Volume 17, Number 1 (January 1964)

(1)

(2)

(3)

IN THIS

ISSUE

Sulfur Fertilization of an Annual-Range Soil During Years of Below-Normal Rainfall

William A. Williams, Cyrus M. McKeZZ and Jack N. Reppert Effects of Early Spring Burning on Yields of Native Vegetation

J. L. Launchbaugh Development and Use of the Esophageal Fisfula: A Review

George M. Van Dyne and D. T. ToreZZ A Comparison of Two Sweefclover Strains and Ladak Alfalfa Alone and

in Mixtures with Crested Wheafgrass for Range and Dryland Seeding F. B. Gomm A Comparative Study of Soils of Selected Creosofebush Sites in

Southern New Mexico... ____ _ ____ ____ __ ____ ___ _____ K. A. Valentine and J. J. Norris Emergence of Cheaigrass and Three Wheafgrasses from Four Seeding

Depths ____ ____ ___________._ ______________ ______ _____________ __________ ____ _.___ ________ .______ ________A. C. Hull, Jr. The Influence of Grazing on the Roofs and Rhizomes of Seacoast

Bluesfem _____________ ___ _______.__ _.______ __________ James E. Bowns, Jr. and Thadis Design Criteria for Rainfall Cafchmenf Areas for Watering Wildlife

and Livestock .______ _____ ___.___________ .___ _._________ ____ _.___________ ___. ___ ____ _______ _____ Joel E. Zonafion of Undersfory Vegetation Around a Juniper Tree

Joseph F. Technical Notes:

A Core Sampler for Excavating Grass Roofs

W. Box

Verner

Arnold

James E.Bowns, Jr. and Thadis W. Box Book Reviews: Careers in Conservation (D. G. Wilson); Approved Practices in Pasture Management (Kling L. Anderson) ; Pasture Economy and Meadow Cultivation (Jack R. Harlan); The Last Horizon (A. L. Hafen- richter); Fertilizer Technology and Usage (J. L. Launchbaugh); Land

and Wafer Use (R. R. Humphrey)___._ ________ _________________________________________~ ______ _ .__..____ _ Current Literature ____ ________-- _-_.._ ____. __ ______ __ _________ _______________ _____ ______.______________._..~ ______ ____ ____ ____

News and Notes __.-- . . . ___________._______ _ .___ _ ________ ____ ____ __ _______ ____ ______ _____ ____ ________________________ ____ ____ Wifh fhe Sections ____. __.__ ___-- _ __._ _ ___._____ _______ _______ _ _______ ____________________ ____ ____________ ____ ________ _____ _ _.__

Society Business: Program Seventeenth Annual Meeting ______ _ ._____________ _______________

Cover Photo- Watering Out

1

5

7

19

23

32

36

39

41

43

45 49 52 57 60

(4)

(5)

Journal of

Volume 17, Number 1 January, 1964

RANGE MANAGEMENT

Sulfur Fertilization of an Annual-Range Soil

During Years of Below-Normal Rainf all1

WILLIAM A. WILLIAMS, CYRUS M. McKELL, and JACK N. REPPERT

Associate Professor of Agronomy and Associate Agron- omist, University of California, Davis and Riverside, and Range Conservationist, Pacific Southwest Forest and Range Experiment Station, Forest Service, USDA, Susanville, California.

Sulfur is a major limiting fac-

tor in forage production on many

range soils in California (Martin,

1958). Leaching loss of available

sulfur in the form of sulfate

from slightly acid, coarse tex-

tured soils has been recognized

as a problem in fertility manage-

ment. Gypsum (calcium sulfate)

is a source of sulfur readily

available to plants, but is highly

susceptible to leaching

loss

(McKell and Williams, 1960).

Elemental sulfur is more slowly

available since it must be oxi-

dized, usually microbially, to sul-

fate before becoming usable by

plants. Elemental sulfur also is

less susceptible to leaching.

A series of lysimeter studies

was started at the San Joaquin

Experimental range in the Sierra

Nevada foothills 25 miles north

of Fresno, California, in 1957. In

the first year of study, various

rates of gypsum labeled with sul-

fur-35 were applied, and the

subsequent distribution of na-

tural and applied sulfur was de-

termined in the plants, soil, air,

1Contribution of the Department of Agronomy, University of California, Davis and Riverside, with the co- operation of the Pacific Southwest Forest and Range Experiment Sta- tion, Forest Service, USDA.

rain and percolate during a sea-

son of above-normal rainfall.

Percolating water carried 77 per-

cent of the applied sulfur out

of the root zone by the end of

the season (McKell and Wil-

liams, 1960)) most of it being lost

before rising temperatures per-

mitted appreciable growth of the

seeded clover (McKell and Wil-

son, 1963).

Lobb and Bennetts (1957) have

recommended the use of elemen-

tal sulfur in preference to gyp-

sum in sulfur-deficient soils of

New Zealand in which excessive

leaching occurs. Hence, an ex-

periment was initiated to com-

pare elemental sulfur and gyp-

sum as sources of sulfur on an

annual-range soil in an environ-

ment typical of much of Califor-

nia’s foothill range country. Sul-

fur-35 was used to distinguish

the applied sulfur from naturally

occurring sulfur.

Methods

The lysimeters, soil, and gen-

eral procedures used in this

study were the same as those

described in detail for the first

experiment (McKell and Wil-

liams, 1960)

.

In brief, the experi-

ment was conducted on Vista

sandy loam soil in lysimeters six

1 feet in diameter and two feet

deep. Major soil characteristics

were pH 6.2, cation exchange

capacity 4.16 me./lOOg., and or-

ganic matter 0.6 percent. At the

conclusion of the first experi-

ment the lysimeters were di-

vided into three stratified groups

based on their residual sulfur

content, with each replication of

the treatments assigned to a

group of lysimeters and the

treatments randomly assigned

within the replicate. The treat-

ments comprised control, finely

ground elemental sulfur applied

at the rate of 60 pounds per acre

and gypsum at 300 pounds per

acre to obtain equivalent sulfur2.

The treatments were applied,

and the lysimeters seeded to rose

clover (Trifolium hirtum All.),

in October 1958. Yield, percolate,

and precipitation were sampled

over a three-year period.

Results Clover Response

During the series of relatively

dry years in which this study

was conducted, clover produc-

tion was influenced greatly by

the amount and distribution of

rainfall. No yield response to sul-

fur or gypsum application was

detectable in 1959 or 1961 when

rainfall was very limited (10.42

and 12.36 inches in the respective

seasons, Figure 1). Dry-matter

production was 1,000 pounds per

acre or less regardless of treat-

(6)

2 WILLIAMS, McKELL, AND REPPERT

Table 1. Yield and sulfur content of rose clover following application of elemental sulfur and gypsum in autumn 1958.

