 acres) 554 1.8

Pasture 5 (250 acres) 341 1.4

* Pasture 6 (316 acres) 509 1.6

Natural area (150 acres) 120 0.8

It will be noted that pastures 1, 2, and 3 are similar in density of
squirrel population, and that pastures 4, 5, and 6 are similar though
they do not support quite half so many squirrels as do pastures 1, 2, and
3. It cannot be stated definitely just why the condition exists, though
several factors might contribute to it. Pastures 4 and 5 have more south
exposure, with probably higher temperatures, and fewer oaks and swales,
both of which may be essential to maintaining squirrels during the sum-
mer. Pasture 6 is lightly grazed, but has a slightly higher population
than 5 which is moderately utilized.

There is no indication here that different intensities of grazing on this
range are responsible for limiting or permitting an increase in squirrel
population. While these grazing-intensity pastures are stocked at dif-
ferent rates, the degree of forage production, and utilization by livestock
is not strictly in accordance with pasture size, as shown in the section
on forage crop management. Pasture 2 was lightly grazed, and supported
3.8 squirrels to the acre. Pasture 6 also lightly grazed had only 1.6
squirrels per acre. Pastures 1 and 5 are similar in that both were moder-
ately grazed and they supported 3.6 and 1.4 squirrels to the acre respec-
tively. Pasture 4 is moderate to moderate-to-light in utilization, and has
1.8 squirrels to the acre. Pasture 3, the closest utilized of all, has only 4.1
squirrels to the acre, and this is not noticeably greater than the 3.6
squirrels to the acre of pasture 2, which is lightly utilized. The natural
area (fig. 1) has been ungrazed since 1935. Old vegetation has piled up,
and the population density of squirrels was only 0.8 to the acre. This
area, however, has been badly disturbed by headquarters activities and
is not truly representative of a totally unutilized area. The only deduc-
tion that can be made from these data is that complete nonuse of the
area might set up a condition that would support only light populations
of ground squirrels. No direct correlation is indicated by these data,
between intensity of utilization and density of squirrel population. Fac-
tors other than grazing appear to be involved, and further study is being
made of these influences.

110 University of California — Experiment Station

Kangaroo Eat. — Though seldom seen because of its nocturnal habits,
the kangaroo rat is so distinctive in appearance that there is little chance
of mistaking it for any other rodent (fig. 17, A). The neck is short, and
the animal appears to be nearly all head, large hind legs (kangaroolike),
and tail. The long, tufted, white-tipped tail ; the broad, flat head with
masklike facial markings; the long vibrissae (whiskers) ; the large, pro-
truding eyes, powerful hind legs and small forelegs ; and the external,
fur-lined pouches, all serve to identify it. The coloration, as seen from
above, is pale brown, while the underparts are pure white. Adults weigh
between 2 and 3 ounces. In its ordinary movements the animal pro-
gresses in short hops using only the large hind legs. At any alarm the
gait is changed to a series of long bounds. The forelegs, though small,
are thick and muscular with long claws, and are well adapted for dig-
ging.

In choice of foods, kangaroo rats are more specialized than ground
squirrels, pocket gophers, or cottontails. Kangaroo rats as a genus are
noted for the habit of storing large quantities of food material in under-
ground nests or caches, although in the species under discussion the
storing habit seems little developed. No special underground chambers
for storage of food were found in any of the burrows (more than 100)
excavated. The ample supply of food on the Range throughout the year
renders extensive storage unnecessary. During the late spring, large
surface caches of soft chess seeds are occasionally dumped in piles at
blind ends of surface runways, and the accumulation may amount to
more than a quart. Much of this harvesting is done before the grass has
fully dried out, and caching aboveground may serve to cure it. During
the winter, when green vegetation is plentiful, the rats feed mainly upon
it. The stomach content of 17 rats trapped during the winter averaged
1.2 grams of green material. The two heaviest stomachs each weighed
2V2 grams, and seemed filled almost to capacity.

Among thousands of food samples collected in winter from burrows
and cheek pouches, filaree leaves were by far the most frequent. Broad-
leaf filaree was taken more often than red-stem filaree. Green foliage
material of many other annual plant species is utilized during the winter
months, and availability seems to be an important factor in determining
the kinds and relative amounts taken. Grass, including common foxtail,
slender oat, and several kinds of brome, together with leaves of red maids
and popcorn flower were found frequently in the small underground
caches. In several instances accumulations of empty shells of pine nuts
and acorns were found in winter burrows. These had been gathered in
the summer or fall and no doubt utilized soon afterward. In the spring
at the height of the growing season, filaree seeds are a favorite food.

Bul. 663] The San Joaquin Experimental Range 111

All data for the five or six months' period of the dry season indicate
that the food consists almost entirely of seeds. Soft chess and filaree
together make up most of the food taken. Common foxtail and fescues
are also staple foods. Popcorn flower, species of red brome, lotus, ground
lupine, and pine bluegrass make up minor parts of the diet. Probably the
greatest kangaroo rat damage to forage is done during late spring, at
the height of the growing season. At that time the animals are beginning
to feed on the ripening grass heads, many of which are out of reach to
them from the ground. Grass stalks are cut through near the base and
mowed down by the rats. This cutting is done both to obtain food, and to
clear away excess thick vegetation which would otherwise impede move-
ments of the animals traveling along their runways.

Food in summer, as indicated by several hundred samples from cheek-
pouch examinations and from fragments mixed with the dirt of burrow
mounds, consists of bromes, common foxtail, soft chess, fescues, and fila-
ree. These made up all but a small portion of the cheek-pouch contents.
Ripgut grass, Australian chess, clovers, and species of lotus were found
in pouches occasionally.

Kangaroo rats are much less uniformly distributed on the Range than
ground squirrels, and are most abundant in well-drained, open situa-
tions where vegetation is sparse. In general, they avoid the larger swales,
and tend to be concentrated on ridges and knolls. The kangaroo rat popu-
lation on the Range has been more subject to fluctuations than the
squirrel population.

During the three months' period of October, November, and December
of 1939, in trapping repeatedly over an 80-acre area in pasture 1, the
number of kangaroo rats caught were 397, or nearly 5 to the acre. Since
this trapped area was somewhat more level and open with greater pro-
portion of swale area than the Range as a whole, population density over
the entire Range on the average considerably exceeded that of the trap-
ping area. Thousands of recaptures of marked kangaroo rats indicate
that these animals are in general rather localized in their movements.
Many individuals were trapped time after time over periods of months,
within a 50-foot radius. A few, however, have been caught from one-
fourth mile to about a mile from the place where they were originally
captured.

Litters of two to five young are produced by each female. Some females
are in breeding condition at all times of the year, except December and
January. So far as known only one litter is produced annually, but such
a long breeding season indicates the possibility of more. In one instance
a young of the year was known to have produced young the same season.
In 1938, young of the year first appeared aboveground in May. In trap-

112 University of California — Experiment Station

ping operations carried on during June, they amounted to 53 per cent
of the total catch. In July they increased to 67 per cent, then decreased
and remained between 50 and 55 per cent through August, September,
and October. The greatest numbers of young first emerge in July. In
October there were still some young so small that they must have left the
nest only a short time before; many others had grown nearly to adult size.
Thus, in its annual cycle the kangaroo rat population reaches a peak in
July, then gradually decreases to a low point (disregarding small young
in the nest) in March or April.

Striking fluctuations in numbers of kangaroo rats from year to year
on the Range have been apparent though no accurate measure of the
change in numbers is as yet available for the entire area. In December,
1935, the rodent enclosures were treated with poisoned grain to remove
all rodents after fences had been installed. On these areas, totaling 1.6
acres, an average of 30 dead rats were found to the acre. This number
represents the minimum which could have been present, since from 25
per cent to 50 per cent of the rodents poisoned on an area die under-
ground and are not found. Probably the kangaroo rat population in 1935
was at least 30 per acre on south-facing slopes and ridges of favorable
habitat. In 1935 and 1936, kangaroo rats were remarkably abundant,
and observers driving over the Range at night could see them in large
numbers on the roads. Counts were made on measured distances of road
while driving at a speed of 15 miles per hour, and as many as 75 rats to
a mile were recorded. In the spring of 1937 the kangaroo rat population
abruptly declined. In March, 1937, not more than 2 or 3 per mile were
usually seen over stretches of road where as many as 75 per mile had
been seen the preceding fall under similar conditions. By the spring of
1938, the population density was probably even lower. By the spring
of 1940, the population had again made substantial increase and a maxi-
mum of 23 per mile were seen in night driving on the Range. In pasture
1, in the fall of 1939, exhaustive live-trapping of the rats indicated a
population density of 5 to the acre, and they were considerably more
abundant on some other parts of the Range.

The causes of such fluctuations in numbers of this rat are as yet ob-
scure. The sudden drop indicated disease to be a possibility and this is
being investigated. The total effect of kangaroo rats on the range is
greatly influenced by these large changes in numbers.

The burrows of the kangaroo rat are short and simple in arrangement
as compared with those of pocket gophers and ground squirrels, and the
rats are poor diggers, as compared with these mammals, or with other
species of this genus. For 118 of their burrow systems excavated, the
average maximum depth was 8.7 inches, and the average length of tunnel

Bul. 663] The San Joaquin Experimental Range 113

was 10.7 feet. There were, on the average, only 1.4 open holes to each
burrow system. Of the burrows excavated, only 1 in 24 was occupied by
a rat. This applies to a population about one sixth that on the area in
1934r-35. Occupied burrow systems contained straw nests, and usually
small amounts of cached food, while other burrows contained neither
nests nor foods. Most of the burrows which had no occupants, neverthe-
less, appeared to be regularly used. Most of the burrows were no doubt
occupied when the rats were at a peak in numbers in 1935-36.

The effect of kangaroo rats upon the soil is much less than that of the
ground squirrel and pocket gopher. There is but a small amount of
digging, and only a little vegetation is admixed with the soil. The culti-
vating effect is low, and such benefits from this kangaroo rat's activity
are at a minimum.

Pocket Gopher. — The species of gopher which occurs on the Range is
smaller than most representatives of the genus, and is peculiar to the
digger pine belt of the foothills. The most conspicuous external features
of the pocket gopher, distinguishing it from other small mammals, are
the slitlike fur-lined cheek pouches, the very large front teeth, protrud-
ing even when the mouth is closed, the short, powerful front limbs with
elongate claws, the small beady eyes set far apart on top of the head,
the rather short, sparsely-haired tail, and the coloration — reddish brown
above and grayish below.

