Changes in smutgrass (Sporobolus poiretii [Roem. and Schult.] Hitchc.) ground cover induced by spraying with molasses an...

MISSING IMAGE

Material Information

Title:
Changes in smutgrass (Sporobolus poiretii Roem. and Schult. Hitchc.) ground cover induced by spraying with molasses and grazing management
Physical Description:
viii, 106 leaves : ill. ; 28 cm.
Language:
English
Creator:
Valle, Leonidas S., 1938-
Publication Date:

Subjects

Subjects / Keywords:
Grazing   ( lcsh )
Pastures   ( lcsh )
Smutgrass
Grazing   ( fast )
Pastures   ( fast )
Agronomy thesis Ph. D
Dissertations, Academic -- Agronomy -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 98-105).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Leonidas S. Valle.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
aleph - 020569566
oclc - 03403502
System ID:
AA00026361:00001


This item is only available as the following downloads:


Full Text














CHANGES IN SMUTGRASS (Sporobolus poiretii [Roem. and Schult.] Hitchc.) GROUND COVER
INDUCED BY SPRAYING WITH MOLASSES AND GRAZING MANAGEMENT















By

LEONIDAS S. VALLE















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA 1977














ACKNOWLEDGMENTS


The author wishes to express his sincere appreciation to Dr, G. 0. Mott, Chairman of the Supervisory Committee, for his technical assistance in all phases of his graduate study including this research and the preparation of this manuscript. Sincere gratitude is extended to Dr. W. R. Ocumpaugh for his supervision and assistance during and after collection of data in the field. Appreciation is also extended to other members of the Supervisory Committee, Drs. J. H. Moore, J. H. Conrad, W. G. Blue, and W. L. Currey, for their valuable assistance during the preparation of this dissertation. The author also wishes to acknowledge Dr. R. C. Littel for his assistance in the statistical analysis.

Sincere gratitude is extended to the Empresa Brasileira de Pesquisi Agropecuaria (EMBRAPA), for providing financial assistance.

The author is greatly indebted to Mr. Fred McKay for his help during the field work. Thanks are due to the personnel of the University of Florida Beef Research Unit for providing research facilities. The author would like to thank Mrs. Maria I. Cruz for typing the preliminary drafts and the final copy.

The author is also grateful to the staff and fellow graduate students of the Agronomy Department for their encouragement and friendship.






ii










The author's children, Rodolfo, Rafael, and Romolo, must be thanked for suffering their father's neglect throughout the study period. However, the greatest debt of gratitude is due to the author' wife, Julia, for her love and devotion, and for her sacrifice during the past three and a half years of often frustrating labor.














































ili











TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS............................................... ii

LIST OF TABLES. ..................... ........... .............. vi

LIST OF FIGURES................................................. v iii

ABSTRACT ........................................................ ix

INTRODUCT ION .................................................... .

REVIEW OF LITERATURE. .............................. 3
Smutgrass.................................... ............... 3
Smutgress Control ............................................ 6
Weed Control by Grazing Management............................ 9
Pasture Loss Due to Weeds.................................... 12
Benefits from Weed Control .... .............................. 14
Measuring Botanical Composition............................... 18
Estimating Forage Yield...................................... 19
Spraying Pasture with Molasses ......... ...................... 24
Effect of Grazing Management to Botanical Composition........ 20

MATERIALS AND METHODS .......................................... 33
General Description.......................................... 33
Experimental Pasture ......................................... 34
Treatments................................... ................. 35
Grazing Management Factors................................... 36
Experimental Design.......................................... 37
Construction of Physical Facilities.......................... 40
Experimental Analysis................................ 40
Treatment Application........................................ 42
Main Treatment 1. Control................................. 42
Main Treatment 2. Molasses Sprayed on the Pastures ....... 42
Main Treatment 3. Dalapon Applied in the Spring Followed
by Mowing and N Fertilization............................ 43
Dalapon Applied in the Spring Followed by Burning and N
Fertilization ..... .............. ..... .................... 43
Dalapon Applied in the Fall Followed by Burning and N
Fert ilization ............................................ 43
Dalapon Applied in the Fall Followed by Mowing and N
Fertil izat ion ............................................ 43
Ground Hawg in the Spring.................................. 44
Ground Hawg in the Spring Followed by N Fertilization..... 44
Ground Hawg in the Spring Followed by Seeding of Bahiagrass and N Fertilization................................. 44
Ground Hawg in the Fall Followed by Seeding of Ryegrass
and N Fertilization ......................................... 44
Burning in the Fall and Dalapon Applied in the Spring
Followed by Mowing and N Fertilization................... 45


iv










Page

Measurements................................................ 45
Smutgrass Ground Cover..................................... 45
Dry Matter Determination After Grazing..................... 48
Parameters Measured........................................... 51

RESULTS AND DISCUSSION............. .............................. 54
Residual Dry Matter........................................... 54
Control Treatment.......................................... 54
Molasses Sprayed Treatment. ......... ........... ....... .. 57
Changes in Smutgrass Ground Cover............................ 59
Control Treatment.......................................... 59
Molasses Sprayed Treatment................................. 67
Animals/ha/day................................................. 73
Control Treatment......................................... 73
Molasses Sprayed Treatment................................ 78
Liveweight/ha/day.............................................. 81
Control Treatment.......................................... 86
Molasses Sprayed Treatment................................. 86

SUMMARY AND CONCLUSIONS ............................................ 94

LITERATURE CITED................................................. 98

BIOGRAPHICAL SKETCH. ................................. 106




























v










LIST OF TABLES

Table Pao"

1 Monthly mean maximum and minimum temperatures and
rainfall at the Beef Research Unit for 1976............. 36

2 Levels of length of rotation cycle and grazing
pressure................................................ 37

3 Combinations of rotation cycle and grazing pressure
of the Central Composite Design......................... 39

4 Residual dry matter left after grazing for different
combinations of length of rotation cycle and grazing
pressure on the control treatment....................... 55

5 Residual dry matter left after grazing for different
combinations of length of rotation cycle and grazing
pressure on the molasses sprayed treatment.............. 58

6 Observed change in smutgrass ground cover for the
different combinations of length of rotation cycle
and grazing pressure on the control treatment, from
April to October 1976.................................... 60

7 Main effect of grazing pressure on the change in
smutgrass ground cover on the control treatment ........ 61

8 Main effect of rotation cycle on the change in
smutgrass ground cover on control treatment ........... 61

9 Fitted equations for treatment and response
variables............................................. 63

10 Values of the stationary point for treatment
and response variables. .............................. 64

11 Observed change in smutgrass ground cover for
the different combinations of rotation cycle and
grazing pressure on the molasses sprayed treatment,
from April to October 1976............................ 68

12 Main effect of grazing pressure on the change in
smutgrass ground cover on the control treatment ...... 69

13 Main effect of rotation cycle on the change in
smutgrass ground cover on sprayed molasses treatment. 69




vi










Table Page

14 Observed animals per hectare per day for the different
combinations of length of rotation cycle and grazing
pressure on the control treatment...................... 74

15 Main effect of grazing pressure on animals per
hectare per day on the control treatment............... 75

16 Main effect of rotation cycle on animals per
hectare per day on the control treatment .............. 75

17 Observed animals per hectare per day for the
different combinations of rotation cycle and grazing
pressure on the sprayed molasses treatment............ 79

18 Main effect of grazing pressure on animals per
-hectare per day on the sprayed molasses treatment..... 80

19 Main effect of rotation cycle on animals per
hectare per day on the sprayed molasses treatment..... 80

20 Observed liveweight per hectare per day for
the different combinations of rotation cycle and
grazing pressure on the control treatment............. 84

21 Main effect of grazing pressure on liveweight
per hectare per day on the control treatment .......... 85

22 Main effect of rotation cycle on liveweight
per hectare per day on the control treatmemt .......... 85

23 Observed liveweight per hectare per day for the different combinations of rotation cycle and
grazing pressure on the sprayed molasses treatment.... 89

24 Main effect of grazing pressure on liveweight per hectare per day on the sprayed molasses
treatment................. ........................... 90

25 Main effect of rotation cycle on liveweight per
hectare per day on the sprayed molasses treatment..... 90












vii












LIST OF FIGURES

Figure Page

1 Response surface design in two factors 13 combinat ions ................................................ 38

2 Field layout of the experimental pastures ............ 41

3 The I m2 frame used to determine the smutgrass ground cover......................................... 46

4 Outlines of smutgrass clumps on acetate transparent sheets................................................ 47

5 -Black circular areas represent the outlines of smutgrass clumps after having been inked ............. 49

6 View of the capacitance meter used in the doublesampling procedure................................... 50

7 Contours of change in smutgrass ground cover (percentage units) on the control treatment........... 66

8 Contours of change in smutgrass ground cover (percentage units) on the molasses sprayed treatment.. 72

9 Contours of animals per hectare per day on the control treatment.................................... 77

10 Contours of animals per hectare per day on the sprayed molasses treatment ............................. 83

11 Contours of liveweight per hectare per day
(metric tons) on the control treatment................ 88

12 Contours of liveweight per hectare per day (metric tons) on the sprayed molasses treatment....... 93













viii










Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirement
for the Degree of Doctor of Philosophy


CHANGES IN SMUTGRASS (Sporobolus poiretii [Roem. and Schult.] Hitchc.) GROUND COVER
INDUCED BY SPRAYING WITH MOLASSES AND GRAZING MANAGEMENT

By

Leonidas S. Valle

August 1977

Chairman: Gerald 0. Mott
Major Department: Agronomy

A grazing experiment was conducted at the Beef Research Unit,

Gainesville, Florida, from April to November 1976. The experimental area was a pasture of 'Pensacola' bahiagrass (Paspalum notatum Flugge), smutgrass (Sporobolus poiretii [Roem. and Schult.] Hitchc.). White clover (Trifolium repens L.) was present in the spring. The initial ground cover of smutgrass was estimated to be between 40 and 50% and bahiagrass was the dominant desirable specie.

The purpose of this research was to determine the effects of

different combinations of lengths of rotation grazing cycles and levels of grazing pressures on the control of smutgrass and to evaluate the effect of spraying molasses upon the palatability of smutgrass, compared to the' unsprayed treatment.

Molasses was diluted in equal parts of water and sprayed at the rate of 320 liters/ha of molasses prior to introducing the animals. Grazing pressures measured in terms of residual dry matter left after grazing were 0.5, 1.3, 2,1, 2.9, and 3.7 metric tons/ha, and lengths of rotation cycles were 0, 14. 28, 42. and 56 days.










Thirteen different combinations of the 2 factors were arranged in a response surface design and superimposed upon the control and molasses sprayed treatments. The axial, center, and corner points were replicated twice, making a total of 22 pastures per main treatment. Smutgrass ground cover was estimated in early April before applying the treatments and again late in October, 1976.

Spraying molasses to the pasture resulted in little increase of palatability of smutgrass when compared with the control treatment. The attractiveness of sprayed smutgrass was lost a few hours after spraying.

In both treatments, smutgrass ground cover decreased with increasin grazing pressure. Smutgrass ground cover was influenced very little by length of rotation cycle.

Stocking rate, when considered as a response variable, expressed

either as animals/ha/day or liveweight/ha/day, increased with increasing grazing pressure. Length of rotation cycle had little effect.















INTRODUCTION


Weeds, whether herbaceous or woody, are undesirable in pastures becais'e they compete with forage plant species for moisture, nutrients, and light. The extent of this competition depends as much on the growth habit and nature of the weed species as on their density and distribution. Also, many weeds harbor some of the worst crop insect pests and are alternate hosts to organisms causing crop diseases. Other weeds may be poisonous, causing reduced weight gains, lowered animal production or even death.

Weed control is a major need in any program of management. The primnar objective of weed control in pastures is to selectively manipulate the canopy to eliminate undesirable plant species while maximizing the production of desirable species. The complexity of weed control requires inputs from a wide variety of specialties spanning plant ecology and physiology, management, and economics. Recommended weed control measures must be implemented as long-term programs without damaging desirable plants. This concern is an integral part of the research effort and a vital consideration.

Most weed problems on grazing lands result from man's activities

with the introduction of exotic plants being a prime cause. For example, smutgrass (Sporobolus poiretii [Roem. and Schult.] Hitchc.) was introduced into the United S*ates from tropical Asia. It is now a serious problem in many pastures of the southeast, particularly on the sandy soils of Florida.






2

Smutgrass, in pasture, spreads through enlargement of original

plants and dissemination of seeds. However, density serious enough to threaten forage production may not be recognized until 5 to 10 years after being introduced into new areas.

Control of smutgrass on heavily infested areas has long been recognized as an effective pasture improvement practice. During the past 15 years several methods have been developed to control this weed. Generally, these methods can be divided into three main categories: mechanical, chemical, and combinations of the two. Mechanical methods, such as mowing and cultivation are not effective in controlling smutgrass and previous chemical control methods have failed to give acceptable selective smutgrass control in established pastures.

In the selection of the control method, not only should degree

of smutgrass kill and cost be considered, but also the effect of the treatment upon the associated forage species.

No information is available on the effect of grazing management systems alone or combined with other weed control practices upon smutgrass. Since many pastures in Florida are infested with smutgrass, it is important to find out how much control of this weed can be reached when grazing management is used with other practices.

The objectives of this study were to determine a) the effects of

different combinations of lengths of rotation cycles and levels of grazing pressure on the control of smutgrass, and b) the applicability of spraying molasses on the pasture in an attempt to enhance the palatability of smutgrass and thereby increase smutgrass acceptability.













REVIEW OF LITERATURE


Smutgrass


Sporobolus poiretii [Roem. and Schult.] Hitchc.is named smutgrass because of a hyphomicete, Bipolaris (Helminthosporium) ravenelii (Curt.) Schoemaker, (Luttrell, 1976), which often infects the panicles and at times is found in patches on the leaves (Currey and Mislevy, 1974). It was observed that seeds produced during the spring were not heavily infested with B. ravenelii while the seeds produced in the fall were (Currey et al., 1973).

Smutgrass has been described on numerous occasions (Hitchcock, 1936, 1950; Swallen, 1955). It is a deep rooted, caespitose perennial of the family Graminea. The grass is glabrous, summer growing, with culms erect, solitary or in small tufts, 30 to 100 cm high. The leaf blade is flat to subinvolute, rather firm, 2 to 5 mm wide at the base, elongated, and tapering to a firm point toward the end. The panicles are 15 to 30 cm long, usually spike-like, plumbeous, dense, the branches appressed floriferous to the base or nearly so. The spikelets are each 1.7 to 2 mm long; glumes are obtuse, the first 0.5 and the second from

0.5 to 0.7 mm long. The seeds are reddish when mature and may remain for some time, sticking to the panicle by the mucilaginous pericarp.

Seeds are spread by water, and wind and by sticking to livestock.

Smutfree seeds may remain on the panicle for some time or shatter quickly depending on the weather (Mislevy and Currey, 1975). Seed production is continuous from May to December, with flowering, immature seed,


3








mature seed, and seed shattering occurring simultaneously on a single inflorescence on the same plant (Currey et al., 1973). A mature plant produces in excess of 45,000 seeds with over 1,400 seeds per panicle. Germination averaged less than 9% while mechanical scarification of the hard seed coat improved germination up to 94%. Currey et al. (1973) concluded that smutgrass exhibits characteristics associated with a very successful weed due to 1) production of large numbers of seed per season, 2) production of seed continually over the entire growing season, 3) variable seed dormancy with germination over an extended period of time, and 4) continual maturation of seed on each inflorescence of the same plant.

On a global scale, smutgrass is found in Asia, Central, South and North America, and West Indies (Hitchcock, 1906, 1936, 1950; Roseveare, 1948; Molinari, 1949; Swallen, 1955; Sacco, 1964) from sea level to approximately 2,700 m. It is adapted to most soil types and especially where rainfall exceeds 40 inches annually (Riewe et al., 1975b). It apparently was introduced into the United States from tropical Asia (Hitchcock, 1950) and occurs along roadsides, lawns, pastures, and wasteland from Virginia to Tennessee and Oklahoma, south to Florida and Texas. It has been reported in New Jersey and also along the Oregon Coast. Smutgrass has become adapted to subtropical and temperate climates. Top growth is killed by frost but regrowth occurs the following spring.

Smutgrass is an undesirable, weedy grass on considerable hectarage of pasture and rangeland across the southeastern United States (Riewe et al., 1975b; Smith et al., 1974; McCaleb et al., 1963). In Florida, McCaleb et al. (1963) first reported smutgrass as a potential threat to forage quality in permanent pastures and a recent survey (Currey and






5


Mislevy, 1974) of some central and south Florida counties indicated that 75% of the improved pasture was infested with smutgrass. The average level of infestation was 38%. It is slow to establish and usually requires several years before a heavy infestation occurs (McCaleb et al., 1963) but if the first few plants in the pasture are not brought under control, they will become the dominant specie in the field. The infestation increases from year to year through extension of the original plants and also from the growth of seedlings. According to Carter (1961), a heavily infested pasture may have as many as 24 plants per square yard and they may vary in size from one to six inches or more in diameter.

The acceptability of smutgrass by livestock as a forage is considered low. McCaleb et al. (1963) stated that lack of palatability of smutgras is particularly evident on the mineral soils of Florida and Georgia and Currey et al. (1973) added that this characteristic helped to account for the high level of infestation presently observed in Florida pastures. In Texas, Riewe (1974) reported that smutgrass produces a low quality forage, palatable to livestock only in the spring, but that after May, cattle will graze it only when forced to. According to Carter (1961), this weed is not palatable to cattle and the carrying capacity of the pasture tends to decrease with increasing smutgrass infestation. In another study (Smith et al., 1974) it was reported that production of high quality forage is greatly reduced as smutgras increases in the pasture.

What causes unpalatability of smutgrass is not known. In this respect (Mislevy and Currey, 1975) suggested that palatability may be limited by the high fiber content (82%) found in the mature plant.





6



Persad (1976) studied the nutritive value of smutgrass and 'Pensacola' bahiagrass. In vitro organic matter digestion and neutral detergent fiber of smutgrass were higher than of bahiagrass after 6 weeks growth. He suggested that after 6 weeks of growth, it is possible that low digestibility associated with high neutral detergent fiber could be the primary factor limiting the acceptability of smutgrass.


Smutgrass Control


A review of the available literature has failed to reveal any information on the control of smutgrass specifically in relation to grazing management. Control methods that have been used include mechanical, cultural, chemical, and combinations of mechanical and chemical (McCaleb et al., 1963; Smith et al., 1975; Currey and Mislevy, 1974).

In 1955, McCaleb et al. (1963) attempted to control smutgrass with mechanical methods. They studied the effect of rotary mowing at a 3inch height at intervals of 1, 2, 3, and 4 weeks. They reported that control of smutgrass by mowing at weekly intervals resulted in some reduction of plant size but that all plants recovered to former density after stopping the treatment. Cultivation and complete renovation gave variable and unsatisfactory results, since new plants grew from seeds already in the soil. In Louisiana, the use of a modified rotary tillage machine resulted in 90 to 95% reduction of smutgrass (Carter, 1961).

Many chemicals have been evaluated to determine their effect on smutgrass. McCaleb et al. (1963) screened 6 different chemicals for their herbicidal efficiency on control of smutgrass. In this preliminary trial several herbicides showed promise, including dalapon (2,2dichloropropionic acid), monuron TCA (3-fp-chlorophenyll-,1 -dimethylurea





7



mono [trichloro acetate]) and monuron (3-[p-chlorophenyll-1, 1dimethylurea). They reported that with dalapon at the rate of 5 lb/acre of active ingredients, the control averaged 85%. Despite the results, they pointed out that additional treatments were necessary to kill surviving plants and new bunches starting from seed. When using a single fall application of either bromacil (5-bromo-3-sec-butyl-6methyluracil) at 2.24 kg/ha or atrazine [2-chloro-4-(ethylamino)-6(isopropylamino)-s-triazinel at 4.48 kg/ha, Smith et al. (1974) reported 83 to 95% control of smutgrass within 36 weeks. In Texas, Riewe et al. (1975a) stated that almost complete smutgrass control prevailed 18 months after spraying with dalapon at a rate of 5.6 kg/ha. However, they added that reinfestation of the treated pasture should be expected in time. In Mississippi (Smith, 1975), spring application of dalapon and diuron controlled 90% of the smutgrass in bermudagrass (Cynodon dactylon [L.] Pers.).

Herbicides may injure desirable species in the pasture, although grass species vary in their reaction to herbicides. Severe leaf damage and root kill of bahiagrass with very slow recovery of surviving plants was observed with a rate of 5 lb/acre of dalapon (McCaleb et al., 1963; McCaleb and Hodges, 1971). On the other hand, Riewe (1974) found that all top growth of common bahiagrass was desiccated at the time of dalapon application, but recovery began 4 to 6 weeks after application. He concluded that bahiagrass seems to be quite tolerant to dalapon at the rate of 4.75 lb/acre of active material. Houston et al. (1975) studied the effect of atrazine and dalapon upon smutgrass control in a dallisgrass (Paspalum dilatatum Poir)-bermudagrass pasture. They






8



found that atrazine was the least damaging to desirable species, while dalapon was the least selective. Dalapon delayed recovery of the two species for approximately 3 weeks. Minimum damage to a dallisgrass and bermudagrass has been achieved with a late October application, after most of the desirable forage had been grazed off by cattle (Riewe, 1975b).

The time of herbicide application seems to be very important to

the control of smutgrass. In Florida, Currey and Mislevy (1974) reported that the best time of the year to apply dalapon is late May, when the plant is most actively growing. Houston et al. (1975) suggested that pastures can be sprayed any time between May and October if conditions are favorable and smutgrass is growing. They also pointed out that if early herbicidal application causes shortage of forage due to damage of the desirable species, dalapon could be applied between late September and early October. When applied in the spring or early summer, dalapon increases the loss of grazing time and other weeds generally become a more serious problem (Riewe et al., 1975b; Schlundt, 1977). In a study to determine the effects of applications of dalapon in March, May, June, August, September, and October, Riewe (1974) concluded that applications in late summer and fall are the most successful. He concluded that application at this time resulted in a) less loss of grazing, b) fewer weeds, and c) fits well into a program of seeding ryegrass into a warm season perennial grass pasture for winter pasture.

Combinations of spraying with mowing have not been explored sufficiently in smutgrass control. In Florida, Currey and Mislevy (1974) studied different cultural treatments before and after application of










dalapon at rates of 4 and 5 lb/acre. They reported that smutgrass can be controlled by mowing to a 2-inch stubble 4 to 5 weeks following the application of dalapon. It was observed that 2.5 years after such a treatment combination, little smutgrass encroachment in 'Pangola' digitgrass (Digitaria decumbens Stent.) and bahiagrass occurred, provided the treated pasture was fertilized and rotationally grazed (Mislevy and Currey, 1975).


Weed Control by Grazing Management


Effective control of weeds may often be obtained by mechanical means such as mowing or cultivation or by treatment with herbicides, but costs will always need to be considered (Leach et al., 1976). Chemical control may be essential where toxic weeds have become established and hazardous to livestock. Michael (1970) pointed out that grazing management studies designed specifically for weed control have as yet barely begun in Australia. He added that this kind of study is much needed, especially in relation to weedy grass and annual grass control. Smith (1968) stated that control of barley grass (Hordeum leporinum Link) by grazing management presents advantages over chemical methods because a) it is cheap, b) the weed is a useful source of forage in its early stage, and c) clover production rather than being lost, may even be increased.

Michalk et al. (1976), in Australia, studied the effect of different stocking rates under 6 grazing management systems on the control of barley grass. Heavy grazing in late winter increased the proportion of barley grass in the pasture and the number of seedheads per unit area.





10




However heavy grazing early in the fall resulted in decrease of the weed and increase in crow-foot (Erodium spp). They concluded that dry matter production for the different treatments was relatively unaffected. However, there was a marked effect on botanical composition and on the development and flowering of the different species particularly in the case of barley grass.

The effect of grazing management on slender thistle (Cardus

pycnocephalus L.) population in an improved pasture was studied in southern Tasmania (Bendall, 1973). Deferring grazing until winter or spring was very effective in reducing slender thistle. Spring grazing favorably altered pasture botanical composition by increasing the frequency of perennial ryegrass (Lolium perenne L.) and subterranean clover (Trifolium subterraneum L.) and reducing the frequency of the weed. He concluded that, of the other successful treatments, deferred fall grazing is the most practical system for incorporation into the farm management system as an alternative to herbicide for the control of slender thistle. He also indicated that such a program has the advantage of being less expensive than chemical treatment and favors general pasture improvement.

Myers and Squires (1970) observed for 3 successive fall seasons the effect of grazing on the control of barley grass in an irrigated pasture sown with subterranean clover. He compared grazing starting 10, 20, and 40 days after the opening of irrigation. Barley grass yield was less on the 20-day treatment than on the 10 and 40 days. Deferment of grazing for 10 days was less successful, because of the tendency of the animals at this early stage to select dead material and warm-season





II


weeds in preference to barley grass. They concluded that substantial reduction in barley grass population in an irrigated pasture can be achieved within 1 year, and that almost complete elimination will occur in two years by deferment of grazing for 20 days followed by continuous grazing management.

In Australia from 1962 to 1966, the influence of pasture topdressin with superphosphate and stocking rate on skeleton weed (Chondrilla juncea L.) in a white clover-ryegrass mixture was measured in a grazing experiment (Kohn and Cuthbertson, 1975). The increase in stocking rate from 5 to 15 sheep per hectare had no effect on final skeleton weed number. Density of skeleton weed increased in rotational grazing (1 week grazing and 2 weeks rest) when compared to continuous grazing. The increase of the weed under rotational grazing was due to the development of satellite plants in the pasture.

Laycock (1970) compared the effect of spring-fall grazing and fall grazing only on a sagebrush-grass range. Heavy spring grazing caused a reduction of grasses and forbs by more than 50% from 1950 to 1964 and increased sagebrush production by 78%. Fall grazing enhanced production of palatable perennial grasses by 36% and reduced production of sagebrush by 22%. He suggested that fall grazing as a method for range improvement is less expensive than mechanical or chemical means of sagebrush control.

For 4 years, Pearce (1972) studied the consequence of different

stocking rate on the control of Patersons's curse (Echium plantagineum). He found that population of Paterson's curse declined by up to 72% with a stocking rate of 3 sheep per acre and up to 80% when the stocking rate was increased to 8 sheep per acre. Pasture sprayed with





12



2,4-D and grazed at a stocking rate of 8 sheep per acre had considerably lower weed population than pasture which was not sprayed.

Southwood (1971) reported control of broomrape (Orobanche minor)

in a Trifolium subterraneum-Hordeum leporinum pasture mixture. Heavy, continuous grazing before the broomrape flowered combined with superphosphate application in autumn, significantly reduced the weed population. The author stated that when continued for a number of years, broomrape could be eradicated by such a treatment combination.


Pasture Loss Due to Weeds


(/Weeds contribute to decrease pasture productivity in several ways: 1) they decrease forage yield due to weed competition, 2) they cause animal discomfort, 3) they result in undesirable flavors in animal products, and 4) they may cause poisoning of the grazing animals (Smith, 1974). The average annual losses due to weeds in pastures and rangelands in the United States from 1951 to 1960 may be estimated at more than $632 millions (Hamill, 1975). In 1973 in Florida, it was estimated that weed damage in pasture and hay crops totaled 28.8 million dollars. This cost resulted from expense of weed control with herbicides and mechanical methods, yield and forage quality loss, lowered land value, and the expense of additional harvesting (Smith 1974). This figure does not include animal losses in decreased weight gains, less milk production, and animal deaths by poisoning.

Weeds interfere with the growth and development of desirable forage species in many ways. They compete for light, water, nutrients,space, and carbon dioxide (Smith, 1974; Hamill, 1975). Crombie (1947) defined





13



competition as the requirement at the same time by more than one living organism for the same resources of the environment in excess of the immediate supply. Daubenmire (cited by Risser, 1969) suggested the following parameters of adaptation which may under specific circumstances, be significant in competiton: 1) time of root penetration. 2) ability to obtain nutrients in short supply, 3) endurance in drought soils, 4) longevity, 5) abundance of seed production, 6) food reserves available to young plants, 7) time of initial growth, 8) nutrient uptake ability, 9) vigor and size of plant, and 10) reproductive potential.