Treatment 1959 1960 1961

Control Elemental S Gypsum Control Elemental S Gypsum Control Elemental S Gypsum

---_ _{(Pounds per Acre) - - - -}

340 3,120’ 620

410 4,490 990

580 4,240 1,040

---_ _{(Total sulfur percent) - - - -}

0.09 0.05” 0.06”

.lO .lO .ll

.26* .lO .ll

- - (Sulfur from fertilizer percent) - -

0 0 0

40 51 26

55” 56 26

*Value significantly different (five percent level) from other treatment means in the same year (zero values were excluded from the analysis of variance).

ment (Table 1). In 1960, the sec-

ond year after application of the

fertilizer, both the elemental sul-

fur and gypsum increased yield

by approximately 1,000 pounds

per acre over the control yield of

3,100 pounds per acre. Substan-

tial rain the previous September,

along with adequate

rain in

March and April 1960 (seasonal

total 15.54 inches, Figure

l),

made moisture conditions more

favorable than in 1959 and 1961,

thus permitting the expression of

a sulfur response.

During each year of the ex-

periment, the sulfur percentage

in the rose clover at harvest time

(bloom stage) was influenced by

treatment.

In 1959 the clover

grown on the gypsum-treated ly-

simeters contained almost three

times as much sulfur as either

the control or elemental sulfur

treatments (Table 1). In the re-

spective treatments the propor-

tion of total sulfur derived from

gypsum was significantly greater

than the sulfur obtained from

the elemental sulfur applied. In

1960 and 1961 the sulfur per-

centage in the clover produced

on lysimeters treated either with

gypsum or elemental sulfur was

nearly double the sulfur percent-

age in the controls. The propor-

tion of sulfur in the plants from

the two sources was not signifi-

cantly differ en

t ,

but declined

from a range of 48 to 56 percent

in 1960 to 24 to 26 percent in 1961.

Loss of Sulfur by Percolation

Percolation

of rain water

through

the soil columns

amounted

to 2.5, 2.4, and 1.7

inches in the first, second, and

third seasons, respectively. The

only major loss of sulfur during

this period occurred in the first

year when 16.0 pounds per acre

appeared in the percolate of the

gypsum treatment (Figure 2). Of

this loss 65 percent came from the

gypsum, but this amounted to

only 17 percent of the total of 61.1

pounds of sulfur per acre applied

in the gypsum. The loss from the

elemental sulfur treatment due

to leaching

in the first year

amounted to only 3.9 pounds of

sulfur per acre, of which less

than one pound was from the ap-

plied total of 59.7 pounds per

acre. The loss of sulfur from the

controls amounted to 3.5 pounds

per acre for the same interval of

time.

Sulfur Balance Sheet

A sulfur balance sheet was

constructed for each treatment

using the data collected for addi-

tions to and losses from the soil

columns (Table 2). Additions

considered were from the fer-

tilizer treatments and from rain.

Previously it has been shown

that additions of sulfur from air

contact and seed are negligible

(McKell and Williams,

1960).

Losses were from clover

re-

moved and deep percolation.

Total sulfur addition

from

rainfall over the three-year pe-

riod was 8.7 pounds per acre.

Sulfur removal in the clover for-

age amounted to: control 2.3, ele-

mental sulfur treatment 5.1, and

gypsum treatment 6.0 pounds per

acre. Sulfur lost in percolating

water amounted to: control 4.7,

elemental sulfur treatment 9.5,

and gypsum

treatment

22.0 pounds per acre. It was calcu-

lated by difference that sulfur

was retained by the soil in the

amount of: control 1.8, elemental

sulfur treatment 53.3, and gyp-

sum 40.9 pounds per acre.

The absorbed sulfate in sam-

ples from the soil columns was

extracted with sodium acetate at

pH 4.8 at the beginning and end

of the experiment. The sulfate-

sulfur content of the soil de-

clined in all treatments in the

following amounts: control 28,

10.42 IN.

15.54 IN.

12.36 IN.

'ASONOJFMAMJ

II958

₁

1959 1960

1

1961

)

(7)

SULFUR FERTILIZATION

SULFUR SOURCE

Cl

NATURAL

_n

lBzi

APPLIED

-

n

59 60 61

CONlROl

m

60 WM. S

EEL

60 GYPSUM

FIGURE 2. Amount and source of sulfur in percolating water from soil columns treated

with elemental sulfur and gypsum.

elemental sulfur 17, and gypsum

18 pounds per acre (Table 3).

The sum of the respective values

and the calculated sulfur-reten-

tion value for the treatment is

the amount of sulfur “apparently

immobilized” in the soil. These

amounts are control

30, ele-

mental sulfur 71, and gypsum 58

pounds per acre.

Discussion

During the first growing sea-

son (1958-59)) a drier one than

normal, some sulfur from the ele-

mental sulfur application was

taken up by the clover plants.

However, the amount was not

sufficient to increase their total

sulfur content relative to the

controls. Neither did appreciable

amounts of sulfate ion percolate

through the soil columns. The en-

vironment

(mainly

low soil

moisture) was not conducive to

microbiological oxidation of ele-

mental sulfur to sulfate, the

form most readily taken up by

plants (Starkey, 1950). However,

clover took up a sizeable amount

of sulfate from the gypsum, as

indicated both by higher total

sulfur content and by the sub-

stantial fraction of labeled sul-

fur present. Also an appreciable

amount of labeled sulfur ap-

peared in the percolate.

Al-

though the initial response to

elemental sulfur was markedly

slower than to gypsum, the mois-

ture limitation was so extreme

that no yield response was ob-

tained from either treatment in

1959.

At another location with more

favorable precipitation Walker

and Williams (1963) observed

that elemental sulfur caused in-

creases in forage yield equal to

that produced by gypsum on an-

nual-type range during the sea-

son the application was made.

Presumably moisture conditions

favorable to microbiological oxi-

3 dation of elemental sulfur are

concomitant with moisture con-

ditions favorable to vigorous an-

nual-range plant growth and vice

versa.

In the second season (1959-60)

the rainfall total, although again

subnormal, was more favorable,

especially as regards distribu-

tion. Clover production improved

greatly. Significant and approxi-

mately equal yield responses

were obtained

from the ele-

mental sulfur and gypsum treat-

ments. Thus the sulfur not lost

nor removed in plant tissue in

the first season was effective in

the second season. Uptake of sul-

fur, both labeled and natural,

was about the same where either

source of sulfur was used, but

much enhanced relative to the

control. Walker (1958) also ob-

served a carry-over effect of gyp-

sum after a dry year, although

with substantially smaller appli-

cations (rates equivalent to five

and 15 pounds of sulfur per

acre).