Pocket gophers spend most of their lives underground, living in elab-
orate tunnels or runways, gathering their food at the ground surface,
carrying it to their nests or storehouses, and eating it there at their will.
The variety of plant materials found cached together underground in-
dicate that during the early stages of growth the gophers exercise little
choice in gathering food, but take whatever is conveniently available.
Broad-leaf filaree is probably the most used food, and the plants are often
taken entire, with leaves, stems, and roots intact, until they reach a size
which makes them difficult to handle whole. Popcorn flower is much used.
Cuttings from other kinds of plants found in the burrow systems in-
cluded tomcat and white-tip clovers, common foxtail, fescue, soft chess,
ripgut grass, red brome, poverty fescue, baby blue-eyes, buckthorn weed,
and tarweeds. Such green material as this is dumped in loose piles at
points along the tunnels or in the nest cavity and is used soon afterward.
In winter and spring, green leaf and stem material make up most of the
food, and gophers do most of their foraging on the surface of the ground.
A large portion of the gophers' food consists of stored material which is
gathered at the season when it is available and cached underground for
use when food is less plentiful. For this purpose, special flask-shaped
chambers usually much smaller than the nest chamber are constructed

114 University of California — Experiment Station

and connected by short tunnels with the main runways in either the
shallow or the deep portion of the burrow system. Acorns of either blue
oak or interior live oak, according to availability, constitute the most
important source of food for storage. In the winter, after acorns have
disappeared from the surface of the ground, quantities may be found in
gopher burrows, and often the ground under oak trees is honeycombed
with holes, indicating unusual activity at such places. Bulbs of blue dicks
and perhaps those of other liliaceous plants also are stored in quantity
and may be of as much importance as acorns in providing a food source.
Occasionally, digger pine seeds are found cached in burrows.

During the dry season roots of tarweed are eaten, probably, because
they provide some degree of succulence when most other vegetation has
dried out.

Only a limited amount of work has been given to population studies of
gophers on this range. In 1938 and 1939, the population in pasture 1
censused late in the breeding season by counting actively used burrow
systems amounted to from 5 to 8 to the acre. In May, 1940, after most
young of the year had left the nests, a 4-acre area in the west end of
pasture 1 was trapped out and it yielded 25 gophers to the acre. This
area was fairly typical of the Eange as a whole in abundance of gopher
mounds.

Gophers construct complex burrow systems. Each system consisted of
a maze of tunnels which may have a total length of several hundred feet.
Tunneling is at a fairly uniform depth of 6 or 8 inches below the soil
surface and also includes a deep portion, from 1 to 5 feet below the sur-
face, ordinarily not more than a few yards long, and which leads to a
globular nest chamber filled with dry, shredded grass leaves and stems.

On the Range the pocket gopher does much more burrowing than all
other rodent species together, and is, therefore, more important as a
cultivator of the soil. During that portion of the year when the soil is
moist enough to be workable, gophers are continually extending their
burrow systems, and usually throw up one or more new mounds daily.
Seasonal movement is directed by varying food supply and soil moisture.
During wet weather of the winter and spring months, swales are unin-
habitable and gophers then work on the higher ground. In the late spring
and early summer, gophers move downhill into the green swales and do
most of their feeding there when the vegetation has dried out elsewhere.
By the time the swales have dried out they may be honeycombed with
burrows, and much of the ground surface may have been covered with
loose earth of the mounds. Much vegetation is covered over, and no new
growth appears on the mounds until the following year.

During the course of the year large quantities of soil are moved. Dur-

Bul. 663] The San Joaquin Experimental Range 115

ing the wet season excavated soil is thrown out upon the surface of the
ground where it covers vegetation, and is admixed with it. During the
dry season when temperatures of the upper soils are high, the gophers
dig deeper, the excavated earth from these lower levels being placed
in the shallow surface runways instead of being thrown out above-
ground. Large quantities of chopped and shredded vegetation is mixed
through all this material. Cattle break into the shallow surface runways,
and the gophers repair the breaks. The net result is constant cultivation
with the admixing of organic matter with the granitic soil. While final
conclusions cannot yet be drawn, it should be stated that such culti-
vating and fertilizing effect on this range may result in increased forage
production to offset the forage eaten and destroyed by the gophers.

Cottontail. — The cottontail is a typical rabbit and is so familiar that
it needs no special description.

Abundant shelter, such as brush, rock piles, or thickets of tall vegeta-
tion interspersed with grassy places, furnish favorable cottontail habitat.
The Eange permits these conditions, though in varying degree in dif-
ferent parts of the area. On the natural area, where vegetation remains
unused to afford thick cover, and a continuous water supply is available,
cottontails are more abundant than on other parts of the Range. It was
not uncommon in the summer of 1939 to see from 10 to "40 on the lawn
about headquarters at dusk.

Cottontails were relatively scarce on the Range in 1934, but no data
are available concerning their abundance previous to that time. There
was evidence of a gradual progressive increase until 1938, and during
1939 and early 1940 their numbers remained fairly stable except for
seasonal fluctuations. During the summer of 1940 the numbers dropped
to less than half of the 1939 summer populations, most of this reduction
occurring over a 6-weeks' period. Extensive live trapping on 80 acres
during the summer of 1939 yielded about one cottontail to the acre.
Probably most parts of the Range are better suited for cottontails than
this trapped area and the figure is conservative.

Breeding is limited to the late winter and early spring, the young
being well grown before the forage crop has dried out. The seasonal peak
of cottontail abundance occurs at a time when forage is maturing and
most heavily used by livestock.

Cottontails use many kinds of vegetation for food and are much less
discriminate in their feeding than are most of the rodents. Being vo-
racious eaters they compete directly with stock for forage, although the
data accumulated thus far are too fragmentary to afford the basis for
a complete account of their annual feeding cycle. During winter and
early spring these rabbits feed on green vegetation, and probably take

116 University of California — Experiment Station

whatever is most readily available, so that filaree and the commoner
grasses make up a major portion of the diet. After the beginning of the
dry season, swales where succulent vegetation persists longer are favor-
ite feeding places. Clover is much used and later such plants as dock,
rush, and tarweed, and even turkey-mullein, which remain green into the
summer, are selected, apparently as a source of succulence. In feeding
on large plants of this type, the foraging rabbit cuts the stem near the
base, and then eats outward toward the tip. Often a large portion of the
cut top is discarded. Heads of soft chess and tarweed plants comprise an
important food source throughout the dry season. Cast seeds of fescues
are often grazed upon by the rabbits during the summer, when the seeds
form a dense mat over the ground in places where these grasses are
abundant.

FACTORS INFLUENCING RODENT AND RABBIT ABUNDANCE

To evaluate changes in rodent pressure on the range resulting from
the marked changes in populations which are apparent, investigations
have been initiated into the causes and frequency of the fluctuations.
Important factors which influence rodent numbers include : changes in
climate, extremes of weather, the scarcity or abundance of food, water,
shelter, predators, parasites, and disease. Some of these factors have
been studied in detail, others more limitedly. Fluctuations of course
result from the interaction of all these factors, and caution must be
exercised in drawing any conclusions from the limited data so far avail-
able. It is hoped that with the completion of studies now in progress
the causes and importance of fluctuations in populations of range ro-
dents will be better revealed.

Rodents serve as food for a number of carnivorous mammals. The
coyote, gray fox, bobcat, and badger are fairly common, but no con-
clusive data are as yet available regarding their abundance or the extent
to which they prey upon rodents. The striped skunk is too scarce to be
a serious factor. Red-tailed hawks and horned owls are the most im-
portant raptors, while the golden eagle, rough-legged hawk, prairie
falcon, and marsh hawk are unimportant, because of their scarcity or
their preference for other prey. The rattlesnake is the most important
reptile, while gopher snakes and king snakes play a lesser part.

Systematic collecting and analysis of "scats" (droppings) of carni-
vores and the pellets of raptors affords data on the food of these preda-
tors. Positive identification of some scats presents a major problem and
where doubt existed the data were discarded from the tabulation by
species.

In analyzing these materials, hair, bones, feathers, scales of reptiles

Bul. 663]

The San Joaquin Experimental Eange

117

and hard parts of insects remain, and may be identified. Such analysis
is in terms of occurrence of the item in each scat or pellet and not on
the basis of volume or weight.

Predation by Coyotes. — Of carnivorous mammals on the Range the |
coyote is probably the most important in its effect on rodents. Coyotes^
are subject to control on the area. About thirty were trapped in each
of the winters of 1937-38 and 1938-39 ; in the winter of 1939-40, thir-
teen were caught. The annual catch undoubtedly has amounted to a
large portion of the total population on the area, and it is the general

TABLE 17
Food Items Identified in Coyote Scats

1939

1940 (to May 11)

Number

Per cent

Number

Per cent

102
28
83
1,185
25

7.2
1.9
5.8
83.3
1.8

34

9

14

336

3

8 6

Birds

2.3

3.5

84.8

0 8

1,423

794

396
219

impression of persons familiar with the Range before control was ap-
plied that coyotes are now held far below their former numbers. Prob-
ably the population originally amounted to at least one to a hundred
acres before control measures were taken. Control is carried out in late
fall and winter and the coyotes are present in lowest numbers in early
spring. From table 17 it is evident that rodents and rabbits make up most
of the food. Ground squirrels were represented in the scats more fre-^»
quently than were any other prey species. The trend of feeding in 1939
and in 1940 is remarkably similar, though the 1940 data are incomplete.^
It is of interest that in this particular habitat the food is made up of
relatively small animals, and rodents and rabbits together constitute
more than four fifths of the food. The effect of coyote predation on rodent
and rabbit populations of the Range is indeterminate with the data so
far at hand, but it is a part of the combined impact of several common
predator species directed simultaneously upon the same prey species.

It is of interest to note that the amount of effort expended on the
Range in the control of coyotes has not eliminated them, nor has it been
accompanied by an increase in squirrels and kangaroo rats. Rather,
these rodents decreased during this period, as stated in an earlier para-

/

118

University of California — Experiment Station

graph, indicating that other factors than coyote predation play a major
part.