In a comprehensive study in Australia to examine the outcome of competition for light between capeweed (Arctotheca calendula) and subterranean clover, Mclvor and Smith (1973) reported that capeweed does not suppress clover growing in association if the two species commence growth together. However, when the capeweed was established 4 weeks before clover, the legume yield was always lower than from pure stand sown on the same day.

Rummell (1946) investigated the competition of Bromus tectorum L. with Agropyron desertorum (Fisch. ex Link) Shult. and A. smithii Rydb. In this study the number of A. desertorum plants was reduced to 50% and A. smithii to about 10%. He concluded that A. desertorum which germinates earlier in the season and makes rapid growth following emergence, competes more successfully with B. tectorum than the slower developing A. smithii.

Wakefield and Skaland (1965) conducted an experiment with alfalfa (Medicago sativa L.) to evaluate effects of intra-species and weed competition on seedling establishment. Seeding rates of alfalfa were approximately 25, 50, and 100 seeds per square foot. Three intensities





14



of weed competition were developed with the aid of chemical treatments. They found that weed control resulted in an increased yield of alfalfa in the seeding year compared to untreated plots. Seeding rates had a marked effect on the average root-crown weights of alfalfa. Smallest weights occurred at the high seeding rate and significantly larger rootcrown weights were observed from the lowest seeding rate.

Grimmett and Weiss (1967) investigated the competition of weeds in sown pasture of 'manawa' ryegrass (Lolium perenne L. x L. multiflorum Lam.) and subterranean clover. They reported that rapid weed growth restricted seedling development of the desirable species resulting in slow and unsatisfactory pasture establishment.

Bryan and McMurphy (1968) reported in their study that caxgrass

(Digitaria sanguinalis [L.1 Scop.) which emerged with the seeded weeping lovegrass (Eragrostis curvula [Schrad.] Nees.) reduced forage yield by more than one-half in the first clipping. The plants which suffered from weed competition were visibly reduced in height and did not produce seedheads the first year.

Scholl and Staniforth (1957) studied the establishment of birdsfoot trefoil (Lotus corniculatus L.) as influenced by competition from weeds. They found that pre-emergence application of monuron TCA or postemergence application of dalapon controlled grassy weeds and enhanced survival and vigor of trefoil seedlings.


Benefits from Weed Control


Beneficial effects of weed control on pastures depend on the efficiency of the weed control method. Klingman (1970) stated that what was most needed to produce good quality pasture was a method to prevent





15



and eliminate weeds. He also suggested that in pasture-forage production, the integration of all beneficial practices into the management system is required to achieve highest efficiency.

Peters and Stritzke (1971) studied the effects of 2,4-D and mowing on the botanical composition and production of a Kentucky bluegrass (Poa pratensis L.). They reported that the average production of broadleaf weeds were 88, 329, and 950 lb/acre for 2,4-D, mowed and untreated plots, respectively. Yields of Kentucky bluegrass were 900 and 1,150 lbs/acre for the control and 2,4-D treatments, respectively. They concluded that the forage yield increase was primarily the result of a decrease in competition from broadleaf weeds.

Alley and Bohmont (1958) studied the control of big sagebrush

(Artemisia tridentata Nutt.) in a native pasture. They indicated that a four-fold increase in yield of native forages was obtained by controlling big sagebrush.

In a four-year study, Morrow and McCarty (1976) observed the influence of green sagewort (Artemisia campestris L.) and other broadleaf weeds on forage production in Nebraska. Chemical treatment increased forage production by 42% and controlled 97% of the weeds in plots receiving two consecutive annual applications. Forage production was increased up to 330 for herbicide alone and 660 lb/acre of dry matter for herbicide followed by nitrogen fertilization. They pointed out that herbicide and fertilizer can be effectively used to increase forage production, but they will not correct the effect of mismanagement which results in weedy pastures.

In Nebraska, Klingman and McCarty (1958) reported that use of

2,4-D was more efficient than mowing for the control of broadleaf weeds.





16



Using three annual applications of 2,4-D reduced the weeds by 70% and mowing alone resulted in a reduction of 30%. Combining annual spraying with plowing and seeding reduced broadleaf weeds more than 90%.

Scholl and Brunk (1962) investigated the competition of weeds with birdsfoot trefoil. They compared no weed control and all weeds removed in the early stage of growth. Where no weeds were controlled, yield on the trefoil on the first year was 389 lb/acre of dry matter and where complete control was practices, 2,342 lb/acre of dry matter was obtained also in the first year. In the second year the yields were 3,480 and 6,533 lb/acre of dry matter for the control and weed-free treatments, respectively. Birdsfoot trefoil plants were shorter on the control treatment than on weed-free treatment. However there was no difference between the two treatments as far as birdsfoot trefoil population was concerned.

Gesink et al. (1972) investigated the control of broom snakeweed

(Gutirrezia sarothrae (Pursh] Britt and Rusky) on the short-grass plains in southeastern Wyoming during 5 years. Control of broom snakeweed increased the desirable species, chiefly blue grama (Bouteloua gracilis (H.B.H.) Lag.). Herbage production was 225 lb/acre of dry matter for untreated areas and 1,200 Ib/acre of dry matter when treated with a herbicide.

Forage quality and intake may be increased by controlling weeds in pastures. Barrett et al. (1973) studied the effect of spraying paraquat in a pasture containing subterranean clover and either silver grass (Vulpia spp.) or ripgut brome (Bromus rigidus Roth). Spraying controlled the grasses and produced pastures containing up to 95% clover.





17

The concentration of N, P, Ca, and Mg were higher in mature herbage on treated plots than on the control treatment.

Monson (1977) studied the effect of paraquat on yield and quality in a pasture of Coastal bermudagrass (Cynodon dactylon L.). He found that the in vitro dry matter digestibility of a sprayed pasture was higher than on the control at 6 weeks after treatment application.

Smith et al. (1974) determined the quality of forage produced after controlling smutgrass in a Paspalum dilatatum Poir.-Cynodon dactylon L. pasture. It was observed that soluble cell contents were significantly higher in the treated plots. The acid detergent fiber and neutral detergent fiber fractions were decreased by application of herbicide. He pointed out that the figures for nutritive value index suggest that intake of digestible dry matter should be higher in pasture where smutgrasZ was controlled when compared to the control.

In most experiments the forage intake has been increased by controlling weeds. After four years of weed control and fertility treatments in a Kentucky bluegrass pasture, Peters and Stritzke (1971) found significant improvement in the intake of forage. Low fertility treatment with 2,4-D increased consumption of edible forage (Kentucky bluegrass, legumes, and weed grasses) about 200 lb/acre, but the medium and high levels of fertilization in combination with herbicide increased intake about'600 and 1,000 lb/acre, respectively.

Klingman and McCarty (1958) compared spraying and mowing for weed control and measured the effect of the treatments on forage intake. They reported that dry matter intake was 1,650 pounds on the sprayed pasture and 1,340 pounds on the mowed as compared to 1,124 pounds on the untreated plots. That represented an increase in forage consumption of 47% for spraying and 19/ for the mowing.





18

Measuring Botanical Composition


The determination of the species in a pasture is important for study ing the changes in botanical composition due to treatment effects. Botanical composition is also essential because individual species may differ in theri reaction to environmental and management factors. (t' Mannetje et al., 1976).

Measurement of botanical composition may be made in terms of the

yield of the species components, the frequency of occurrence of different species and the number of plants on the area covered by different specie(t' Mannetje et al., 1976).

The area ground covered by the aerial parts of the plants has proven to be a valuable measure of botanical composition and its change. Brown (1954) defined cover as the vertical projection of the above-ground parts of the plants on the ground, and it is expressed as a percentage of the total area (Winkworth et al., 1962). Terms used to express the area covered are 1) density, 2) basal area, 3) herbage area, 4) foliage density, 5) cover, and 6) leaf area index.

Pasture research had led to the development of a large variety of methods for estimating percent cover (Brown, 1954). The methods fall into four basic categories, depending on the type of observation made and the dimensions of the sampling unit. These are charting, ocular estimated, line intercepts, and point methods.

The chart quadrat method is one of the earliest techniques used to determine changes in botanical composition (Hill, 1920). Basically, the ground position of each plant is drawn in its relative position on a recorded sheet (Weaver and Clements, 1938), usually at yearly intervals (Wright, 1972). However, Brown (1954) stated that the interval between





19

charting depends on the purpose of the study. He emphasized that in areas where weather varies considerably from year to year an annual charting is important to determine the changes in the vegetation.

Hutchings and Pase (1962) stated that chart quadrat method is

particularly well suitable to grass and other types of low vegetation. However, he added that this technique presents better results when the plants are clearly defined in bunches or tufts. It becomes more difficult as the growth becomes closer and is quite impossible in dense pastures (Brown, 1954).

Heady et al. (1959) compared the accuracy and practicability of

charting, line intercept, and line point in the sampling of two shrub communities. They reported that means and confidence intervals obtained by the three methods gave reliable estimates of the population mean. Line intercept and line point yielded less variable data and were sampled adequately with fewer plots and less time than the charting methods. However, Ellison (1942) found that accuracy in charting method depends on the ability and care of the operators and to some extent on the charting device used.

Although the chart quadrat method can provide a reasonably accurate record of the vegetation (Holscher, 1959), it is always time consuming. The individual culms as well as the area occupied by clumps must be carefully drawn into a small scale quadrat map with the aid of a pantograph or directly by means of numerous wires crossing the frame (Anderson, 1942).


Estimating Forage Yield


Sampling to estimate yield is one of the most difficult procedure involved in pasture research. The methods generally used for





20



measuring the production of a pasture involve the cutting of areas or spot samples of the pastures. As stated by Campbell et al. (1962), the use of grazing animals in pasture management studies increases the difficulty of estimating yield from sample-cutting techniques due to a) the need to cut sufficient samples to give an accurate estimate of the yield, because of the very variable nature of grazed pasture, b) the requirement of cutting as few samples as possible, which arises from the physical limitations on cutting large number of samples, and c) the need to ensure that the area cut is not so large that it acts as a treatment in itself.

Many efforts have been devoted to the search of a quantitative method for determining the yield of pasture in situ. Bransby et al. (1977) emphasized that a rapid, indirect, in situ, nondestructive technique for making accurate estimates of pasture dry matter yield would benefit grazing experiments. The literature describing the methods that have been developed which do not require the harvesting of herbage usually refer to the term "density." The use in each instance is in accordance with the use of the term for the specific method and is based on the relationship as stated by Mott (1962).

Yield/Unit Area = f (density, height) One of the more promising methods of estimating pasture dry matter yield in.situ is the use of electronic capacitance meter (Alcock and Lovett, 1967). Fletcher and Robinson (1956) proposed the use of a capacitance mete-. It is based upon the fact that herbage has a high dielectric constant and air has a low dielectric constant. Readings were taken from plants under conditions ranging from dry grasses to soggy, wet sedges in swamp land. They demonstrated that the





21



capacitance meter showed promise of being faster than clipping and more accurate than other estimation methods. They reported that the slight decrease in precision per determination is more than offset by reduction in sampling error. Campbell et al. (1962) with a modified capacitance meter stated that if the errors of prediction were solely random error, the greater efficiency of sampling by the instrument would compensate for any moderate increase in error per estimate compared with a cutting technique. Johns and Watkin (1965) reported that the great advantage of using a capacitance meter is that once the apparatus has been calibrated for a particular pasture, a very large number of readings can be taken in a short time with no detrimental effect on the pasture.

In New Zealand, Campbell et al. (1962) found that within pastures the capacitance meter allows an estimate of sample yield to be made with considerable accuracy. The instrument predicted about 90% of the variation in forage weight either as wet, dry or organic matter. For different pastures significant differences existed between prediction equations, although certain pastures can be combined. They concluded that for different pastures, individual prediction equations would be required. Johns and Watkin (1965) studied the relationship between capacitance readings and yield of pastures. They found that in all pastures'except for native pasture, the regression of dry matter yield, fresh material and total water on the meter reading were highly significant. However, the dry matter regressions were generally slightly inferior to both the fresh material and total water regressions. They suggested that dry matter is still considered adequate for use in individual experiments where the particular regression could be reliably





22



established, even though regression based on yield of fresh material would be more accurate. In Colorado, Carpenter et al. (1973), studied the influence of woody stems on the relationship of the capacitance meter to herbage weight. A single meter reading of the plot estimated weight of total herbaceous material is more accurate than total herbaceous material plus woody stems. They reported that excluding wood would probably improves the regression because woody material has little capacitance relative to herbaceous material and the amount of wood on the plots varied greatly relative to the amount of herbaceous material.

It is apparent that the height of the forage and the above stubble height affect capacitance meter readings. Hydy and Lawrence (cited by Alcock and Lovett, 1967) found that a change in capacitance of the probe is caused by the dielectric properties of the herbage and plant acting as a circuit when leaves come into contact or near contact with the electrodes. Lovett and Bofinger (1970) used a capacitance meter to measure growth of 30 cultivars of rape (Bassica spp.). The height of some cultivars exceeded the height of the probe, resulting in plant material touching the top plate of the instrument. Under such conditions probe readings were affected by changes both in capacitance and a tactile factor, the latter occurring when material was crushed down in down in obtaining readings.

When measuring a short stubble the main factor influencing change in capacitance is the dielectric properties of the herbage (Hydy and Lawrence; cited by Alcock and Lovett, 1967). Back et al. (1969) sampled plots sown with Lolium multiflorum Lam. and a grass-white clover (Trifolium repens L.) mixture. On each plot. 5 samples were cut to ground level and a further five to approximately 2 cm above





23




ground. They reported higher relative efficiency of the samples cut to 2 cm in comparison with those taken to ground level.

A critical factor in obtaining a close relation between capacitance meter readings and dry matter yield of cut forage is the water content. Johns and Watkin (1965) studied the effect of dew on capacitance meter readings. They pointed out that the presence of heavy dew on the pasture enhanced the dial reading by approximately 1.2 units. This increase would represent an apparent increase of approximately 125 lb/acre of dry matter in the estimate of pasture yield. They concluded that this over-estimation of dry matter yield varies with the amount of dew, the moisture collecting and retaining ability of the species present in the pasture and the amount of surface moisture when the readings are taken. In further investigations, Alcock and Lovett (1967) determined the influence of surface-water upon capacitance meter readings. Plots of Italian ryegrass were sprayed at the rates of 0, 2,750, 5,500, and 11,000 gallons of water per acre. Readings of the meter increased with the level of water added to the pasture. However, they reported that estimated dry matter yield varied very little. In a similar study, Jones and Haydock (1970) reported that application of water resulted in increased readings followed by a decline as the water dripped off the plants. They concluded that unlike the marked effect of water applied to the plants, there was little effect from adding water to the soil beneath the probes. Flooding the area and allowing the water to soak-in had very little effect provided the soil surface was level.






24



Neal and Neal (1973) reviewed the use of electronic capacitance meters to estimate weight of standing vegetation. They concluded that variation in site and phenology generally have more influence on meter performance than does meter design. In their explanation, the meter reading is the sum of numerous external influences and internal characteristics of the instrument. They emphasized that the technique is accurate, rapid and nondestructive when used properly. However, Nichols (1973) stated that any apparatus for outdoor use must be designed so that adverse conditions do not contribute to further inaccuracy. Shaw et al. (1976) pointed out that the capacitance meters have not lived up to their early promise due to limitations such as a) it measures water yield in the plant tissues and not dry matter yield, b) it is influenced by plant species, and c) it is relatively insensitive where the herbage consists mainly of dead material. He concluded, however, that the above limitations apply much more often in the case of tropical pasture than temperate pastures.


Spraying Pasture with Molasses


It has long been known that consumption of unpalatable forage is influenced by adding molasses. The direct application of molasses on standing pasture, seems to have first been developed in South Africa (Loosli and McDonald, 1968). Plice (1952) advised farmers to take advantage of this technique and get rid of weeds, unpalatable and poorquality forage by spraying them with molasses or other similar product, and turning grazing animals on them.





25



Change in intake from preferred to initially non-preferred components of the pasture is enhanced by inducing the animals to graze heavily on the sprayed area in preference to grazing lightly over the whole pasture (Willoughby and Axelsen, 1960). In this respect, Coombe and Tribe (1962) also stated that only small areas of the pasture should be sprayed with molasses at a time. Native pastures with considerable amount of unpalatable annual grasses were sprayed with urea-molasses mixture during the summer (Pope et al., 1955). They reported that spraying a small area led to very intensive grazing and removal of all top growth, while spraying a large area resulted in an incomplete consumption of the sprayed forage.

According to Coombe and Tribe (1962) spraying molasses on standing herbage presents several disadvantages. They emphasized that the most serious is the high proportion of spray which falls on the soil and is washed off by rain. In addition, they added that the technical difficulties of spraying forage may be considerable where a) pasture areas are large, b) stocking rates are low, and c) the ground surface is rough. Mostert (1959) emphasized the importance of spraying only thick grass stands or patches, otherwise much of the sprayed mixture will be wasted. In detailed tests on one cattle pasture, it was found that only 16.5% of the spray could be recovered from the herbage immediately after spraying (Loosli and McDonald, 1968).

In Oklahoma, Plice (1952) compared table sugar, black-strap molasses, sorghum molasses, and corn syrup when sprayed on weeds and grasses which are seldom, or never, touched by grazing animals. The order of preference for the different materials was as follows: black-strap molasses,






26




sorghum molasses, table sugar, and corn syrup. He observed that the animals did not take very long to discover the sprayed plants and then consume them completely.

In Australia, Willoughby and Axelsen (1960) tested four spray treatments, containing urea, molasses, and urea plus molasses on a grass-legume mixture which had been ungrazed in the spring and had matured and dried in early summer. The pastures sprayed with molasses were consumed most rapidly by the animals. The application of sprays, mainly molasses, resulted in increased removal of the most abundant component, Phalaris, an initially non-preferred component of the pasture. He concluded that spraying affected intake in three ways: a) it provided a supplement, b) it altered the amount of forage consumed, and.c) it increased or decreased the quality of the material consumed, depending on whether the non-preferred components are higher or lower than the preferred in nutritive value.

O'Bryan (1960) examined the utilization by cattle of carpet grass (Axonopus affinis Chase) following foliar application of urea, molasses, and monosodium phosphate in southeastern Queensland. The pastures provided a complete ground cover and were sprayed in strips at weekly intervals. He reported that the animals showed preference for the treated strips immediately after spraying. The treatment failed to prevent selective grazing, so that the animals selected pasture containing 10 to 12% crude protein in the first year and 8 to 10% during the second year. He pointed out that frequency of spraying (weekly), heavy dews, intermittent rainfall during the experimental period, low palatability of mature carpet qrass and the high degree and selectivity





27


by the animals were some of the factors that contributed to the low effect of spraying. He concluded that, for Queensland conditions, more frequent spraying would be uneconomical and impracticable.

Coombe and Tribe (1962) studied the value of spraying urea and

molasses on dry, standing forage. It was observed that practically all animals (sheep and cattle) were attracted to pasture treated with molasses. In.general, cattle showed a greater preference for sprayed forage than did sheep.

In Uruguay, Christiansen (1965) studied the effect of spraying

molasses and urea on a native pasture. The mixture was supplied twice weekly and 1/8 of the total area was sprayed each time until the entire pasture had been sprayed. He found that the palatability of dry coarse grasses Paspalum quadrifarium and Schyzachyrium paniculum was improved and that the animals grazed extensively in the sprayed areas.

In California, Wagnon and Goss (1961) compared a) dry rank forage sprayed with molasses-urea mixture,,b) sprayed only with molasses, and c) untreated forage. Weekly application of 14 pounds of molasses or molasses-urea mixture per animal were made. They found that rank, dry forage of low palatability was completely utilized by the animals after spraying with molasses or molasses-urea mixture, however, similar unsprayed forage was mostly left ungrazed. They also reported that there was no loss of the sprayed material until light dews occurred 43 and 63 days after initial spraying. Practically all sprayed mixture was washed from the forage by 0.72 inches of rain that occurred 90 days after the initial treatment.

In another study Tulloh et al. (1963) compared a) molasses fed in a trough, b) standing forage sprayed with urea and molasses, and






28




c) not supplemented. They reported that all supplemented treatments significantly increased forage intake.

It appears that a sudden change in cattle grazing habits occurs

when green forage became available after rain. Coombe and Tribe (1962) reported that if a "green pick" occurred due to unseasonal rains, spraying molasses and urea had no effect on forage intake. They stated that once sufficient rain had stimulated growth of green shoot, the animals preferred to graze the green material rather than concentrate on sprayed areas. On the other hand, Wagnon and Goss (1961) observed that the animals on a sprayed treatment continued to eat the old forage that had been sprayed with molasses-urea mixture, instead of the new regrowth. However, in the same experiment it was observed that animals on the control treatment started to graze the young green plants, rejecting as much as possible the old leached forage.

The preference for molasses sprayed forage appears to vary with the palatability of the mixture sprayed, the rate of molasses, and the weather. Sprayed pastures were strongly preferred by grazing animals, however, preference was not evident 24 hours after spraying, O'Bryan (1960) observed that the preference for the sprayed forage was maintained for two days and Mostert (1959) reported that the sprayed area should be grazed off within 3 to 4 days.

In studies of different spraying densities, Bishop (1959) compared three rates of a mixture molasses-urea: a) one gallon per 10 to 15 square yards, b) one gallon per 50 square yards, and c) one gallon per 100 square yards. Spraying at the rate of one gallon to 50 square




29



yards was found to give the best results. At a density of one gallon per 100 square yards much of the grass remained ungrazed, so that the mixture absorbed by this ungrazed grass was merely wasted, Densities of one gallon to 10 to 15 square yards were found to be unsuitable because animals then grazed too short. He emphasized that this is particularly undesirable on sandy soils where grass roots can be pulled out easily.


Effect of Grazing Management
to Botanical Composition


One of the most important factors to consider in grazing experiments is the change in botanical composition. The presence of grazing animals has effects upon the pasture through defoliation, excretion, and trampling (Coaldrake et al., 1976). T' Mannetje et al. (1976) stated that even a so called pure stand will usually contain varying amounts of other species and that botanical composition is important because individual species or cultivars vary in feeding value, in content of harmful substances, and in their reaction to environmental and management factors.

Changes in botanical composition as a result of grazing management systems are well documented. Bryan (1970) studied the effects of low and high stocking rates on the botanical composition of mixed pasture. He found that high stocking rate increased Paspalum dilatatum Poir., and total weeds from 24 to 33% and 14 to 22%, respectively, and reduced Chloris gayana Kunth. from 18 to 7%. In this experiment heavily stocked pastures were reduced to a closely clipped lawn while the lightly grazed ones usually carried a considerable bulk of material 0.3 to 1 m high.





30



In Australia, Cameron and Cannon (1970) observed changes in botanical composition resulting from increased stocking rates of 4.9, 7.4, 9.9, 12.4, 14.8, 17.3, and 19.8 sheep/ha, from 1963 to 1968. Trifolium subterraneum L. during this period increased from 30 to 70% at 4.9 sheep/ha and decreased from 30 to 10% at 19.8 sheep/ha. Lolium perenne L. which in 1963 was the main grass component, representing 20 to 40%; by 1968 declined to a trace at all levels of stocking rates. They also pointed out that the decline of L. perenne was more rapid at high than low stocking rates and that by 1965 it had decreased to less than 10% at 19.8 sheep/ha. Poa annua L. not initially present in any of the pastures, in 1968 ranged from a trace at 7.4 and 9.9 sheep/ha, and to 30% at 17.3 and 19.8 sheep/ha.

Rodel (1970) compared the effect of two stocking rates upon different grasses. He found that high stocking rate caused marked changes in basal cover of the grasses. Chloris gayana Kunth. decreased from 3.8 to 0.7% while Cynodon plectostachyum increased from 3.0 to 11.5%.

Serrao (1976) studied the response of Desmodium intortum (Mill)

Urb-'Coastcross-l' bermudagrass mixture to different levels of grazing period, rest period, and grazing pressure. He reported that the grass percentage in the mixture increased with heavy grazing pressure, while the legume percentage increased with long rest periods associated with medium to light grazing pressure.

Ritson et al. (1971) studied the changes in botanical composition of a mixture of Townsville stylo (Stylosanthes humilis H.B.K,), perennial grasses, and annual grasses at two stocking rates, namely I cow per 1.2 and 2.4 hectare. They found that stocking rate





31


significantly influenced the botanical composition of the mixture. With a stocking rate of I cow per 2.4 hectare the pasture was dominated by perennial grasses. At a stocking rate of I cow per 1.2 hectare the pasture became dominated by Townsville stylo and annual grasses.

Campbell and Beale (1973) studied the botanical composition of

natural pastures stocked at 2.5, 3.7, and 4.9 sheep/ha. Higher stocking rates resulted in a lower contribution by barley grass (Hordeum leporinum Link.) to the pasture, but an increased contribution by silver grass (Vulpia myuros K. Gmel.) and naturalized medics (Medicago spp.).

Johnston et al. (1971), in a long-term grazing experiment conducted over a 17-year period, compared light, moderate, heavy, and very heavy stocking rates on the botanical composition of a fescue grassland range. Percent basal area of vegetation in lightly grazed plots changed from dominance by Danthonia parryi Scribn. to dominance by Festuca scabrella Torr. F. scabrella was largely eliminated by very heavy grazing and the plots were invaded by various species, including Taraxacum officinale Weber. Populus tremuloids Michx. encroached upon grassland in the lightly and moderately grazed treatment but the same was prevented in the heavily and very-heavily grazed pastures.

Pearson and Whitaker (1974) presented changes in botanical composition by cattle grazing yearlong at light, moderate or heavy stocking rates. The most abundant grasses in the pasture were slender bluestem (Andropogon tener (Ness) Kunth.), pinehill bluestem (A. divergens (Hack,) Anderss. ex Hitchc.), panicums (Panicum spp.), and paspalums (Paspalum




32




spp.) They reported that although overall herbage composition was not changed, grazing intensity did affect individual species. Pinehill bluestem, the principal specie on the range declined from 57% under light grazing to 17% under heavy grazing. Carpetgrass was I and 26% under light and heavy grazing, respectively.

In Nebraska, McCarty et al. (1974) studied the effect of rotational and continuous grazing upon change in botanical composition of a pasture, from 1.946 to 1969. They reported that under rotational grazing, relatively few weeds invaded the pastures and only a small amount of blue grama (Bouteloua gracilis [H.B.H.] Lag. ex Stend) and sand lovegrass (Eragrostis trichodes [Nutt.] Wood) persisted. In 1969, the mixture consisted primarily of big bluestem (Andropogon gerardi Vitman), indiangrass (Sorghastrum nutans [L.] Nash) and switchgrass (Panicum virgatum L.). In the continuously grazed warm season grass plots the main desirable specie was blue grama.

Ottosen et al. (1975) studied the change in botanical composition of a mixture of tropical grass-legume pasture by comparing strip and continuous grazing. Both grazing management systems caused marked changes in botanical composition of the pasture. Legumes decreased from 24 to 16% in the strip grazed plots while that in the continuously grazed treatment, the legumes increased to 38%. They emphasized that after the experiment, both grasses and legumes regrew without any differences resulting from the previous grazing systems.















MATERIALS AND METHODS


General Description


A grazing experiment was conducted from April to November 1976

at the Beef Research Unit which is located approximately 21 kilometers northeast of Gainesville, Florida.

The soil at the experimental site is underlain by limestone of the Ecocene age having an overlay of acid, sandy, and loamy marine sediments (U.S.D.A., 1954). These Myakka fine sand soils are somewhat poorly drained and contain a spodic horizon (organic hardpan) which is an accumulation or organic matter, iron, and aluminum (Carlisle and Pritchett, 1971). This hardpan increases the severity of both wet and dry periods by retarding the vertical movement of water to and from lower levels. Data reported by Koger et al. (1961) shows that the average organic matter content of this soil is 2.258. The average pH is 4.9, varying from 4.5 to 5.2, depending on the percentage of organic matter.

The predominant native vegetation on the experimental site consisted mainly of longleaf pine (Pinus australis Michx. f.), wiregrass (Aristida spp. and Sporobolus spp.), saw palmetto (Serenoa repens Bartr. Small), gallberry (Ilex glaba L.), runner oak (Quercus minima Sarg.), and cypress (Taxodium asendens Brongn.)