Unfortunately

in the third

growing season (1960-61) rain-

fall shortage again imposed a

Table 2. Sulfur balance sheet for Vista sandy loam freafed wiih elemental sulfur and gypsum.

Sulfur disposition Elemental

Source Control Sulfur Gypsum

--- (Pounds per acre) - - - Sulfur added from:

Fertilizer 1958 0 59.7 61.1

Rain 1958-59 3.2 3.2 3.2

1959-60 4.1 4.1 4.1

1960-61 1.4 1.4 1.4

Total sulfur added 8.7 68.4 69.8

Sulfur removed in:

Clover 1959 0.3 0.4 1.5

1960 1.6 4.2 4.3

1961 0.3 1.0 1.1

Clover total 2.2 5.6 6.9

Percolate 1959 3.5 3.9 16.0

1960 .8 1.3 1.6

1961 .4 4.3 4.4

Percolate total 4.7 9.5 22.0

Total sulfur removed 6.9 15.1 28.9

Calculated sulfur retention by soil: ~~

(8)

4 _{WILLIAMS, McKELL, AND REPPERT}

Table 3. Extractable sulfur confenf

of Vista sandy loam ireafed with

elemental sulfur and gypsum, and apparent immobilizafion of sulfur.

Extractable sulfur Calculated retention Apparent immobilization Treatment 1958 1961 Decrease (from balance sheet) (decrease -I- retention) --- _{(Pounds per acre) - - - -}

Control 47 19 28 1.8 30

Elemental S 41 23 18 53.3 71

Gypsum 41 24 17 40.9 58

very low ceiling on clover pro-

duction, but sulfur uptake con-

tinued at an enhanced level from

both sulfur sources, indicating

that the applied materials were

at least somewhat available to

the plants at that time.

The approximate equivalence

of uptake of sulfur from the two

sources in the second and third

seasons after application is at

variance with the observations of

Jordan and Baker (1959) in a

field experiment

with alfalfa.

Their results showed

an in-

creased uptake of sulfur from

gypsum relative to elemental sul-

fur persisting over a three-year

period after application.

The

mean uptake of fertilizer-sulfur

as a percent of total plant uptake

was 15 percent from gypsum and

4 percent from elemental sulfur.

The peak uptake from both ma-

terials occurred in the second

year of their experiment.

The inference

that can be

made concerning the fate of the

applied sulfur remaining in the

soil at the conclusion of the pres-

ent experiment is worthy of note.

The sulfur balance sheet calcu-

lations show that a very high

proportion of the applied ele-

mental sulfur and a somewhat

lesser, yet substantial, amount of

gypsum-sulfur was retained by

the soil at the end of the three-

year period. Extraction by so-

dium acetate at pH 4.8 (Ens-

minger, 1954 and 1958; Kamprath

et al., 1956 and 1957; and Bards-

ley and Jordan, 1957) demon-

strated that the adsorbed sulfur

content of the soil columns was

low, even lower than at the be-

ginning of the experiment. It is

hypothesized,

therefore, that

either immobilization through

biological activities, reduction to

metallic sulfides, or reduction to

hydrogen sulfide occurred. The

two reduction processes seem to

be unlikely possibilities in view

of the limited rainfall of the pe-

riod, the well-drained character

of the soil profile, and the con-

sequent low probability of the

existence of reducing conditions

during the experiment.

Immobilization appears to be

a more likely possibility. Walker

(1957) has pointed out that sul-

fur may be immobilized through

the activities of microorganisms

in a manner similar to the im-

mobilization

of nitrogen and

phosphorus. Some sulfur might

have been immobilized

during

decomposition of root tissue re-

maining in the soil from the 1957

experiment and from the size-

able clover crop produced in the

second season of the experiment

reported here. The supposition

that substantial immobilization

occurred in the second season is

supported by leaching evidence

of the gypsum treatment. Very

little of the applied sulfate ap-

peared in the percolate in the

second season, although about as

much percolation occurred in the

season as in the first, and about

80 percent of the applied gypsum

remained in the soil at the be-

ginning of the season.

Since (1) only small amounts

of sulfur were removed in the

clover, (2) relatively

small

amounts occurred in the perco-

late, and (3) adsorbed sulfate

sulfur in the soil declined in the

course of the experiment, it was

deduced that substantial immo-

bilization occurred in all treat-

ments. The apparent immobiliza-

tion was greatest for the ele-

mental sulfur treatment, inter-

mediate for the gypsum treat-

ment, and least for the control.

It is evident that sulfur immo-

bilization is a factor requiring

further study, especially as it is

influenced by rate and frequen-

cy of application of various sul-

fur sources under various cli-

matic and soil conditions.

Applicafion of Results

Under excessively

deficient

moisture conditions response of

clover to sulfur fertilization on

a sulfur deficient soil may be nil.

Even so, some sulfate from gyp-

sum may be lost through deep

percolation, as demonstrated by

the results in 1959 in this experi-

ment. However, in contrast to

the great loss of gypsum in a

high rainfall season shown pre-

viously (McKell and Williams,

1960,) loss in percolating water

in a series of dry years was low.

Percolation loss from elemental

sulfur was inconsequential.

Sulfur carried over to the sec-

ond year either as the elemental

form or as sulfate may be util-

ized in the event of more favor-

able moisture conditions for the

enhancement of forage produc-

tion as was illustrated by the

1960 results.

No significant difference

in

yield response to the two forms

of sulfur was detected in this ex-

periment.

LITERATURE CITED

BARDSLEY, C. E., JR. AND H. V. JORDAN. 1957. Sulfur availability in seven southeastern soils as measured by growth and composition of white clover. Agron. Jour. 49: 310-312.

ENSMINGER, L. E. 1954. Some factors

affecting the adsorption of sulfate by Alabama soils. Soil Sci. Sot. Amer. Proc. 18: 259-264.

ENSMINGER, L. E. 1958. Sulfur in relation to soil fertility. Ala. Agr. Exp. Sta. Bull. 312.

(9)

SULFUR FERTILIZATION 5

KAMPRATH, E. J., W. L. NELSON, AND J. W. FITTS. 1956. The effect of pH, sulfate and phosphate concentrations on the adsorption of sulfate by soils. Soil Sci. Sot. Amer. Proc. 20: 463-466.

KAMPRATH, E. J., W. L. NELSON AND J. W. FITTS. 1957. Sulfur removed from soils by field crops. Agron. Jour. 49: 289-293.

LOBB, W. R. AND R. L. BENNETTS. 1957. Improvement of tussock grassland in Canterbury. New Zeal. Jour. Agr. 96: 537-549.