Predation by Bed-tailed Hawks. — Red-tailed hawks are year-round
residents on the Range. During the winter of 1939-40, part of the Range,
comprising an area of about 3,500 acres, was found to be occupied by
26 hawks, each restricted to a definite territory. In the spring of 1939,
f 71 per cent of prey items brought to young in the nests by adult hawks,
[were ground squirrels ; the remainder consisted of rabbits, pocket go-
phers, wood rats, kangaroo rats, mice, and several species of birds,

TABLE 18
Food Items Identified in Ked-tailed Hawk Pellets

1939

1940

Number

Per cent

Number

Per cent

84
29
149
629

9.4
3.3
16.7
70.6

31

11

107

312

6.7

Birds

2.4

23.2

Rodents and rabbits

67.7

891
525

...

461
266

snakes, and lizards. Squirrels were the preferred food. When the prey
at a nest included both a squirrel and another animal, the latter was
usually left uneaten and the young hawks fed from the squirrel exclu-
sively. A study of feeding at times of year other than the nesting season,
based on analysis of pellets picked up under territorial perches, revealed
that squirrels continue to be the principal food throughout the year.
From table 18 it will be seen that rodents and rabbits make up most of
the food, but that insects, birds, and reptiles also are frequently eaten.
In preying on rodent-eating snakes the hawks tend to partly offset the
effect of their own predation on rodents. Analysis of 525 pellets shows
that during the entire year of 1939, 70 per cent of the food of these
hawks consisted of rodents and rabbits.

Predation by Horned Owls. — Horned owls occur at about 1 to each 80
acres on the Range, and their habits of feeding mainly at night give them
some importance as predators on nocturnal rodents. Table 19 shows
the occurrence of food items, from analysis of pellets for the years 1938,
1939, and 1940.

Rodents and rabbits constitute the major items of diet, and insects
were second in importance. Of these rodents, wood rats comprised about
31 per cent; kangaroo rats, 23 per cent; pocket gophers, 15 per cent;

Bul. 663] The San Joaquin Experimental Range 119

ground squirrels, less than 5 per cent; and rabbits, 23 per cent. The
wood rats and kangaroo rats are active at night, while rabbits and pocket
gophers feed both day and night, and squirrels only during the daytime.
A study of horned owl nests {2) during the spring of 1939 shows that
of the food brought to the nest to feed the young, all the 67 items were
rodents and rabbits. Of these 23 (34 per cent) were wood rats (nearly
^11 of them adults) ; 17 (25 per cent), kangaroo rats ; 13 (19 per cent),
cottontails; 11 (16 per cent), ground squirrels; and 3 (5 per cent),
pocket gophers.

TABLE 19
Food Items Identified in Horned Owl Pellets

1938

1939

1940

Number

Per cent

Number

Per cent

Number

Per cent

Insects

1

0

0

22

4.3

0.0

0.0

95.7

141
22
32

606

801
520

17.6
2.7
4.0

75.6

102
11
5

258

27 1

Birds

2 9

Reptiles

1 4

Rodents and rabbits

68.6

Total food items in pellets

Pellets examined

23
14

376
250

Of all the animals on the area near the nests upon which owls could
feed, pocket gophers and kangaroo rats were most abundant. Ground
squirrels were less common; cottontails and wood rats were far less
numerous. The relative abundance of wood rats in the nests and pellets
indicated a definite selection by the owls. No birds of any kind were
brought to the nests, and the pellets show only 2.7 per cent of birds, on
the average. These data indicate that the horned owls are beneficial and
that their pressure as predaceous birds is upon wood rats, cottontails,
and kangaroo rats.

Predation by Rattlesnakes. — Over the three-year period from 1935
to 1937, all rattlesnakes found on the Range were killed in an attempt
to exterminate them on the area. About an equal number were killed
also in the 1938 and 1939 seasons and data were actually recorded for
a total of 899 killed for the five-year period; perhaps as many more
were killed for which no records were made. It became evident that
despite this constant drain on the population, rattlesnakes were not
becoming noticeably scarcer.

Past information together with preliminary work at the Range shows
rodents to be an important article of diet of these snakes. Likewise,
rattlesnakes are, numerically, the most abundant predator of rodents

120 University of California — Experiment Station

on the Range. Because of these facts, and the necessity of determining
the function of predators in influencing the population of various ro-
dents, it was essential that more complete understanding of the rattle-
snake be gained. At the start of the 1938 season, these investigations were
inaugurated on that part of the Range (pasture 1) set aside for wildlife
research purposes.

On this intensive-study area all captured rattlesnakes were marked,
released, and records made of recaptures. All captured snakes were
caused to regurgitate their food ; this added further to the data of prey
taken. Except on this study area, snakes were killed when and where
found, and their stomach contents examined. Such study seemed highly
desirable because despite the abundance and widespread range of the
Pacific rattlesnake, relatively little is known regarding it as a rodent
predator. As the most numerous rodent predator it could have a pro-
nounced effect on rodent populations, as well as its effect on livestock
and other domestic animals. As a result of its bite, cattle occasionally die
(see the section "Experimental Herd Management"), or at best suffer
loss of weight, entailing a financial loss to the stockman. In the region
of the Range there are no human fatalities from snake bite on record,
and the only cases of persons being bitten have resulted from handling
of the snakes. Nevertheless, the potential hazard involved in the presence
of the snakes, and the resultant worry and inconvenience to humans
must be taken into account.

For an evaluation of the effect of rattlesnakes on rodent populations,
it is necessary to have a fairly accurate idea of the number present,
their seasonal habits, frequency of feeding, and the species preyed upon
with the relative numbers of each. If, at any time or place, it should
be desirable to take control measures against a rattlesnake population,
or exterminate them locally, knowledge of the population density, extent
of normal movement and range of individuals, rate of population re-
placement by reinvasion and reproduction would all be essential to
plan a control program more intelligently.

In the 1938 and 1939 seasons, and to June 30, 1940, a total of 515
rattlesnakes were caught alive on the Range, marked, and released after
measurements, weight, and other data had been recorded for each. Sub-
sequently, many of the marked snakes were again found and were caught
or killed.

In the spring of 1939 the population on pasture 1 was computed to be
1 snake to the acre and in the spring of 1910 a figure of 1.3 was obtained.
This computation was made on an area not especially favorable as habi-
tat to the snakes and where large numbers had been killed in the pre-
ceding years. Therefore, it is probable that the population density for

Bul. 663], The San Joaquin Experimental Range 121

the Range as a whole was somewhat higher, and for the entire area of
4,700 acres a population of 5,000 rattlesnakes seems conservative.

Rattlesnakes are sluggish and inactive in their habits, but wander
indefinitely with no apparent tendency ever to return to a given spot
after once having left it. Time and extent of movement is exceedingly
variable, but on the average the snakes move less than 10 feet a day, and
may move only a few yards over periods of months.

Young rattlesnakes are born in the fall, and averaged 7.6 per litter in
a small sampling. The population is probably more than doubled each
fall by birth of the annual crop of young. The 300 or so killed annually
on the Range and its vicinity by human intervention, make up only a
small part of the annual increase. Hence, this must be only a minor cause
of mortality which may have no significant effect on the population from
year to year.

In a study of rattlesnake feeding during the 1938 and 1939 seasons
it was found that nearly all the local species of rodents, cottontails, two
kinds of birds, several kinds of lizards, and the spadefoot toad were all
included in the diet ; but the number of ground squirrels exceeded that
of any other prey species, making up about one third of the total kinds
of prey. The effectiveness of the rattlesnake as a ground squirrel predator
is increased through its habit of occupying the burrow systems of the
squirrels and striking any which come within reach. Frequently, dead
squirrels showing symptoms of rattlesnake bite were found in the field.
It is not known how often these snakes feed.

Presence of Diseases and Parasites. — The violent fluctuations in num-
bers of kangaroo rats, and the slower but certain change in numbers of
ground squirrels, cottontails, and quail, together with the finding of
sick animals, strongly indicated that disease and parasites are present
in some of these wildlife populations. Reduction of kangaroo rats from
not less than 30 per acre to only a small fraction of that number, over a
period of less than two months, can more reasonably be ascribed to dis-
ease than to predation or habitat change. Some of the sick and dying
ground squirrels found during the 1938 and 1939 rainy season were
heavily parasitized and others displayed a pneumonic condition of the
lungs. Examination of a large number of ground squirrels on the Range
by the California Department of Public Health failed to show sylvatic
plague to be present.

Sick rabbits have been found, but those examined did not show tulare-
mia (rabbit fever), one of the diseases most commonly associated with
rapid reductions of these animals.

Birds, as well as mammals are subjected to the attacks of parasites and
the ravages of disease. Quail are known to have a blood-inhabiting pro-

122 University of California — Experiment Station .

tozoan, but it is not known how these organisms affect numbers of quail.
Glading (see the section "Studies on Valley Quail") found quail on the
Range to be infested with this parasite. Possibly the disease reaches
epidemic proportions under certain conditions, causing a reduction in
the quail population. Young red-tailed hawks are attacked by parasitic
flies (Eusimulium clarum) which in the 1940 season were an important
cause of nest mortality.

The whole problem of disease and parasites as factors contributing to
the regulation of numbers of wild creatures is a vital one. A number of
prominent workers in this field have stated that disease and parasites are
probably the most important factors in regulating the numbers of rapid-
breeding animals when violent fluctuations do occur. Preliminary work
on this phase has been done, and is being continued in an intensified
study.

EFFECT OF RODENTS ON VEGETATION IN THE ENCLOSURES

The rodent enclosure installed on the Range in 1935 and 1936 was
designed to give further information on the effect of ground squirrels,
pocket gophers, and kangaroo rats, on the composition and density of the
forage, and on the amount eaten and destroyed during a season by a
definite number of each of these rodents. The initial installation con-
sisted of a large rectangular pen divided into four compartments. One
of these, comprising a half acre, was stocked with six ground squirrels ;
another, of one-quarter acre, was stocked with eight pocket gophers;
one-quarter acre was kept as a control area with nothing using the
vegetation, and one-quarter acre was stocked with eight kangaroo rats.
This stocking was about equal to the maximum population densities
known to have existed on the Range at the start of this experiment. A
strip around the outside of the enclosures was fenced against livestock,
but was accessible to all rodents and rabbits occurring in the vicinity.
In addition to these enclosures, a series of fenced plots in the various
pastures create areas completely protected from all rodents and live-
stock, and others exclude livestock but permit use by all rodents. (See
the section "The Forage Crop and Its Management.") An original forage
inventory and clip, prior to stocking the enclosures, affords the starting
point on measures of production, composition, and density of forage.