The climate is subtropical and humid, with a frost-free season averaging 276 days and an average annual precipitation of 1,300 mm


33




34



U.S.D.A., 1954). Table 1 presents mean monthly maximum and minimum temperatures and rainfall during 1976 at the Beef Research Unit. For 1976, total rainfall was recorded to be 1,312 mm. Table 1. Monthly mean maximum and minimum temperatures and rainfall
at the Beef Research Unit for 1976.



Month Temperature Rainfall Max Min. 1976 Normal

----- ------ ------- mm-------January 18.3 1.8 53 65 February 23.7 5.5 34 82 March 27.2 10.7 46 103 April 27.7 11.0 77 93 May 29.6 15.5 141 87 June 31.5 18.5 215 166 July 34.4 20.4 51 187 August 33.3 19.9 100 192 September 31.2 19.0 231 124 October 26.0 10.8 55 106 November 20.8 5.5 56 44 December 18.1 4.1 126 63 t Average precipitation for Gainesville (Fla.) from 1931 to 1960.


Experimental Pasture


The experiment was conducted in a pasture of bahiagrass (Paspalum notatum Flugge), smutgrass (Sporobolus poiretii [Roem. and Schult.]





35


Hitchc.), and white clover (Trifolium repens L.) appearing in the spring, This pasture was selected because the smutgrass infestation is typical of many pastures in Florida. The initial ground cover was between 40 and 50% smutgrass. Bahiagrass was the predominant desirable specie. This area had previously been grazed as one pasture for several years and mowed annually late in the fall.


Treatments


The main treatments in this experiment consisted of

1. Control

2. Molasses sprayed on the pasture to increase palatability of

the smutgrass.

3. Dalapon applied in the spring followed by mowing and N fertilization.

In addition to the three main treatments presented above, eight extra treatments were included, making a total of 11 treatments. The eight extra treatments included the following:

4. Dalapon applied in the spring followed by burning and N fertilization.

5. Dalapon applied in the fall followed by burning and N fertilization.

6. Dalapon applied in the fall followed by mowing and N fertilization.

7. Ground Hawgt in the spring.

8. Ground Hawg in the spring followed by N fertilization,



A rototiller-type cultivator.




36


9. Ground Hawg in the spring followed by seeding of bahiagrass

and N fertilization.

10. Ground Hawg in the fall followed by seeding annual ryegrass

(Lolium multiflorum Lam.) and N fertilization.

11. Burning in the fall and dalapon applied in the spring

followed by mowing and N fertilization.


Grazing Management Factors


Within the first 3 main treatments listed above, length of rotation cycle (grazing period and rest period combined), and grazing pressure were also experimental variables. Each factor was studied at 5 levels. Length of rotation cycle was expressed in days and grazing pressure was defined as residual dry matter in metric tons/ha left after grazing. The combinations of rotation cycle and grazing pressure, each at 5 levels, were arranged in a response surface type of experiment.

The factors studied with their respective levels are presented

in Table 2. A range of plus or minus 0.2 metric tons/ha was established for the projected residual dry matter left after grazing, Grazing periods lasted from I to 4 days for all treatments and were included in the number of days in the rotation cycle. In the continuously grazed pastures the animals were not maintained in the pastures continuously. In an effort to simulate continuous grazing, animals were moved in and out of the pastures every few days to achieve the projected level of residual dry matter.





37



Table 2. Levels of length of rotation cycle and grazing pressure.



Length of rotation Grazing pressure
cycle projected RDM


-------days------- -metric tons/ha0+ 0.50.2 14 1.30.2 28 2.110.2 42 2.90.2 56 3.70.2 tResidual dry matter left after grazing. +Continuous grazing simulated.


Experimental Design


The experimental design used was a modified central composite in

2 factors (rotation cycle and grazing pressure) each at 5 levels arranged in a Response Surface Design (Fig. 1).

Combinations of length of rotation cycle and grazing pressure were superimposed upon each of the first 3 main treatments. The combinations of the 5 levels of each factor made up thirteen different grazing management combinations or design points. The arrangement of the thirteen combinations consisted of 4 factorial, 4 axial, I center, and 4 corner points (Table 3). The factorial points were not replicated. However, the axial, center, and corner points were replicated twice. making a total of 22 pastures (experimental units) per main treatment.




38














0.5




4.


-o
I v,-a



N
L .M


2.9
U
1




3.7o
0 14 28 42 56 Rotation Cycle (Days)


Factorial Corner Points


O Star Points Center Point

Figure 1. Response surface design in two factors 13 combinations.




39









Table 3. Combinations of rotation cycle and grazing pressure of
the Central Composite Design.




Length of Grazing pressure Points Number of rotation cycle projected RDMt replications


-----days----- -metric tons/ha42 2.9
42 1.3 Factorial
14 2.9 14 1.3

28 3.7 28 0.5
56 2.1 Axial 2
56 2.1

28 2.1 Central 2

56 3.7
56 0.5
Corner 2 O0 3.7 0o 0.5


Residual dry matter left after grazing.

Continuous grazing simulated




40




In this study, the central treatment (center point) was the combination of a length of rotation cycle of 28 days and a grazing pressure equal to 2.1 metric tons/ha of residual dry matter left after grazing. The central treatment was the combination of rest period and grazing pressure imposed upon the 8 extra treatments, each replicated twice. There were a total of 82 combinations of treatments and grazing managements and they were assigned to the pastures units completely at random.


Construction of Physical Facilities


The experimental area was selected in December 1975 on the basis of uniform smutgrass ground cover. In early January 1976, 4.1 hectares were allocated to the experiment and surveyed for the location of the fence lines. The area was subdivided into 82 pastures (Fig. 2), 50 m in length and 10 m in width (500 m2). Five-strand barbed wire line fences and electric division fences were built. A water supply system was installed underground using 1" plastic pipe with risers along the fence to provide water for the animals. Mineral boxes and water containers were provided. The water level was controlled in each container by a float valve.


Experimental Analysis


From May to July, 36 Brown Swiss-Angus heifers and cows (cows

with calves until August, when the calves were weaned), with an average weight of 351 kg were used to graze the experimental pastures.





41





























0N-O0










-NO



L (0(0



E
L

x 4J




0


0 000 0 0








I
o
------ -- O --U
NN v, tIo (Y- 0 M CDIf- Win .. pn C14










N O 1-..
*-




42




In August, due a higher forage production, the number of animals was increased to 50.

The put-and-take technique was used to stock the pastures to the desirable grazing pressure levels. When needed, the animals were assigned at random to graze the pastures. The animals were weighed every 28 days in order to estimate the stocking rate.

From 20 September through the conclusion of the experiment, all animals were supplemented with molasses at an average rate of 2.2 kg/animal/day. A complete mineral mixture was made available freechoice throughout the study.

On 12 December 1976, all experimental pastures were fertilized with 14.8 kg of P and 55.8 kg of K/ha.


Treatment Application


Main Treatment I. Control

On the control treatments, only different grazing management systems were imposed.

Main Treatment 2. Molasses Sprayed on the Pastures

On the rotationally grazed pastures, molasses was sprayed on the standing forage before each grazing period to increase the palatability of the smutgrass. In the continuously grazed treatments, molasses was sprayed at 2-week intervals. In each molasses treatment it was applied immediately prior to introducing the animals to the pastures.

Molasses was diluted in equal parts of water and sprayed on the pastures at the rate of 320 liters/ha. Molasses before diluting with water weighed 1.34 kg/liters. Spraying was carried out with a tractor




43




mounted, boomless sprayer. A spray pump was operated at 2.5 kg/cm2 pressure and driven at 4 km/hour.

Main Treatment 3. Dalapon Applied in the Spring Followed by Mowing and N Fertilization

On 28 April, when the smutgrass plants were growing quite vigorously, dalapon (2,2-dichloropropionic acid) at the rate of 5.6 kg/ha was sprayed on the pastures. It was applied in water at 277 liters/ha with a tractor-mounted sprayer. On 19 May, three weeks after the dalapon application, all pastures were mowed to a height of 6 to 7 cm with flail mower. On 7 June, 56 kg N/ha was applied using a Gandy fertilizer spreader. Ammonium sulfate (21% N) was used as the source of N.

Dalapon Applied in the Spring Followed by Burning and N Fertilization

Dalapon at the rate of 5.6 kg/ha was applied to the pastures on 28 April, 1976. On 1 June, the top kill of smutgrass plants became evident due to the application of dalapon. The pastures were then burned. One week later the pastures were fertilized with 56 kg N/ha as ammonium sulfate.

Dalapon Applied in the Fall Followed by Burning and N Fertilization

The fall application of dalapon was completed on 4 October,

when the smutgrass was still actively growing. The rate and method of application was the same.as used for the spring treatment (5.6 kg/ha). The pastures were burned on 15 February 1977. Dalapon Applied in the Fall Followed by Mowing and N Fertilization

In this treatment, dalapon (5.6 kg/ha) application was carried

out on 4 October 1976. The pastures were mowed to a 6 to 7-cm stubble on 10 January 1977.





44




Ground Hawg in the Spring

On 6 May, the Ground Hawg was used to till the pastures. Depth of cultivation was 12 to 13 cm below the soil surface. The cutting blades broke up and lossened the surface soil and destroyed the shallow rooted plants. This operation usually destroys the canopy and reseeding is necessary unless a good supply of viable seed exists in the soil.

Ground Hawg in the Spring Followed by N Fertilization

On 6 May, the Ground Hawg was passed over the pastures. On 5 June, 56 kg/ha of N fertilizer was applied as ammonium sulfate. Ground Hawg in the Spring Followed by Seeding of Bahiagrass and N Fertilization

The tillage operation with Ground Hawg was carried out on 6 May. On the following day bahiagrass, at the rate of 16 kg/ha, was sown, covered and packed with a cultipacker seeder. One month later a N application at the rate of 56 kg N/ha was made using a Gandy fertilizer spreader. Due to high temperatures and lack of sufficient rain, a poor stand resulted from this seeding. Bahiagrass was seeded again on 10 August at the same rate, after tilling the soil again with the Ground Hawg.

Ground Hawg in the Fall Followed by Seeding of Ryegrass and N Fertilization.

On 8 October, Ground Hawg tillage operation was carried out. On the following day 'Tetrablend' ryegrass was seeded at the rate of 16 kg/ha. One month later the pastures were fertilized with 56 kg N/ha. Pastures seeded to ryegrass in early October were fully sodded by midDecember of 1976.




45



Burning in the Fall and Dalapon Applied in the Spring Followed by Mowing and N Fertilization

Burning was carried out later than anticipated due to heavy rainfall during the fall season.


Measurements


Smutgrass Ground Cover

In order to study the effect of the treatments on the change of smutgrass ground cover, 4 non-randomly selected I m2 permanent quadrats were set up in each pasture. The area for each quadrat was chosen in such a way that all quadrats contained similar smutgrass ground cover before starting the experiment. A wooden stake was driven in each of 2 predetermined corners of the quadrat.

The equipment used to determine the smutgrass ground cover consisted of a 2.0 m x 0.5 m ( m 2) frame constructed of aluminum (Fig. 3). It was subdivided by thin aluminum bars in order to give 100 equal squares each 10 cm x 10 cm, to facilitate the estimation of plant ground cover. The frame was supported above the canopy by four legs, each equipped with a lock screw which allowed the height of the frame to vary from 20 to 30 cm above the ground.

The quadrat charting method was used to evaluate the change in smutgrass ground cover. Ground cover was estimated in early April before applying the treatments and again late in October 1976. The frame was placed over each permanent quadrat and outlines of smutgrass clumps were drawn on acetate transparent sheets (Fig, 4) at a scale of I mm on the transparency to 10 mm on the frame. Later, areas




46










































Fig. 3. The I m2 frame used to determine the smutgrass ground
cover.




47




























4 2 3 5 6 7 8i9401 2 34T56 7 8 9 V ii- i Fig. 4. Outlines of smutgrass clumps on acetate transparent
sheets.




48




outlined on the transparency were inked (Fig. 5) with india ink and the areas were determined, by passing the transparency through an electronic leaf area meter.

The change in ground cover of smutgrass for each pasture was determined by the difference between the ground cover estimates made in October and April 1976, and are expressed as percentage units of change.

Dry Matter Determination After Grazing

Residual dry matter/ha was estimated after each grazing period on the rotationally grazed pastures. On the continuously grazed pasture residual dry matter was estimated every 28 days. All estimates were made by a double sampling procedure, using a Neal Electronic Model 18-200 Herbage Meter (Fig. 6).

In this procedure the capacitance meter was first adjusted to zero, with the probes standing on bare soil in the field where measurements were to be made. The instrument was then taken to the pasture for sampling. Twenty random readings were taken in each pasture. From these twenty, 4 were randomly selected to be cut. A metal frame measuring 0.186 m2 (2 sq. ft,), which fit around the probes, was placed in position. The capacitance meter was then removed and the forage within the frame was cut to ground level using a grass clipper. The forage cut was gathered into a paper bag and ovendried at 700 C for 24 hours and weighed.

The capacitance meter readings and dry weight yields/0.186 m2 were used as independent and dependent variables respectively, in a regression model forced through the origin. The regression coefficient was obtained from the equation:




49











































Fig. 5. Black circular areas represent the outlines of smutgrass
clumps after having been inked.





50

















A--E



























Fig. 6. View of the capacitance meter used in the doublesampling procedure.





51




Y=b X

whe re:

Y = estimated dry matter, metric tons/ha

b = regression coefficient of capacitance meter reading on dry

matter yield of harvested samples.

X = value of capacitance meter reading.

The forage dry matter left after each grazing period was estimated, by the following formula:



RDMA =
2

where:

RDMA = residual dry matter adjusted

Y = mean dry matter of the harvested samples (4 samples/pasture/

rotation cycle)

b = regression coefficient

X = mean of the capacitance meter reading of the unharvested

samples (16 readings/pasture/rotation cycle).


Parameters Measured


The response of the pasture to the main treatments and grazing management was measured in terms of the following response variables:

1. Change in smutgrass ground cover expressed as percentage

units

2. Stocking rate

a. Animals/ha/day

b. Liveweight/ha/day





52




For each main treatment the response variables were fitted using a second order polynomial of the type:
2 2
Y = b + b X + b X + b X + b X + b XX
o 1 1 2 2 11 1 22 x 12 1 2 where:

Y = estimated response

X1 = length of rotation cycle (days)

X2 = grazing pressure (metric tons/ha of RDMA)

b = intercept

b. = linear regression coefficient

b.. = quadratic regression coefficient
II
b.. = regression coefficient of the interaction


The following analysis of variance was performed in order to test the fit of the model: Analysis of Variance



Source of Variance df MS F Total 21 Due to regression 5

RC

GP RC x RC 1

GP x GP

RC x GP I Residual 16

Lack of fit 7 Error 9




53


The estimation of the error was generated from the sums of

squares of the nine replications of the axial, center, and corner points. The lack of fit sum of squares was calculated by difference, subtracting the error sum of square from the residual sum of square.

The statistical analysis was done using the function GLM of the Statistical Analysis System (Barr et al., 1976) and the APL function iwas used to locate the stationary point.














RESULTS AND DISCUSSION


Results and discussion will be presented only for the control and sprayed molasses treatments. The spring applied dalapon retarded and/or killed both the smutgrass and the bahiagrass to such an extent that regrowth of bahiagrass was not enough to allow grazing during the 1976 growing season. Treatments 3 through 11 (see page 35) were grazed from time to time but no data were collected during the first year of this multiyear study.

The following notations will be used on the presentation of the results and discussion.

RDM = projected residual dry matter

RDMH = residual dry matter based on harvested samples only

RDMA = residual dry matter adjusted to forage meter readings


Residual Dry Matter


Control Treatment

Projected residual dry matters (RDM) left after grazing by combination of rotation cycle and grazing pressure on the control treatment are presented in Table 4. Also, included in the table are the seasonal averages of residual dry matters based on harvested samples (RDMH) and residual dry matters based on the harvested samples and adjusted to meter readings (RDMA).





54





55


Table 4. Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the
control treatment.


Combinations RDMH RDMA

RC Projected RDM X SDM CV X SD b CV r


days --------metric tons/ha------- % -metric tons/ha- %

56 0.50.2 .785 .221 29 .716 .147 .050 19 .84 56 0.50.2 .838 .301 36 .837 .204 .047 24 .90 42 2.90.2 2.893 .660 23 2.841 .362 .049 12 .87 14 1.30.2 1.390 .447 32 1.339 .283 .045 20 .81 28 0.50.2 .733 .220 30 .743 .139 .051 19 .81 28 0.50.2 .748 .356 47 .739 .126 .049 17 .94 0 3.70.2 3.594 .871 24 3.579 .419 .046 12 .90 0 3.70.2 3.577 1.053 29 3.722 .480 .046 13 .91 0 2.10.2 2.193 .801 36 2.088 .377 .045 17 .88 0 2.10.2 2.258 .778 34 2.449 .460 .046 20 .82 14 2.90.2 3.101 .840 27 3.059 .449 .045 14 .86 56 3.70.2 3.727 .712 19 3.419 .411 .046 11 .92 56 3.70.2 3.628 .862 24 3.665 .315 .047 9 .93 42 1.30.2 1.436 .519 36 1.437 .260 .043 18 .91 28 3.70.2 3.769 1.008 26 3.729 .455 .047 12 .90 28 3.70.2 3.652 1.086 29 3.795 .376 .047 10 .94 28 2.10.2 2.135 .697 33 2.125 .331 .048 15 .88 28 2.10.2 2.221 .789 35 2.205 .290 .048 13 .93 0 0.50.2 .783 .282 36 .789 .144 .050 18 .88 0 0.50.2 .779 .259 33 .813 .149 .045 19 .88 56 2.10.2 2.135 .819 38 2.092 .250 .046 12 .95 56 2.10.2 2.212 .804 36 2.072 .392 .042 18 .95


RC = Rotation cycle
X = mean
SDM = Standard deviation from mean SDR = Standard deviation from regression




56



The estimated RDMH in almost all combinations of rotation cycle and grazing pressure were within the range of projected residual dry matter/ha. The values of the RDMH ranged from 0,733 to 3,769 metric tons/ha. Residues greater than those projected occurred in all combinations in which grazing pressure was projected at 0.5 metric tons/ha of residual dry matter left after grazing. The relative high residues in this case may have been due in part to high dung spot concentration on the pastures as a consequence of the heavy stocking rate required to have a residue of 0.5 metric tons/ha. /This observation is in agreement with Greenhalgh and Reid (1969), who indicated that fouling of less than 3% of the surface of a ryegrass pasture with dung resulted in over 20% rejection at a heavy grazing intensity.

The coefficients of variation for RDMH ranged from 19 to 47%. The coefficients of variation increased with decreasing amount of residue left after grazing (Table 4), This is apparently due to the lower mean value for the residual dry matter and the very low variation in the standard deviation among the different levels of grazing pressure,

Residual dry matter based on harvested samples and adjusted to

meter readings (RDMA) tended to be lower than the RDMH. The RDMA ranged from 0.716 to 3.795 metric tons/ha. The data show that like RDMH, the values of RDMA for grazing pressure equal to 0.5 metric tons/ha of residue, are also outside of the projected ranges (RDM).

The regression coefficients (b) were always positive and varied from 0.043 to 0.051 metric tons/ha per unit of forage meter reading. Although the regression coefficients were similar, heavier grazing pressure tended to give larger regression coefficients.




57



The standard deviations of the regression of RDMA ranged from

0.144 to 0.480 metric tons/ha. It may be noted that larger standard deviations from regression occurred at lighter grazing pressures (high RDM) than at heavier grazing pressures. The lower standard deviations from regression observed at heavier grazing pressures were probably due to a lower residual dry matter yield (RDMA),

All correlation coefficients (r) were highly significant (P<0,01), ranging from 0.81 to 0.95 (Table 4). However, the correlation coefficients were generally slightly lower at heavier grazing pressures.

The most noticeable effect of adjusting the harvested sample to meter reading was to reduce the coefficients of variation. The coefficients of variation of RDMA ranged from 9 to 24%, which means a decrease in sample variation of about 40% when compared to the variation of the RDMH.

Molasses Sprayed Treatment

Table 5 presents the seasonal average residual dry matters left

after grazing by combinations of rotation cycle and grazing pressure on the molasses sprayed treatment.

As in the case for control treatment, most of the RDMH were within the range of projected residual dry matter/ha (RDM/ha). They ranged from 0.734 to 3.931 metric tons/ha. However, exceptions occurred for all grazing management systems with 0,5 metric tons/ha of residual dry matter, which were greater than the projected RDM left after grazing. Similarly to the control treatment, the high frequency of dung spots on the pastures seems to have had an effect upon the amount of RDM left.




58




Table 5. Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the
molasses sprayed treatment.



Combinations RDMH RDMA

RC Projected RDM X SDM CV X SDR b CV r


days -----metric tons/ha------- % --metric tons/ha- %

56 0.50.2 .762 .255 33 .803 .137 .051 18 .89 56 0.50.2 .853 .338 40 .814 .149 .043 17 .95 42 2.90.2 3.164 1.066 34 2.970 .770 .048 24 .74 14 1.30.2 1.367 .469 34 1.282 .239 .043 17 .89 28 0.50.2 .734 .228 31 .712 .162 .047 22 .82 28 0.50.2 .750 .220 29 .747 .103 .048 14 .88 0 3.70.2 3.931 1.012 25 3.915 .291 .047 7 .96 0 3.70.2 3.841 .970 25 3.819 .378 .046 10 .92 0 2.10.2 2.123 .811 38 2.196 .553 .048 26 .75 0 2.10.2 2.110 .623 29 2.136 .299 .045 14 .88 14 2.90.2 3.057 1.021 33 2.988 .564 .050 18 .83 56 3.70.2 3.722 1.038 28 3.692 .282 .046 7 .97 56 3.70.2 3.902 .737 19 3.851 .338 .047 8 .93 42 1.30.2 1.240 .491 39 1.233 .251 .045 20 .86 28 3.70.2 3.726 .977 26 3.732 .384 .046 10 .92 28 3.70.2 3.664 .723 20 3.516 .549 .045 15 .77 28 2.10.2 2.151 .743 34 2.194 .300 .047 14 .92 28 2.10.2 2.283 .861 38 2.274 .518 .046 23 .80 0 0.50.2 .926 .336 36 .932 .197 .040 21 .88 0 0.50.2 .815 .325 39 .804 .178 .049 22 .93 56 2.10.2 2.094 .815 39 2.033 .279 .048 13 .14 56 2.10.2 2.031 .904 44 1.997 .183 .050 9 .98

RC = Rotation cycle.
X = Mean.
SDM = Standard deviation from mean. SDR = Standard deviation from regression.




59




The coefficients of variation for RDMH are also given in Table 5.

They varied from 19 to 44%, and were higher for heavier grazing pressure (low RDM).

For most of the combinations of rotation cycle and grazing pressures, RDMA were slightly lower than the RDMH.

The regression coefficients ranged from 0.040 to 0.051, As in the control treatment, heavier grazing pressures tended to result in larger regression coefficients. The standard deviation of the regressions of RDMA ranged from 0.137 to 0.770, and tended to decrease with heavier grazing pressures (low RDM).

In all combinations of rotation cycle and grazing pressure the

correlation coefficients of meter reading on residual dry matter were highly significant (P<0.01) and ranged from 0,74 to 0.98, Lower correlation coefficients for heavier grazing pressures were evident (Table 5).


Changes in Smutgrass Ground Cover


Control Treatment

Changes in smutgrass ground cover for the different combinations of length of rotation cycle and grazing pressure from April to October 1976, are given in Table 6. In the control treatment, the change in smutgrass ground cover ranged from -22.2 to 33.8 percentage units.

The main effects of grazing pressure and length of rotation cycle on the change in smutgrass ground cover are presented in Tables 7 and 8, respectively. Changes in smutgrass ground cover indicated an increase (P<0.05) with decreasing grazing pressure (low RDM). On the





60




Table 6. Observed change in smutgrass ground cover for the different
combinations of length of rotation cycle and grazing pressure
on the control treatment, from April to October 1976


Length of rotation Grazing pressure Observed change in
cycle RDMA' smutgrass ground cover

------days------- -metric tons/ha- -percentage units56 0.771 2.8 56 0.837 9.1 42 2.841 3.8 14 1.339 4.6 28 0.743 5.0 28 0.738 -22.2 0 3.579 33.8 0 3.722 11.7 0 2.088 12.2 0 2.449 3.4 14 3.059 4.2 56 3.419 15.6 56 3.665 27.4 42 1.437 21.7 28 3.729 20.8 28 3.795 5.1 28 2.125 4.9 28 2.206 16.3 0 0.789 4.9 0 0.814 0.2 56 2.092 6.6 56 2.073 6.1


Residual dry matter adjusted.




61




Table 7. Main effect of grazing pressure on the change in smutgrass
ground cover on the control treatment,



Grazing pressure Observed change in
RDMt smutgrass ground cover


-metric tons/ha ---percentage units--0.5 1.8 1.3 8.8 2.1 12.7 2.9 0.2 3.7 19.1


Projected residual dry matter.






Table 8. Main effect of rotation cycle on the change in smutgrass
ground cover on control treatment.



Length of rotation Observed change in
cycle smutgrass ground cover


-------days------- ---percentage units--0 8.2 14 0,1 28 4,9 42 8.5 56 7.0




62




other hand, changes in smutgrass ground cover were not influenced by the length of the rotation cycle.

The fitted equation for change in smutgrass ground cover on control treatment is presented in Table 9. It shows a significant (P<0.05) linear response in change of smutgrass ground cover as a result of changes in grazing pressure. The other terms of the equation were not significant.

The calculated stationary point for change in smutgrass ground

cover on the control treatment was found to be located at 11 days for length of rotation cycle and 0.3 metric tons/ha for the adjusted residual dry matter, with a response at this point of -2.4 percentage units (Table 10). The stationary point is outside of the experimental area with relation to grazing pressure, since the heaviest grazing pressure studied was 0.5 metric tons/ha. However, the response at the stationary point (-2.4 percentage units) is between the minimum and maximum values found for change in smutgrass ground cover for control treatment (Table 6).

The contours of predicted change in smutgrass ground cover on the control treatment are presented in Fig. 7. It can be seen from Fig. 7 that smutgrass ground cover decreases with heavier grazing pressures (low RDM). It is evident from the contours that very small changes in smutgrass ground cover occur due to the effect of length of the rotation cycle. The contours emphasize the importance of grazing pressure since there is little effect due to length of rotation cycle on smutgrass ground cover.






63







X a X

I X 0 r- u 0 0 x

o .0 0 X %D
a N *

1 (',4 'I ',-
o 0 c0 0

N 0 N4 Co
1 + -.- 0o
C 0 0 o


+ 0 j. + +

a. co 0o 0 r _I- -T
SN 0 0) 0 0


o N O (' o + r + (0 O N N ON

u 0 u 0
+ w 0 + cc 0
0 o


So I
0 N 0 0 0 .0 co I -l + m0 aa % A o L 0 -, Lt\ 0

O "" II N. I 0 0 o* % (n c L m C -T N N 0 L O 4 I "a
S 0 0 0 c 0 I '0 0


N N 0 *4 *

C O O I E O .a 0
- 1 I 0
- + N I L 4-' <>- N < >- CO 0 C
00N 0
O ** II ** 1 ,


C 0 1 0 11 4O u ** 0 0 CU C C v 0 O > O 0 -v0 (1

& C) t f 1 C

(n -C L -C -a)
0N 0 0 L
LL c) 3 C 0 0)


L -) .. I 0 L a. aI) I

(a NNOO S3SSV1OW u aC







64










m 0n 1


.
no

3 0..


..J












0
9 --+











IA
V C LE t C > 4J O m an ) a








Su
L 0

C N 4- a

0E 0 m 0
& *- 0 I. .-c c 4 0 Q.


E 0 n


0 0 4L44 EQ oi L















(0 u 0
O U C
C LJ O E L 0
> in -1 ..

EN C EQ V L i I I .CI > i
- C 0 0) O C O c I O L- 0 >
*- c a Eo In -- 0 0 > IE EQ ( 0 0 0- u4 < L


4 1- E m o t- 0 I a 0


n 3 x I


EQ C .
C V C 4 *- a a C
* 44 E N C E i OC IA 4 EQ EoQ
) V O L .C C *
E O Io a* eo < .