MARTIN, W. E. 1958. Sulfur defici- ency widespread. Calif. Agric. 12

(11) : 10-12.

MCKELL, C. M. AND A. M. WILSON. 1963. Effects of temperature on S35 uptake and translocation by rose and subterranean clovers. Agron. Jour. 55: 134-137.

MCKELL, C. M. AND W. A. WILLIAMS. 1960. A lysimeter study of sulfur fertilization of an annual-range soil. Jour. Range Mangt. 13: 113- 117.

STARKEY, R. L. 1950. Relations of microorganisms to transformations

Part of a lightly grazed native pasture located on the Fort Hays Branch of the Kansas Agricul-

tural Experiment Station at

Hays, Kansas, was burned by

wildfire March 18, 1959. The

heavy accumulation of dormant

vegetation as well as the soil

surface were extremely dry,

and the fire consumed the plant material to ground level. A ra- vine running at right angles to

the wind direction caused the

fire to divide into two fronts, re- sulting in an unburned area between two burned strips on upland with two to three percent

slope. The arrangement of un-

burned areas in relation to the burned afforded an opportunity to make a replicated study of the effects of the fire on herbage yields on a clay upland range

site supporting a mixture of

shortgrasses, buffalograss (Buch-

Zoe dactyloides [Nutt.] Engelm.)

and blue grama (Bouteloua gra-

cilis [H.B.K.] Lag. ex Steud.) with frequent stands of western

wheatgrass (Agropyron smithii

Rydb.) in the shortgrass matrix. The site is part of the shortgrass

Contribution No. 182, Fort Hays Branch, Kansas Agricultural Experi-

ment Station, Hays, Kansas.

of sulfur in soils. Soil Sci. 70: 55- 65.

WALKER, T. W. 1957. The sulfur cycle in grassland soils. Jour. Brit. Grassl. Sot. 12: 10-18.

WALKER, T. W. AND A. F. R. ADAMS. 1958. Competition for sulfur in a grass-clover association. Plant and Soil 9: 353-366.

WALKER, C. F. AND W. A. WILLIAMS. 1963. Responses of annual-type range vegetation to sulfur fertilization. Jour. Range Mangt. 16: 64- 69.

Effects of Early Spring Burning

on Yields of Native Vegetation’

J. L. LAUNCHBAUGH

Pasture Management Specialist, Fort Hays Branch, Kansas Agricultural Experiment Station, Hays, Kansas.

habitat described by Albertson

(1937). The physical characteristics of the area and vegetative

composition under light grazing

are discussed in a previous re-

port (Launchbaugh, 1957).

Burned and unburned vegetation were protected from grazing

and yield measurements were

made in late summer each year during 1959, 1960, and 1961 us-

ing ten 3.1 square-foot, clipped

BUFFALOGRASS-BLUE GRAMA MIXTURE

1960 1961-

BUFFALOGRASS WESTERN AND BLUE GRAMA WHEATGRASS

subsamples per treatment in

each of the two grass mixtures. The central unburned strip and nearby burned area made up one

replication. The nearest un-

burned margin and adjacent

burned area were considered an-

other replication. Yields were

measured separately in the

shortgrass alone and in the west-

ern wheatgrass-shortgrass mix-

ture. Weeds were separated from

the subsample clippings and

composition estimates were made

of the remaining plot material

into categories of buffalograss

and blue grama combined, western wheatgrass, and old growth.

The harvested material was

oven-dried at 170 O F. for 72 hours

prior to being weighed. Plant

WESTERN WHEATGRASS-SHORTGRASS MIXTURE

WEEDS

EJ

OLO GROWTH

FIGURE 1. Vegetation yields in two native grass mixtures on a clay upland range site dur- ing three growing seasons following a March 18, 1959, wildfire. U-unburned;

(10)

SULFUR FERTILIZATION 5

KAMPRATH, E. J., W. L. NELSON, AND J. W. FITTS. 1956. The effect of pH, sulfate and phosphate concentrations on the adsorption of sulfate by soils. Soil Sci. Sot. Amer. Proc. 20: 463-466.

KAMPRATH, E. J., W. L. NELSON AND J. W. FITTS. 1957. Sulfur removed from soils by field crops. Agron. Jour. 49: 289-293.

LOBB, W. R. AND R. L. BENNETTS. 1957. Improvement of tussock grassland in Canterbury. New Zeal. Jour. Agr. 96: 537-549.

MARTIN, W. E. 1958. Sulfur defici- ency widespread. Calif. Agric. 12

(11) : 10-12.

MCKELL, C. M. AND A. M. WILSON. 1963. Effects of temperature on S35 uptake and translocation by rose and subterranean clovers. Agron. Jour. 55: 134-137.

MCKELL, C. M. AND W. A. WILLIAMS. 1960. A lysimeter study of sulfur fertilization of an annual-range soil. Jour. Range Mangt. 13: 113- 117.

STARKEY, R. L. 1950. Relations of microorganisms to transformations

Part of a lightly grazed native pasture located on the Fort Hays Branch of the Kansas Agricul-

tural Experiment Station at

Hays, Kansas, was burned by

wildfire March 18, 1959. The

heavy accumulation of dormant

vegetation as well as the soil

surface were extremely dry,

and the fire consumed the plant material to ground level. A ra- vine running at right angles to

the wind direction caused the

fire to divide into two fronts, re- sulting in an unburned area between two burned strips on upland with two to three percent

slope. The arrangement of un-

burned areas in relation to the burned afforded an opportunity to make a replicated study of the effects of the fire on herbage yields on a clay upland range

site supporting a mixture of

shortgrasses, buffalograss (Buch-

Zoe dactyloides [Nutt.] Engelm.)

and blue grama (Bouteloua gra-

cilis [H.B.K.] Lag. ex Steud.) with frequent stands of western

wheatgrass (Agropyron smithii

Rydb.) in the shortgrass matrix. The site is part of the shortgrass

Contribution No. 182, Fort Hays Branch, Kansas Agricultural Experi-

ment Station, Hays, Kansas.

of sulfur in soils. Soil Sci. 70: 55- 65.

WALKER, T. W. 1957. The sulfur cycle in grassland soils. Jour. Brit. Grassl. Sot. 12: 10-18.

WALKER, T. W. AND A. F. R. ADAMS. 1958. Competition for sulfur in a grass-clover association. Plant and Soil 9: 353-366.

WALKER, C. F. AND W. A. WILLIAMS. 1963. Responses of annual-type range vegetation to sulfur fertilization. Jour. Range Mangt. 16: 64- 69.