It was inevitable that soil depth and fertility, areas of rock, slope, and
tree shading varied somewhat in each one of the enclosures, and each
one of these factors must have influenced the kind and amount of vegeta-
tion produced. The original location of the enclosures, however, was
carefully selected on an area where the vegetation appeared as uniform
in height, density, and general composition as it was possible to deter-

Bul. 663] The San Joaquin Experimental Range 123

mine, and where squirrels, kangaroo rats, and gophers already were
present in normal abundance. After the first year of rodent use, dif-
ferences in total quantities of some particular kinds of plants were
easily seen. While these differences may have been brought about by a
complex of factors, some of which are mentioned above, it seems obvious
that the main difference in treatment of these areas is that of rodent use.

Each spring at the height of the growing season an "inventory" of the
vegetation in the enclosures was taken by estimating the density and
composition of the plants on 100 one-foot-square sample quadrats se-
lected by random sampling in each enclosure. Late in the fall, just prior
to the start of rain, the quadrats on these plots were clipped, and the
vegetation from them was air-dried and weighed to determine the forage
residue after the annual crop had been subjected to a year of rodent use.

In the spring of 1939, a clipping and weighing of vegetation on 100
one-foot square plots was made in each enclosure, at the height of the
growing season, to aid in interpreting the effect of rodents on forage
yield.

These studies have pointed out needed changes in past methods of
study of rodent pressure, and of the need of more intensive methods for
these studies on annual- type ranges. It would be premature to draw hard
and fast conclusions from present data, and further study now under-
way and contemplated during the next few years will, without doubt,
alter at least some of the conclusions that could be reached now. With
this caution it may be permissible to point out a few trends of this study.

In order to measure the effect of rodents upon the vegetation with
precision, it would be necessary to measure the effects of all the other
factors that affect forage production. It has, of course, been impossible
to do all this, up to this time.

These rodent plots were placed in operation during the winter of
1935-36.

Wide fluctuations in the composition and density of plants occur in
all of these plots from year to year, just as they do on the range as a
whole. In some cases, still greater changes in certain forage species occur
in the enclosure occupied by different rodents.

Tables 20, 21, and 22 set forth the percentage of forage plants in each
enclosure that are the principal food of each rodent species in that en-
closure. In the squirrel enclosure (table 20) the amount of broad-leaf
filaree was consistently less for the years 1936 through 1939 than it was
in either the control, the gopher, or the kangaroo rat enclosures. Filaree,
in 1935, was more abundant in the squirrel plot than in the gopher plot,
but less abundant by 4.94 per cent than in the control. During the period,
1936 through 1939, filaree averaged 13.83 per cent less than on the con-

pH

P

u

u

o

k1

w

a

s

^

Of

02

w

K

H

«

g

g

m

DQ

J

O

w

h-l

3

u

g

te

P

ft

o-

J

02

o

P

B

P

o

o

o

rh

w

pa

o

&

©

G

fe

o

o

W

r/j

H

U
In

Q
O

o

ft

0

p

a

<1

o

o

1-1
K

1

ft

ta

o

g

H

£

ft

W

O

W

o

<}

ft

fc

O

w

o

<

H

fe

"S&

5 o

9 c«

CD o

n-3

coco->»<»o<MOt^cN»cNi<M

CNOOOiC^OiCOi-HO©
©uiO^OO^HOOC-1

+7 i +T i + i + +

^00i-ccocot^f.-<cMco
0»o©<ni«ococo©oco

oi^©<-icnicnico©oo

i 7 i+7 i + i + +

O (M O CM

O OS o o o

I + + + I

O CO O

o o o

I I I

O CM Oi o o o

+ + + +

MfOOfOOOOOOOi
tiOii— irf.— cO»0©005

Sfr-icqoOfOOO

11+11 I +

oo t- co ■>* co in

CO O © O

CO CO t~ CM O O CO

O W O * N

CO O O CO

OOOO-HCNlOlOCO©
O « O t>. O -< -h o o
O*0Ot^^HfC0OO

o o o o

O "* O CO o o

cm os o cm

© 00 ~4 O

o co o o o

o o o o o

©lOCOOOCMCOCMOiCOO

coo©co©iococooa>

O00OfOO»0OOCM

O "5 O CO o o

o cm o m

© "*

© co

o o o o

©C0O00OCMCN©©©

OfO'-iOCMOOOO
CM H f

U) oo oo o

© CO © © 05
© Oi © © <M

© "3 O O N

%

T3 O

Co

•91

2 « g 12 -s^

3 S £

at?*

.,-. I_ C5 O *« *^ O - ^ I.

Bul. 663] The gAN Joaquin Experimental Range 125

trol plot. Since filaree is one of the principal foods of ground squirrels,
their feeding effect may be responsible for this difference.

Ground lupine, a choice squirrel food, is without doubt held down by
squirrel pressure. Never abundant in the forage, it is avidly sought out
by the squirrels ; on the average it was only 3.0 per cent as abundant
in the squirrel enclosure as in the control. Despite this fact, ground
lupine does not show any progressive decrease in the squirrel enclosure ;
in fact it increased in 1938, but only slightly as compared with the con-
trol (0.04 per cent as against 12.6 per cent).

Soft chess also showed consistent change in the squirrel enclosure,
and it increased rather than decreased in abundance. This common grass
is eaten by the squirrels only during the early stages of growth in the
fall months. Later in the growing season squirrel use of filaree and other
rosette-type plants may favor growth of soft chess by thinning out these
competing rosette-type species.

Other favorite squirrel foods, including ripgut grass, birdseye gilia,
yellow tarweed, Wright's tarweed, catchfly (or campion), foxtail fescue,
popcorn flower, red maids, and godetia, showed no consistent increase
or decrease in percentages in the squirrel enclosure during the four-year
period, but they showed marked fluctuations from year to year.

In the pocket gopher enclosure (table 21) filaree was consistently
more abundant than on the control area despite the fact that during the
growing season gophers take considerable quantities of this plant for
food. Ripgut grass was consistently less abundant in the gopher en-
closure ; there was on the average only 20 per cent as much over the four
years as occurred on the control plot. Yellow tarweed, soft chess, red
brome, turkey-mullein, and all clovers varied from year to year, but
there was no great difference in their relative abundance on the gopher
plot and on the control.

In the kangaroo rat enclosure (table 22) broad-leaf filaree (even
though it is one of the principal foods) was consistently more abundant
than in the check enclosure. Filaree was more abundant in the rat en-
closure in 1935. Soft chess, the most used rat food, was 61 per cent, and
ripgut grass was only 79 per cent, as abundant in the rat enclosure as
in the control enclosure. Some of these differences probably are due to
differences in the areas themselves and not to kangaroo rat activity.

Data from the dry clippings of the rodent enclosures indicate that
the poundage of vegetation destroyed by a given population of rodents
on a given area in a given time will vary greatly, the amount depending
on the bulk of the forage crop, its composition, the feeding peculiarities
of the rodent, and doubtless many other factors.

At the start of these enclosure studies in 1935 an inventory of forage

p

£ S

S §

8 H

S £

e> g

&h £

« O

M O

g a

»H S £

„ o
Ph a

h
o

g

H

&

w
o

^ _ ^ h a>

13

© »-< i-~

><-

i-

«»

o-

oc

a

OS

M M 0O N C

OS

O0 O <M <-l O t-

CM ©

© ©

o*

O 0)

"~<

+ 1 1 + 4- 1

+ + 1

2 =

1 o

o £5

00

■* oo o n t>
© ,_; o i-i C

CM

CO 1C
CO t^

© 00

© T-i

to <"

"82

^ 43

+ 1 1 1 1

1

+ + 1 1

« C

2 o

N O 00 Ol oc

CO-

co ©

© b-

ft o

CO
OS

TJ< o O »H —

CM

CO CM

O tH

J3A

+ + + + 1

1

1 + +

i's

o S

© CM

© ©

£ <fl

CO

■* o oo o) a

C

Tf< t>

OS

•* O ^ N «

c

^H o

© ©

+ + 1 + 1

1

+ 1

1

c a

OS r~ 00 OS l>

co i-

»C

OS CO CO -5ft OC

t~- CO

• «5

«2

CO

as

00 S M CM C

• oo

to

s

1 1 + 1 I

1 1

+

os
co

O N N O) OC

<*

OS OS

OO OO

M H « N N
CM H

CO CM

CO

© ©

O0

O (O ^ OO N

ec

oo ©

co co

i-H CM CO CO CO

CM i-H

© CM

4>

OS

M M lO rt M
CM

CM

S 3

© CO

P

s

t^

o

CM O t~ rt CC

© ©

os

io o t~- o oc

M-

co o

© ©

O

-t-^

d

o

O

CO
CO

or

os o i-H co m

c

© t-

© ©

os

"<*< O CO i-H I>
CO CM

c

S M

© ©

«o

OS ■»*« CM OS CO

f W 0O Tt< CC

CO ^H
Tf CO

• CO

cn>

if} ■<& OO b- t-
CM CM

O ifl

OS

•«*< Tf oo co a-

CC

© lO

lO •<*<

OS

H 11 CO t «
CO

e

CO CO

© o

00
CO

K5 * N O C

<_

co co

© CO

2

CO H <* O «
CM

e

os ^
i-i CM

© T-*

3

o

a
S3

t^

t- t~ IC © "*

O O *C o oc

£

S ^H

00 t~

© t^

OS

O © OO CM CC

CM CM

CO

© »-<

ft

o

O

to

CM CO
© ©

CO © CM CM CC

c

■* ©

OS © 00 Tt< CO

CO T-i

C

CM CC
CM

© ©

U3

iO © lO O c

co ©

• CO

• oo

OS

CO tO (M lO r-

«-H r-l CM

© ©

• OS

J
a

C

• 1

|

c

0

6^

1

c

1

a

s

9

e

<1> ^ «*- cj C

T3 CD c8 O JS

1 of 1*5

j2
•Si

a &
hi

M o

pq O fr

c

PC

s

^

^

K
>-

9 ^

o 5

O (U

2-d
St

^^

53-2

+ + + + I l

CO O CO CO CO

»0 OS Tt< i-H t-

H O0 tO O N

+ + + + I

OO © CM i-i CM
CM

+ + + I +

OS CM CM CO ©
lO CO © CN ©

+ I I I I

CO © CM ©
+ + I +

OO CO CO t^ CO CM uo

O N O) CO •* O Ol

CO CM CM CM .-l © CO

CO UJ H M

00 i— I © Tf( lO • lO
CM t^ i-i eo ^h . ^

io n o oo n< i co

H CO ID O

OS CM OS CO
■**< OO t}1 oo

«3 CO N i-i

ii * to CO i* H «
N DiTlt HOON
S OS N CO O O O

tO lO CO «

CO 00 CO © CO

(ON*
00 00 ©

© 'OS ^H

CM OS lO CM

U O cu

3«s« 155 8

i '3 1 -^ s -s ^
2? &^.&i«s

Cq fx, PL, rt Pi 02 CQ

.5 c

si

S5.S

* c

c3
,J3

Bul. 663] The San Joaquin Experimental Range 127

and a clipping and weighing of vegetation was made at the height of the
growing season.