V C
I O O ( 0. -. t F- 0T T 0*





















E 4







0 L.






-C
O





4-)
c



0



4-1
r
















Cl
U















L
U




Cu



(1

L


-1

E in

c
C






u


4
0 Vn
L

0 4-I
CD








0




L
C,




66








C'









cr)
xf
LI /







co
C-!,










0
CC3 ... -* .
at I 0















. ... "

+C
a
~I




o I













I
o m a m .










a








" a=rr













Th OL 't Ob'2 "1'. OE' l 000





67




A general effect of increasing the grazing pressure (low RDM) wa to reduce in smutgrass ground cover in the pasture. The results obtained in this experiment indicate that heavy grazing pressures are needed to control smutgrass. Similar results were reported by Campbell and Beale (1973), who reported that increased grazing pressure throughout the growing season reduced the proportion of barley grass in the pasture. In this study it was observed that smutgrass was more attractive to animals in its early growth stages than in later stages, especially after flowering. This was consistent with the findings of Riewe et al. (1975b) who reported that grazing preference for smutgrass was higher during early spring than after May. Molasses Sprayed Treatment

Table II presents the changes in smutgrass ground cover for the

different combinations of length of rotation cycle and grazing pressure on the molasses sprayed treatment, from April to October. The change in smutgrass ground cover varied from -20.0 to 31.8 percentage units.

The main effects of grazing pressure and length of rotation cycle upon the change in smutgrass ground cover are shown in Tables 12 and 13, respectively. Smutgrass ground cover increased linearly (P<0.01) with decreasing grazing pressure (high ROM). It was not affected (P>O.05) by the length of rotation cycle but smutgras increased slightly as the length of rotation cycle increased.

Smutgrass ground cover on the sprayed molasses treatment responded (P<0.01) to grazing pressure (Table 9). The other terms of the model were not significant (P>0.05).




68




Table 11. Observed change in smutgrass ground cover for the different
combinations of rotation cycle and grazing pressure on the
molasses sprayed treatment, from April to October 1976.


Length of rotation Grazing pressure Observed change in
cycle RDMAt smutgrass ground cover

------days------- -metric tons/ha- -percentage units56 0.803 6.4 56 0.814 6.7 42 2.970 9.2 14 1.282 8.2 28 0.712 9.3 28 0.747 9.8 0 3.915 18.0
0 3.819 9.4 O 2.196 15.8 O 2.136 3.9 14 2.988 0.1 56 3.692 31.8 56 3.851 29.3 42 1.233 3.7 28 3.732 15.6 28 3.516 12,9 28 2.194 18,9 28 2.274 1,7 0 0.932 -20,0 0 0.804 5.5 56 2.033 23.2 56 1.997 8.9


Residual dry matter adjusted.





69




Table .12. Main effect of grazing pressure on the change in smutgrass
ground cover on the 4ent--rol treatment.


Grazing pressure Observed change in
RDMt smutgrass ground cover


-metric tons/ha- ---percentage units--0.5 5.2 1.3 3.9 2.1 5.8 2.9 3.7 3.7 19.5


Projected residual dry matter.







Table 13. Main effect of rotation cycle on the change in smutgrass
ground cover on sprayed molasses treatment



Length of rotation Observed change in
cycle smutgrass ground cover


-------days------- ---percentage units--0 -2.7 14 3.6 249 5.0 42 2.5 56 11.4




70




The stationary point for change in smutgrass ground cover on the molasses sprayed treatment (Table 10) was determined to be at 5 days for rotation cycle and 0.4 metric tons/ha of RDMA. The response at that point was found to be -8.5 percentage units. It can be seen that the stationary point is very close to the experimental region and that the response (-8.5 percentage units) is between the lower and higher values observed for change in smutgrass ground cover (Table 11).

The contours of change in percentage units of smutgrass ground

cover on the molasses sprayed treatment are plotted in Fig. 8. It can be observed that significant decreases in smutgrass ground cover may be achieved with heavier grazing pressures. Another trend is that for any level of grazing pressure, smutgrass ground cover increased with an increase in length of rotation cycle. However, a much greater decrease may be attained with heavier grazing pressures than with shorter length of rotation cycle. The contours suggest a fairly rapid reduction in smutgrass ground cover with a combination of heavy grazing pressure and short rotation cycle.

Statistical analysis revealed no differences (P>0.05) between the response surfaces for the control and molasses treatments. In this experiment, initially the animals showed a preference for the sprayed pastures; however, the preference did not persist for more than 24 hours. This fact appears to be similar to that reported by O!Bryan (1960) in which the preference for the sprayed pastures did not persist for more than 2 days. The lack of response to the molasses sprayed treatment may have been due to the low rate of molasses used. This observation agrees with those obtained by Bishop (1959) using a similar rate of molasses.





















-u










0
r









E


Q)




I


4








1U






0
L.










L
D
O
L





4


E

C
IA



u
01
C

U
-c


O
4-1
0


- O
4E
Lv


CO







01 Li,




72







C






a /*


00















C,.


C--)
a.. a












0 C LLI
C.








c


0C





















h O "0
II .











h OL E0 l' L 10 UWO SHSGH' W 7




73




Animals/ha/day


Control Treatment

Observed number of animals/ha/day for the different combinations of rotation cycle and grazing pressure on the control treatment are shown in Table 14. Animals/ha/day varied from 4.3 to 28.7.

The main effects of grazing pressure and length of rotation cycle on animals/ha/day are presented in Tables 15 and 16, respectively. Number of animals/ha/day declined linearly (P<0,0l) with decreasing levels of grazing pressure (increasing RDM). Number of animals/ha/day was not affected by length of rotation cycle.

There was both a linear and quadratic effect (P<0.01) of grazing pressure upon the number of animals/ha/day on the control treatment (Table 9 and Fig. 9).

The stationary point was located at 10 days for length of rotation cycle and 3.2 metric tons/ha of RDMA. The response of the stationary point at this combination is equal to 6.5 animals/ha/day, The stationary point in this case is located within the experimental region; however, it is closer to the lower observed values for animals/ha/day (Table 14).

Figure 9 represents the contours of the response of number of

animals/ha/day on the control treatment. Under this treatment, number of animals/ha/day increased with increasing grazing pressure (decreasing RDM). At heavier grazing pressure, length of rotation cycle has no effect upon animals/ha/day,




74



Table 14. Observed animals per hectare per day for the different
combinations of length of rotation cycle and grazing
pressure on the control treatment.



Length of rotation Grazing pressure Observed
cycle RDMAt animals/ha/day


-------days------ -metric tons/ha56 0.771 20.1 56 0.837 20.5 42 2.841 7.4 14 1.339 14.6 28 0.743 28.7 28 0.738 23.5 0 3.579 6.6 0 3.722 7.3 0 2.088 9.5 0 2.449 11.2 14 3.059 9.6 56 3.419 5.8 56 3.665 7.1 42 1.437 15.3 28 3.729 4.3 28 3.795 5.5 28 2.125 6.7 28 2.206 7.2
0 0.789 20.7 0 0.814 20.9
56 2.092 6.7 56 2.073 7.9

Residual dry matter adjusted.




75



Table 15. Main effect of grazing pressure on animals per hectare per
day on the control treatment.



Grazing pressure Observed
RDM' animals/ha/day


-metric tons/ha0.5 22.4 1.3 14.9 2.1 8.2 2.9 *8.5
3.7 6.1


Projected residual dry matter.






Table 16. Main effect of rotation cycle on animals per hectare per
day on the control treatment.




Length of rotation Observed
cycle an imal s/ha/day


-------days------0 12.7 14 12.1 28 12.6 42 11.4 56 11.3












































C

4J


L


0
L

O
C
0
U


41

C
0














.
4










0
L an











'4



C
o





77








O




I 1








0












+ I Ca







LL









o ci




o




0
INC'
e






















C .. .... I


err a IO I b Ot1Ob20'2G'1O'




78




Molasses Sprayed Treatment

The number of animals/ha/day for the different combinations of

length of rotation cycle and grazing pressure on sprayed molasses treatment is presented in Table 17. Animals/ha/day ranged from 3.9 to 22.8. The main effects of grazing pressure and length of rotation cycle on animals/ha/day are listed in Tables 18 and 19, respectively. The number of animals/ha/day declined (P<0.01) with decreasing grazing pressure. Number of animals/ha/day was not affected by length of rotation cycle.

The fitted equation (Table 9 and Fig. 10) indicates that there was a quadratic effect (P<0.01) upon animals/ha/day due to grazing pressure and that there was no interaction between grazing pressure and length of rotation cycle (P>0.05). The number of animals/ha/day was greatly affected only by grazing pressure and there was no response to length of rotation cycle.

The stationary point for animals/ha/day was found to be located at 60 days and 3.4 metric tons/ha of RDMA (Table 10), with a response at this point of 5.8 animals/ha/day. The stationary point is located outside of the experimental region with relation to length of rotation cycle, because 56 days was the longest cycle studied, However, it is well within the experimental region of grazing pressure used in the experiment, The response at the stationary point is within the observed values for the different combinations of grazing pressure and length of rotation cycle (Table 17).

The contours of predicted number of animals/ha/day on the molasses sprayed treatment are presented in Fig. 10, The contours show that




79



Table 17,. Observed animals per hectare per day for the different
combinations of rotation cycle and grazing pressure on
the sprayed molasses treatment.



Length of rotation Grazing pressure Observed
cycle RDMAt animals/ha/day


-------days------- -metric tons/ha56 0.803 18.0 56 0.814 17.6 42 2.970 7.8 14 1.282 14.9 28 0.712 21.5 28 0.747 22.8
0 3.915 6.2 0 3.819 7.1 0 2.196 8.5 0 2.136 10.1 14 2.988 8.7 56 3.692 6.1 56 3.851 5.2 42 1.233 12.0 28 3.732 3.9 28 3.516 5.1
28 2.194 5.4 28 2.274 6,8 0 0.932 22.3 0 0.804 20,2 56 2.033 10.1 56 1.997 10,3


Residual dry matter adjusted.





80




Table 18. Main effect of grazing pressure on animals per hectare per
day on the sprayed molasses treatment.



Grazing pressure Observed
RDMt animals/ha/day


-metric tons/ha0.5 20.4 1.3 13.5 2.1 8.5 2.9 8.3 3.7 5.6


Projected residual dry matter.






Table 19. Main effect of rotation cycle on animals per hectare per day
on the sprayed molasses treatment.



Length of rotation Observed
cycle animals/ha/day


-------days------0 12.4 14 11.8 28 10.9 42 9.9 56 11.2




81




animals/ha/day increased with combination of heavier grazing pressures and shorter length of rotation cycle. However, Fig. 10 indicates that very small decreases in number of animals/ha/day is obtained however, if we maintain a constant grazing pressure and increase the length of the rotation cycle.


Liveweight/ha/day


Control Treatment

Liveweight/ha/day for the different combinations of length of rotation cycle and grazing pressure on the control treatment, ranged from.

1.437 to 9.084 metric tons/ha/day (Table 20).

Liveweight/ha/day shows a decrease (P<0,01) with decreasing (Table 21) grazing pressure (high RDM). The amount of liveweight/ha/day was not influenced by length of rotation cycle (Table 22).

The fitted equation for liveweight/ha/day on the control treatment

shows a highly significant (P<0.01) linear and quadratic effect of grazing pressure. The other terms of the equation were not significant (Table 9 and Fig. 11).

The computed stationary point was detected at 31 days for length of rotation cycle and 3.3 metric tons/ha of residual dry matter (Table 10), with a response at that point of 2.2 metric tons/ha/day of liveweight. The stationary point for liveweight/ha/day on the control treatment is located within the experimental region and is close to the lower values observed in the experiment (Table 20).

Contours of response of liveweight/ha/day are presented in Fig. 11. The contours show that amount of liveweight/ha/day increases with heavier































r


E 4J tu






0 an









it
L








4-j 4, C.

CU




L

a



0





L.

Q.






C 4
o




0 4








Cm L.
0o




83







cnp















00
I 0 Lo I




(-)












o f



o


+















U .. H N, 3
L)>











- I








I...Io
co

0 a 1i I I










OICL W
* 1Jc] HoJO)~]j bU




84




Table 20. Observed liveweight per hectare per day for the different
combinations of rotation cycle and grazing pressure on the
control treatment.



Length of rotation Grazing pressure Observed
cycle RDMAt liveweight/ha/day


-------days------- -metric tons/ha --metric tons--56 0.771 5.987 56 0.837 6.889 42 2.841 2.512 14 1.339 5.061 28 0.743 8.627 28 0.738 9.084 0 3.579 2.209 0 3.722 2.519 0 2.088 3.196 0 2.449 3.532 14 3.059 3.104 56 3.419 2.361 56 3.665 2.340 42 1.437 5.172 28 3.729 1.437 28 3.795 1.772 28 2.125 2.291 28 2.206 2.506
0 0.789 7.105 0 0.814 6.917 56 2.092 2.890 56 2.073 4.039

Residual dry matter adjusted.




85



Table 21. Main effect of grazing pressure on liveweight per hectare
per day on the control treatment.



Grazing pressure Observed
RDMt liveweight/ha/day

-metric tons/ha- ---metric tons--0.5 7,435 1.3 5.116 2.1 3.075 2.9 2,808 3.7 2.106


Projected residual dry matter.







Table 22. Main effect of rotation cycle on liveweight per hectare
per day on the control treatment.



Length of rotation Observed
cycle liveweight/ha/day

-------days------- ---metric tons--0 4.246 14 4.082 28 4.286 42 3.842 56 4.084





86




grazing pressures (low RDM). The figure also points out that for a constant grazing pressure the liveweight/ha/day is not influenced by length of rotation cycle.

Molasses Sprayed Treatment

Liveweight/ha/day for the different combinations of length of rotation cycle and grazing pressure on the molasses sprayed treatment is listed in Table 23. The values ranged from 1.305 to 7.988 metric tons/ha.

Tables 24 and 25 present the effects of grazing pressure and length of rotation cycle, respectively, upon liveweight/ha/day. Liveweight/ha/day declined linearly (P<0.01) with decreasing grazing pressure. Liveweight/ha/day shows a tendency to decrease with increasing length of rotation cycle (Table 25).

The fitted equation (Table 9 and Fig. 12) for liveweight/ha/day on the molasses sprayed treatment shows a significant (P<0.01) linear effect of grazing pressure on liveweight. The interaction between grazing pressure and length of rotation cycle was also significant (P<0.1) for liveweight/ha/day.

The stationary point was located at 42 days for length of rotation cycle and 3.7 metric tons/ha of residual dry matter. At that point the liveweight/ha/day was found to be equal to 1.9 metric tons/ha. The stationary point is located inside of the experimental area and is approximately in the middle of the values observed for liveweight/ ha/day (Table 23).

The contours obtained for liveweight/ha/day on the sprayed molasses treatment is shown in Fig,. 12. They indicate that liveweight/ha/day

























4.



c








4-J

0

4-J
c
E















L-i

0


0





88







C)


IC)



I


o o





0 +I
N




O






o

o I I I


0
o a

a





















II -f.a






I 0






(UW'd H/S01 93d 'ZUUO o&




89




Table 23. Observed liveweight per hectare per day for the different
combinations of rotation cycle and grazing pressure on the
sprayed molasses treatment.


Length of rotation Grazing pressure Observed
cycle RDMAt liveweight/ha/day


-------days------- -metric tons/ha- --metric tons-56 0.803 6.102 56 0.814 6.144 42 2.970 2.814 14 1.282 4.938 28 0.712 7.385 28 0.747 7.988
O 3.915 2.044 O 3.819 2.375 0 2.196 2.793 0 2.136 3.478 14 2.988 3.081 56 3.692 1.973 56 3.851 1.932 42 1.233 4.103 28 3.732 1.305 28 3.516 1.844 28 2.194 1.937 28 2.274 2.342
0 0.932 7.414 0 0.804 7.005 56 2.033 3.432 56 1.997 3.496


t
Residual dry matter adjusted.






90




Table 24. Main effect of grazing pressure on liveweight per hectare
per day on the sprayed molasses treatment.



Grazing pressure Observed
RDMt liveweight/ha/day

-metric tons/ha- ---metric tons--0.5 7.006 1.3 4.520 2.1 2.912 2.9 2.916 3.7 1.912


Projected residual matter.






Table 25. Main effect of rotation cycle on liveweight per hectare per
day on the sprayed molasses treatment.



Length of rotation Observed
cycle liveweight/ha/day


-------days------- ---metric tons--0 4.184 14 4.009 28 3.799 42 3.458 56 3.846




Full Text

PAGE 1

CHANGES IN SMUTGRASS ( Sporobolus poiret i i [Roem. and Schult.] Hitchc.) GROUND COVER INDUCED BY SPRAYING WITH MOLASSES AND GRAZING MANAGEMENT By LEON I DAS S. VALLE A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1977

PAGE 2

ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Dr, G. 0. Mott, Chairman of the Supervisory Committee, for his technical assistance in all phases of his graduate study including this research and the preparation of this manuscript. Sincere gratitude is extended to Dr. W. R. Ocumpauqh for his supervision and assistance during and after collection of data in the field. Appreciation is also extended to other members of the Supervisory Committee. Drs. J. H. Moore, J. H. Conrad, W. G. Blue, and W. L. Currey, for their valuable assistance during the preparation of this dissertation. The author also wishes to acknowledge Dr. R. C. Littel for his assistance in the statistical analysis. Sincere gratitude is extended to the Empresa Brasileira de Pesquis^ Agropecuaria (EMBRAPA) for providing financial assistance. The author is greatly indebted to Mr. Fred McKay for his help during the field work. Thanks are due to the personnel of the University of Florida Beef Research Unit for providing research facilities. The author would like to thank Mrs. Maria I. Cruz for typing the preliminary drafts and the final copy. The author is also grateful to the staff and fellow graduate students of the Agronomy Department for their encouragement and friendship.

PAGE 3

The author's children, Rodolfo, Rafael, and Romolo, must be thanked for suffering their father's neglect throughout the study period. However, the greatest debt of gratitude is due to the author' ^ wife, Julia, for her love and devotion, and for her sacrifice during the past three and a half vears of often frustrating 'labor. J

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGMENTS ii -v. LIST OF TABLES vi LIST OF FIGURES vi i t ABSTRACT fx INTRODUCTION 1 'If' ; REVIEW OF LITERATURE 3 Smutgrass 3 Smutgrass Control 6 ;' Weed Control by Grazing Management 9 Pasture Loss Due to Weeds 12 Benefits from Weed Control lit Measuring Botanical Composition 18 Estimating Forage Yield tq Spraying Pasture with Molasses 2h Effect of Grazing Management to Botanical Composition 2'^ MATERIALS AND METHODS 33 | General Description 33 Experimental Pasture Treatments 35 Grazing Management Factors 36 Experimental Design 37 Construction of Physical Facilities hO Experimental Analysis hO Treatment Application k7 ^ Main Treatment 1. Control k2 Main Treatment 2. Molasses Sprayed on the Pastures ^42 Main Treatment 3. Dalapon Applied in the Spring Followed by Mowing and N Fertilization Dalapon Applied in the Spring Followed by Burning and N Fert i 1 izat ion Dalapon Applied in the Fall Followed by Burning and N Fert i 1 i za t i on /43 Dalapon Applied in the Fall Followed by Mowing and N i Fert i 1 izat ion 1)3 Ground Hawg in the Spring kh 1 Ground Hawg in the Spring Followed by N Fertilization ii'l l Ground Hawg in the Spring Followed by Seeding of Bahiagrass and N Fertilization h>^ Ground Hawg in the Fall Followed by Seeding of Ryegrass and N Fert f 1 izat ion l^k Burning in the Fall and Dalapon Applied in the Spring Followed by Mowing and N Fertilization ^45 Iv

PAGE 5

Measurements Smutgrass Ground Cover 'i5 Dry Matter Determination After Grazing Parameters Measured 51 RESULTS AND DISCUSSION 5h Residual Dry Matter 5^ Control Treatment 5^ Molasses Sprayed Treatment 57 Changes in Smutgrass Ground Cover 59 Control Treatment 59 Molasses Sprayed Treatment 67 Animals/ha/day 73 Control Treatment 73 Molasses Sprayed Treatment 7? Liveweight/ha/day 81 Control Treatment 86 Molasses Sprayed Treatment 86 SUMMARY AND CONCLUSIONS 3h LITERATURE CITED 98 BIOGRAPHICAL SKETCH 106 V

PAGE 6

LIST OF TABLES Table Paq^' ] Monthly mean maximum and minimum temperatures and rainfall at the Beef Research Unit for 1976 Z Levels of length of rotation cycle and grazing pressure 37 3 Combinations of rotation cycle and grazing pressure of the Central Composite Design 39 k Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the control treatment 55 5 "Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the molasses sprayed treatment 58 6 Observed change in smutgrass ground cover for the different combinations of length of rotation cycle and grazing pressure on the control treatment, from April to October 1976 60 ^7 Main effect of grazing pressure on the change in smutgrass ground cover on the control treatment 61 ^8 Main effect of rotation cycle on the change in smutgrass ground cover on control treatment 61 9 Fitted equations for treatment and response variables 63 10 Values of the stationary point for treatment and response variables 6'( n Observed change in smutgrass ground cover for the different combinations of rotation cycle and grazing pressure on the molasses sprayed treatment, from April to October 1976 68 12 Main effect of grazing pressure on the change in smutgrass ground cover on the control treatment 69 13 Main effect of rotation cycle on the change in smutgrass ground cover on sprayed molasses treatment. 69 vl

PAGE 7

Table Page \k Observed animals per hectare per day for the different combinations of length of rotation cycle and grazing pressure on the control treatment 7i 15Main effect of grazing pressure on animals per hectare per day on the control treatment 75 16 Main effect of rotation cycle on animals per hectare per day on the control treatment 75 17 Observed animals per hectare per day for the different combinations of rotation cycle and grazing pressure on the sprayed molasses treatment 79 18 Main effect of grazing pressure on animals per -hectare per day on the sprayed molasses treatment 80 19 Main effect of rotation cycle on animals per hectare per day on the sprayed molasses treatment 80 20 Observed liveweight per hectare per day for the different combinations of rotation cycle and grazing pressure on the control treatment Qh 21 Main effect of grazing pressure on liveweight per hectare per day on the control treatment 85 22 Main effect of rotation cycle on liveweight per hectare per day on the control treatmemt 85 23 Observed liveweight per hectare per day for the different combinations of rotation cycle and grazing pressure on the sprayed molasses treatment,... 89 2k Main effect of grazing pressure on liveweight per hectare per day on the sprayed molasses treatment gO 25 Main effect of rotation cycle on liveweight per hectare per day on the sprayed molasses treatment 90 V i i

PAGE 8

LIST OF FIGURES Figure Page 1 Response surface design in two factors 13 combinations 38 2 Field layout of the experimental pastures k\ 2 3 The 1 m frame used to determine the smutgrass ground cover Outlines of smutgrass clumps on acetate transparent sheets kj 5 -Black circular areas represent the outlines of smutgrass clumps after having been inked kS 6 View of the capacitance meter used in the doublesampling procedure 50 7 Contours of change in smutgrass ground cover (percentage units) on the control treatment 66 8 Contours of change in smutgrass ground cover (percentage units) on the molasses sprayed treatment.. 72 9 Contours of animals per hectare per day on the control treatment 77 10 Contours of animals per hectare per day on the sprayed molasses treatment 83 11 Contours of liveweight per hectare per day (metric tons) on the control treatment 88 12 Contours of liveweight per hectare per day (metric tons) on the sprayed molasses treatment 93 v i i i

PAGE 9

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requiremen' for the Degree of Doctor of Philosophy CHANGES IN SMUTGRASS ( Sporobolus poiret i i [Roem. and Schult.j Hitchc.) GROUND COVER INDUCED BY SPRAYING WITH MOLASSES AND GRAZING MANAGEMENT By Leon i das S Va 1 1 e August 1977 Chairman: Gerald 0. Mott Major Department: Agronomy A grazing experiment was conducted at the Beef Research Unit, Gainesville, Florida, from April to November 1976. The experimental area was a pasture of 'Pensacola' bahiagrass (Paspalum notatum Flugge), smutgrass (Sporobol us po i ret i i [Roem. and Schult.] Hitchc). White clover ( Tr i f ol i um repens L ) was present in the spring. The initial ground cover of smutgrass was estimated to be between 40 and SQ% and bahiagrass was the dominant desirable specie. The purpose of this research was to determine the effects of different combinations of lengths of rotation grazing cycles and levels of grazing pressures on the control of smutgrass and to evaluate the effect of spraying molasses upon the palatability of smutgrass, compared to the' unsprayed treatment. Molasses was diluted in equal parts of water and sprayed at the rate of 320 liters/ha of molasses prior to introducing the animals. Grazing pressures measured in terms of residual dry matter left after grazing were 0.5, 1.3, 2.1, 2.9, and 3-7 metric tons/ha, and lengths of rotation cycles were 0. \k. 28, ^2, and 56 days.

PAGE 10

Thirteen different combinations of the 2 factors were arranged in a response surface design and superimposed upon the control and molasses sprayed treatments. The axial, center, and corner points were replicated twice, making a total of 22 pastures per main treatment. Smutgrass ground cover was estimated in early April before applying the treatments and again late in October, 1976. Spraying molasses to the pasture resulted in little increase of palatability of smutgrass when compared with the control treatment. The attractiveness of sprayed smutgrass was lost a few hours after spraying. In both treatments, smutgrass ground cover decreased with increasin: grazing pressure. Smutgrass ground cover was influenced very little by length of rotation cycle. Stocking rate, when considered as a response variable, expressed either as animals/ha/day or 1 i vewe i ght /ha/day increased with inciC'isinq grazing pressure. Length of rotation cycle had little effect.

PAGE 11

INTRODUCTION Weeds, whether herbaceous or woody, are undesirable in pastures becansr they compete with forage plant species for moisture, nutrients, and light. The extent of this competition depends as much on the growth habit and nature of the weed species as on their density and distribution. Also, many weeds harbor some of the worst crop insect pests and are alternate hosts to organisms causing crop diseases. Other weeds may be poisonous, causing reduced weight gains, lowered animal production or even death. Weed control is a major need in any program of management. The priniarv objective of weed control in pastures is to selectively manipulate the canopy to eliminate undesirable plant species while maximizing the production of desirable species. The complexity of weed control requires inputs from a wide variety of specialties spanning plant ecology and physiology, management, and economics. Recommended weed control measures must be implemented as long-term programs without damaging desirable plants. This concern is an integral part of the research effort and a vital consideration. Most weed problems on grazing lands result from man's activities with the introduction of exotic plants being a prime cause. For example, smutgrass (Sporobolus poiret i i [Roem. and Schult.] Hitchc.) was introduced into the United S-ates from tropical Asia. It is now a serious problem in many pastures of the southeast, particularly on the sandy soils of Florida. I

PAGE 12

2 Smutgrass, in pasture, spreads through enlargement of original plants and dissemination of seeds. However, density serious enough to threaten forage production may not be recognized until 5 to 10 years after being introduced into new areas. Control of smutgrass on heavily infested areas has long been recognized as an effective pasture improvement practice. During the past 15 years several methods have been developed to control this weed. Generally, these methods can be divided into three main categories: mechanical, chemical, and combinations of the two. Mechanical methods, such as mowing and cultivation are not effective in controlling smutgrass and previous chemical control methods have failed to give acceptable selective smutgrass control in established pastures. In the selection of the control method, not only should degree of smutgrass kill and cost be considered, but also the effect of the treatment upon the associated forage species. No information is available on the effect of grazing management systems alone or combined with other weed control practices upon smutgrass. Since many pastures in Florida are infested with smutgrass, ft is important to find out how much control of this weed can be reached when grazing management is used with other practices. The objectives of this study were to determine a) the effects of different combinations of lengths of rotation cycles and levels of grazing pressure on the control of smutgrass, and b) the applicability of spraying molasses on the pasture in an attempt to enhance the palatability of smutgrass and thereby increase smutgrass acceptability.