Effects of Early Spring Burning

on Yields of Native Vegetation’

J. L. LAUNCHBAUGH

Pasture Management Specialist, Fort Hays Branch, Kansas Agricultural Experiment Station, Hays, Kansas.

habitat described by Albertson

(1937). The physical characteristics of the area and vegetative

composition under light grazing

are discussed in a previous re-

port (Launchbaugh, 1957).

Burned and unburned vegetation were protected from grazing

and yield measurements were

made in late summer each year during 1959, 1960, and 1961 us-

ing ten 3.1 square-foot, clipped

BUFFALOGRASS-BLUE GRAMA MIXTURE

1960 1961-

BUFFALOGRASS WESTERN AND BLUE GRAMA WHEATGRASS

subsamples per treatment in

each of the two grass mixtures. The central unburned strip and nearby burned area made up one

replication. The nearest un-

burned margin and adjacent

burned area were considered an-

other replication. Yields were

measured separately in the

shortgrass alone and in the west-

ern wheatgrass-shortgrass mix-

ture. Weeds were separated from

the subsample clippings and

composition estimates were made

of the remaining plot material

into categories of buffalograss

and blue grama combined, western wheatgrass, and old growth.

The harvested material was

oven-dried at 170 O F. for 72 hours

prior to being weighed. Plant

WESTERN WHEATGRASS-SHORTGRASS MIXTURE

WEEDS

EJ

OLO GROWTH

FIGURE 1. Vegetation yields in two native grass mixtures on a clay upland range site dur- ing three growing seasons following a March 18, 1959, wildfire. U-unburned;

(11)

6 LAUNCHBAUGH

Table 1. Average heights of important plant species af fhe end of the first growing season on early-spring burned and unburned plots in shorf- grass and western wheatgrass-shorfgrass mixtures combined.

Treatment Species

Western wheatgrass Blue grama

Buffalograss Western ragweed Horseweed -

Burned 3-18-59 Unburned _---- (Inches) - - - - -

11.0 17.0

6.7 11.9

4.1 6.9

6.1 9.7

10.3 14.2

heights and number of weeds

were recorded with the first-sea-

son sampling. Subsample plots

were relocated each year.

Figure 1 shows mean yields

from burned and unburned plots.

The 1959 grass yields in the

buffalograss-blue grama mixture

were reduced 65 percent by the

fire. First-season yields of west-

ern wheatgrass and shortgrass

were reduced 82 and 48 percent,

respectively. Second-season grass

yields were 39 percent below the

check plot yields in the short-

grass alone and 73 and 19 percent

lower respectively for western

wheatgrass and shortgrass

in

combination. Third-season herb-

age production

on the previ-

ously burned areas was not sig-

nificantly different from that on

unburned sites.

Weeds consisted primarily of

western

ragweed

(Ambrosia pdostachya

DC.) on both the

burned and unburned areas, and

yields were relatively low each

year regardless

of treatment.

Lowest yields, however, were on

the burned areas. First-season

weed numbers averaged 9.9 west-

ern ragweed and 0.7 horseweed

(Conyxa canadensis

[L.] Cron.)

per square foot in the burned

plots compared with 8.5 western

ragweed and 1.9 horseweed per

square foot in the unburned

plots. The differences in weed

numbers between treatments

were not significant. Old growth

increased annually on the burned

areas, but did not equal amounts

present on the unburned areas

by the end of the third growing

season.

The reductions in vegetation

yields during the first and second

growing seasons following the

fire were significant at the one

percent level and apparently

were caused by two main effects

of burning. First, there were ob-

viously fewer first-year grass

tillers compared with numbers

of dead remnants,

indicating

partial killing of plants. Second,

plant heights recorded at the

close of the first growing season

(Table 1) indicated stunting of

growth in the burned

areas,

which might be attributed to

lower vigor of the perennials

brought about directly by fire

damage-or to lower amounts of

moisture entering the soil in the

burned areas.

Previous studies by Hopkins

et

al.

(1948) on the effects of

late fall and early spring burn-

ing on a similar upland short-

grass site showed that where

litter accumulations were heavy

at the time of burning, ground

cover of living vegetation and

subsequent yields were greatly

reduced. Also western ragweed

plant numbers and yields were

increased considerably by early

spring burning. Such weed in-

creases were not observed in the

present study.

That horseweed,

an annual,

was reduced in stature on the

burned areas suggests that soil

moisture conditions were not so

favorable in the burned as in the

unburned plots. Several studies

have shown that a cover of vege-

tation both living and dead en-

hanced moisture intake in grass-

land soils, and removal of the

cover reduced soil moisture in-

filtration (Duley and Domingo,

1949; Hopkins, 1954; and Rauzi,

1960). Thus when yields of pe-

rennial vegetation are reduced

by burning, it appears that re-

covery is brought about jointly

by revegetation

to reclaim

ground cover losses and accumu-

lation of a protective cover to

increase soil moisture intake.

LITERATURE CITED ALBERTSON, F. W. 1937. Ecology of

mixed prairie in west-central Kansas. Ecol. Monog. 7: 481-547. DTJLEY, F. L. AND C. E. DOMINGO.

1949. Effect of grass on intake of water. Nebr. Agr. Exp. Sta. Res. Bul. 159.

HOPKINS, HAROLD H. 1954. Effects of mulch upon certain factors of the grassland environment. Jour. Range Mangt. 7: 255-258.

HOPKINS, HAROLD, F. W. ALBERTSON, AND ANDREW RIEGEL. 1948. Some effects of burning upon a prairie in west-central Kansas. Trans. Kan. Acad. Sci. 51: 131-141. LAUNCHBAUGH, J. L. 1957. The effect

of stocking rate on cattle gains and on native shortgrass vegetation in west-central Kansas. Kan. Agr. Exp. Sta. Bul. 394.

(12)

Development and Use of the Esophageal Fistula:

A Review

GEORGE M. VAN DYNE AND D. T. TORELLl Research Nutritionist and Specialist, Animal Husbandry Department, University of California, Davis, California

One of the foremost problems in range and pasture nutrition is making an accurate assessment

of the chemical and botanical

composition of the diet of graz-

ing livestock. In recent years the esophageal fistula has been used to obtain samples of the forage

grazed by ruminants. These

studies have been conducted

under a wide variety of conditions in at least six countries on four continents.

The purpose of this article

is

to review the development and

use of the esophageal fistula and to discuss present problems and practices. The medical literature is replete with references to hu- man and animal esophageal fis-

tulas due to congenital abnor-

malities, disease, and accidental

injury. However, this article is

restricted to experimental stud-

ies with domestic animals and

reviews the literature through

mid-1963.