The forage produced by these plots in 1935 is as follows :

Pounds
per acre

Plot 1 (ground squirrels) 1,854

Plot 2 (pocket gophers) 1,196

Plot 3 (control) 1,145

Plot 4 (kangaroo rat) 1,128

Plot 5 (all rodents, no livestock) 1,202

Plot 6 (all rodents and livestock) 1,419

'35 '37 '38 \

Squirrel
area

'35 '37 '38 '39

Stock and Three-year averages
rodents area

Fig. 18. — Pounds per acre of forage remaining in the fall on comparable areas
subjected to use by different kinds of animals in 1937-1939. The data for 1935 are
forage yield at the height of the growing season before the plots were stocked with
the rodents.

These figures represent the forage produced with no usage, and clip-
pings made at the height of the growing season. Analysis of these data
show that only plot 1 (squirrel enclosure) was significantly different
from plot 3 (the control plot). In all subsequent years, clippings were
made at the close of the dry season, and figures represent residue after
the rodents had been active all year. Plot 6 is somewhat different.

Figure 18 sets forth, graphically, the forage yield at the start of the
experiment in 1935, and also the forage residue in each rodent enclosure
at the close of the forage season, just prior to fall rain and the start of
new growth.

128 University of California — Experiment Station

The seasonal fluctuation in forage production was discussed in detail
in the section "The Forage Crop and Its Management." The data from
the control enclosure of this table may be taken, therefore, as an ap-
proximation of the seasonal forage yield of these four plots. It portrays
the trend, if not the actual poundage of each plot. On this basis it is
evident that there is a wide annual difference between the forage pro-
duced and the forage residue on a plot that has been grazed by the same
number of the same rodent species each year.

In 1938, the difference between forage produced in the control plot
and residue left by eight kangaroo rats in the rat enclosure was more
than six times this difference in 1939. The forage yield of the control and
the kangaroo rat enclosures was about equal, showing that dissimilarity
of the plots would not be the most logical contributing factor. Presum-
ably the animals eat about the same amount from year to year, and the
observed variation in poundage of residue and annual yield must result
mainly from variation in the amount cut or trampled, but not actually
eaten. The amount destroyed beyond the actual food requirement is
probably somewhat proportional to the total amount available. Nat-
urally, when there is a thick cover of vegetation, the rodents, in the
course of their normal trampling and cutting, destroy a much larger
amount than they would if the vegetation were extremely sparse. The
poundage destroyed also might be expected to vary according to the
kinds of plants being used. At certain times, in their foraging, rodents
are destroying a much greater bulk of vegetation than they are actually
eating. For instance, squirrels take filaree seeds by plucking off the
entire fruits and discarding them after the fine seeds have been removed.
After being dropped by the squirrel the fruit is no longer available to
grazing stock, and its bulk is much greater than the minute seeds which
are actually eaten by the squirrel. At other times squirrels are eating
fine foliage material and all that is cut is actually consumed. Thus, the
proportion destroyed but uneaten varies according to the particular item
for which the rodent is foraging, and on the stage of development of the
plants. The kind of items used depend partly on availability and there-
fore variation in the composition would in part determine the amount
destroyed from year to year if the total bulk of the forage crop produced
should remain constant.

It cannot be expected that the amount destroyed under a given rodent
pressure will be exactly the same for any two years. For the three years
over which our records extend, and for the areas and plant compositions
on which they are based, our figures are probably representative. How-
ever, to obtain an accurate figure for the average amount of vegetation
destroyed under average conditions, it would be necessary to continue

Bul. 663] The San Joaquin Experimental Range 129

the records over a longer period of years, to be assured of statistically
sound results.

It will be noted that the unused vegetation in the squirrel enclosure
has been consistently lower than in the control plot, or than in any of
the other rodent plots. At the start of the study the squirrel enclosure
yielded one third more forage than the control, and more than any of
the other plots. The indications here are that squirrels eat and destroy
more forage than any of the other species of rodents studied.

In the gopher enclosure in the 1939 season, despite the activity of the
animals, the amount of dry vegetation remaining at the end of the year
was slightly greater than in the control enclosure. This suggested that
the cultivating effect of the gophers on the soil might be sufficiently bene-
ficial in stimulating plant growth to more than compensate for the actual
destruction of vegetation by them. This seems especially true since each
year's residue from the gopher enclosure is nearer to the residue of the
control plot. Since gophers do much of their feeding underground,
hardly any vegetation is trampled by them, and the underground por-
tions of the plants used by gophers do not show up in the forage residue,
and have little effect unless underground cutting kills the plants.

The discussion on unused forage, after a season of rodent use, serves
also to show the need of further studies upon which to base conclusions
of the amount of forage consumed or destroyed by each species of rodent.
Briefly, the present data indicate that the forage consumed and de-
stroyed by pocket gophers is largely compensated for by the increased
production of forage brought about by their cultivating activities on
the soils of this region. Caution should be used in assuming that a
similar effect would be found on all soils in all other areas. Ground
squirrels and kangaroo Tats are detrimental ; they without doubt took
15 per cent or more of the forage produced on these plots, under the
conditions of these experiments. Additional data for analysis are needed
before it can be said that a certain number of squirrels or kangaroo rats
take enough vegetation to support a steer.

STUDIES ON VALLEY QUAIL

T. I. Storer, F. P. Cronemiller, E. E. Horn,
and Ben Glading

The California valley quail (Lophortyx calif ornica vallicola) is the
principal upland game bird of the Sierra Nevada foothills, and provides
choice shooting for many sportsmen each hunting season. In early days
quail were extremely abundant, often occuring in huge flocks. While
still common, their numbers have been reduced by settlement, hunting,
and other factors. Because of their value as game, there is an increasing
interest in means for conserving and improving the crop of quail.

Since the major use of the foothill region is for grazing, the manage-
ment of land there involves an adjustment between the needs for game
and livestock. This relation will probably become of greater impor-
tance in the future, since proper management for quail may enable land
holders to derive some additional income from the game crop. In turn,
this makes a better understanding of the requirements for quail neces-
sary, so that means may be taken toward increasing their numbers.

Association of upland game birds and livestock on public and private
foothill ranges presents questions and problems of management. In an
attempt to obtain necessary fundamental information for answers and
solutions, a quail study on the San Joaquin Experimental Range was
undertaken.

The Range is representative of the foothill range lands of California
and is a particularly favorable site for studying the fundamental prob-
lems of quail biology, because (1) it is under strict control for research
purposes, (2) the plant cover has been carefully mapped and annual
inventories of the forage are available, (3) it is grazed by beef cattle
and an accurate record of grazing intensity and other factors is at hand,
and (4) studies are in progress on the biology and the relation of
rodents and other wildlife.

In the autumn of 1936 the Cooperative Quail Study Committee was
organized to plan and direct work on quail. The membership was drawn
from several public agencies having special interests in the quail prob-
lem.7

The Committee has met with the biologist to plan the study, give tech-

7 The Committee membership as of 1940 was as follows: E. I. Kotok, M. W.
Talbot, J. W. Nelson, and H. H. Biswell, California Forest and Eange Experiment
Station; F. P. Cronemiller and I. Sack, U. S. Forest Service, Eegion 5; E. E. Horn
and Henry Fitch, U. S. Fish and Wildlife Service; G. H. True, Jr., California Divi-
sion of Fish and Game; J. T. Emlen, Jr. and T. I. Storer (chairman), Division of
Zoology, University of California. Ben Glading was engaged as resident biologist
to conduct the field studies, beginning in December, 1936.

[130]

Bul. 663] The San Joaquin Experimental Range 131

nical advice, review the findings, and aid in preparing publications that
report the results of the study.

The studies to date have included a determination of the total annual
quail populations on the Range ; kinds of cover used by quail for escape,
roosting, and nesting ; sex and age ratios ; seasonal changes in weight ;
seasonal program as to pairing, nesting, and flocking ; history of pairs
and their nests, eggs, and broods, together with causes of failure ; ene-
mies of nests, eggs, young, and adults ; food throughout the year, espe-
cially in relation to kinds available ; importance of drinking water
through the dry season and means of providing it where needed ; dis-
eases, especially blood parasites ; effect of man, livestock, rodents, and
other animals on quail ; and experimental "harvest" of part of the quail
crop by shooting on a portion of the Range, the hunter effort involved,
and crippling losses.

Methods of Study. — To determine the factors influencing quail, it is
necessary to learn the activities and behavior of flocks, pairs and in-
dividuals, to follow them through seasonal changes in the weather and
the plant growth of the environment, to determine the annual cycle of
nesting, and to find the effects of food, shelter, competitors, enemies, and
disease. Since methods for the- quantitative study of animals in their
natural habitats are still in process of development, it has been necessary
to try experimentally various means to obtain the data which are wanted
regarding the quail.

Special traps have been designed to capture quail so that individuals
may be marked with leg bands for subsequent identification in order
to obtain weights, blood samples, and other periodic data. On each
trapped bird was placed a numbered aluminum band and one or more
colored celluloid bands for later sight identification. Nests were located
and marked and the history of each followed to final success or fail-
ure. For over two years quail were collected on outlying portions of
the Range or adjacent lands to provide stomachs for careful determina-
tion of the foods used. Observation blinds have been used at some nests
and watering devices to study in detail the activities of individuals and
small groups of quail.