PAGE 13

REVIEW OF LITERATURE Smutgrass Sporobolus poiretfi [Roem. and Schult.] Hitchc. is named smutgrass because of a hyphomicete, B i po 1 a r i s ( Helm i nthospor ium ) ravenelii (Curt.) Schoemaker, (Luttrell, 1976), which often infects the panicles and at times is found in patches on the leaves (Currey and Mislevy, \37'^) It was observed that seeds produced during the spring were not heavily infested with ravenel i i wh i le the seeds produced in the fall were (Currey et al 1973) • Smutgrass has been described on numerous occasions (Hitchcock, 1936, 1950; Swallen, 1955). It is a deep rooted, caespitose perennial of the family Graminea. The grass is glabrous, summer growing, with culms erect, solitary or in small tufts, 30 to 100 cm high. The leaf blade is flat to subinvolute, rather firm, 2 to 5 mm wide at the base, elongated, and tapering to a firm point toward the end. The panicles are 15 to 30 cm long, usually spike-like, plumbeous, dense, the branches appressed floriferous to the base or nearly so. The spikelets are each 1.7 to 2 mm long; glumes are obtuse, the first 0.5 and the second from 0.5 to 0.7 mm long. The seeds are reddish when mature and may remain for some time, sticking to the panicle by the mucilaginous pericarp. Seeds are spread by water, and wind and by sticking to livestock. Smutfree seeds may remain on the panicle for some time or shatter quickly depending on the weather (Mislevy and Currey, 1975). Seed production is continuous from May to December, with flowering, immature seed. 3

PAGE 14

mature seed, and seed shattering occurring simultaneously on a single Inflorescence on the same plant (Currey et al.. 1973)A mature plant produces in excess of ^iS^OOO seeds with over ] ,kOO seeds per panicle. Germination averaged less than 3% while mechanical scarification of the hard seed coat improved germ inat ion up to 3h%. Currey et al. (1973) concluded that smutgrass exhibits characteristics associated with a very successful weed due to 1) production of large numbers of seed per season, 2) production of seed continually over the entire growing season, 3) variable seed dormancy with germination over an extended period of time, and h) continual maturation of seed on each inflorescence of the same plant. On a global scale, smutgrass is found in Asia, Central, South and North America, and West Indies (Hitchcock, I906, 1936, 1950; Roseveare, 19^48; Mol inari, 19^*9) Swallen, 1955; Sacco, 196^) from sea level to approximately 2,700 m. It is adapted to most soil types and especially where rainfall exceeds 'O inches annually (Riewe et al., 1975b). It apparently was introduced into the United States from tropical Asia (Hitchcock, 1950) and occurs along roadsides, lawns, pastures, and wasteland from Virginia to Tennessee and Oklahoma, south to Florida and Texas. It has been reported in New Jersey and also along the Oregon Coast. Smutgrass has become adapted to subtropjcal and temperate climates. Top growth is killed by frost but regrowth occurs the following spring. Smutgrass is an undesirable, weedy grass on considerable hectarage of pasture and rengeland across the southeastern United States (Riewe et al., 1975b; Smith et al., I97'4; McCaleb et al.. I963). In Florida, McCaleb et al. (I963) first reported smutgrass as a potential threat to forage quality in permanent pastures and a recent survey (Currey and

PAGE 15

I Mislevy, \37h) of some central and south Florida counties indicated \ that 75% of the improved pasture was infested with smutgrass. The averagR level of infestation was 38%. It is slow to establish and usually re^ quires several years before a heavy infestation occurs (McCaleb et al., 1963) but if the first few plants in the pasture are not brought under control, they will become the dominant specie in the field. The infestation increases from year to year through extension of the original plants and also from the growth of seedlings. According to Carter (I96I), a heavily infested pasture may have as many as 2h plants per square yard and they may vary in size from one to six inches or more in diameter. The acceptability of smutgrass by livestock as a forage is considered low. McCaleb et al. (I963) stated that lack of pa 1 atab i 1 i ty of smutgrass is particularly evident on the mineral soils of Florida and Georgia and Currey et al. (1973) added that this characteristic helped to account for the high level of infestation presently observed in Florida pastures. In Texas, Riewe (197') reported that smutgrass produces a low quality forage, palatable to livestock only in the spring, but that after May, cattle will graze it only when forced to. According to Carter (I96I), this weed is not palatable to cattle and the car rying capacity of_the pasture tends to decrease with increasing smutgrass infestaLL^^^ In another study (Smith et al \S7^) it was reported that production of high quality forage is greatly reduced as smutgras increases in the pasture. What causes unpa 1 atab i 1 i ty of smutgrass is not known. In this respect (Mislevy and Currey, 1975) suggested that palatability may be limited by the high fiber content {82%) found in the mature plant. ^

PAGE 16

Persad (1976) studied the nutritive value of smutgrass and 'Pensacola' bahiagrass. J_n^ v i t ro organ i c matter digestion and neutral detergent fiber of smutgrass were higher than of bahiagrass after 6 weeks growth. He suggested that after 6 weeks of growth, it is possible that low digestibility associated with high neutral detergent fiber could be the primary factor limiting the acceptability of smutgrass. Smutgr a ss Control A review of the available literature has failed to reveal any information on the control of smutgrass specifically in relation to grazing management. Control methods that have been used include mechanical, cultural, chemical, and combinations of mechanical and chemical (McCaleb et al., 1963; Smith et al., 1975; Currey and Mislevy, 197^). In 1955, McCaleb et al. (1963) attempted to control smutgrass with mechanical methods. They studied the effect of rotary mowing at a 3inch height at intervals of 1, 2, 3, and h weeks. They reported that control of smutgrass by mowing at weekly intervals resulted i n so me reduction of plant size but that all plants recovered to former density after stopping the treatment. Cultivation and complete renovation gave variable and unsatisfactory results, since new plants grew from seeds already in the soil. In Louisiana, the use of a modified rotary tillage machine resulted in 90 to 35% reduction of smutgrass (Carter, 1961). Many chemicals have been evaluated to determine their effect on smutgrass. McCaleb et al. (1963) screened 6 different chemicals for their herbicidal efficiency on control of smutgrass. In this preliminary trial several herbicides showed promise, including dalapon (2,2dichloropropionic acid), monuron TCA (3[p-ch loropheny 1 ] 1 1 -d imethy 1 urea

PAGE 17

7 mono [trichloro acetate]) and monuron (3[p-ch1 oropheny 1 ] 1 1dimethy 1 urea) They reported that with dalapon at the rate of 5 Ib/acrr of active ingredients, the control averaged 8S%Despite the results, they pointed out that additional treatments were necessary to kill surviving plants and new bunches starting from seed. When using a single fall application of either bromacil (5-bromo-3-sec-buty 1 -6methylurac i 1 ) at 2.2h kg/ha or atrazine [2-chloro-'+(ethylamino) -6(isopropylamino)-s-triazine] at 4.'48 kg/ha, Smith et al (197') reported 83 to SSZ control of smutgrass within 36 weeks. In Texas, Riewe et al. (1975a) stated that almost complete smutgrass control prevailed 18 months after spraying with dalapon at a rate of 5.6 kg/ha. However, they added that re inf estat ion of the treated pasture should be expected in time. In Mississippi (Smith, 1975), spring application of dalapon and diuron controlled 30% of the smutgrass in bermudagrass ( Cynodon dactylon [L.] Pers.) Herbicides may injure desirable species in the pasture, although grass species vary in their reaction to herbicides. Severe leaf damage and root kill of bahiagrass with very slow recovery of surviving plants was observed with a rate of 5 lb/acre of dalapon (McCaleb et al 1963; McCaleb and Hodges, 1970On the other hand, Riewe (197') found that all top growth of common bahiagrass was desiccated at the time of dalapon application, but recovery began to 6 weeks after application. He concluded that bahiagrass seems to be quite tolerant to dalapon at the rate of '.75 lb/acre of active material. Houston et al. (1975) studied the effect of atrazine and dalapon upon smutgrass control in a dallisgrass (Paspalum d i 1 atatum Poi r) -bermudagrass pasture. They

PAGE 18

8 found that a traz ine was the Jjeas_t_^amaging, to, ..desJjr-ai>-U^^ while dalapon was the least selective. Dalapon delayed recovery of the two species for approximately 3 weeks. Minimum damage to a dallisgrass and bermudagrass has been achieved with a late October application, after most of the desirable forage had been grazed off by cattle (Riewe, 1975b). The time of herbicide application seems to be very important to the control of smutgrass. In Florida, Currey and Mislevy (197^) reported that the best time of the year to apply dalapon is late May, when the plant is most actively growing. Houston et al. (1975) suggested that pastures can be sprayed any time between May amd Qctpber if conditions are favorable and smutgrass is growing. They also pointed out that if early herbicidal application causes shortage of forage due to damage of the desirable species, dalapon could be applied between late September and early October. When applied in the spring or early summer, dalapon increases the loss of grazing time and other weeds generally become a more serious problem (Riewe et al 1975b; Schlundt, 1977). In a study to determine the effects of applications of dalapon in March, May, June, August, September, and October, Riewe (197') concluded that applications in late summer and fall are the most successful. He concluded that application at this time resulted in a) less loss of grazing, b) fewer weeds, and c) fits well into a program of seeding ryegrass into a warm season perennial grass pasture for winter ^^ture. Combinations of spraying with mowing have not been explored sufficiently in smutgrass control. In Florida, Currey and Mislevy (197't) studied different cultural treatments before and after application of

PAGE 19

dalapon at rates of h and 5 lb/acre. They reported that smutgrass can be controlled by mowing to a 2inch stubble ^ to 5 weeks following the application of dalapon. It was observed that 2.5 years after such a treatment combination, little smutgrass encroachment in 'Pangola' digitgrass (P igitaria decumbens Stent.) and bahiagrass occurred, provided the treated pasture was fertilized and rotational ly grazed (Mislevy and Currey, 1975). Weed Control by Grazing Management Effective control of weeds may often be obtained by mechanical means such as mowing or cultivation or by treatment with herbicides, but costs will always need to be considered (Leach et al 1976). Chemical control may be essential where toxic weeds have become established and hazardous to livestock. Michael (1970) pointed out that grazing management studies designed specifically for vieed control have as yet barely begun in Australia. He added that this kind of study is much needed, especially in relation to weedy grass and annual grass control. Smith (1968) stated that control of barley grass ( Hordeum ^ lepor inum Link) by grazing management presents advantages over chemical methods because a) it is cheap, b) the weed is a useful source of forage in its early stage, and c) clover production rather than being lost, may even be increased. Michalk et al. (1976), in Australia, studied the effect of different stocking rates under 6 grazing management systems on the control of barley grass. Heavy grazing in late winter increased the proportion of barley grass in the pasture and the number of seedheads per unit area.

PAGE 20

10 However heavy grazing early in the fall resulted in decrease of the weed and increase in crow-foot ( Erodium spp) They concluded that dry matter production for the different treatments was relatively unaffected. However, there was a marked effect on botanical composition and on the development and flowering of the different species particularly in the case of barley grass. The effect of grazing management on slender thistle (Cardus pycnocepha 1 us L.) population in an improved pasture was studied in southern Tasmania (Bendall, 1973). Deferring grazing until winter or spring was very effective in reducing slender thistle. Spring grazing favorably altered pasture botanical composition by increasing the frequency of perennial ryegrass ( Lol ium perenne L.) and subterranean clover ( Trifolium subterraneum L.) and reducing the frequency of the weed. He concluded that, of the other successful treatments, deferred fall grazing is tjie most practical system for incorporation into the farm management system as an alternative to herbicide for the control of slender thistle. He also indicated that such a program has the advantage of being less expensive than chemical treatment and favors general pasture improvement. Myers and Squires (1970) observed for 3 successive fall seasons the effect of grazing on the control of barley grass in an irrigated pasture sown with subterranean clover. He compared grazing starting 10, 20, and hO days after the opening of irrigation. Barley grass yield was less on the 20-day treatment than on the 10 and ^0 days. Deferment of grazing for 10 days was less successful, because of the tendency of the animals at this early stage to select dead material and warm-season

PAGE 21

n weeds in preference to barley grass. They concluded that substantial reduction in barley grass population in an irrigated pasture can be achieved within 1 year, and that almost complete elimination will occur in two years by deferment of grazing for 20 days followed by continuous grazing management. In Australia from 1962 to 1966, the influence of pasture topdressinq with superphosphate and stocking rate on skeleton weed ( Chondr i 1 la juncea L.) in a v/hite clover-ryegrass mixture was measured in a grazing experiment (Kohn and Cuthbertson, 1975). The increase in stocking rate from 5 to 15 sheep per hectare had no effect on final skeleton weed number. Density of skeleton weed increased in rotational grazing (1 week grazing and 2 weeks rest) when compared to continuous grazing. The increase of the weed under rotational grazing was due to the development of satellite plants in the pasture. Laycock (1970) compared the effect of spring-fall grazing and fall grazing only on a sagebrush-grass range. Heavy spring grazing caused a reduction of grasses and forbs by more than 50% from 1950 to ]3(>h and Increased sagebrush production by 78%. Fall grazing enhanced production of palatable perennial grasses by 36% and reduced production of sagebrush by 22%. He suggested that fall grazing as a method for range improvement is less expensive than mechanical or chemical means of sagebrush control. For k years, Pearce (1972) studied the consequence of different stocking rates on the control of Patersons's curse ( Echium pi antag i neum ) He found that population of Paterson's curse declined by up to 72% with a stocking rate of 3 sheep per acre and up to 80% when the stocking rate was increased t:o 8 sheep per acre. Pasture sprayed with

PAGE 22

12 2,k-D and grazed at a stocking rate of 8 sheep per acre had considerably lower weed population than pasture which was not sprayed. Southwood (1971) reported control of broomrape (Orobanche minor) in a Tr i fol i um subter raneum Hordeum lepor i num pasture mixture. Heavy, continuous grazing before the broomrape flowered combined with superphosphate application in autumn, significantly reduced the weed population. The author stated that when continued for a number of years, broomrape could be eradicated by such a treatment combination. Pasture Loss Due to Weeds (^eeds contribute to decrease pasture p roduct i v i ty i n several ways: 1) they decrease forage yield due to weed competition, 2) they cause animal discomfort, 3) they result in undesirable flavors in animal products, and h) they may cause poisoning of the grazing animals (Smith, 197^)^ The average annual losses due to weeds in pastures and rangelands in the United States from 1951 to I96O may be estimated at more than $632 millions (Hamill, 1975). In 1973 in Florida, it was estimated that weed damage in pasture and hay crops totaled 28.8 million dollars. This cost resulted from expense of weed control with herbicides and mechanical methods, yield and forage quality loss, lowered land value, and the expense of additional harvesting (Smith ]37h) This figure does not include animal losses in decreased weight gains, less milk production, and animal deaths by poisoning. Weeds interfere with the growth and development of desirable forage species in many ways. They compete for light, water, nu t r i ent s, space and carbon dioxide (Smith, \37hHamill, 1975). Crombie (19^*7) defined

PAGE 23

13 competition as the requirement at the same time by more than one livinq organism for the same resources of the environment in excess of the immediate supply. Daubenmire (cited by Risser, 1969) suggested the following parameters of adaptation which may under specific circumstances, be significant in competiton: 1) time of root penetration. 2) ability to obtain nutrients in short supply, 3) endurance in drought soils, k) longevity, 5) abundance of seed production, 6) food reserves available to young plants, 7) time of initial growth, 8) nutrient uptake ab i 1 i ty 9) vigor and size of plant, and 10) reproductive potential. In a comprehensive study in Australia to examine the outcome of competition for light between capeweed ( Arctotheca calendula ) and subterranean clover, Mclvor and Smith (1973) reported that capeweed does not suppress clover growing in association if the two species commence growth together. However, when the capeweed was established k weeks before clover, the legume yield was always lower than from pure stand sown on the same day. Rummell (19^6) investigated the competition of Bromu s tectorum L. with Agropyron desertorum (Fisch. ex Link) Shult. and A. smi th i i Rydb. In this study the number of A. desertorum plants was reduced to 5QZ and A. smithi i to about \0%. He concluded that A. desertorum which germinates earlier in the season and makes rapid growth following emergence, competes more successfully with B^. tectorum than the slower developing A^. smi th i i Wakefield and Skaland (1965) conducted an experiment with alfalfa ( Medicago sativa L.) to evaluate effects of int ra-spec ies and weed competition on seedling establishment. Seeding rates of alfalfa were approximately 25, 50, and 100 seeds per square foot. Three intensities

PAGE 24

of weed competition were developed with the aid of chemical treatments. They found that weed control resulted in an increased yield of alfalfa in the seeding year compared to untreated plots. Seeding rates had a marked effect on the average root-crown weights of alfalfa. Smallest weights occurred at the high seeding rate and significantly larger rootcrown weights were observed from the lowest seeding rate. Grimmett and Weiss (1967) investigated the competition of weeds in sown pasture of 'manawa' ryegrass ( lol ium perenne L. x L_. mu 1 1 i f lorum Lam.) and subterranean clover. They reported that rapid weed growth restricted seedling development of the desirable species resulting in slow and unsatisfactory pasture establishment. Bryan and McMurphy (I968) reported in their study that crab^tass ( Pigitaria sanguinalis [L.] Scop.) which emerged with the seeded weeping lovegrass ( Erag rostis curvula [Schrad.] Nees.) reduced forage yield by more than one-half in the first clipping. The plants which suffered from weed competition were visibly reduced in height and did not produce seedheads the first year. Schol 1 and Staniforth (1957) studied the establishment of birdsfoot trefoil ( Lotus corniculatus L.) as influenced by competition from weeds. They found that pre-emergence application of monuron TCA or postemergence application of dalapon controlled grassy weeds and enhanced survival and vigor of trefoil seedlings. Ben efits from Weed Contro l Beneficial effects of weed control on pastures depend on the efficiency of the weed control method. Klingman (1970) stated that what was most needed to produce good quality pasture was a method to prevent

PAGE 25

15 and eliminate weeds. He also suggested that in pasture-forage production, the integration of all beneficial practices into the management system is required to achieve highest efficiency. Peters and Stritzke (1971) studied the effects of 2,'4-D and mowing on the botanical composition and production of a Kentucky bluegrass ( Poa pratens is L. ) They reported that the average production of broadleaf weeds were 88, 329, and 950 lb/acre for 2,'4-D, mowed and untreated plots, respectively. Yields of Kentucky bluegrass were 900 and 1,150 lbs/acre for the control and 2,h-D treatments, respectively. They concluded that the forage yield increase was primarily the result of a decrease in competition from broadleaf weeds. Alley and Bohmont (1958) studied the control of big sagebrush ( Artemisia tridentata Nutt.) in a native pasture. They indicated that a four-fold increase in yield of native forages was obtained by controlling big sagebrush. In a four-year study. Morrow and McCarty (1976) observed the influence of green sagewort ( Artemisia campestris L.) and other broadleaf weeds on forage production in Nebraska. Chemical treatment increased forage production by h2% and controlled S7Z of the weeds in plots receiving two consecutive annual applications. Forage production was increased up to 330 for herbicide alone and 660 lb/acre of dry matter for herbicide followed by nitrogen fertilization. They pointed out that herbicide and fertilizer can be effectively used to increase forage production, but they will not correct the effect of mismanagement which results in weedy pastures. In Nebraska, Klingman and McCarty (1958) reported that use of 2,4-D was more efficient than mowing for the control of broadleaf weeds.

PAGE 26

16 Using three annual applications of 2,'+-D reduced the weeds by 70% and mowing alone resulted in a reduction of 30?;. Combining annual spraying with plowing and seeding reduced broadleaf weeds more than 30%. Scholl and Brunk (1962) investigated the competition of weeds with birdsfoot trefoil. They compared no weed control and all weeds removed in the early stage of growth. Where no weeds were controlled, yield on the trefoil on the first year was 389 lb/acre of dry matter and where complete control was practices, 2,3^2 lb/acre of dry matter was obtained also in the first year. In the second year the yields were 3,^80 and 6,533 lb/acre of dry matter for the control and weed-free treatments, respectively. Birdsfoot trefoil plants were shorter on the control treatment than on weed-free treatment. However there was no difference between the two treatments as far as birdsfoot trefoil population was concerned Gesink et al. (1972) investigated the control of broom snakeweed (Gutirrezia sarothrae [Pursh] Britt and Rusky) on the short-grass plains in southeastern Wyoming during 5 years. Control of broom snakeweed increased the desirable species, chiefly blue grama ( Souteloua grac ills (H.B.H.) Lag.). Herbage production was 225 lb/acre of dry matter for untreated areas and 1,200 lb/acre of dry matter when treated with a herbicide. Forage quality and intake may be increased by controlling weeds in pastures. Barrett et al (1973) studied the effect of spraying paraquat in a pasture containing subterranean clover and either silver grass (Vulpia spp.) or ripgut brome ( Bromus rigidus Roth). Spraying controlled the grasses and produced pastures containing up to 35% clover

PAGE 27

17 The concentration of N, P, Ca, and Mg were higher in mature herbage on treated plots than on the control treatment. Monson (1977) studied the effect of paraquat on yield and quality In a pasture of Coastal bermudagrass (Cynodon dactyl on L.). He found that the in vitro dry matter digestibility of a sprayed pasture was higher than on the control at 6 weel
PAGE 28

18 Measuring Botanical Composition The determination of the species in a pasture is important for study ing the changes in botanical composition due to treatment effects. Botanical composition is also essential because individual species may differ in theri reaction to environmental and management factors, (t' Mannetje et al 1976) Measurement of botanical composition may be made in terms of the yield of the species components, the frequency of occurrence of different species and the number of plants on the area covered by different specie-(t' Mannetje et al 1976) The area ground covered by the aerial parts of the plants has proven to be a valuable measure of botanical composition and its change. Brown (195^) defined cover as the vertical projection of the above-ground parts of the plants on the ground, and it is expressed as a percentage of the total area (Winkworth et al., 1962). Terms used to express the area covered are 1) density, 2) basal area, 3) herbage area, h) foliage density, 5) cover, and 6) leaf area index. Pasture research had led to the development of a large variety of methods for estimating percent cover (Brown, 195'f). The methods fall into four basic categories, depending on the type of observation made and the dimensions of the sampling unit. These are charting, ocular estimate^, line intercepts, and point methods. The chart quadrat method is one of the earliest techniques used to determine changes in botanical composition (Hill, 1920). Basically, the ground position of each plant is drawn in its relative position on a recorded sheet (Weaver and Clements, 1938), usually at yearly intervals (Wright, 1972). However, Brown (igS't) stated that the interval between

PAGE 29

19 charting depends on the purpose of the study. He emphasized that in areas where weather varies considerably from year to year an annual charting is important to determine the changes in the vegetation. Hutchings and Pase (1962) stated that chart quadrat method is particularly well suitable to grass and other types of low vegetation. However, he added that this technique presents better results when the plants are clearly defined in bunches or tufts. It becomes more difficult as the growth becomes closer and is quite impossible in dense pastures (Brown, 1954). Heady et al. (1959) compared the accuracy and practicability of charting, line intercept, and line point in the sampling of two shrub communities. They reported that means and confidence intervals obtained by the three methods gave reliable estimates of the population mean. Line intercept and line point yielded less variable data and were sampled adequately with fewer plots and less time than the charting methods. However, Ellison (19^2) found that accuracy in charting method depends on the ability and care of the operators and to some extent on the charting device used. Although the chart quadrat method can provide a reasonably accurate record of the vegetation (Holscher, 1959), it is always time consuming. The individual culms as well as the area occupied by clumps must be carefully drawn into a small scale quadrat map with the aid of a pantograph or directly by means of numerous wires crossing the frame (Anderson, I9't2). Estimating Forage Yield Sampling to estimate yield is one of the most difficult procedure involved in pasture research. The methods generally used for

PAGE 30

20 .measuring the production of a pasture involve the cutting of areas or spot samples of the pastures. As stated by Campbell et al. (1962), the use of grazing animals in pasture management studies increases the difficulty of estimating yield from sample-cutting techniques due to a) the need to cut sufficient samples to give an accurate estimate of the yield, because of the very variable nature of grazed pasture, b) the requirement of cutting as few samples as possible, which arises from the physical limitations on cutting large number of samples, and c) the need to ensure that the area cut is not so large that it acts as a treatment in itself. Many efforts have been devoted to the search of a quantitative method for determining the yield of pasture jjt^ situ Bransby et al. (1977) emphasized that a rapid, indirect, in situ nondestructive technique for making accurate estimates of pasture dry matter yield would benefit grazing experiments. The literature describing the methods that have been developed which do not require the harvesting of herbage usually refer to the term "density." The use in each instance is in accordance with the use of the term for the specific method and is based on the relationship as stated by Mott (I962). Yield/Unit Area = f (density, height) One of the more promising methods of estimating pasture dry matter yield in situ is the use of electronic capacitance meter (Alcock and Lovett, 1967). Fletcher and Robinson (1956) proposed the use of a capacitance mete". it is based upon the fact that herbage has a high dielectric constant and air has a \ovi dielectric constant. Readings were taken from plants under conditions ranging from dry grasses to soggy, wet sedges in swamp land. They demonstrated that the

PAGE 31

21 capacitance meter showed promise of being faster than clipping and more accurate than other estimation methods. They reported that the slight decrease in precision per determination is more than offset by reduction in sampling error. Campbell et al. (1962) with a modified capacitance meter stated that if the errors of prediction were solely random error, the greater efficiency of sampling by the instrument would compensate for any moderate increase in error per estimate compared with a cutting technique. Johns and Watkin (1965) reported that the great advantage of using a capacitance meter is that once the apparatus has been calibrated for a particular pasture, a very large number of readings can be taken in a short time with no detrimental effect on the pasture. in New Zealand, Campbell et al (1962) found that within pastures the capacitance meter allows an estimate of sample yield to be made with considerable accuracy. The instrument predicted about 90^ of the variation in forage weight either as wet, dry or organic matter. For different pastures significant differences existed between prediction equations, although certain pastures can be combined. They concluded that for different pastures, individual prediction equations would be required. Johns and Watkin (1965) studied the relationship between capacitance readings and yield of pastures. They found that in all pastures except for native pasture, the regression of dry matter yield, fresh material and total water on the meter reading were highly significant. However, the dry matter regressions were generally slightly inferior to both the fresh material and total water regressions. They suggested that dry matter is still considered adequate for use in individual experiments where the particular regression could be reliably

PAGE 32

22 established, even though regression based on yield of fresh material would be more accurate. in Colorado, Carpenter et al. (1973), studied the influence of woody stems on the relationship of the capacitance meter to herbage weight. A single meter reading of the plot estimated weight of total herbaceous material is more accurate than total herbaceous material plus woody stems. They reported that excluding wood would probably improves the regression because woody material has little capacitance relative to herbaceous material and the amount of wood on the plots varied greatly relative to the amount of herbaceous material. It is apparent that the height of the forage and the above stubble height affect capacitance meter readings. Hydy and Lawrence (cited by Alcock and Lovett, 1967) found that a change in capacitance of the probe is caused by the dielectric properties of the herbage and plant acting as a circuit when leaves come into contact or near contact with the electrodes. Lovett and Bofinger (1970) used a capacitance meter to measure growth of 30 cultivars of rape ( Bass i ca spp.). The height of some cultivars exceeded the height of the probe, resulting in plant material touching the top plate of the instrument. Under such conditions probe readings were affected by changes both in capacitance and a tactile factor, the latter occurring when material was crushed down in down in obtaining readings, When measuring a short stubble the main factor influencing change in capacitance is the dielectric properties of the herbage (Hydy and Lawrence; cited by Alcock and Lovett, 1967). Back et al. (1969) sampled plots sown with LoHum multiflorum Lam. and a grass-white clover ( Tr i f ol i urn re pens L.) mixture. On each plot. 5 samples were cut to ground level and a further five to approximately 2 cm above

PAGE 33

23 ground. They reported higher relative efficiency of the samples cut to 2 cm in comparison with those tal
PAGE 34

Neal and Neal (1973) reviewed the use of electronic capacitance meters to estimate weight of standing vegetation. They concluded that variation in site and phenology generally have more influence on meter performance than does meter design. In their explanation, the meter reading is the sum of numerous external influences and internal characteristics of the instrument. They emphasized that the technique is accurate, rapid and nondestructive when used properly. However, Nichols (1973) stated that any apparatus for outdoor use must be designed so that adverse conditions do not contribute to further inaccuracy. Shaw et al. (1976) pointed out that the capacitance meters have not lived up to their early promise due to limitations such as a) it measures water yield in the plant tissues and not dry matter yield, b) it is influenced by plant species, and c) it is relatively insensitive where the herbage consists mainly of dead material. He concluded, however, that the above limitations apply much more often in the case of tropical pasture than temperate pastures. Spraying Pasture with Molasses It has long been known that consumption of unpalatable forage is influenced by adding molasses. The direct application of molasses on standing pasture, seems to have first been developed in South Africa (Loosli and McDonald, I968). PI ice (1952) advised farmers to take advantage of this technique and get rid of weeds, unpalatable and poorquality forage by spraying them with molasses or other similar product, and turning grazing animals on them.