Historical Development

The esophageal fistula tech-

nique is not new. It was reported

early by the famous French

physiologist, Claude Bernard

(1855). In fact, it was used by

his teacher, Magendie, several

years earlier in the horse (Ma- gendie and Ryer, 1847). Pavlov’s classic studies involving esopha-

geal-fistulated dogs were initi-

ated in 1889 (Pavlov, 1897).

Some of his dogs lived for many years on food fed directly into the stomach. However, it is only in the last decade that the tech-

1The following are acknowledged for their review or suggestions: Drs. G. P. Lojgreen, J. H. Meyer, and W. C. Weir (California), C. W. Cook (Utah), and V. R. Bohman (Nevada).

nique has been used widely in

ruminants. A review of some

early uses of the esophageal fistula in various species is out- lined in Table 1. Most studies

have been physiological or psy-

chological rather than nutri-

tional in nature.

Surgical Techniques

At best, surgery of the esophagus is difficult; this accounts for numerous losses and for skepti-

cism regarding the technique.

Saint (1929) pointed out the fol-

lowing as major reasons why

surgery of the esophagus is difficult: (1) the esophagus lacks a serosa, (2) in order to expose the esophagus in surgery it is necessary to open the mediastinal structures and fascial planes, (3) there is a very poor blood supply to the esophagus, (4) the esopha-

gus cannot be restricted from

movement, (5) there is an ab-

sence of the greater omentum.

Further details concerning

esophageal surgery are given by Saint and Mann (1929).

Early physiologists often con-

sidered it necessary to accom-

plish fistulation in a two-step

operation because of these diffi- culties (Markowitz, 1954). Drag-

stedt and Mullenix (1931) re-

ported over 50 percent mortality in making a one-stage fistula of the esophagus in dogs. In the two-stage operation the esopha-

gus is first exteriorized in the

neck; later, the esophagus may be sectioned without danger of

mediastinitis. A two-stage pro-

cedure recently has been used in ruminants in establishing saliva

collection cannulae (Whitmore

et al., 1963 unpublished manu-

script).

Descriptions of surgical tech-

niques for large animals have been published by Tore11 (1954)) Cook et al. (1958)) Hamilton et

al. (1960)) McManus (1960)2,

McManus et al. (1962b), Chap-

man and Hamilton (1962) and

Cook et al. (1963). The following

is a brief description of an un-

2McManus, W. R. 1960. The develop- ment and use of oesophageal jis- tulae in sheep. Ph.D. Diss., U. New South Wales (Australia) 338 pp. Table 1. Early uses of the esophageal fisfula in domestic animals.

Worker (s) Year SDecies

Bernard 1855

Colin 1856

Lobasov 1896

Pavlov 1897

Best and Cohnheim 1910

Nikovlina 1919

Karpov 1919

Collip 1922

Haberland 1926

Komarov 1926

Dragstedt and Mullenix 1931

Wilder and Stokes 1931

Bellows and Van Wagenen 1938

Heyenga 1938

Adolph 1939

Bellows 1939

Goldman 1939

Wise et al. 1940

Janowitz and Grossman 1949

Tore11 1954

Mook 1962

(13)

8 VAN DYNE AND TORELL

FIGURE 1. An animal positioned and restrained for esophageal fistulation. The stipuled area is clipped and disinfected. A steel rod with a hard rubber ball on the end

is passed down the esophagus and used to position the esophsgus in surgx-y.

published technique used suc-

cessfully in both cattle and

sheep3. The animal is adjusted to pelleted feed or green herbage for several weeks prior to surgery, but feed and water are withheld for 24 hours immediately prior to surgery. After the animal is anesthesized “to effect” by means of a general anesthetic, it is placed in a right lateral re- cumbancy and the rear legs are

extended and tied (Figure 1).

The forelegs are doubled back

under the body and tied by

means of a butterfly (or “tom-

fool”) knot. The surgical area is

disinfected and clipped. The

head is slightly elevated and

held by means of a halter rope. A steel rod with a hard rubber ball on the end is passed down the esophagus and manipulated to aid in making the incision and blunt dissection through the tissues (Figure 1). After the rod is

passed, the supporting block

(Figure 1) is moved back under the neck so that fluids will drain from the nose and mouth. The fistula should be located as near as possible to the ventral midline of the neck and about midway between the jaw and the brisket. After removal of an oval-shaped piece of skin (the size of the fistula), the tissues and muscles are separated and the esophagus is exposed. A short longitudinal incision is made in the esophagus and the sides of the incision are

sutured. Sutures pass through

the esophagus, submucosa, and

inner layers of the skin. Incision

3A detailed outline of surgical pro-

cedure and care of animals is avail- able on request.

and suturing is continued until the entire perimeter of the fistula is sutured. The cannula or plug is then inserted and the area disinfected and treated with fly repellant. The animal is kept off coarse feed and water for 24

hours postoperative, or is re-

turned to grass. Antibiotics are

given until the incision is healed. The sutures may be removed in seven to ten days. An experi- enced surgeon and two assistants

can complete a fistula

in a steer

in about one hour.

McManus (1962b) suggested

putting animals on green grass as soon as possible after surgery. Cook et al. (1958) recommended

continuation of pellet feeding.

However, care must be taken to prevent fistulated animals being

fed pellets from consuming

straw bedding or wood shavings which may become lodged and

compacted in the esophagus

(Goldman, 1939; Van Dyne,

1960). Tribe and Peel (1963)

allow lambs to resume grazing

and nursing immediately after

surgery.

FisNfulafion Success

Success with large animals

varies widely. In early develop- mental stages losses were great. Tore11 (1954) first reported success in only one out of four sheep. Later, he established an esophageal fistula in a heifer but she died of a digestive disorder before use. Cook et al. (1958) re-

ported on fistulation of four

sheep, all of which survived surgery. However, one lost its cannula due to necrosis within two

months after installation. Four

other sheep were fistulated but

did not survive for various rea-

sons. Lesperance (1959) 4 fistu-

lated four steers, but none of

these animals survived a year

and some survived only a few weeks. McManus (1962a, 1962b) reported on esophageal fistulation of 35 sheep; he found 14 percent suitable for field studies. Some sheep were suitable for

pen studies although not for

field use. More than half of the fistulated sheep were not suit-

able for either pen or field

studies. Nelson (1962) estab-

lished stainless steel cannulae in four steers, but all were unsatis- factory over any extended period. He lists the following com- plications: 1) loss of cannula due to pressure necrosis; 2) recur- rent lack of appetite; and 3) ul- ceration of the rumen and retic- ulum.