Quail Census. — Each year a late winter count of the quail on the
Range has been made by "horseback census," with three riders traveling
in parallel courses back and forth over strips until all birds are enu-
merated. The method is practicable on open rolling range land with
scattered cover. This type of census provides a determination of the
population per acre and reveals the fluctuations which have occurred.
The method and results have been reported elsewhere in detail (5).
Counts of quail to date are given in table 23.

132

University of California — Experiment Station

Cover. — Valley quail require brush, rocks, or small trees as escape
cover to avoid their enemies. Piles of cut brush plants are an adequate
substitute where natural cover is scant. Small live oaks and certain of
the larger shrubs are highly important as cover in which the quail roost.
Destruction or removal of such growth on range lands results in a les-
sened quail population by reducing the "housing facilities" for the birds.
Types of cover used by quail for nesting are listed in table 24.

Sex and Age Ratios. — In 1939-40, a total of 703 birds were trapped
(including repeats), from July to February inclusive. These included
254 individuals previously trapped and banded and 193 not previously

TABLE 23

Censuses of Valley Quail on the Range

Year and dates of count

Acres in
area of
count

Number
of quail

Birds
per acre

Acres
per bird

1935, February 27

3,600
3,600
3,600
3,600
3,412
3,412

1,600*
1,637
2,123
1,545
1,293
873

0.44f
.45
.59
.43
.38

0.26

2.3

1936, February 27 to March 4

1937, February 24 to March 2

1938, March 9-17

2.2
1.7
2 3

1939, March 13-17

2 6

1940, March 6-10

3 8

* Estimated.

t Based on sample strip count.

handled. The sex ratio was 103 males for each 100 females in the birds
retrapped, and 105 for each 100 in the new birds. The age ratio naturally
fluctuated, immatures of the current season being more numerous in
early autumn. The older group showed 55 immatures per 100 adults and
the newer 87 per 100 ; by February both groups contained about 63 im-
matures per 100 adults. Of birds banded in earlier years, there were 12
from 1936-37, 3 from 1937-38, and 46 from 1938-39. The oldest in-
dividuals handled were four to four and a half years of age.

Seasonal Changes in Weight. — Over 2,000 quail weights have been
taken, 598 of these in the 1939-40 season. Records on young of the year
show a gradual rise in weight from about 120 grams (4.2 ounces) in
July to nearly 160 grams (5.6 ounces) in December. Adult males remain
about the same from July to October, then increase sharply to over 170
grams (6 ounces) and decline but slightly until February. Adult females
decline in weight until October, then follow the immature males as to
increase. During the winter season there is an abundance of food, so
there is no decline in weight by the local quail such as experienced by
upland game in regions of severe cold winters.

Nesting. — During the four seasons, 1937-1940, a total of 289 nests
were studied. Summaries of the results of these studies are given in

Bul. 663]

The San Joaquin Experimental Range

133

tables 24 and 25 ; details of the results for 1937 have been published
elsewhere (4).

Enemies of Quail. — Since many sportsmen believe that wild enemies
are the major factor in reducing quail populations, particular effort has
been made to determine the cause of losses. Some data on losses at times

TABLE 24
Principal Types of Cover Used by Quail for Nesting

Cover type

Percentages in the years given*

1937

1938

1939

1940

32.3
19.8
13.5
10.4
8.3
6.2
3.1
2.1
1.0
3.1

29.8
17.9
16.4
10.4
2.9
1.5
10.4
2.9
1.5
6.0

40.0
23.0
10.7
4.6

4.6
1.5
6.1
4.6
4.5

65 5

Fallen dead brush with grass or weeds

13.1
11 5

1.6

1.6

6 7

* Based on numbers of nests as follows: 96 in 1937; 67 in 1938; 65 in 1939; and 61 in 1940.

TABLE 25
Summary of Nesting Data for Valley Quail at the Experimental Eange

Year and nesting period

Total
nests
found

Incubated
(at least
in part)

Eggs per

nest
(average)

Broods
hatched

Young
per nest
(average)

1937; April 22 to July 26

96
67
65
61

40
24
34
35

11.0
11.7
11.1
12.9

16
12
16
15

10.2

1938; April 15 to July 28

9.8

1939; April 5 to July 5

9.5

1940; April 9 to July 22

9.5

other than during the nesting season have been obtained, but these are
still too few for reliable appraisal. Nesting losses can usually be indenti-
fied more accurately, although for some it is very difficult to determine
the cause. The rather heavy losses in which there is no obvious disturb-
ance of the nest have been ascribed to ground squirrels, but results ob-
tained in 1940 suggest that a considerable percentage of these may be
due to snakes. A summary of the record of nesting losses for 1937-1940
is given in table 26.

Food Taken by Quail. — Detailed analysis of the items taken by quail
on the Range as food has been made for one year (6) .

The food consisted of 99.5 per cent plant and 0.5 per cent animal
material. The plant food consisted of 72 per cent seeds and 27.5 per cent
leaves.

134

University of California — Experiment Station

Some plants such as Spanish clover and yellow tarweed were readily
accepted as seeds and ignored as leaves ; others, such as birdseye gilia
and blue dicks, were highly acceptable as leaves and hardly ever taken
as seeds.

Green plants, such as turkey-mullein, yellow tarweed, spurge, and
Spanish clover, which were present in considerable amounts during the

TABLE 26

Losses in Nesting of Valley Quail

Fate of nests

Successful

Abandoned before layingf

Abandoned after laying

Robbed by:

Ground squirrel

House cat

Coyote or dog

Skunk

Bobcat

Gray fox

Raccoon

California jay

Snaked

Unknown predators

Trampled by cattle

Trampled by horse

Accidentally destroyed by man .
Robbed for experiment

Percentages for the years given *

1937 1938 1939

17.7
13.5
5.2

31.2
5.2
5.2
4.1
2.1
1.0
0.0
2.1
0.0
6.2
3.1
1.0
1.0
1.0

17.9
16.4
11.9

26.8
1.5
7.4
4.4
0.0
0.0
1.5
1.5
4.4
5.9
0.0
0.0
0.0
0.0

24.6
15.3

7.7

30.7
0.0
6.1
1.5
0.0
1.5
0.0
0.0
6.1
3.0
1.5
0.0
1.5
0.0

1940

24.2
14.7

16.4
1.6
1.6
0.0
0.0
0.0
0.0
3.3
9.8

22.9
0.0
1.6
4.9
0.0

* Based on same number of nests as in table 25.
t Nests of this category not included in the 1940 count.

J This was segregated in 1940 into gopher snake, 6.5 per cent, and king snake, 3.3
per cent.

dry summer, did not serve to any appreciable degree as sources of suc-
culent vegetation in lieu of drinking water ; the local birds depended
chiefly on free water. Unpublished observations show that distribution
of the birds in summer was definitely influenced by watering places.

A wide discrepancy was found between the relative amounts of certain
items present in the plant cover and the proportions in which they were
taken by the quail. Some species, though sparse in the plant cover, for
example, clovers and ground lupine, made up a large percentage of the
quail diet ; whereas other common species, notably soft chess and foxtail
chess, contributed only small percentages of quail food.

Ground litter samples were taken to determine the abundance of seeds
available for quail food. It was revealed that there was a surprisingly
large quantity of choice seeds remaining, even after quail had fed con-
tinually during the dry season ; sample plots indicated that there was

Bul. 663] The San Joaquin Experimental Range 135

about 334 pounds of seed per acre on such areas in autumn before new
growth of plants had started. This suggests that food alone may not be
a very critical factor in the size of the quail population on the Range.

Drinking Water. — Location of available water has an important in-
fluence on the distribution of quail coveys. This is in agreement with
other investigations on western dry ranges. Experiments are being con-
ducted on the efficiency of various types of artificial watering devices at
the Range. Information is being obtained on the amount of use other
birds and mammals make of the same water supply, as also any inter-
ference by predators and range cattle. Eighteen water developments for
quail have been established in dry locations on the range.

Disease. — About 500 blood smears have been taken from trapped birds
and those collected for stomach analysis. Preliminary examination of
some of these has shown that quail malaria is present, but detailed study
of the material has not been completed. Additional studies are planned
in an effort to ascertain the relation, if any, of disease to fluctuations in
the general quail population.

Quail and Other Life. — Results of nesting losses show something of
the relative importance of various animals, livestock, and man on the
quail population. Additional data of other kinds are still too meager
for analysis. General observation suggests that sizable quail populations
may exist successfully on range lands devoted to other uses.

Experimental Hunting. — In wildlife management the ultimate ob-
jective is a harvest that can be placed on a sustained-yield basis. After
numbers were determined, factors of environment analyzed, and life
history studied, a regulated experimental hunt was arranged in 1938. A
hunting management unit of 720 acres was established. A harvest of
26.5 per cent of the birds on the controlled area was accomplished by
hunters selected by the Madera County Sportsmen's Association. A total
of 136 quail was taken by 21 hunters, spending 96.8 hours of effort; or
an average of 4.6 hours per man. The effort resulted in an average of 1.4
birds per man hour. It was continued on the same basis in 1939 and 1940.

Pinal conclusions and definite management recommendations must of
course be based on further study, but the foregoing statements of prog-
ress illustrate the trend of results to date.

OTHER STUDIES AND EXPERIMENTS IN THE PROGRAM

OF THE SAN JOAQUIN EXPERIMENTAL RANGE

M. W. Talbot, H. H. Biswell, P. B. Rowe, and A. W. Sampson

The preceding sections have dealt in considerable detail with those
studies of management of forage, cattle, and wildlife which have pro-
ceeded far enough to warrant progress-report statements. In addition,
there are two studies which are not yet at publication stage and one
which has been reported on in a separate paper. By including a brief
description of the purpose, nature, and scope of these, it is believed that
the reader may obtain a more complete picture of the whole program.