PAGE 35

25 Change in intake from preferred to initially non-preferred component? of the pasture is enhanced by inducing tlie animals to graze heavily on the sprayed area in preference to grazing lightly over the whole pasture (Willoughby and Axelsen, I96O). In this respect, Coombe and Tribe (1962) also stated that only small areas of the pasture should be sprayed with molasses at a time. Native pastures with considerable amount of unpalatable annual grasses were sprayed with urea-molasses mixture during the summer (Pope et al., 1955)They reported that spraying a small area led to very intensive grazing and removal of all top growth, while spraying a large area resulted in an incomplete consumption of the sprayed forage. According to Coombe and Tribe (1962) spraying molasses on standing herbage presents several disadvantages. They emphasized that the most serious is the high proportion of spray which falls on the soil and is washed off by rain. In addition, they added that the technical difficulties of spraying forage may be considerable where a) pasture areas are large, b) stocking rates are low, and c) the ground surface is rough. Mostert (1959) emphasized the importance of spraying only thick grass stands or patches, otherwise much of the sprayed mixture will be wasted. In detailed tests on one cattle pasture, it was found that only 16.5% of the spray could be recovered from the herbage immediately after spraying (Loosli and McDonald, I968). In Oklahoma, Pi ice (1952) compared table sugar, black-strap molasses, sorghum molasses, and corn syrup when sprayed on weeds and grasses which are seldom, or never, touched by grazing animals. The order of preference for the different materials was as follows: black-strap molasses,

PAGE 36

26 sorghum molasses, table sugar, and corn syrup. He observed that the animals did not take very long to discover the sprayed plants and then consume them completely. In Australia, Willoughby and Axelsen (I96O) tested four spray treatments, containing urea, molasses, and urea plus molasses on a grass-legume mixture which had been ungrazed in the spring and had matured and dried in early summer. The pastures sprayed with molasses were consumed most rapidly by the animals. The application of sprays, mainly molasses, resulted in increased removal of the most abundant component, Pha lar i s an initially non-preferred component of the pasture He concluded that spraying affected intake in three ways: a) it provided a supplement, b) it altered the amount of forage consumed, and c) it increased or decreased the quality of the material consumed, depending on whether the non-preferred components are higher or lower than the preferred in nutritive value. O'Bryan (I96O) examined the utilization by cattle of carpet grass ( Axonopus affinis Chase) following foliar application of urea, molasses, and monosodium phosphate in southeastern Queensland. The pastures provided a complete ground cover and were sprayed in strips at weekly intervals. He reported that the animals showed preference for the treated strips immediately after spraying. The treatment failed to prevent selective grazing, so that the animals selected pasture containing 10 to ]2% crude protein in the first year and 8 to 10^ during the second year. He pointed out that frequency of spraying (weekly), heavy dews, intermittent rainfall during the experimental period, low palatability of mature carpet grass and the high degree and selectivity

PAGE 37

27 by the animals were some of the factors that contributed to the low effect of spraying. He concluded that, for Queensland conditions, more frequent spraying would be uneconomical and impracticable. Coombe and Tribe (1962) studied the value of spraying urea and molasses on dry, standing forage. It was observed that practically all animals (sheep and cattle) were attracted to pasture treated with molasses. In general, cattle showed a greater preference for sprayed forage than did sheep. In Uruguay, Christiansen (I965) studied the effect of spraying molasses and urea on a native pasture. The mixture was supplied twice weekly and 1/8 of the total area was sprayed each time until the entire pasture had been sprayed. He found that the palatability of dry coarse grasses Paspalum quadrifarium and Schyzachy r i um paniculum was improved and that the animals grazed extensively in the sprayed areas. In California, Wagnon and Goss (I96I) compared a) dry rank forage sprayed with molasses-urea mixture, b) sprayed only with molasses, and c) untreated forage. Weekly application of ]h pounds of molasses or molasses-urea mixture per animal were made. They found that rank, dry forage of low palatability was completely utilized by the animals after spraying with molasses or molasses-urea mixture, however, similar unsprayed forage was mostly left ungrazed. They also reported that there was no loss of the sprayed material until light dews occurred 43 and 63 days after initial spraying. Practically all sprayed mixture was washed from the forage by 0.72 inches of rain that occurred 90 days after the initial treatment. In another study Tulloh et al. (I963) compared a) molasses fed in a trough, b) standing forage sprayed with urea and molasses, and

PAGE 38

28 c) not supplemented. They reported that all supplemented treatments significantly increased forage intake. It appears that a sudden change in cattle grazing habits occurs when green forage became available after rain. Coombe and Tribe (1962) reported that if a "green pick" occurred due to unseasonal rains, spraying molasses and urea had no effect on forage intake. They stated that once sufficient rain had stimulated growth of green shoot, the animals preferred to graze the green material rather than concentrate on sprayed areas. On the other hand, Wagnon and Goss (I96I) observed that the animals on a sprayed treatment continued to eat the old forage that had been sprayed with molasses-urea mixture, instead of the new regrowth. However, in the same experiment it was observed that animals on the control treatment started to graze the young green plants, rejecting as much as possible the old leached forage. The preference for molasses sprayed forage appears to vary with the palatability of the mixture sprayed, the rate of molasses, and the weather. Sprayed pastures were strongly preferred by grazing animals, however, preference was not evident 2k hours after spraying, O'Bryan (i960) observed that the preference for the sprayed forage was maintained for two days and Mostert (1959) reported that the sprayed area should be grazed off within 3 to 4 days. In studies of different spraying densities. Bishop (1959) compared three rates of a mixture molasses-urea: a) one gallon per 10 to 15 square yards, b) one gallon per 50 square yards, and c) one gallon per 100 square yards. Spraying at the rate of one gallon to 50 square

PAGE 39

29 yards was found to give the best results. At a density of one gallon per 100 square yards much of the grass remained ungrazed, so that the mixture absorbed by this ungrazed grass was merely wasted, Densities of one gallon to 10 to 15 square yards were found to be unsuitable because animals then grazed too short. He emphasized that this is particularly undesirable on sandy soils where grass roots can be pulled out easily. Effect of Grazing Management to Botanical Composition One of the most important factors to consider in grazing experiments is the change in botanical composition. The presence of grazing animals has effects upon the pasture through defoliation, excretion, and trampling (Coaldrake et al., 1976). T' Mannetje et al, fl976) stated that even a so called pure stand will usually contain varying amounts of other species and that botanical composition is important because individual species or cultivars vary in feeding value, in content of harmful substances, and in their reaction to environmental and management factors. Changes in botanical composition as a result of grazing management systems are well documented. Bryan (.1970) studied the effects of low and high stocking rates on the botanical composition of mixed pasture. He found that high stocking rate increased Paspalum dilatatum Poir., and total weeds from 2h to 33^ and ]k to 22%, respectively, and reduced Chloris gayana Kunth. from 18 to 1%. In this experiment heavily stocked pastures were reduced to a closely clipped lawn while the lightly grazed ones usually carried a considerable bulk of material 0.3 to 1 m high.

PAGE 40

30 In Australia, Cameron and Cannon (1970) observed changes in botanica composition resulting from increased stocking rates of ^4.9, 7-^, 9-9, 12. i*, 1^.8, 17.3, and 19-8 sheep/ha, from 1963 to I968. Trifol ium subterraneum L. during this period increased from 30 to 70% at ^.9 sheep/ha and decreased from 30 to 10% at 19-8 sheep/ha. Lol ium perenne L which in 1963 was the main grass component, representing 20 to hO%; by 1968 declined to a trace at all levels of stocking rates. They also pointed out that the decline of L_. perenne was more rapid at high than low stocking rates and that by I965 it had decreased to less than 10% at 19.8 sheep/ha. Poa annua L. not initially present in any of the pastures, in I968 ranged from a trace at 7.^ and 9.9 sheep/ha, and to 30% at 17.3 and I9.8 sheep/ha. Rodel (1970) compared the effect of two stocking rates upon different grasses. He found that high stocking rate caused marked changes in basal cover of the grasses. Chloris gayana Kunth. decreased from 3.8 to 0.7% while Cynodon p 1 ectostachyum increased from 3.0 to 11.5%. Serrao (1976) studied the response of Desmodium intortum (.Mill) Urb' Coastcross1 bermudagrass mixture to different levels of grazing period, rest period, and grazing pressure. He reported that the grass percentage in the mixture increased with heavy grazing pressure, while the legume percentage increased with long rest periods associated with medium to light grazing pressure. Ritson et al. (1971) studied the changes in botanical composition of a mixture of Townsv i 1 le sty lo ( Sty losanthes humi 1 is H B K. ) perennial grasses, and annual grasses at two stocking rates, namely 1 cow per 1.2 and 2.k hectare. They found that stocking rate

PAGE 41

31 significantly influenced the botanical composition of the mixture. With a stocking rate of 1 cow per 2.h hectare the pasture was dominated by perennial grasses. At a stocking rate of 1 cow per 1.2 hectare the pasture became dominated by Townsville stylo and annual grasses. Campbell and Beale (1973) studied the botanical composition of natural pastures stocked at 2.5, 3-7, and ^4.9 sheep/ha. Higher stocking rates resulted in a lower contribution by barley grass (Hordeum lepor inum Link.) to the pasture, but an increased contribution by silver grass ( Vulpia myuros K. Gmel.) and naturalized medics ( Med icago spp.). Johnston et al. (1971), in a long-term grazing experiment conducted over a 17-year period, compared light, moderate, heavy, and very heavy stocking rates on the botanical composition of a fescue grassland range. Percent basal area of vegetation in lightly grazed plots changed from dominance by Danthonia parryi Scribn. to dominance by Festuca scabrella Torr. F_. scabrel la was largely eliminated by very heavy grazing and the plots were invaded by various species, including Taraxacum officinale Weber. Populus tremuloids Michx. encroached upon grassland in the lightly and moderately grazed treatment but the same was prevented in the heavily and very heavily grazed pastures. Pearson and Whitaker (197') presented changes in botanical composition by cattle grazing yearlong at light, moderate or heavy stocking rates. The most abundant grasses in the pasture were slender bluestem ( Andropogon tener (Ness) Kunth.), pinehill bluestem (A. divergens [Hack,) Anderss. ex Hitchc), panicums ( Pan icum spp. ) and paspalums (Paspalum

PAGE 42

32 spp.) They reported that although overall herbage composition was not changed, grazing intensity did affect individual species. Pinehill bluestem, the principal specie on the range declined from 57% under light grazing to ]7% under heavy grazing. Carpetgrass was 1 and 26% under light and heavy grazing, respectively. ^J\n Nebraska, McCarty et al, (197'+) studied the effect of rotational and continuous grazing upon change in botanical composition of a pasture, from VSkS to I969. They reported that under rotational grazing, relatively few weeds invaded the pastures and only a small amount of blue grama ( Bouteloua graci 1 is [H.B.H.] Lag. ex Stend) and sand lovegrass ( Eragrostis trichodes [Nutt.] Wood) persisted. In I969, the mixture consisted primarily of big bluestem ( Andropogon gerardi Vitman), indiangrass (Sorghastrum nutans [L.] Nash) and switchgrass ( Panicum virgatum L.). In the continuously grazed warm season grass plots the main desirable specie was blue grama. Ottosen et al. (1975) studied the change in botanical composition of a mixture of tropical grass-legume pasture by comparing strip and continuous grazing. Both grazing management systems caused marked changes in botanical composition of the pasture. Legumes decreased from 2k to 16^ in the strip grazed plots while that in the continuously grazed treatment, the legumes increased to 38^. They emphasized that after the experiment, both grasses and legumes regrew wi thout any differences resulting from the previous grazing systems.

PAGE 43

MATERIALS AND METHODS G eneral Description A grazing experiment was conducted from April to November 1976 at the Beef Research Unit which is located approximately 21 kilometers northeast of Gainesville, Florida. The soil at the experimental site is underlain by limestone of the Ecocene age having an overlay of acid, sandy, and loamy marine sediments (U.S.D.A., )35^) These Myakka fine sand soils are somewhat poorly drained and contain a spodic horizon (organic hardpan) which is an accumulation or organic matter, iron, and aluminum (Carlisle and Pritchett, 1971)This hardpan increases the severity of both wet and dry periods by retarding the vertical movement of water to and from lower levels. Data reported by Roger et al. (1961) shows that the average organic matter content of this soil is 2. 25/1. The average pH is '.9, varying from 5 to 5-2, depending on the percentage of organic matter. The predominant native vegetation on the experimental site consisted mainly of longleaf pine (PJnus austr al is Michx. f.), wiregrass (A r i st i da spp. and Sporobol us spp.), savi palmetto (Serenoa repens Bartr. Small), gallberry ( I 1 ex gl aba L ) runner oak (Quercus minima Sarg.), and cypress (T axod ium ase ndens Brongn.) The climate is subtropical and humid, with a frost-free season averaging 276 days and an aver.jge annual precipitation of 1,300 mm 33

PAGE 44

3k U.S.D.A., }35^) Table 1 presents mean monthly maximum and minimum temperatures and rainfall during 1976 at the Beef Research Unit. For 1976, total rainfall was recorded to be 1,312 mm. Table 1. Monthly mean maximum and minimum temperatures and rainfall at the Beef Research Unit for 1976. Temperature Rainfall Max Min. 1976 Normal Oq January 18.3 1.8 53 65 February 23.7 5.5 Ik 82 March 11.1 10.7 '6 103 April 11.1 11.0 11 93 May 29.6 15.5 \k\ 87 June 31.5 18.5 215 166 July Ik.k l^.k 51 187 August 33.3 19.9 100 192 September 31 .2 19.0 231 Mh October 26.0 10.8 55 106 November 20.8 5.5 56 kk December 18.1 k.\ 126 63 Average precipitation for Gainesville (Pla.) from 1931 to I960. Experimental Pasture The experiment was conducted in a pasture of bahiagrass ( Paspa lum notatum Flugge), smutgrass (Sporobolus poiretii [Roem. and Schult.]

PAGE 45

35 Hitchc), and whfte clover ( Tr ifol ium repens L.) appearing (n the spring This pasture was selected because the smutgrass infestation is typical of many pastures in Florida. The initial ground cover was between 40 and 50% smutgrass. Bahiagrass was the predominant desirable specie. This area had previously been grazed as one pasture for several years and mowed annually late in the fall. Treatments The main treatments in this experiment consisted of 1. Control 2. Molasses sprayed on the pasture to increase palatability of the smutgrass. 3. Dalapon applied in the spring followed by mowing and N fer' t i 1 izat ion. In addition to the three main treatments presented above, eight extra treatments were included, making a total of 11 treatments. The eight extra treatments included the following: Dalapon applied in the spring followed by burning and N fert i 1 izat ion 5. Dalapon applied in the fall followed by burning and N fertilization. 6. Dalapon applied in the fall followed by mowing and N fertilization. t 7. Ground Hawg in the spring. 8. Ground Hawg in the spring followed by N fertilization. A rotot i 1 1 er-type cultivator.

PAGE 46

36 9Ground Hawg in the spring followed by seeding of bahiagrass and N fert i 1 izat ion. 10. Ground Hawg in the fall followed by seeding annual ryegrass ( LoI ium mul t if lorum Lam. ) and N fertilization. 11. Burning in the fall and dalapon applied in the spring followed by mowing and N fertilization. Grazing Management Factors Within the first 3 main treatments listed above, length of rotation cycle (grazing period and rest period combined), and grazing pressure were also experimental variables. Each factor was studied at 5 levels. Length of rotation cycle was expressed in days and grazing pressure was defined as residual dry matter in metric tons/ha left after grazing. The combinations of rotation cycle and grazing pressure, each at 5 levels, were arranged in a response surface type of experiment. The factors studied with their respective levels are presented in Table 2. A range of plus or minus 0.2 metric tons/ha was established for the projected residual dry matter left after grazing. Grazing periods lasted from 1 to days for all treatments and were included in the number of days in the rotation cycle. In the continuously grazed pastures the animals were not maintained in the pastures continuously. In an effort to simulate continuous grazing, animals were moved in and out of the pastures every few days to achieve the projected level of residual dry matter.

PAGE 47

37 Table 2. Levels of length of rotation cycle and grazing pressure. Length of rotation cycle metric tons/ha0.5+0.2 1.3+0.2 2.1+0.2 2.90.2 3.7+0.2 'Residual dry matter left after grazing. ^Continuous grazing simulated. Experimental Design The experimental design used was a modified central composite in 2 factors (rotation cycle and grazing pressure) each at 5 levels arranged in a Response Surface Design (Fig. l). Combinations of length of rotation cycle and grazing pressure were superimposed upon each of the first 3 main treatments. The combinations of the 5 levels of each factor made up thirteen different grazing management combinations or design points. The arrangement of the thirteen combinations consisted of h factorial, h axial, 1 center, and h corner points (Table 3)The factorial points vjere not replicated. However, the axial, center, and corner points were replicated twice, making a total of 22 pastures (experimental units) per main treatment. Grazing pressure projected RDM days 0-^ 28 k2 56

PAGE 48

J8 0.5 if) in (U i_ Q. Ol C N ro u C3 0) E T3 m 0) o c o o t. V 1.3 I I I I -o2.12.9 6 3.70 I ^ 14 42 28 Rotation Cycle (Days) Factorial Star Points ^ Center Point Figure 1. Response surface design in two factors 56 Corner Points 13 combinations.

PAGE 49

39 Table 3. Combinations of rotation cycle and grazing pressure of the Central Composite Design. Length of Grazing pressure Points Number of rotation cycle projected RDM"^ replications days -metric tons/ha'*2 2.9 ^2 1 .3 Factorial 1 1^ 2.9 1.3 '28 3.7 28 0.5 56 2.1 ^^'^^ 2 Ot 2.1 25 2.1 Central 2 56 56 ot 0+ 3.7 0.5 3.7 0.5 Corner t Residual dry matter left after grazing. Continuous grazing simulated

PAGE 50

AO In this study, the central treatment (center point) was the combination of a length of rotation cycle of 28 days and a grazing pressure equal to 2.1 metric tons/ha of residual dry matter left after grazing. The central treatment was the combination of rest period and grazing pressure imposed upon the 8 extra treatments, each replicated twice. There were a total of 82 combinations of treatments and grazing managements and they were assigned to the pastures units completely at random. Construction of Physical Facilities The experimental area was selected in December 1975 on the basis of uniform smutgrass ground cover. In early January 1976, h.] hectares were allocated to the experiment and surveyed for the location of the fence lines. The area was subdivided into 82 pastures (Fig. 2), 50 m in length and 10 m in width (500 m^) Five-strand barbed wire line fences and electric division fences were built. A water supply system was installed underground using 1" plastic pipe with risers along the fence to provide water for the animals. Mineral boxes and water containers were provided. The water level was controlled in each container by a float valve. Experimental Analysis From May to July, 36 Brown Swiss-Angus heifers and cows (cows with calves until August, when the calves were weaned), with an average weight of 351 kg were used to graze the experimental pastures.

PAGE 51

ill o 01 00 N 00 CO IS N s N N to Sin M 0) L. ZJ *J v> re Q. ID c Q. l V JZ 3 o lU 3

PAGE 52

k2 In August, due a higher forage production, the number of animals was increased to 50. The put-and-take technique was used to stock the pastures to the desirable grazing pressure levels. When needed, the animals were assigned at random to graze the pastures. The animals were weighed every 28 days in order to estimate the stocking rate. From 20 September through the conclusion of the experiment, all animals were supplemented with molasses at an average rate of 2.2 kg/an ima 1 /day A complete mineral mixture was made available freechoice throughout the study. On 12 December 1976, all experimental pastures were fertilized with 14.8 kg of P and 55.8 kg of K/ha. Treatment Application Ma in Treatment 1 Control On the control treatments, only different grazing management systems were imposed. Main Treatment 2. Molasses Sprayed on the Pastures On the rotational ly grazed pastures, molasses was sprayed on the standing forage before each grazing period to increase the palatability of the smutgrass. in the continuously grazed treatments, molasses was sprayed at 2-week intervals. in each molasses treatment it was applied immediately prior to introducing the animals to the pastures. Molasses was diluted in equal parts of water and sprayed on the pastures at the rate of 320 liters/ha. Molasses before diluting with water weighed 1.3^ kg/liters. Spraying was carried out with a tractor

PAGE 53

^3 mounted, boomless sprayer. A spray pump was operated at 2.5 kg/cm^ pressure and driven at 4 km/hour. Main Treatment 3 Dalapon Applied in the Spring Followed by Mowing and N Pert i 1 izat ion On 28 April, when the smutgrass plants were growing quite vigorously, dalapon (2,2-dichloropropionic acid) at the rate of 5.6 kg/ha was sprayed on the pastures. It was applied in water at 277 liters/ha with a tractor-mounted sprayer. On 19 May, three weeks after the dalapon application, all pastures were mowed to a height of 6 to 7 cm with flail mower. On 7 June, 56 kg N/ha was applied using a Gandy fertilizer spreader. Ammonium sulfate (21^ N) was used as the source of N. Dalapon Applied in the Spring Followed by Burning and N Fertilization Dalapon at the rate of 5.6 kg/ha was applied to the pastures on 28 April, 1976. On 1 June, the top kill of smutgrass plants became evident due to the application of dalapon. The pastures were then burned. One week later the pastures were fertilized with 56 kg N/ha as ammonium sulfate. Dalapon Applied in the Fall Followed by Burning and N Fertilization The fall application of dalapon was completed on October, when the smutgrass was still actively growing. The rate and method of application was the same as used for the spring treatment (5.6 kg/ha) The pastures were burned on 15 February 1977. D alapon Applied in the Fall Followed by Mowing and N Fertilization in this treatment, dalapon (5.6 kg/ha) application was carried out on / October 1976. The pastures were mowed to a 6 to 7-cm stubble on 10 January 1977.

PAGE 54

Ground Hawg in the Spring On 6 May, the Ground Hawg was used to till the pastures. Depth of cultivation was 12 to 13 cm below the soil surface. The cutting blades broke up and lessened the surface soil and destroyed the shallow rooted plants. This operation usually destroys the canopy and reseeding is necessary unless a good supply of viable seed exists in the so i 1 Ground Hawg in the Spring Followed by N Fertilization On 6 May. the Ground Hawg was passed over the pastures. On 5 June, 56 kg/ha of N fertilizer was applied as ammonium sulfate. Ground Hawg in the Spring Followed by Seeding of Bahiagr ass an d N Fert i 1 izat ion The tillage operation with Ground Hawg was carried out on 6 May. On the following day bahiagrass, at the rate of 16 kg/ha, was sown, covered and packed with a cultipacker seeder. One month later a N application at the rate of 56 kg N/ha was made using a Gandy fertilizer spreader. Due to high temperatures and lack of sufficient rain, a poor stand resulted from this seeding. Bahiagrass was seeded again on 10 August at the same rate, after tilling the soil again with the :Ground Hawg. Ground Hawg in the Fall Fol lowed by Seeding of Ryegrass and N Fertilizat ion On 8 October, Ground Hawg tillage operation was carried out. On the following day 'Tetrablend' ryegrass was seeded at the rate of 16 kg/ha. One month later the pastures were fertilized with 56 kg N/ha. Pastures seeded to ryegrass in early October were fully sodded by midDecember of 1976.

PAGE 55

i45 Burning in the Fall and Dalapon Applied in the Spring Followed by Mowing and N Fertilization Burning was carried out later than anticipated due to heavy rainfall during the fall season. Measurements Smutgrass Ground Cover In order to study the effect of the treatments on the change of 2 smutgrass ground cover, ^ non-randomly selected 1 m permanent quadrats were set up in each pasture. The area for each quadrat was chosen in such a way that all quadrats contained similar smutgrass ground cover before starting the experiment. A wooden stake was driven in each of 2 predetermined corners of the quadrat. The equipment used to determine the smutgrass ground cover consisted of a 2.0 m x 0.5 m (l m ) frame constructed of aluminum (Fig. 3) • It was subdivided by thin aluminum bars in order to give 100 equal squares each 10 cm x 10 cm, to facilitate the estimation of plant ground cover. The frame was supported above the canopy by four legs, each equipped with a lock screw which allowed the height of the frame to vary from 20 to 30 cm above the ground. The quadrat charting method was used to evaluate the change in smutgrass ground cover. Ground cover was estimated in early April before applying the treatments and again late in October 1976. The frame was placed over each permanent quadrat and outlines of smutgrass clumps were drawn on acetate transparent sheets (Fig. k) at a scale of 1 mm on the transparency to 10 mm on the frame. Later, areas

PAGE 56

46 ig. 3. The 1 m frame used to determine the smutgrass ground cover.