Recently, greater success due

to development of more efficient

closure devices has been re-

ported. Van Dyne (1962) used five steers and seven sheep fis-

tulated by the techniques de-

scribed above. All the steers survived surgery and field use; all sheep survived surgery, but two were killed after several months

use. Tore11 and Bredon (1961)

established fistulas in 18 Ankole and Zebu cattle which survived a six-month study in good condi- tion and at last report were still

usable. Cook et al. (1961, 1962)

have reported on use of fistulated sheep under range condi-

tions. Several of their sheep

have been used in more than one season with some now collecting in their fifth year. Wethers and calves have been fistulated suc-

cessfully at the Grassland Re-

search Institute in England

(Lambourne, 1963, and personal

correspondence), lambs and

wethers in Australian and New Zealand studies (McManus et al. 1962b; Tribe and Peel, 1963; Ar-

-

(14)

THE ESOPHAGEAL FISTULA

FIGUR’E 2. Examples of long term esophageal fistulated animals: The esophageal fistula was installed in the steer over two and a half years prior to the photograph and the rumen fistula for about two years; the ewe was fistulated in 1960 and has since been used in grazing studies.

nold et al. 1963; and W. H.

Bishop, personal correspon-

dence), and dairy cattle in the southern United States (G. H. Rollins and L. L. Rusoff, personal correspondence).

Greater surgical success has

been obtained with young animals than with mature animals and fistulation is probably more successful in cattle than in, sheep. Still, probably about ten percent

of the animals fistulated by current techniques will not be use- ful over a long period of time,

due to various operative and

post-operative losses. Yet, some

animals are serviceable for sev-

eral years. Heady and Tore11

(1959) reported on an esophageal-fistulated wether which had been in use four years at the time of their study. Rusoff and

Foote (1961b) reported using

9

esophageal-fistulated dairy cows

for two years. The bifistulated steer in Figure 2 has both an esophageal and a ruminal fistula. The esophageal fistula had been

established about 2% years at

the time of the photograph, and the ruminal fistula for about two

years. Fistulated ewes which

raised lambs have been used

(Van Dyne, unpublished data;

Arnold et al., 1963).

Types of Closure Devices

Various devices have been

used for closing the esophageal fistula. Tore11 (1954) described the use of two stainless steel pins

inserted into imbedded poly-

ethylene tubing. The exposed

end of the pins was held together by a cord or rubber bands. More recent closure devices are sche- matically illustrated in Figure 3. Type A represents a device used by Tore11 (1954) and by Van

Dyne (unpublished). Two plas-

tic plates are drawn together by nylon cord. Usually, a piece of cork or foam rubber is placed between the two plates of this completely removable plug. Les- perance (1959) 4 illustrates a sim-

ilar device wherein the inner

plate was wired to a metal outer plate. The outer plate was held in place by a strap around the animal’s neck.

The nonremovable type of can-

nula (Figure 3B) has been used

by various workers; it is con-

structed of lucite or acrylic plastic or stainless steel (Cook et al., 1958; Van Dyne and Van Horn,

1959; Lesperance et al., 1959,

1960a; Rusoff and Foote, 1961a, b; and others). The cannula in Fig- ure 3D (Van Dyne, unpublished) has three important advantages over that in Figure 3B: 1) less esophagus need be incised to in- stall the cannula; 2) it can be removed completely if necessary; and 3) when the plug is in place, the plate of the cannula does not have a hole in which forage may become lodged.

(15)

10 has been in use for several years

under a wide variety of condi-

tions (Van Dyne, 1962; Tore11

and Bredon, 1961; and Tribe and

Peel, 1963). Various size plugs

can be interchanged to accom-

modate changes in the fistula. In

use, both closure devices C and

D are made so that the plug

portion is “off center.” This

allows periodic switching of the

long and short ends of the plate

and aids in mainteance of a

healthy fistula. Often there is a

tendency for a “pouch” to pull

down anterior to the fistula;

periodic switching prevents this.

The molded latex plug in Fig-

ure 3E is one of several types de-

scribed by McManus

et al.

(1962b). They also describe split

plug stoppers made from surgi-

cal rubber. One disadvantage of

this type plug under range con-

ditions is that it pulls out rela-

tively easily when caught in

fences, brush, or in the animal’s

rear hooves when scratching.

Nelson (1962) has used two “L-

shaped” pieces of plastic held in

place by two bolts for a closure

device (Figure 3F). Spacers can

be used in such a device to

adjust its length. Wise et al.

(1940) described a double-open-

ing fistula in a dairy calf which

was fitted with a rubber tube

inserted into the exposed ends of

the esophagus to serve as a con-

duit for normal milk feeding.

The type of closure device is not

well described in many early ex-

periments. Many of the early in-

vestigations were probably acute

studies.

A removable cannula has some

advantages over a permanently

fixed cannula. The permanently

fixed lucite or stainless steel can-

nula eventually may cause the

development of a pocket or blind

pouch anterior to the fistula and

eventually may be expelled. All

the plugs or cannulae in Figure

3 can be removed, interchanged,

and modified except B. The lu-

cite cannula has some advantage

in cold-weather sampling be-

VAN DYNE AND TORELL

cause it can be used while the

operator is wearing gloves.

Inside openings in cannulae

are usually about three cm in

diameter in sheep and four cm

or more in cattle. Openings

smaller than this may permit

plugging of the cannula

and

compaction of the feed within

the esophagus, or may limit the

percent of forage collected. The

size of the opening is important

because the size of bolus varies

with the type of feed eaten (Bai-

ley, 1961).

Removable

cannulae

are

placed in Bistulae which may

vary considerably in length and

width. In the author’s studies

the cattle fistulae are oval and

four to five cm long and about

three cm wide. The sheep fistu-

lae are correspondingly smaller.

It is desirable to establish a uni-

form size and shape of cannula

or plug which may be inter-

changed among animals, thus

eliminating the necessity

of

maintaining individual animal

equipment.

Collecfion Apparatus

Sample collection apparatus

include plastic bags (McManus,

B

II

FIGURE 3. Schematic illustration of various types geal fistulae in cattle and sheep (see text).

(16)

THE ESOPHAGEAL FISTULA

bag used on cattle (see also Figure 2 for sheep forage collection bag).

1962a; Lusk et al., 1961)) canvas bags (Torell, 1954; Cook et al., 1958; Cook et al., 1961)) rubber-

ized canvas (Lesperance, 1959) 4

and screen bottom bags (Van Dyne and Van Horn, 1959). The plastic bag without canvas pro- tection is not suitable for most

range investigations. The water-

proof canvas bag can be a disadvantage under range conditions because a considerable weight of saliva and forage may accumu- late and thus affect the grazing

performance of the animal. The

screen bottom bag allows saliva to drip off the sample (see Fig- ure 4D). Thus, samples collected with screen bottom bags are less

affected by salivary contamina-

tion than samples collected in

plastic bags. This would be dis-

advantageous if nutrients were

leached from the sample by saliva. However, McManus (1961b)

has demonstrated by in vitro

studies there is no significant leaching of nitrogen from suc-

culent or roughage plant ma-

terial.