ARTIFICIAL RESEEDING

Stockmen throughout the granite area are becoming increasingly in-
terested in range reseeding as a possible means of improving range
forage. Increasing attention is being given to reseeding as an integral
part of improved practices and systems of grazing. More and more they
are asking such questions as : Can the present forage cover be improved
through artificial seeding ? How long will the new cover last, and what
will be the cost? This is a very well- justified interest, because of the
well-known deficiencies in the present foothill forage during nearly
half of the year, from the time these annual plants dry to a crisp in May
or June until fall rains bring green forage again in November or Decem-
ber. So, very logically the foothill operator asks : Can artificial reseeding
help us bridge this gap of poor feed that is a source of great loss to the
industry every year ?

The question is complicated because of the practically rainless sum-
mer period of five months or more, which makes this problem more
difficult of solution than in many other western range regions. Still, the
values at stake are high, and the reseeding studies are widely regarded
as an exceedingly important part of the investigations on this area.

The work has been planned as a three-step program. First, a study in
the forage-plant nursery to test the adaptability of various species to
the prevailing climate — on the better sites and with a minimum of
competition from other plants. This comparison of many different kinds
of plants growing side by side in weeded rows sorts out those that sur-
vive and grow best under the severe climatic conditions of this region.
To date this trial has included some 225 species, including grasses, herbs,
and shrubs — annual and perennial, native and foreign. Seeds for test-
ing have been generously furnished by cooperators in the Bureau of
Plant Industry, the Commonwealth of Australia, the Soil Conservation
Service, and various other agencies.

[136]

Bul. 663] The San Joaquin Experimental Range 137

In the next step of this group of studies, the few plant species that
show most promise in the nursery, in good soil, are seeded or trans-
planted to larger range plots, in more typical range soil, and also fenced
against livestock and for a time against rabbits while the plants are
becoming established. Thereafter, they are to be left alone to compete
as best they can with the existing range vegetation. This second step will
further sort out the comparatively few species that show promise in this
test against both climate and competing vegetation. Approximately 50
different kinds of plants have been thus seeded or transplanted to date.

The third planned step, not yet undertaken, will consist of large-scale
seeding on the range of those few species that best meet the first two
tests ; also the development of satisfactory methods of seeding.

This reseeding project has been limited largely to survival testing in
the nursery plus preliminary testing on the range in competition plots.
These tests are yielding needed information on behavior of the various
plant species or strains under varying conditions of the rigorous foothill
climate. In this area perennial grasses and herbs generally are at a
disadvantage, for they are subjected almost every year to extreme
drought conditions during the long hot dry summer period. Results to
date of these incomplete studies do not yet justify recommending any
plant species for extensive seeding on the range. Nevertheless, the
promise shown by some, as briefly sketched in the following comments,
warrants their further and wider testing.

Subterranean Clover. — Unlike our common bur-clover, the species
Trifolium subterraneum produces burs both above and below the soil
surface. The flower stems curve downward, which brings the flower
clusters in contact with the soil. If the ground is moist and sufficiently
soft, rootlike growths develop which draw the ripening fruit (the so-
called "bur") into the soil — hence the common name, subterranean
clover. There are many different strains, some of which are doing well
at the Range. In 1939, some of the runners were 3 feet long and a heavy
bur crop was produced. During the period 1938 to 1940, this plant re-
seeded itself well in the nursery where a good stand of volunteer plants
appeared. Seed of this plant was obtained from Australia where it is
regarded as an excellent forage plant.

Creeping -Booted Alfalfa. — The species of Medicago, known as creep-
ing-rooted alfalfa, has been tested for only one year, but the promise
shown in this time warrants its mention. It was introduced from Turkey
by the Bureau of Plant Industry from an area where the annual rainfall
is about 12 inches and the elevation is 4,000 feet. This is a very deep-
rooted perennial which spreads rapidly by underground stems. Cuttings
can be made readily. Seed is not yet available.

138 University of California — Experiment Station

Other Alfalfas. — About 25 different species of alfalfas have been re-
ceived from the Bureau of Plant Industry for testing at the Range. Most
of these are from India, Turkey, Chile, Manchuria, Afghanistan, China,
Ecuador, Russia and Turkestan; two are native to dry areas of this
country. Some of these alfalfas planted in the fall of 1939 are producing
a heavy crop of seed. These are being tested in the hope that some may
prove more drought-resistant than our common cultivated varieties
and for that reason offer some possibilities locally in the more fertile
range areas. Seeds of these plants are not yet available.

Bur-Clovers. — Several different kinds of bur-clover are included in
the nursery tests. Some of these have reseeded naturally in the nursery
and in other respects show promise, but their testing has not been car-
ried far enough to show whether they are superior to the California bur-
clover, which is an excellent range plant but has two weaknesses : spotted
occurrence and uncertainty of yield. Some of these new bur-clovers pro-
duce very large burs, and if they should prove adapted to local conditions
they would add to the value of the dry summer forage.

Garden Burnet. — This is another perennial plant that has survived
the hot dry summers. It has reseeded itself in the plant nursery each
year since 1936 and many new seedlings appear there every year. A
planting was made in the competition plots out on the Range in 1937 and
the plants have matured and produced seed every year.

Wimmera Rye Grass. — The species, Lolium rigidum var. strictum,
grows, matures and produces seed in a short period of time. It is re-
ported to be very palatable to livestock. During the two years of test in
the nursery it has done well. Most of the seed crop was harvested but
a scattered remnant produced many seedlings in the nursery the next
fall. From a seeding made in the competition plots out in the field in
1939, this grass made a fine growth and produced much seed, but it is
not yet known whether or not it will spread from the plots, or maintain
itself.

Smilo. — The species, Oryzopsis miliacea, known as smilo or rice grass,
is rated as about the most promising of the perennial grasses thus far
tested. It makes a good winter growth and stays green late in the season.
During the summer many of the grass blades dry and fall, which enables
the plant to withstand extreme drought periods. This plant produces a
good seed crop in most years. Abnormally dry weather during spring
may prevent seed development.

Saltbushes. — Two large saltbushes (also called shadscales) now being
tested are showing promise. One is a common forage shrub (Atriplex
canescens) of arid sites in the southwestern and intermountain regions
of the United States; the other (A. nummularia) is reported to be a

Bul. 663] The San Joaquin Experimental Range 139

good forage plant in Australia. An important object of the reseeding
experiments is to determine whether some of these better browse species
can be established in the granite area, to supplement in some degree the
inferior native shrubs on this area and thus furnish more green feed
during the summer dry-feed period.

In addition to testing of native and introduced plant species, plans
for this project include assistance in field testing of improved forms of
plants developed through plant-breeding work by the Bureau of Plant
Industry and the California Agricultural Experiment Station.

RAINFALL RUNOFF AND EROSION

Sierra Nevada foothill lands are chiefly valuable for grazing. This is
their primary use. These same lands have important watershed values
also, since with the adjoining higher mountains they comprise the
drainage areas which supply municipal and irrigation water to exten-
sive irrigated sections below. The effect of grazing on watershed values
is a subject on which there has been little definite information for this
particular type of land. In the summer of 1934, a special set of plots
was established to furnish information on some of the numerous water
problems, more particularly those concerned with surface runoff. No
particular concern was felt about erosion, for the locality represented
by the Range is not a critical range erosion area under reasonably good
management.

The field setup for this experiment consists essentially of a group of
9 rectangular plots %0 acre in size and equipped with automatic measur-
ing and recording devices. The runoff from each plot is measured by
catching the water in a concrete trough at the base of the plot, then
running it through a settling tank and a specially designed tipping
bucket which fills and empties as long as the runoff continues ; with each
tipping an electrical contact is made and automatically recorded. An
intensity rain gage is also connected with the recorder. A continuous
record is thus obtained for the total rainfall, the period during which
it fell, and the total amount of resulting runoff.

The plots were left ungrazed for the seasons of 1934-35, 1935-36,
1936-37, and until March of the 1937-38 season, in order to determine
their natural differences. During this period there was no appreciable
erosion, and surface runoff was less than 1 per cent of precipitation,
which indicates an infiltration capacity of this soil under ungrazed
forage cover equal to or in excess of the highest rainfall intensities that
occurred during that period. Precipitation on these particular plots
for the first three seasons was 29.4, 22.7, and 23.3 inches, respectively.
By March of the fourth season, precipitation amounted to 28.5 inches.

140 University of California — Experiment Station

After March 1938, and during 1939, three of the plots were grazed
closely, three lightly to moderately, and three were left ungrazed. Pre-
cipitation for the rainy season of 1938-39 was 12.1 inches ; for 1939-40,
21.7 inches. Under conditions that prevailed following these treatments,
there was no significant increase in surface runoff from any of the plots
as a result of grazing, on these particular gentle slopes. Surface runoff
as here reported should not be confused with total water yield, which
would include percolation, which was not measured in this study. The
results indicate rather that most of the runoff from this granite-soil site
of rolling foothill country with gentle slopes comes from seepage along
the drainage lines rather than from overland flow down the slopes. This
study, of course, has been running for too short a time to justify any
definite conclusions.

CHEMICAL COMPOSITION OF THE IMPORTANT RANGE PLANTS

Seasonal trends in the chemical composition of forage plants common
to the San Joaquin foothills have been presented by Gordon and Samp-
son (7) following a three-year study. The composition of the plants
studied showed a rather close correlation within the different groups of
plants, such as the grasses, the deciduous trees and shrubs, and the
grasslike plants. The closest correlation, however, was found with the
cyclic march to maturity as noted in distinct "growth stages." Species
growing side by side may differ widely in composition; likewise, the
same species in distinct growth stages may differ more widely in chem-
ical values than the differences noted between various species of similar
habit. Also plant samples collected on the same calendar date in two
different years without regard to growth stages may fluctuate widely in
chemical composition; but samples collected at the same growth stage
are consistent in composition. Consequently, it was concluded that ana-
lytical data of a species whose sample is not representative of a specific
growth stage has little or no bearing on the nutritional value of the
plant. Moreover, to contribute to the question of the relative worth of
a species as a range plant, chemical analyses at different growth stages,
must be included.

The percentages of crude protein, and of calcium, phosphorus, and
potassium in herbaceous species were found to decrease continuously
in an orderly manner from the earliest appearance of leaf blades to
plant maturity. With respect to crude protein, phosphorus, and potas-
sium values, this downward trend, associated with development toward
maturity, was found to be true also in deciduous shrubs and trees. In
sharp contrast, the total silica-free ash and calcium in deciduous shrubs
and trees increased as development progressed toward maturity. Crude

Bul. 663] The San Joaquin Experimental Range 141

fiber also increased with development in all plants except the deciduous
shrubs and trees, the nondeciduous shrubs, and the half-shrubs. Crude-
fiber content remained constant in the deciduous shrubs and trees,
whereas in the half -shrubs this constituent fluctuated much less.