PAGE 57

g. k. Outlines of smutgrass clumps on acetate transparent sheets

PAGE 58

1^8 outlined on the transparency were inked (Fig. 5) with india ink and the areas were determined, by passing the transparency through an electronic leaf area meter. The change in ground cover of smutgrass for each pasture was determined by the difference between the ground cover estimates made in October and April 1976, and are expressed as percentage units of change. Dry Matter Determination After Grazing Residual dry matter/ha was estimated after each grazing period on the rotational ly grazed pastures. On the continuously grazed pasture residual dry matter was estimated every 28 days. All estimates were made by a double sampling procedure, using a Neal Electronic Model 18-200 Herbage Meter (Fig. 6). In this procedure the capacitance meter was first adjusted to zero, with the probes standing on bare soil in the field where measurements were to be made. The instrument was then taken to the pasture for sampling. Twenty random readings were taken in each pasture. From these twenty, ^4 were randomly selected to be cut. 2 A metal frame measuring 0.186 m (2 sq. ft.), which fit around the probes, was placed in position. The capacitance meter was then removed and the forage within the frame was cut to ground level using a grass clipper. The forage cut was gathered into a paper bag and ovendried at 70 C for 2^4 hours and weighed. The capacitance meter readings and dry weight yierds/0.l86 m^ were used as independent and dependent variables respectively, in a regression model forced through the origin. The regression coefficient was obtained from the equation:

PAGE 59

^9

PAGE 60

50 Fig. 6. View of the capacitance meter used in the doublesampl ing procedure.

PAGE 61

51 Y = b X where: Y = estimated dry matter, metric tons/ha b = regression coefficient of capacitance meter reading on dry matter yield of harvested samples. X = value of capacitance meter reading. The forage dry matter left after each grazing period was estimated by the following formula: nr,uA Y + b I RDMA = 2 where: RDMA = residual dry matter adjusted Y = mean dry matter of the harvested samples {h samples/pasture/ rotation cycle) b = regression coefficient X = mean of the capacitance meter reading of the unharvested samples (l6 read ings/pasture/rotat ion cycle). Parameters Measured The response of the pasture to the main treatments and grazing management was measured in terms of the following response variables: 1. Change in smutgrass ground cover expressed as percentage units 2. Stocl
PAGE 62

52 For each main treatment the response variables were fitted using a second order polynomial of the type: 9=b^.b,X, -b2X2-b,,xJ.b22xJ.b,2X,X2 where: Y = estimated response Xj = length of rotation cycle (days) X^ = grazing pressure (metric tons/ha of RDMA) b = intercept o b. = linear regression coefficient b.. = quadratic regression coefficient b.. = regression coefficient of the interaction ij The following analysis of variance was performed in order to test the fit of the model: Analysis of Variance Source of Variance df MS F Total 21 Due to regression 5 RC 1 GP • 1 RC X RC t GP X GP 1 RC X GP 1 Res idual 16 Lack of fit 7 Error 9

PAGE 63

53 The estimation of the error was generated from the sums of squares of the nine replications of the axial, center, and corner points. The lack of fit sum of squares was calculated by difference, subtracting the error sum of square from the residual sum of square. The statistical analysis was done using the function GLM of the Statistical Analysis System (Barr et al., 1976) and the APL f uhct FonfT fwas used to locate the stationary point

PAGE 64

RESULTS AND DISCUSSION Results and discussion will be presented only for the control and sprayed molasses treatments. The spring applied dalapon retarded and/or killed both the smutgrass and the bahiagrass to such an extent that regrowth of bahiagrass was not enough to allow grazing during the 1976 growing season. Treatments 3 through 11 (see page 35) were grazed from time to time but no data were collected during the first year of this multiyear study. The following notations will be used on the presentation of the results and discussion. RDM = projected residual dry matter RDMH = residual dry matter based on harvested samples only RDMA = residual dry matter adjusted to forage meter readings Res idua 1 Dry Matter Control Treatment • Projected residual dry matters (RDM) left after grazing by combination of rotation cycle and grazing pressure on the control treatment are presented in Table if. Also, included in the table are the seasonal averages of residual dry matters based on harvested samples (RDMH) and residual dry matters based on the harvested samples and adjusted to meter readings (RDMA) Si*

PAGE 65

55 Table k. Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the control treatment. Comb inat ions RDMH RDMA RC Projected RDM X CV X ^^R b CV r days metr ic tons/ha% -metric tons/ha% 56 0.50.2 .785 .221 29 .716 .147 .050 19 .84 56 0.50.2 .838 .301 36 .837 ,204 047 24 .90 k2 2.9+0.2 2.893 .660 23 2.841 .362 .049 12 .87 ]h 1 3 0 2 1 390 .447 32 1 .339 283 .045 20 .81 28 0 5 0 2 .733 .220 30 .743 1 39 051 19 .81 28 0 5 0 2 .748 .356 47 .739 .126 049 17 .94 0 3 7 0 2 3.594 .871 24 3.579 .419 .046 12 .90 0 3.7b.2 3.577 1 .053 29 3 722 .480 .046 13 .91 0 2. 10.2 2. 193 .801 36 2.088 377 045 17 .88 0 2.10.2 2.258 .778 34 2 449 .460 .046 20 .82 ]k 2.90.2 3.101 .840 27 3.059 .449 .045 14 .86 56 3.70.2 3.727 .712 19 3.419 .411 .046 1 1 .92 56 3 7 0 2 3.628 .862 24 3.665 .315 .047 q .93 1 .3t0.2 1 .436 .519 36 1 .437 .260 .043 18 .91 28 3.70.2 3.769 1 .008 26 3.729 .455 .047 1 2 28 3.70.2 3.652 1 .086 29 3.795 .376 .047 10 .94 28 2. 10.2 2.135 .697 33 2. 125 .331 .048 15 .88 28 2. 10.2 2.221 .789 35 2.205 .290 .048 13 .93 0 0.5t0.2 .783 .282 36 .789 .144 .050 18 .88 0 0.5i:0.2 .779 .259 33 .813 .149 .045 19 .88 56 2 1 0 2 2.135 .819 38 2.092 .250 .046 12 .95 56 2. 10.2 2.212 .804 36 2.072 .392 .042 18 .95 RC = Rotation cycle X = mean SD|^ = Standard deviation from mean SD„ = Standard deviation from regression

PAGE 66

56 The estimated RDMH in almost all combinations of rotation cycle and grazing pressure were within the range of projected residual dry matter/ha. The values of the RDMH ranged from 0,733 to 3.769 metric tons/ha. Residues greater than those projected occurred in all combinations in which grazing pressure was projected at 0.5 metric tons/ha of residual dry matter left after grazing. The relative high residues in this case may have been due in part to high dung spot concentration on the pastures as a consequence of the heavy stocking rate required to have a residue of 0.5 metric tons/ha. This observation is in agreement with Greenhalgh and Reid C1969), who indicated that fouling of less than 3% of the surface of a ryegrass pasture with dung resulted in over 201 rejection at a heavy grazing intensity. The coefficients of variation for RDMH ranged from 19 to h7%. The coefficients of variation increased with decreasing amount of residue left after grazing (Table h) This is apparently due to the lower mean value for the residual dry matter and the very low variation in the standard deviation among the different levels of grazing pressure. Residual dry matter based on harvested samples and adjusted to meter readings (RDMA) tended to be lower than the RDMH. The ROMA ranged from 0.716 to 3.795 metric tons/ha. The data show that like RDMH, the values of RDMA for grazing pressure equal to 0.5 metric tons/ha of residue, are also outside of the projected ranges (RDM). The regression coefficients (b) were always positive and varied from 0.0'3 to 0.051 metric tons/ha per unit of forage meter reading. Although the regression coefficients were similar, heavier grazing pressure tended to give larger regression coefficients.

PAGE 67

57 The standard deviations of the regression of RDMA ranged from 0.1^'* to O.A8O metric tons/ha. It may be noted that larger standard deviations from regression occurred at lighter grazing pressures (high RDM) than at heavier grazing pressures. The lower standard deviations from regression observed at heavier grazing pressures were probably due to a lower residual dry matter yield (RDMA) All correlation coefficients (r) were highly significant CP<0,01), ranging from O.8I to 0.95 (Table A). However, the correlation coefficients were generally slightly lower at heavier grazing pressures. The most noticeable effect of adjusting the harvested sample to meter reading was to reduce the coefficients of variatfon. The coefficients of variation of RDMA ranged from 9 to Zh%, which means a decrease in sample variation of about ^0% when compared to the variation of the RDMH. Molasses Sprayed Treatment Table 5 presents the seasonal average residual dry matters left after grazing by combinations of rotation cycle and grazfng pressure on the molasses sprayed treatment. As in the case for control treatment, most of the RDMH were within the range of projected residual dry matter/ha (RDM/ha). They ranged from 0.73'* to 3.931 metric tons/ha. However, exceptions occurred for all grazing management systems with 0,5 metric tons/ha of residual dry matter, which were greater than the projected RDM left after grazing. Similarly to the control treatment, the high frequency of dung spots on the pastures seems to have had an effect upon the amount of RDM left.

PAGE 68

58 Table 5. Residual dry matter left after grazing for different combinations of length of rotation cycle and grazing pressure on the molasses sprayed treatment. Combinat ions RDMH RDMA RC Projected RDM X ^^M CV X ^^R b CV r days MIC L 1 \ \^ % — metric tons/ha% 56 0.5+0.2 255 33 .803 .137 .051 18 .89 56 0.50.2 .853 338 40 .814 .149 .043 17 95 42 2.90.2 3.164 1 .066 34 2.970 .770 .048 24 .74 \k 1 .3+0.2 1 .367 .469 34 1 .282 .239 .043 1 7 .89 28 0.5+0.2 .734 .228 31 .712 .162 .047 22 .82 28 0.5+0.2 .750 220 29 .747 .103 .048 14 .88 0 3.7+0.2 3.931 1.012 25 3.915 .291 .047 7 96 0 3.7+0.2 3.841 970 3.819 .378 .046 1 0 • j*0 2. 10.2 2.123 .81 1 38 2. 196 .553 .048 26 0 2. 10.2 2.110 .623 29 2. 136 .299 .045 14 88 2.9+0.2 3.057 1 021 33 2.988 .564 .050 18 83 56 3. /0.2 0 TOO 3. /22 1 0^8 28 3.692 .282 .046 7 1 97 56 3.7+0.2 3.902 • 1 J 1 1 q 3.851 .338 .047 8 9"? kl 1 .30.2 1 .240 1 .233 .251 .045 • OD 28 3.7+0.2 3.726 .977 26 3.732 .384 .046 10 .92 28 3.70.2 3.664 .723 20 3.516 .549 .045 15 .77 28 2 1 0 2 2. 151 .743 34 2. 194 .300 .047 14 .92 28 2.1+0.2 2.283 .861 38 2.274 .518 .046 23 .80 0 0.50.2 .926 .336 36 .932 .197 .040 21 .88 0 0.50.2 .815 .325 39 .804 .178 .049 22 .93 56 2. 10.2 2.094 .815 39 2.033 .279 .048 13 .14 56 2. 10.2 2.031 .904 44 1 .997 .183 .050 9 .98 RC = Rotation cycle. X = Mean. SD^ = Standard deviation from mean. SD = Standard deviation from regression.

PAGE 69

59 The coefficients of variation for RDMH are also given in Table 5They varied from 19 to h^%, and were higher for heavier grazing pressure (low RDM). For most of the combinations of rotation cycle and grazing pressures, RDMA were slightly lower than the RDMH. The regression coefficients ranged from 0,0^0 to 0.051, As in the control treatment, heavier grazing pressures tended to result in larger regression coefficients. The standard deviation of the regressions of RDMA ranged from 0.137 to 0.770, and tended to decrease with heavier grazing pressures (low RDM). In all combinations of rotation cycle and grazing pressure the correlation coefficients of meter reading on residual dry matter were highly significant (P<0.01) and ranged from 0.7^ to 0.98. Lower correlation coefficients for heavier grazing pressures were evident (Table 5). Changes in Smutgrass Ground Cover Control Treatment Changes in smutgrass ground cover for the different combinations of length of rotation cycle and grazing pressure from April to October 1976, are given in Table 6. In the control treatment, the change in smutgrass ground cover ranged from -22.2 to 33.8 percentage units. The main effects of grazing pressure and length of rotation cycle on the change in smutgrass ground cover are presented in Tables 7 and 8, respectively. Changes in smutgrass ground cover indicated an increase (P<0.05) with decreasing grazing pressure (low RDM). On the

PAGE 70

60 Table 6. Observed change in smutgrass ground cover for the different combinations of length of rotation cycle and grazing pressure on the control treatment, from April to October 1976 Length of rotation Grazing pressure Observed change in cycle RDMAt smutgrass ground cover -metric tons/ha-percentag 56 0.771 2.8 56 0.837 9. 1 42 2.841 3.8 1.339 4.6 28 0.743 5.0 28 0.738 -22.2 0 3.579 33.8 0 3.722 11.7 b 2.088 12.2 0 2.449 3.4 ]h 3 059 56 3.419 15.6 56 3.665 27.4 k2 1 .437 21 .7 28 3.729 20.8 28 3.795 5.1 28 2.125 4.9 28 2.206 16.3 0 0.789 -4.9 0 0.814 0.2 56 2.092 • 6.6 56 2.073 6.1 Residual dry matter adjusted.

PAGE 71

61 Table 7. Main effect of grazing pressure on the change in smutgrass ground cover on the control treatment. Grazing pressure RDM+ Observed change in smutgrass ground cover -metric tons/ha percentage units 0.5 1.8 1-3 8.8 2.1 12.7 2.9 0.2 3.7 19.1 ^ Projected residual dry matter. Table 8. Main effect of ground cover on rotat ion cont rol cycle on the change in smutgrass treatment. Length of rotation cycl e Observed change in smutgrass ground cover percentage units-'-0 8.2 ]k 0,1 28 i.9 k2 8.5 56 7.0

PAGE 72

62 other hand, changes in smutgrass ground cover were not influenced by the length of the rotation cycle. The fitted equation for change in smutgrass ground cover on control treatment is presented in Table 9It shows a significant (P<0,05) linear response in change of smutgrass ground cover as a result of changes in grazing pressure. The other terms of the equation were not significant. The calculated stationary point for change in smutgrass ground cover on the control treatment was found to be located at 11 days for length of rotation cycle and 0.3 metric tons/ha for the adjusted residual dry matter, with a response at this point of -2. '4 percentage units (Table 10). The stationary point is outside of the experimental area with relation to grazing pressure, since the heaviest grazing pressure studied was 0.5 metric tons/ha. However, the response at the stationary point (-2.^4 percentage units) is between the minimum and maximum values found for change in smutgrass ground cover for control treatment (Table 6). The contours of predicted change in smutgrass ground cover on the control treatment are presented in Fig. 7. it can be seen from Fig. 7 that smutgrass ground cover decreases with heavier grazing pressures (low RDM). It is evident from the contours that very small changes in smutgrass ground cover occur due to the effect of length of the rotation cycle. The contours emphasize the importance of grazing pressure since there is little effect due to length of rotation cycle on smutgrass ground cover.

PAGE 73

^"^ QQCD X X C_3 *(_) CD C_D V ) CO Q_ X CO X •~~ CD o VO O X CSl o o o c_) 00 CO O 1 csj 1 o o CO 0 CN o CM OO D_ + Q. 0 o CD O J' CO • + CM 1 CO O vD CM OA CM ^ 1 — 4. QJ* CD CT% CM CM CM Q_ OO .1 CD CO + CD O CM LO o C_) CO 0 O + • + ro O CM 0 Cs| o + 0 + CJ 0^ CM C3 CM O 0 + Cd O + ci^ 0 O 0 o -ic 0 o O -IJ 0 0 Q. o CL. 0 VI I CD • 1 11% o o CO + LO I— (_D CD Q) O LO 4J QLO LO 4-J f (_3 1 CD vD (D CO OO OA r*^ E V O CJ vD LO QC CO fO >c O O 0 — Cd ^ \D 0 1 00 CM ro "D c ) O 1 CJ LO c O Cd 0 t/t (D \0 0) LO O CM vO 0 i_ 1 f CNJ LO r*^ C o I 0 • — 1 U— 4) 0 C o CO CO 0 CO 4J 1 sD fO nu + 1 LO tl CM 11 0 Lr\ LO • (/) < >CM < >CO 0 c CO CM — 0 LO M II II ij 1OA oO < >d) CM < >> > C o If 0 11 • -M /-\ \J CJ • • I/) (U < >> < >>>E "D fO (D 1 fu c c ~o T3 0) 3 \ 1 VO" o >^ 0 > 4-J >0) (U (/I jC if) (\3 JZ U 14-1 V) a> in x: ai Q. 4-* fO \ OJ \ c — ^ L. 0) u ^ (U 0 01 LA u. O) 5 CD fO 2 — c 0 0 *J E 0) M H • • > 3 — > 0 E c E C m V V CO < CO < —J 0 Q. Q. 0£ C3 1 1 1 1 ab "loyiNoo S3SS\/10W <_) Q. 1 it

PAGE 74

> -J (_) q: -3•E < t/> c o Q. ui 0) a. o DC 4-1 L. +c 0) > o o c a o o Q. >D Wl 1. C OJ TO 3 a: C O o u 4-1 (D 1/1 4-" 1/) a. 1/1 re D1 4-) 4-1 3 E 14to a: CM -JLTV -3O O — vD I O OO I -3O c 0) E 4-> re c o <_) i/i 0) 1/1 re i_ (U 4J 4-> re • 0) E L. re >. 4-1 i_ c u -a 3 0) X 0) re cn L. 3 re (U T3 4-( a c l/l 0) (/I (U o c 1_ i_ o (U 4-1 UQ. o I u L. re I> 4J \ 0 (U 1/1 o E c 1 o •o >4-1 c re 3 u o o >1re (U cn XJ i_ 4-1 re ~ E 1/1 u o re 1 re re OJ 1/1 i_ o \ Q. C Q. -C c O) o Ol Vl c (U (U 4-1 O) re re N c E *j re re > o >_ jr c QC 1 tD o < O 1 CL 1 +1 H1 CD

PAGE 75

cn

PAGE 76

66

PAGE 77

67 A general effect of increasing the grazing pressure (low RDM) via-^ to reduce in smutgrass ground cover in the pasture. The results obtained in this experiment indicate that heavy grazing pressures are needed to control smutgrass. Similar results were reported by Campbell and Beale (1973), who reported that increased grazing pressure throughf)ijl the growing season reduced the proportion of barley grass in the pasture. In this study it was observed that smutgrass was more attractive to animals in its early growth stages than in later stages, especially after flowering. This was consistent with the findings of Riewe et al. (1975b) who reported that grazing preference for smutgrass was higher during early spring than after May. Molasses Sprayed Trea tment Table 11 presents the changes in smutgrass ground cover for the different combinations of length of rotation cycle and grazing pressure on the molasses sprayed treatment, from April to October. The chanqe in smutgrass ground cover varied from -20.0 to 31.8 percentage units. The main effects of grazing pressure and length of rotation cycle upon the change in smutgrass ground cover are shown in Tables 12 and 13, respectively. Smutgrass ground cover increased linearly (P<0.01) with decreasing grazing pressure (high RDM). It was not affected (P>0.05) by the length of rotation cycle but smutgras increased slightly as the length of rotation cycle increased. Smutgrass ground cover on the sprayed molasses treatment responded (P<0.01) to grazing pressure (Table 9)The other terms of the model were not significant (P>0.05)'

PAGE 78

68 Table 11. Observed change in smutgrass ground cover for the different combinations of rotation cycle and grazing pressure on the molasses sprayed treatment, from April to October 1976. Length of rotation Grazing pressure Observed change in cycle RDMAt smutgrass ground cover -days — -metric tons/ha-percentage 0. 803 6.4 56 0.8l'4 6.7 k2 2.970 9.2 \k 1.282 8.2 28 0.712 y • J 28 0 Ikl '98 0 18.0 0 3 819 9.4 0 2. 196 1 ^ 8 0 2. 1 36 3.9 \k 2.988 -0.1 po 0 too ? 1 fl ^1.0 56 3.851 29.3 42 1.233 3.7 28 3.732 15.6 28 3. 516 12.9 28 2.19'< 18.9 28 2.274 1.7 0 0.932 -20.0 0 0.804 5.5 56 2.033 23.2 56 1.997 8.9 Residual dry matter adjusted.

PAGE 79

69 Table .12. Main effect of grazing pressure on the change in smutgrass ground cover on the eentrol treatment. Grazing pressure Observed change in I RDM"!" smutgrass ground cover -metric tons/ha0.5 1-3 2.1 2.9 3.7 percentage units 5.2 3.9 5.8 3.7 19.5 Projected residual dry matter. Table 13. Main effect of rotation cycle on the change in smutgrass ground cover on sprayed molasses treatment Length of rotation Observed change in ^ycle smutgrass ground cover days percentage units-0 2.7 I'' 3.6 2^ 5.0 ^2 2.5 56 11.4

PAGE 80

70 The stationary point for change in smutgrass ground cover on the molasses sprayed treatment (Table 10) was determined to be at 5 days for rotation cycle and 0.4 metric tons/ha of RDMA. The response at that point was found to be -8.5 percentage units. It can be seen that the stationary point is very close to the experimental region and that the response (-8.5 percentage units) is between the lower and higher values observed for change in smutgrass ground cover (Table 11). The contours of change in percentage units of smutgrass ground cover on the molasses sprayed treatment are plotted in Fig. 8. It can be observed that significant decreases in smutgrass ground cover may be achieved with heavier grazing pressures. Another trend is that for any level of grazing pressure, smutgrass ground cover increased with an increase in length of rotation cycle. However, a much greater decrease may be attained with heavier grazing pressures than with shorter length of rotation cycle. The contours suggest a fairly rapid reduction in smutgrass ground cover with a combination of heavy grazing pressure and short rotation cycle. Statistical analysis revealed no differences (P>0.05) between the response surfaces for the control and molasses treatments. In this experiment, initially the animals showed a preference for the sprayed pastures; however, the preference did not persist for more than 2k hours. This fact appears to be similar to that reported by O'Bryan (i960) in which the preference for the sprayed pastures did not persist for more than 2 days. The lack of response to the molasses sprayed treatment may have been due to the low rate of molasses used. This observation agrees with those obtained by Bishop (1959) using a similar rate of molasses.

PAGE 81

a 9) > 10 u a in lA (/> 4> JZ c o (A c 3 0) C7) (0 M C 0) u I. 0) a u 0) > O u o c 3 o u a> I/I (A 10 1O) 4-) 3 E I/) c (U cn c to -C u o 4J (/I C L. 4) 3 B o *j 4-) Q C 0) o 1_ _> 4-> CO C7

PAGE 82

72

PAGE 83

73 An ima 1 s/ha/day Control Treatment Observed number of animals/ha/day for the different combinations of rotation cycle and grazing pressure on the control treatment are shown in Table 1^. Animals/ha/day varied from ^4.3 to 28.7. The main effects of grazing pressure and length of rotation cycle on animals/ha/day are presented in Tables 15 and 16, respectively. Number of an ima I s/ha/day declined linearly (P<0,01) with decreasing levels of grazing pressure (increasing RDM). Number of animals/ha/day was not affected by length of rotation cycle. There was both a linear and quadratic effect (P<0.01) of grazing pressure upon the number of animals/ha/day on the control treatment (Table 9 and Fig. 9) The stationary point was located at 10 days for length of rotation cycle and 3.2 metric tons/ha of RDMA. The response of the stationary point at this combination is equal to 6.5 animals/ha/day. The stationary point in this case is located within the experimental region; however, it is closer to the lower observed values for animals/ha/day (Table ]k) Figure 9 represents the contours of the response of number of an ima 1 s/ha/day on the control treatment. Under this treatment, number of animals/ha/day increased with increasing grazing pressure (decreasing RDM). At heavier grazing pressure, length of rotation cycle has no effect upon animals/ha/day.

PAGE 84

7^ Table I'*. Observed animals per hectare per day for the different combinations of length of rotation cycle and grazing pressure on the control treatment. Length of rotation Grazing pressure Observed cycle RDMA+ animals/ha/day —mp f r 1 r fnn^/ha— ill^LI i \^ lid 56 0.771 20. 1 56 0.837 20.5 k2 2.8'1 ]k 1.339 14.6 28 0.7^3 28.7 28 0.738 23.5 0 3.579 6.6 0 3.722 7.3 0 2.088 9.5 0 l.kks 1 1 .2 3.059 9.6 56 3.^19 5.8 56 3.665 7.1 42 1.^*37 15.3 28 3.729 ^.3 28 3.795 5.5 28 2.125 6.7 28 2.206 7.2 0 0.789 20.7 0 0.8li 20.9 56 2.092 6.7 56 2.073 7.9 Residual dry matter adjusted.

PAGE 85

75 Table 15. Main effect of grazing pressure on animals per hectare per day on the control treatment. Grazing pressure Observed RDM^ an imal s/ha/day — m^i" r i r frin q / Ha — NIC LI 11.* LVJilo/i lO 0.5 22. ii 1.3 2.1 8.2 2.9 8.5 3.7 6.1 ^ Projected residual dry matter. Table 16. Main effect of rotation cycle on an ima 1 s per hectare per dr?v on f hp rnnf ml t rf>!^fmf^nt Length of rotation Ob<^p rvpfl ^ J ^ 1 V cycle an imal s/ha/day days 0 12.7 ]k 12.1 28 12.6 ^2 11.4 56 11.3

PAGE 87

77 o

PAGE 88

78 Molasses Sprayed Treatment The number of animals/ha/day for the different combinations of length of rotation cycle and grazing pressure on sprayed molasses treatment is presented in Table 17. Animals/ha/day ranged from 3.9 to 22.8. The main effects of grazing pressure and length of rotation cycle on animals/ha/day are listed in Tables I8 and I9, respectively. The number of animals/ha/day declined (P<0.01) with decreasing grazing pressure. Number of animals/ha/day was not affected by length of rotation cycle. The fitted equation (Table 9 and Fig. 10) indicates that there was a quadratic effect (P<0.01) upon animals/ha/day due to grazing pressure and that there was no interaction between grazing pressure and length of rotation cycle (P>0.05). The number of animals/ha/day was greatly affected only by grazing pressure and there was no response to length of rotation cycle. The stationary point for animals/ha/day was found to be located at 60 days and 3.^ metric tons/ha of RDMA (Table 10), with a response at this point of 5.8 an ima I s/ha/day The stationary point is located outside of the experimental region with relation to length of rotation cycle, because 56 days was the longest cycle studied, However, it is well within the experimental region of grazing pressure used in the experiment, The response at the stationary point is within the observed values for the different combinations of grazing pressure and length of rotation cycle (Table 17) The contours of predicted number of animals/ha/day on the molasses sprayed treatment are presented in Fig. 10, The contours show that

PAGE 89

79 Table 17,. Observed animals per hectare per day for the different combinations of rotation cycle and grazing pressure on the sprayed molasses treatment. Length of rotation Grazing pressure Observed cycle RDMA^ an ima 1 s /ha/day WW 7 -J -metric tons/ha56 0. 803 1 0 0 56 O.oi^ 1 / 0 l.n H2 2.970 7 fl /.O \k 1 .282 1 1 9 28 0.712 21.5 28 00 Q 11 0 0 3.915 0,2 0 3.0I9 7. 1 0 2. 196 ^ ft C 0.5 0 2. 1 36 10.1 \k 2.988 0 t 0.1 56 3.692 6.1 56 3.851 5.2 42 1.233 12.0 28 3.732 3,9 28 3.516 5,1 28 2.194 s,u 28 2.274 6,8 0 0.932 22.3 0 0.804 20,2 56 2.033 10.1 56 1.997 10.3 Residual dry matter adjusted.

PAGE 90

80 Table 18. Main effect of grazing pressure on animals per hectare per day on the sprayed molasses treatment. Grazing pressure Observed RDM+ animals/ha/day -metric tons/ha20.^4 13.5 8.5 8.3 5.6 Projected residual dry matter. Table 19. Main effect of rotation cycle on animals per hectare per day on the sprayed molasses treatment. 0.5 1.3 2.1 2.9 3.7 Length of rotation Observed cycle animals/ha/day days 0 12.4 1^ 11.8 28 10.9 9.9 56 11.2

PAGE 91

81 an ima 1 s/ha/day increased with combination of heavier grazing pressures and shorter length of rotation cycle. However, Fig. 10 indicates that very small decreases in number of an ima 1 s/ha/day is obtained however, if we maintain a constant grazing pressure and increase the length of the rotation cycle. Liveweight /ha/day Control Treatment Liveweight/ha/day for the different combinations of length of rotation cycle and grazing pressure on the control treatment, ranged from 1.^37 to 9.084 metric tons/ha/day (Table 20). Liveweight/ha/day shows a decrease (P<0,01) with decreasfng (Table 21) grazing pressure (high RDM). The amount of liveweight/ha/day was not influenced by length of rotat ion cycle (Table 22). The fitted equation for liveweight/ha/day on the control treatment shows a highly significant (P<0.01) linear and quadratic effect of grazing pressure. The other terms of the equation were not significant (Table 9and F ig. 1 1 ) The computed stationary point was detected at 31 days for length of rotation cycle and 3.3 metric tons/ha of residual dry matter (Table 10), with a response at that point of 2.2 metric tons/ha/day of liveweight. The stationary point for liveweight/ha/day on the control treatment is located within the experimental region and is close to the lower values observed in the experiment (Table 20). Contours of response of liveweight/ha/day are presented in Fig. 11. The contours show that amount of liveweight/ha/day increases with heavier

PAGE 92

iZ

PAGE 93

83

PAGE 94

8k Table 20. Observed liveweight per hectare per day for the different combinations of rotation cycle and grazing pressure on the control treatment. Length of rotation Grazing pressure Observed cycle RDMA"^ 1 i vewe ight/ha/day -days -me trie tons/ ha --metric tons 56 0.771 5.987 56 0.837 6.889 k2 2.8'tl 2.512 14 1.339 5. 061 28 0.7't3 8.627 28 0.738 9.08^4 0 • 3.579 2.209 0 3.722 2.519 0 2.088 3.196 0 l.hks 3.532 \h 3.059 3. lOi* 56 3.^19 2.361 56 3.665 2.3^0 kl 1 .437 5.172 28 3.729 1.^37 28 3.795 1.772 28 2.125 2.291 28 2.206 2.506 0 0.789 7.105 0 0.8\h 6.917 56 2.092 2.890 56 2.073 4.039 Residual dry matter adjusted.