The type of collection bag in

Figures 4D and 2B can be rapidly attached to either cattle or sheep. Two adjustable straps with “D” rings are snapped over the neck. A small snap on the front of the bag attaches to the back of the halter. A strap on the back of

11

the bag passes along the brisket

and between the forelegs,

through a “D” ring in a sur- tingle, and returns to snap to the bag on the other side. These bags are not displaced whether

full or empty or whether the

animal is browsing on high

shrubs or grazing on low grass. A wire screen bottom is prefer- able to a nylon screen one because it holds the bag open and does not rip as easily in brush.

Similar sheep collection bags

may not require the straps at the front and back of the bag because the wool prevents slippage

(see Cook et al., 1961) .

Length of Sampling

Sample volumes from one pint to one quart for sheep and half to one gallon for cattle are col-

lected easily. McManus (1960) 2

reported 86 per cent of his sheep samples were s 20 g dry matter. Collection time depends upon the species, size of the fistula, rate of grazing, and type of forage. Bath et al. (1956) suggested a 30-min-

ute maximum for collecting

samples on irrigated pasture.

Cook et al. (1958) reported two to four hours were required to collect an adequate sample under winter desert range forage

conditions. Two-hour collection

periods either in morning or

evening grazing were adequate for sampling open foothill winter range (Van Dyne and Van Horn,

1959). Van Dyne (1963)) in

studying grazing of esophageal- fistulated steers and wethers on a common dry annual foothill

range, used collection times of

1.4 to 2.4 hours. Sampling dura- tion as short as ten to 15 minutes

has been reported in forage

studies with esophageal-fistu-

lated sheep on small fenced

plots (Lusk et al., 1961) and

with steers fed roughages or con-

centrates (Nelson, 1962) . Collec-

tion periods up to one-hour dura- tion resulted in 30-200 g samples from sheep (Arnold et al., 1963).

(17)

12 VAN DYNE AND TORELL

his data that probably collection time should not exceed one hour in order to maintain normal conditions in the rumen. Extended loss of saliva may cause changes

in digestive activity by allow-

ing the accumulation of rela-

tively large concentrations of

volatile fatty acids in the rumen.

If this accumulation leads to ab-

normally low pH levels, ruminal stasis could result. Whether this

occurs under extended range

sampling has not been demonstrated.

Chemical Analyses

Forage samples taken from

esophageal-fistulated animals

have been subjected to a wide

variety of analyses. Various

workers have analyzed samples for the proximate components-

crude protein, ether extract,

total ash, crude fiber, and nitro-

gen free extract (by difference).

Other commonly determined

constituents are lignin, cellulose,

other carbohydrates (by differ-

ence) , silica, phosphorus, cal-

cium, and gross energy. Potas-

sium (Rusoff and Foote, 1961b)

and plant chromogens (Van

Dyne, unpublished, 1959) also

have been determined. Renet

coagulation time. surface and

body curd tension, pH, and lipo- lytic activity have been determined in milk samples from a

sham-fed esophageal-fistulated

calf (Wise et al., 1940) . No spe-

cial changes in laboratory pro-

cedures have been reported for analyzing esophageal fistula forage samples. However, salivary

contamination has been consid-

ered in some investigations (see

next section).

MacDougall and DeLong

(1942) and ‘?Jan Soest (1962)

have reported on effect of heat drying on lignin content in for- ages and in cattle and sheep

feces. This heat drying effect

may be important in drying fis-

tula forage samples. Bohman

(1958) found hay samples collected by the rumen evacuation

technique had greater lignin

content than the hay had before feeding. However, Sharp (1962) 5 found crude fiber in the forage sample to be greater than crude

fiber in the rumen-collected

sample. Ensalivation of the

s amp 1 e and high-temperature drying may bias lignin values.

MacDougall’s and DeLong’s

(1942) and Van Soest’s work (1962) show water and high drying temperature cause a non- enzymatic browning reaction in

which carbohydrate degradation

products condense with protein. This dark colored polymeric material is insoluble in 72 percent H2SG4 and thus would cause a positive bias in the lignin deter- mination. Recent work has indicated this material may be modi-

fied xylans (Gaillard, 1962).

Therefore, it may be desirable to keep the forage sample as dry as possible by use of screen bottom

collection bags and to dry the

samples at relatively low tem-

peratures. More work is needed

to establish the importance of

drying conditions on lignin de-

terminations in fistula forage

samples.

Salivary Contamination

Fistula forage samples contain

differential amounts of salivary

contaminants, depending upon

the type of collection bag. Bath

et al. (1956) indicated salivary

contamination increased the ash

content of the sample while not

appreciably affecting other con-

stituents. McDougall (1948) re-

ported sheep saliva contains

about 0.8 percent ash, Lesper- ante et al. (1960a) found bovine

saliva to contain 0.85 percent

ash, and Bailey and Balch (1961) found bovine saliva dry matter

and ash to be respectively 1.02

percent and 0.89 percent. The

calculations of Lesperance et al. (1960a) indicate that ash, phosphorus, or calcium contamination from saliva would increase significantly the content of those

components in the forage sam-

ple. They considered it doubtful that regression equations could be established to relate salivary contamination and feed composi-

tion when comparing samples

from stall-fed animals and samples from animals grazing on pasture. Cook et al. (1961,1962), however, have attempted to cor- rect fistula forage sample composition for added nitrogen, ash,

and phosphorus from salivary

contamination. This procedure

is as follows (C. Wayne Cook,

1963, personal correspondence) :

simulate grazing by hand collecting samples, determine moisture in these; collect saliva sam-

ples and analyze for moisture

and chemical constituents; sub-

tract plant moisture from the total moisture in the fistula sample to determine amount of sal-

ivary contamination; deduct

amount of various constituents added by the saliva. However, this procedure is subject to crit-

icism because of the assump-

tions: 1) that forage grazed can be sampled adequately by hand;

2) that saliva composition or

secretion rate is invariable; and 3) that the fistula samples are completely saturated with either

plant or salivary moisture, or

both (screen bottom collection bags are used).

McManus (1961b) reported on collection of roughage feedstuffs from small fistulas and found recoveries generally exceeded 35 percent of the material fed. He stated that chemical composition

of the extruded material was

similar to that fed except for an increase in ash content. He suggested that results always should

be expressed on an ash-free

basis and similar results were obtained by Nelson (1962). Be- cause trampled forage or seeds

picked from the ground may

BSharp, G. D. 1962. An evaluation of the rumen klearance technique for measuring the nutritive value

.