All grass species investigated were high in caloric values, and in min-
erals, from the early leaf stage until they were in full bloom. The grass-
like plants, such as the rushes, are relatively low in protein values and
high in crude fiber by midsummer, but maintain value as late forage
because they retain high succulence after grasses have dried up. The
broad-leaved herbs fall into two forage categories: species having a
short vegetative period but which maintain a high level of protein and
ash contents well into the late flowering stage ; and those having a long
vegetative period and characterized by a low level of ash and protein
when the flowering period is reached. Compared with other plants, the
crude protein content of the leafage of deciduous shrubs and trees is high
in the early leaf stage ; and even when the foliage is mature this con-
stituent is maintained at higher levels than in most species studied.
Despite the fact that the nondeciduous shrubs are at no time especially
high in protein or ash constituents, they, as well as the more nutritious
deciduous woody plants, are important as browse. Both these plant
groups remain succulent, and some maintain a fairly high level of pro-
tein and essential inorganic constituents late in the season when the
herbaceous cover is at its lowest nutritional ebb ; for this reason trees
and shrubs supply important livestock feed.

Since grasses constitute an important part of the forage provided on
the lands typified by the Range, the following criteria are suggested for
their evaluation from the standpoint of chemical composition : ( 1 ) the
interval between the beginning of growth and the flowering period
must be long ; (2) the crude protein content must remain at a relatively
high level after the blossoming period, and well into the time that the
seeds attain the dough stage; even after seed dissemination it is de-
sirable that a portion of the basal leaves and stems remain succulent
until late in the summer — a characteristic of many perennial grasses ;
(3) in the species whose herbage matures early a large proportion of
the seeds must be retained after the leaf blades have become dry.

General criteria for judging the forage value of a range plant were
found to group around high nutritive value through the life cycle
together with a long period of vegetative growth. Most of the perennial
grasses investigated have both these desirable characteristics. However,
the importance of other range plants lower in nutritive values, yet hav-
ing relatively long life cycles, was not discounted on an annual-type
range.

142 University of California — Experiment Station

Although the forage is richest in most constituents in the early leaf
stage, grazing intensively, or to the point of full utilization of the forage
early in the season is regarded as undesirable. Many workers have shown
that continuous close cropping reduces the season's total forage crop.

A supplementary study indicated that a short growing season, such
as not uncommonly occurs in the foothill and valley regions of the state,
may result in forage production particularly low in phosphorus, and in
some species, in calcium.

CONTRIBUTORS AND GENERAL ACKNOWLEDGMENTS

Contributors of the several subject-matter sections in this bulletin are
grouped, by major agencies, as follows :

Agricultural Experiment Station, University of California

Adams, E. L., Professor of Farm Management, Agricultural Economist in the Ex-
periment Station, and Agricultural Economist on the Giannini Foundation.

Carpenter,, G. A., Eesearch Assistant on the Giannini Foundation; resigned Jan-
uary 16, 1940.

Crawford, L. A., Assistant Professor of Agricultural Economics, Assistant Agricul-
tural Economist in the Experiment Station, and Assistant Agricultural Econo-
mist on the Giannini Foundation ; resigned January 1, 1941.

Guilbert, H. R., Associate Professor of Animal Husbandry and Associate Animal
Husbandman in the Experiment Station.

Hart, G. H., Professor of Animal Husbandry and Animal Husbandman in the
Experiment Station.

Sampson, A. W., Professor of Forestry and Plant Ecologist in the Experiment
Station.

Storer, T. I., Professor of Zoology and Zoologist in the Experiment Station.

Storie, R. E., Lecturer in Soil Technology and Associate Soil Technologist in the
Experiment Station.

Voorhies, E. C, Professor of Agricultural Economics, Agricultural Economist in
the Experiment Station, and Agricultural Economist on the Giannini Founda-
tion.

Wagnon, K. A., Animal Husbandry Specialist at the San Joaquin Experimental
Range.

California Forest and Range Experiment Station, Forest Service,
United States Department of Agriculture
Biswell, H. H., Associate Forest Ecologist ; transferred September 1, 1940.
Nelson, J. W., Superintendent, San Joaquin Experimental Range.
Rowe, P. B., Associate Silviculturist.
Talbot, M. W., Chief, Division of Range Research.

Region 5, Forest Service, United States Department of Agriculture
Cronemiller, F. P., Assistant Regional Forester, in charge of Division of Wildlife

and Range Management.
Glading, Ben, Junior Biologist ; resigned July 31, 1940.

Fish and Wildlife Service, United States Department of the Interior

Fitch, H. S., Junior Biologist, Division of Wildlife Research ; f urloughed for mili-
tary service May 22, 1941.
Horn, E. E., Biologist, Division of Wildlife Research.

Credit is especially due F. G. Renner, formerly Forest Ecologist of
the California Forest and Range Experiment Station, who was in field
charge of planning, establishing, and supervising the forage experi-
ments for the first two years, 1934 and 1935. Similarly, credit is due

[143]

144 University of California — Experiment Station

J. W. Nelson, present Superintendent of the Experimental Range, for
administrative supervision of the whole cooperative program since 1935.

A. L. Hormay and J. R. Bentley, Associate Forest Ecologists, assisted
in analyzing the results; and various students and other temporary field
assistants, to whom individual credit cannot be given, helped from time
to time in gathering the data.

CCC labor and various emergency funds have been of material as-
sistance in the improvement and development of the area, including
construction of fences, buildings, livestock watering places, corrals, and
other facilities required for the different experiments.

Acknowledgment is due the Work Projects Administration for the
original drawings of figures 5, 6, 7 and 8, made by Pat Haskey, under
WPA Official Project No. 65-2-08-369 ; and for assistance rendered on
various phases of the work on interrelations of rodents and other wild-
life of the range, under WPA Official Project No. 165-2-08-225.

Throughout the period the workers have had the advantage of a
stockman's advisory committee composed of producers of wide experi-
ence. They have thus been able to utilize the rich experience of this very
representative cross section of the industry, to develop and foster a
better understanding of the problems. This committee consists at present
of the following men whose assistance is gratefully acknowledged : Har-
vey A. Russell, chairman; Hubbard Russell, American National Live-
stock Association ; P. P. Cronemiller, M. A. Benedict, Region 5, U. S.
Forest Service; E. L. Garthwaite, California Agricultural Extension
Service; Theodore Chamberlin, W. Hugh Baber, Phil Klipstein, H.
Clay Daulton, Eugene Mallory, Brooke Mordecai, Joe Urrutia, Cali-
fornia Cattlemen's Association ; H. E. Black, State Division of Fish and
Game; Charles P. Campbell, State Division of Forestry; Craig Cun-
ningham, Madera County Chamber of Commerce ; Ernest Mason, Will
Ryan, Madera County Farm Bureau; Charles Marsh, E. E. Nelson,
Madera County Fish and Game Protective Association; Neils Over-
gaard, Madera County Agricultural Commissioner; C. T. Carver, C. L.
Guthrie, Kern and Tulare counties ; Irvine Armstrong, Monterey and
San Benito counties; Charles Ellinwood, John Grohl, Sacramento, San
Joaquin and Stanislaus counties; Walter Buell, Santa Barbara and San
Luis Obispo counties ; Jesse Bigelow, Herrick Brown, Neil McDougald,
Jackass Cattlemen's Association ; Archibald Shaw, Jr., Will Footman,
Ahwahnee Farm Center. In addition, the cooperating staff members are
ex-officio members of this committee.

The Bureau of Plant Industry and the Soil Conservation Service of
the United States Department of Agriculture and various state and
foreign agencies cooperate in the range reseeding project.

Bul. 663] The San Joaquin Experimental Range 145

LITERATURE CITED

1. Bis well, H. H.

1939. A change in our California forage plants. California Cattleman. Septem-
ber, p. 4-5.

2. Fitch, H. S.

1940. Some observations on horned owl nests. Condor 42:73-75.

3. Fluhartt, L. W.

1939. Beef cattle enterprise efficiency studies. A four-year summary, 1935-1938
inclusive. California Agr. Ext. Leaflet. 48 p. (Mimeo.)

4. Glading, Ben.

1938. Studies on the nesting cycle of the California valley quail in 1937.
California Fish and Game 24:318-40.

5. Glading, Ben.

1940. Valley quail census methods and populations at the San Joaquin Experi-
mental Eange. California Fish and Game 27:33-38.

6. Glading, Ben, H. H. Bis well, and C. F. Smith.

1940. Studies on the food of the California quail in 1937. Jour. Wildlife Mangt.
4:128-44.

7. Gordon, A., and A. W. Sampson.

1939. Composition of common California foothill plants as a factor in range
management. California Agr. Exp. Sta. Bul. 627:1-95.

8. Guilbert, H. E., and A. McDonald.

1934. Weight records on purebred beef cattle during growth, gestation, and
lactation, together with data on reproduction. Amer. Soc. Anim. Prod.
1933 Proc. 26:244-53.

9. Guilbert, H. E., S. W. Mead, and H. C. Jackson.

1931. The effect of leaching on the nutritive value of forage plants. Hilgardia
6:13-26.

10. Guilbert, H. E., and L. H. Eochford.

1940. Beef production in California. California Agr. Ext. Cir. 115:1-125.

11. Hart, G. H., and H. E. Guilbert.

1928. Factors influencing percentage calf crop in range herds. California Agr.
Exp. Sta. Bul. 458:1-43.

12. Hart, G. H., and H. E. Guilbert.

1933. Vitamin-A deficiency as related to reproduction in range cattle. California
Agr. Exp. Sta. Bul. 560:1-30.

13. Hart, G. H., H. E. Guilbert, and H. Goss.

1932. Seasonal changes in the chemical composition of range forage and their
relation to nutrition of animals. California Agr. Exp. Sta. Bul. 543 : 1-62.

14. Talbot, M. W., H. H. Biswell, and A. L. Hormay.

1939. Fluctuations in the annual vegetation of California. Ecology 20:394-402.

17m-5, '42(6975)