PAGE 95

85 Table 21. Main effect of grazing pressure on 1 ivewerght per hectare per day on tfie control treatment. Grazing pressure RDM"}" Observed ivewe ight/ha/day -metr ic tons/hametric tons 0 5 IMS 1 3 5.116 2 1 3.075 2 9 2,8o8 3. 7 2.106 Projected residual dry matter, Table 22. Main effect of rotation cycle on liveweight per hectare per day on the control treatment. Length of rotat ion eye 1 e Observed iveweight/ha/day days— metric tons 0 \k 28 kl 56 i.246 i.082 4.286 3.8i2

PAGE 96

86 grazing pressures (low RDM). The figure also points out that for a constant grazing pressure the 1 iveweight/ha/day is not influenced by length of rotation cycle. Molasses Sprayed Treatment Liveweight/ha/day for the different combinations of length of rotation cycle and grazing pressure on the molasses ^,prayed treatment is listed in Table 23. The values ranged from 1.305 to 7-988 metric tons/ha. Tables 2k and 25 present the effects of grazing pressure and length of rotation cycle, respectively, upon liveweight/ha/day. Liveweight/ha/day declined linearly (P^O.Ol) with decreasing grazing pressure. Liveweight/ha/day shows a tendency to decrease with increasing length of rotation cycle (Table 25). The fitted equation (Table 9 and Fig. 12) for 1 i vewe ight/ha/day on the molasses sprayed treatment shows a significant (P<0.01) linear effect of grazing pressure on liveweight. The interaction between grazing pressure and length of rotation cycle was also significant (P<0.1) for liveweight/ha/day. The stationary point was located at hi days for length of rotation cycle and 3-7 metric tons/ha of residual dry matter. At that point the liveweight/ha/day was found to be equal to 1.9 metric tons/ha. The stationary point is located inside of the experimental area and is approximately in the middle of the values observed for liveweight/ ha/day (Table 23) The contours obtained for liveweight/ha/day on the sprayed molasses treatment is shown in Fig. 17. They indicate that liveweight/ha/day

PAGE 98

88 SMOij -S^Ud 'ZdUQ

PAGE 99

89 Table 23. Observed liveweight per hectare per day for the different combinations of rotation cycle and grazing pressure on the sprayed molasses treatment. Length of rotation Grazing pressure Observed cycle RDMA"'' 1 iveweight/ha/day Ua y b lllcLi 1 LUIIb/lld"56 0.803 6.102 56 0.8]k 6. 144 m 2.970 2.814 ]k 1.282 4.938 28 0.712 7.385 28 0.7A7 7.988 0 3.915 2.044 0 3.819 2.375 0 2.196 2.793 0 2. 136 3.478 ]k 2.988 3.081 56 3.692 1.973 56 3.851 1 .932 it2 1 .233 4.103 28 3.732 1 .305 28 3.516 1.844 28 2.19^ 1 .937 28 2.27^ 2.342 0 0.932 7.414 0 0.80^* 7.005 56 2.033 3.432 56 1.997 3.496 t Residual dry matter adjusted.

PAGE 100

90 Table 2k. Main effect of grazing pressure on liveweiqhtper hectare per day on the sprayed molasses treatment. Grazing pressure RDM+ Observed 1 i vewe ight/ha/day -metric tons/ha-metric tons0.5 1.3 2.1 2.9 3.7 7.006 '4.520 2.912 2.9I6 1.912 Projected residual matter, Table 25. Main effect of rotation cycle on liveweight per hectare per day on the sprayed molasses treatment. Length of rotat ion eye 1 e days metric tons 0 k.]Sh ]h k.003 28 3.799 m 3.'*58 56 3.Sh6 Observed 1 iveweiqht/ha/day

PAGE 101

91 Increased with heavier grazing pressures and shorter length of rotation cycle. From Fig. 12 also can be seen that the length of rotation cycle when shorter than 14 days had little influence upon I iveweight/ha/day. However, 1 ivewe ight/ha/day decreased by increasing length of rotation cycle. Liveweight/ha/day is reduced with lighter grazing pressures and longer lengths of rotation cycle.

PAGE 102

in (U I/) I/) ID MO M E 0) E o u 4-1 fD c (U o l_ <_3 4-> CM C31

PAGE 103

93

PAGE 104

SUMMARY AND CONCLUSIONS A grazing experiment was conducted from April to November 1976, at the Beef Research Unit, which is located approximately 21 kilometers northeast of Gainesville, Florida. The experimental pasture was a mixture of bahiagrass ( Paspalum notatum Flugge) smutgrass ( Sporobol us po i ret i i [Roem. and Schult.] Hitchc.)and white clover ( Tr ifol ium repens L.). The experimental area had an initial smutgrass ground cover between ^jO to 50%, which is typical of many pastures in Florida. The -ma in purpose of the research was to determine the effects of different combinations of lengths of rotatj^on cycle and Jevel^s of grazing pressure, and to evaluate the applicability of spraying molasses on the pasture to increase the palatability of smutgrass. Length of rotation cycle was studied at 5 different levels (O, ]h, 28, hZ, and 56 days) and grazing pressure at 5 levels of residue left after grazing (0.5, 1.3, 2.1, 2.5, and 3-7 metric tons/ha of dry matter). The combinations of rotation cycle and grazing pressure, each at 5 levels, were arranged in a response surface type of experiment and superimposed upon control and sprayed molasses treatments. The combinations of the 2 factors each at 5 levels made up thirteen different grazing combinations. The thirteen combinations were arranged as ^4 factorial, ^ axial, 1 center, and h corner points. The axial, center, and corner points were replicated twice, making a total of 22 pastures per treatment. The central treatment was the 3k

PAGE 105

95 combination of a length of rotation cycle of 28 days and a grazing pressure equal to 2.1 metric tons/ha of residual dry matter left after grazing. The 22 pastures were assigned at random within each of the main treatments. The thirteen combinations of length of rotation cycle and grazing pressure only were superimposed upon the control treatment. The molasses treatment differed from the control in that molasses was sprayed on the fol iage before animals were given access to the pasture at the beginning of each cycle. Molasses was diluted in equal parts of water and sprayed at the rate of 320 liters/ha. Residual dry matter/ha was estimated after each grazing period by a double-sampling procedure. Twenty random readings (0.186 m ) were taken in each pasture with a capacitance meter. From these twenty, k were randomly selected and harvested. Capacitance meter readings and dry weight of the cut samples were used as independent and dependent variables, respectively, in a regression analysis forced through the or i g i n In order to study the effect of the treatments on the change in j 2 smutgrass ground cover, h non-randomly selected 1 m permanent quadrats were set up in each pasture. A charting method was used to evaluate the change in smutgrass ground cover. Ground cover was estimated in early April before applying the treatments and again late in October. The change in ground cover of smutgrass for each pasture was determined by the difference between the ground cover estimates made in October and Apr i 1976.

PAGE 106

96 The response of the pasture to the main treatments and qrazinq management was measured in terms of the following response variables: change in smutgrass ground cover, animals/ha/day, and 1 i vewe i ght /ha/ day. (spraying molasses in the pasture resulted in very little increase of palatability of smutgrass when compared to the control treatment^ Despite the small effect of molasses, it was observed that the animals showed preference for the sprayed pastures for a few hours after spraying. The lack of a molasses effect may be due to the low rate of molasses used in this experiment. (^mutgrass ground cover on the control treatment decreased primarily with heavier grazing pressure. Little effect of length of rotation cycle was observed^ Smutgrass ground cover on the sprayed molasses treatment decreased with heavier grazing pressure associated with shorter rotation cycle. However, grazing pressure was much more important than length of rotation cycle. Animals/ha/day on the control treatment increased with increasing grazing pressure. Length of rot at ion cycle had almost^o effe ct upon animals per hectare pe r d£y The number of animals/ha/day on the molasses treatment was mainly a function of grazing pressure. However, the number was maximum at heavier grazing pressures associated with shorter rotation cycles. Liveweight/ha/day on the control treatment increased with heavier grazing pressure. Very little effect of length of rotation cycle was observed.

PAGE 107

97 LIvewe ight/ha/day on the sprayed molgsses treatment wag also primarily a function of grazing pressure. Higher 1 iveweights/ha/day were found at heavier grazing pressures in combination with shorter rotation cycles. The results suggested that grazing management may be a successful way to control smutgrass in pasture. One season of grazing resulted in measurable _d e£r,ea s,e.5_-.. ili..-.S.raw tg r a s s g round cove r^ e„. t o.„h.e^ viej:., .ar a z i n g pressures. Heavier grazing pressure improved vigor and production of desirable species (mainly white clover) by allowing less competition by smutgrass. Under the conditions studied, the animals were stressed and lost weight. The loss. of weight observed seems to be related to the very low acceptability of the smutgrass but also in part to the initial large ground cover percentage, and the high quantity of accumulated mature smutgrass at the beginning of the experiment. This suggests that other control methods should be applied in combination with grazing management in order to eliminate the old material before using grazing animals. Further studies are suggested; 1) increasing the rate of molasses used; 2) changing the time at which grazing starts, in terms of the growth stage of smutgrass (early in the spring); 3) applying other control methods before superimposing grazing management systems, and studying different grazing management combinations to determine animal response (weight changes) for each system.

PAGE 108

LITERATURE CITED Alcock, M. B. and J. V. Lovett. 1967. The electronic measurement of the yield of growing pasture. I. A statistical assessment. J, Agric. Sci., Camb. 68:27-38. Alley, H. P. and D. W. Bohmont. 1958. Big sagebrush control. Wyoming Agric. Expt. Sta. Bull. SS'*. 6 p. Anderson, K. L. 19'*2. A comparison of line transects and permanent quadrats in evaluating composition and density of pasture vegetation of the tall prairie grass type. J. Amer. Soc. Agr. 3'*:805822. Back, H. L., F. E. Alder, and B. G. Gibbs. 1969. An evaluation of an electronic instrument for pasture yield estimation. J. Brit. Grassld. Soc. 2^4:168-172. Barr, A. J., J. H. Goodnight, J. P. Sail, and J. T. Helwig. 1976. A user's guide to Statistical Analysis System. Sparks Press. Raleigh, N. C. Barrett, D. W. G. W. Arnold, and N. A. Campbell. 1973. Effects of time and rate of paraquat application on yield and botanical composition of annual pastures containing subterranean clover in a Mediterranean climate. Aust. J. Exp. Agric. Anim, Husb. 13:556-562. Bendall, G. M. 1973. The control of slender thistle, ( Carduus pycnocephalus L.) and C^. tenuiflorus Curt, (compos i tae) in pasture by grazing management. Aust. J. Agric. Res. 2^:831-837. Bishop, E. J. B. 1959The value of urea and molasses proved at Dohne. Farming South Afr. 35:27-29. Bransby, D. I A. G. Matches, and G. F. Krause. 1977. Disk meter for rapid estimation of herbage yield in grazing trials. Agron. J. 69:393-396. Brown, D. ]3S^Methods of surveying and measuring vegetation. Commonw. Bur. Past. Fid. Crops. Hurley, England. Bull. No. hi. Bryan, G. G. and W. E. McMurphy. I968. Competition and fertilization as influences on grass seedlings. J. Range Manage. 21:98-101. Bryan, W. W. 1970. Changes in botanical composition in some subtropical sown pastures. Int. Grasland Cong., Proc. 11th. Surfers Paradise, Australia. 636-639. 98

PAGE 109

99 Cameron, I. H. and D. J. Cannon. 1970. Changes in the botanical composition of pasture in relation to rate of stocking with sheep, and consequent effects on wool production. Int. Grassland Cong. 11th. Surfers Paradise, Australia. 6A0-643. Campbell, A. G. D. S. M. Phillips, and E. D. O'Reilly. 1962. An electronic instrument for pasture yield estimation. J. Brit. Grassld. Soc. 17:89-100. Campbell, R. J. and J. A. Beale. 1973. Evaluation of natural annual pastures at Trangie in central western New South Wales. 2. Botanical composition changes with particular reference to Hordeum lepor inum Aust. J. Exp. Agr. Anim. Husb. 13:662-668"! Carlisle, V. W. and W. L. Pritchett. 1971. Soil identification handbook: Selected soils in the thermic temperature zone of the lower coastal plains. Univ. of Florida. Soil Sci. Dept, in coop, with USDA Soil Conservation Serv. Carpenter, L. H., 0. C. Wallmo, and M. J. Morris. 1973. Effect of woody stems on estimating herbage weights with a capacitance meter. J. Range Manage. 26:151-152. Carter, W. H. I96I. Smutgrass control with a rotary tiller. Louisiana State Univ. Agric. Expt. Sta. Cir. 70. 7 p. Christiansen, W. C. I965. The influence of molasses and urea foliage sprayed and/or tank-fed upon performance of cattle grazing unpalatable grass in Uruguay. Int. Grassland Cong., 9th., Sao Paulo, Sao Paulo, Brazil. 1631-1634. Coaldrake, J. E., J. C. Tothill, and P. Gillard. I976. Natural vegetation and pasture research. In: N. H. Shaw and W. W. Bryan (ed.) Tropical pasture research, principles and methods. Commonw. Bur. Past. Fid. Crops. Hurley, England. Bull. No. 51. Coombe, J. B. and D. E. Tribe. I962. The feeding of urea supplements to sheep and cattle: The results of penned feeding and grazing experiment. J. Agric. Sci. 55:125-l'tl. Crombie, A. C. 19^*7. Interspecific competition. J. Anim, Ecol. 16: 44-73. Currey, W. L. and P. Mislevy. 1974. Smutgrass control in Florida. Down to Earth. 29:6-9. Currey, W. L., R. Parrado, and D. W. Jones. 1973. Seed characteristics of smutgrass ( Sporobolus po i ret i i ) Soil and Crop Sci. Soc, of Fla. Proceedings. 32:53-54" ison, L. 1942. A comparison of methods of quadratting short-grass vegetation. J. Agric. Res. 64:595-614.

PAGE 110

100 1^ Fletcher, J. E. and M. E. Robinson. 1956. A capacitance meter for estimating forage weight. J. Range Manage. 9:96-97, Gesink, R. W. H. P. Alley, and G. A. Lee. 1972. Chemical control of broom snakeweed and its effect on the short-grass plains in Southeastern Wyoming. Proc. W. Weed. Sci. Soc. 25:36-37. Greenhalg, J. F. D. and G. W. Reid. I969. The effects of grazing intensity on herbage consumption and animal production. ill. Dairy cows grazed at two intens it ies on clear or contaminated pasture. J. Agric. Sci. 72:223-228. Grimmett, S. G. and P. W. Weiss. I967. Response of grasslands manawa ryegrass (L ol ium perenne L. x U mult if lorum Lam.) and tallarook subterranean clover ( Trifolium subterraneum L.) to four herbicides. Weed. Res. 7:360-363. ~~ Hamill, A. S. 1975. The dollars and no sense of crop losses from weeds. Phytoprotect ion 56:121-13'4. u Heady, H. F. R. P. Gibbens, and R. W. Powell. 1959. A comparison of the charting, line intercept, and line point methods of sampling shrub types of vegetation, j. Range Manage. 12:l8o-l88. ^ ^'"'2^^ ^' '320Charting quadrats with a pantograph. Ecol 1:270Hitchcock, A. S. I906. The grasses of Cuba. U.S. National Herb. 22:18^4-258. Hitchcock, A. S. I936. Manual of the grasses of the West India. USDA Misc. Publ 2^3. Hitchcock, A. S. I950. Manual of grasses of the United States. 2nd. Edition. USDA Misc. Publ. 200. / Holscher, C. E. I959. Plant and composition. In: Techniques and methods of measuring understory vegetation. USDA, Forest Service Houston, W., V. Watson, and J. W. McKie. 1975Control of smutgrass in Mississippi. Down to Earth, 31:13-15. ^/ Hutchings, S. S. and C. P. Pase. I962. Measurement of plant coverbasal, crown, leaf area. In: Range Research Methods. USDA Forest Service, Miscellaneous Publ. No. 9^10. ^ Johns, G. G. and B. R. Watkin. I965. A modified capacitance probe technique for estimating pasture yield, 2. The effect of different pastures, soil types, and dew on the calibration. J Brit Grassld. Soc. 20:217-226.

PAGE 111

101 Johnson, B. J. 1975. Smutgrass control with herbicides in turfgrasses. i^o Weed Sci 23:87-90. Johnston, A., J. F. Dormaar, and S. Smol iak. 1971. Long-term grazing effects on fescue grassland soils. J, Range Manage. 2':l85-l88. "jones, R. J. and K. P. Haydock. 1970. Yield estimation of tropical and temperate pasture species using an electronic capacitance meter. J. Agric. Sci,, Camb. 75:27-36. Klingman, D. L. 1970. Brush and weed control on forage and grazing lands. In: FAO International Conference on Weed Control, 401-^2'*. Klingman, D. L. and M. K. McCarty. 1958. Interrelations of methods of weed control and pasture management at Lincoln, Nebraska, 1955. USDA Tech, Bui 1 1 l80. Koger, M., W. G. Blue, G. B. Killinger, R. E. L. Greene, H. C. Harris, J. M. Myers, A. C. Warnick, and N. Gammon, Jr. 1961, Beef production, soil, and forage analysis, and economic returns from eight pasture programs in north central Florida. Univ. Florida I FAS Agric. Exp. Sta. Bui. No. 631. Kohn, G. D. and E. G. Cuthbertson. 1975. Response of skeleton weed (C hondr i 1 la juncea ) to applied superphosphate and grazing management. Aust. J. Exp. Agr. Anim. Husb. 15:102-10^. Laycock, W. A. 1970. The effects of spring and fall grazing on sagebrush-grass ranges in eastern Idaho, Int, Grassland Cong., Proc. 11th. Surfers Paradise, Australia, p. 52-54. Leach, G. J., R. M. Jones, and R. J. Jones. 1976. The agronomy and ecology of improved pastures. In: N. H. Shaw and W. W. Bryan (ed.) Tropical pasture research, principles and methods. Commonw, Bur. Past. Fid. Crops. Hurley, England. Bull, No. 51, Loosli, J. K. and I. W. McDonald. I968. Nonprotein nitrogen in the nutrition of ruminants. FAO Agric. Studies No. 75. 9^ p. Lovett, J. V. and V. J. Bofinger. 1970. The electronic measurement of the yield of rape ( Brass ica napus) J. Brit. Grassld. Soc, 25:119-124. Luttrell, E. S. 1976. Ovarian infection of Sporobolus poiretii by Bipolar is ravenel i i Phytopathology. 66:260-268. McCaleb, J. E. and E. M. Hodges. 1971. Smutgrass control at Range Cattle Station, Ona, Florida. Proc. S. Weed Sci. Soc. 24:182-186. McCaleb, J. E., E. M. Hodges, and W. G. Kirk. 1963. Smutgrass control Fla. Agric. Expt. Sta. Circular S-149. 10 p./y,

PAGE 112

102 McCarty, M. K. D. L. Klingman, and L. A. Morrow. 197^. Interrelations of weed-control and pasture management methods at Lincoln, Nebraska 19^9-1969. USDA Tech. Bull. No. 1^73. Mclvor, J. G. and D. F. Smith. 1973. Competitive growth of capeweed ( Arctotheca calendula ) and some annual pasture species. Aust. J. Exp. Agric. Anim. Husb. 13:185-189. Michael, P. W. I97O. Weeds of grasslands. In: R. M. Moore (ed.) Australian grasslands. Aust. Nat, Univ, Press. Canberra, Australia Michalk, D. L., C. C. Byrnes, and G. E. Robards. 1976. Effects of grazing management on natural pastures in a marginal area of southeastern Australia. J. Range Manage. 29:380-383, Mislevy, P. and W. L. Currey. 1975. Smutgrass control in Florida pastures. Sunshine State Agricultural Research Report. 20:28-29. Molinary, 0. G. 19'*9. Succession of grasses in Puerto Rico. Rev. Agric Puerto Rico. 39:200-217. Monson, W. G. 1977. Effects of paraquat on yield and quality of 'Coastal' bermudagrass Agron. J. 69:323^32'*. Morrow, L. A. and M. K. McCarty. 1976. Effect of weed control on forage production in the Nebraska sandhills. J. Range Manage. 29: 1 ^jO-I 43, Mostert, J. W. C. 1959. Remove that surplus grass material. Farming South Afr. 35:^2-kk. Mott, G. 0. 1962. Evaluating forage production. In: M, E. Heath, D. S. Metcalfe, and R. E. Barnes (ed.) Forages. Iowa State Univ. Press. Myers, L. F. and V. R. Squires. 1970. Control of barley grass ( Hordeum leporinum) by grazing management in irrigated pastures, Aust. J. Exp. Agr. Anim. Husb. 10:151-155. Neal, D. L. and J. L. Neal. 1973. Uses and capabilities of electronic capacitance instruments for estimating standing herbage. I. History and development. J. Brit. Grassld. Soc. 28:81-89. Nichols, G. M. I973. Pasture yield instrument using a radio frequency bridge. J. Brit. Grassld. Soc. 28:27-29, O'Bryan, M. S. I96O. Observations on the utilization by cattle of Axonopus affinis following foliar application of urea, molasses, and monosodium phosphate. Qsd. J. Agric, Sci. 17:135-1^45.

PAGE 113

103 Ottosen, E. M. G. W. Brown, and M. R. Maraske. 1975Str ip-graz ingadvantage or disadvantage. Qsd. Agric. J. 101:569-570. Pearce, G. A. 1972. "Spray-grazel' The answer to weeds in pastures. J. Agric, Western Australia. 13:16-19. Pearson, H. A. and L. B. Whitaker. 197^. Yearlong grazing of slash pine ranges: Effects on herbage and browse. J. Range Manage. 27:195-197. Persad, N. K. 1976. Nutritive value components and yield of smutgrass and its influence on Pensacola bahiagrass. M. S. Thesis, Univ. Flor ida. Peters, E. J. and J. F. Stritzke. 1971. Effects of weed control and fertilization on botanical composition and forage yields of Kentucky bluegrass pasture. USDA Tech. Bull. 1^*30. PI ice, M. J. 1952. Sugar versus the intuitive choice of foods by livestock. J. Range Manage. 5:69-75. Pope, L. S. R. D. Humph rey and R. MacVicar. 1955. Urea-molasses and ammoniated molasses as supplements for beef cattle on native grass. Oklahoma Agric. Expt. Sta. Misc. Publ. No. 43. Riewe, M. E. 197^*. Field experience in smutgrass control. PastureBeef Cattle Research Open House. Texas A and M Univ. Agric. Res. Sta. Anglenton, Texas. A l-**. Riewe, M. E., G. W. Evers, and G. Merkle. 1975a. Smutgrass control and winter grazing establishment with herbicides. Proc. S. Weed Sci. Soc. 28:218 (Abstr.). Riewe, M. E., G. W. Evers, and G. Merkle. 1975b. Smutgrass control and sod-seeded winter pasture. Pasture-Beef Cattle Research Open House. Texas A and M Univ. Agric. Res. Sta., Anglenton, Texas. B 1-2. Risser, P. G. 1969Competitive relationships among herbaceous grassland plants. Bot. Rev. 35:251-28'. Ritson, J. B., L. A. Edye, and P. J. Robinson. 1971. Botanical and chemical composition of a Townsville stylo-spear grass pasture in relation to conception rate of cows. Aust. J. Agric. Res. 22:993-1007. Rodel, M. G. W. 1970. Herbage yields of five grasses and their ability to withstand intensive grazing. int. Grassland Cong., Proc, 11th. Surfers Paradise, Australia. 618-621. Roseveare, G. M. 19^*8. The grasslands of Lat in America. Imp. Bureau of Past and Field Crop. Bull. 36.

PAGE 114

Rummel 1 R. S. ]Sh(). Some effects of competition from cheatgrass brome on crested wheatgrass and bluestem wheatgrass. Ecol 27: 159-167. Sacco, J. C. 1964. Os nomes populares das principais invasoras do Rio • Grande do Sul. An. V. Sem. Bras. Herb. Erv. Dan., Bahia, Brazil. 271-292. Schlundt, A. F. 1977. Effects of management treatments on the botanical composition of a bahiagrass ( Paspalum notatum Flugge) smutgrass ( Sporobolus poiretii Roem. and Schult., H itchc. ) and white clover ( Trifolium repens L.) pasture. M. S. Thesis, Univ. Florida. Scholl, J. M. and R. E. Brunk. 1962. Birdsfoot trefoil stand establishment as influenced by control of vegetative competition. Agron. J. 5^:]^2-]kh. Scholl, J. M. and D. W. Staniforth. 1957. Establishment of birdsfoot trefoil as influenced by competition from weeds and companion crops. Agron. J. '49:^32-435. Serrao, E. A. S. 1976. The use of a response surface design in the agronomic evaluation of a grass-legume mixture under grazing. Ph.D. Dissertation. Univ. Florida. Shaw, N. H. L. t'Mannetje, R. M. Jones, and R. J. Jones. 1976. Pasture measurements. In: N. H. Shaw and W. W. Bryan (ed.). Tropical pasture research, principles and methods. Commonw. Bur. Past. Fid. Crops. Hurley, England. Bull. No. 51. >Smith, A. E. 1974. How weeds influence the pasture environment. Proc. S. Weed Sci Soc. 27:35-41 Smith, D. F. I968. The growth of barley grass (Hordeum leporinum) in annual pasture. 4. The effect of some management practices on barley grass content. Aust. J. Exp. Agr, Anim. Husb. 8:706-711. Smith, J. E., A. W. Cole, and V. H. Watson. 1974. Selective smutgrass control and forage quality response in bermudagrass-da 1 1 isgrass pastures. Agron. J. 66:424-426. Smith, J. E., A. W. Cole, and V. H. Watson. 1975, Carbohydrate response of bermudag rass dall isgrass and smutgrass to atrazine, bromacil, and MSMA. Weed Sci. 23:383-385. Southwood, 0. R. 1971. The effect of superphosphate application, 2,4-DB and grazing on broomrape ( Orobanche minor) in a subterranean clover pasture. Weed Res. 1 1 :240-246. Swallen, J. R. 1955. Flora of Guatemala. 11. Grasses of Guatemala. Chicago Natural History Museum. 24:1-390.

PAGE 115

105 t' Mannetje, L., R. J. Jones, and T. H. Stobbs. 1976. Pasture evaluation by grazing experiments. In: N, H. Shaw and W. W. Bryan (ed.) Tropical pasture research, principles and methods. Commonw. Bur. Past. Fid. Crops. Hurley, England. Bull. No. 51. Tulloh, N. M., M. J. Watson, and D. A. Burnell. I963. A system for the trough feeding of supplements to grazing sheep. Aust. J. Exp. Agr. Anim. Husb. 3:276-279. U.S. Dep. Agr. 195'*. Soil survey-Alachua County, Florida. USDA. Soil Conservation Serv. in coop, with Univ. Florida Agr. Exp, Sta. Series I9A0. Wagnon, K. A. and H. Goss. I96I. The use of molasses to increase the utilization of rank, dry forage and molasses-urea as a supplement for weaner calves. J. Range Manage. l'4:5-9. Wakefield, R. C. and N. Skaland. I965. Effects of seeding rate and chemical weed control on establ ishment and subsequent growth of alfalfa ( Med icago sat iva L. ) and birdsfoot trefoil ( Lotus corniculatus L.) Agron. J. 57:5^7-550. Weaver, J. E. and F. E. Clements. 1938. Plant Ecology. 2nd ed. New York, McGraw-Hill Book. Col., Inc. Willoughby, W. M. and A. Axelsen. I96O, Selective consumption of dry pasture by sheep as affected by spraying with urea or molasses or both. Aust. J. Agric. Res. 11:828-835. Winkworth, R. E., R. A. Perry, and C. 0. Rossetti. I962. A comparison of methods of estimating plant cover in an arid grassland community J. Range Manage. 15:19^-196. Wright, R. G. Jr. 1972. Computer processing of chart quadrat maps and their use in plant demographic studies, J. Range Manage. 476-^78,

PAGE 116

BIOGRAPHICAL SKETCH Leonidas da Costa Schalcher Valle was born in Coroata, Maranhio, Brazil, on 21 August I938. In I962 he enrolled as a student in the Universidade Federal Rural do Rio de Janeiro, Brazil. He received his Engenheiro Agronomo degree in I965. in January I966 he joined the staff of the Institute de Pesquisas Agropecuar ias do Centro Sul (IPEACS), working in tropical pastures. In September 1967, he was granted a scholarship by the Ministerio da Agricultura, Brazil/USAID contract to study toward the degree of Master of Science at the Tropical Agriculture Research and Training Center, Turrialba, Costa Rica, where in September I969, he was awarded the degree. He worked as technical assistant in pastures of the Divisao de Pesquisa Zootecnica of the Departamento Nacional de Pesquisa Agropecuaria from I97O to 1974. In March 197'*, he entered the Graduate School of the University of Florida to pursue the Doctor of Philosophy degree in agronomy, under the sponsorship of the Empresa Brasileira de Pesquisa Agropecuaria (EMBRAPA). He is married to Julia W, Valle and they have three children, Rodolfo, Rafael, and Romolo. He is a member of the Sociedade Brasileira de Zootecnia, and Gamma Sigma Delta (Florida Chapter). 106

PAGE 117

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /terald 0. Mott, Chairman Professor of Agronomy i certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /ohn E. Moore 'rofessor of Animal Science 1 certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy, ^ V JOX-c.^JosepM H. Conrad Professor of Animal Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, m scope and quality, as a dissertation for the degree of Doctor of Philosophy.

PAGE 118

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Wi 1 1 i am G B 1 ue Professor of Soil Science I certify that i have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. William R. Ocumpaugh u f Assistant Professor of Agronomy This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1977 Dean, Graduate School