Citation
Growth response of tropical forage legumes to inoculation with VA mycorrhizal fungi and phosphorus application

Material Information

Title:
Growth response of tropical forage legumes to inoculation with VA mycorrhizal fungi and phosphorus application
Creator:
Medina, Onesimo A., 1953-
Publication Date:
Language:
English
Physical Description:
xii, 91 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Dissertations, Academic -- Soil Science -- UF
Forage plants ( fast )
Legumes ( fast )
Soil Science thesis Ph. D
Vesicular-arbuscular mycorrhizas ( fast )
Tropics ( fast )
Miami metropolitan area ( local )
Fungi ( jstor )
Inoculation ( jstor )
Plant roots ( jstor )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Includes bibliographical references (leaves 83-90).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Onesimo A. Medina.

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:
023247219 ( ALEPH )
19045147 ( OCLC )

Downloads

This item has the following downloads:

growthresponseof00medi.pdf

EWSUT5ZQB_TQAXHH.xml

AA00026465_00001.pdf

growthresponseof00medi_0031.txt

growthresponseof00medi_0048.txt

AA00026465_00001_0084.txt

AA00026465_00001_0006.txt

growthresponseof00medi_pdf.txt

growthresponseof00medi_0094.txt

AA00026465_00001_0055.txt

AA00026465_00001_0076.txt

AA00026465_00001_0021.txt

growthresponseof00medi_0039.txt

growthresponseof00medi_0055.txt

AA00026465_00001_0103.txt

growthresponseof00medi_0081.txt

growthresponseof00medi_0075.txt

growthresponseof00medi_0099.txt

AA00026465_00001_0082.txt

AA00026465_00001_0022.txt

AA00026465_00001_0085.txt

AA00026465_00001_0092.txt

growthresponseof00medi_0084.txt

AA00026465_00001_0024.txt

growthresponseof00medi_0041.txt

growthresponseof00medi_0021.txt

growthresponseof00medi_0038.txt

AA00026465_00001_0002.txt

growthresponseof00medi_0093.txt

AA00026465_00001_0081.txt

AA00026465_00001_0075.txt

growthresponseof00medi_0046.txt

growthresponseof00medi_0037.txt

growthresponseof00medi_0060.txt

growthresponseof00medi_0002.txt

AA00026465_00001_0036.txt

growthresponseof00medi_0061.txt

AA00026465_00001_0086.txt

growthresponseof00medi_0036.txt

growthresponseof00medi_0011.txt

growthresponseof00medi_0032.txt

AA00026465_00001_0090.txt

growthresponseof00medi_0092.txt

AA00026465_00001_0058.txt

growthresponseof00medi_0069.txt

AA00026465_00001_0027.txt

AA00026465_00001_0063.txt

AA00026465_00001_0091.txt

AA00026465_00001_0050.txt

AA00026465_00001_0070.txt

growthresponseof00medi_0080.txt

AA00026465_00001_0010.txt

AA00026465_00001_0083.txt

growthresponseof00medi_0090.txt

AA00026465_00001_0064.txt

AA00026465_00001_0017.txt

AA00026465_00001_0015.txt

AA00026465_00001_0099.txt

growthresponseof00medi_0079.txt

AA00026465_00001_0047.txt

AA00026465_00001_0051.txt

growthresponseof00medi_0013.txt

AA00026465_00001_0053.txt

growthresponseof00medi_0104.txt

AA00026465_00001_0093.txt

growthresponseof00medi_0029.txt

growthresponseof00medi_0017.txt

growthresponseof00medi_0078.txt

growthresponseof00medi_0047.txt

growthresponseof00medi_0070.txt

growthresponseof00medi_0049.txt

growthresponseof00medi_0071.txt

growthresponseof00medi_0095.txt

growthresponseof00medi_0024.txt

AA00026465_00001_0004.txt

AA00026465_00001_0043.txt

growthresponseof00medi_0040.txt

AA00026465_00001_0011.txt

growthresponseof00medi_0098.txt

growthresponseof00medi_0034.txt

growthresponseof00medi_0063.txt

growthresponseof00medi_0001.txt

growthresponseof00medi_0052.txt

AA00026465_00001_0105.txt

growthresponseof00medi_0006.txt

AA00026465_00001_0052.txt

growthresponseof00medi_0003.txt

AA00026465_00001_0035.txt

growthresponseof00medi_0016.txt

AA00026465_00001_0072.txt

AA00026465_00001_0031.txt

growthresponseof00medi_0066.txt

growthresponseof00medi_0100.txt

AA00026465_00001_0037.txt

growthresponseof00medi_0077.txt

AA00026465_00001_0069.txt

AA00026465_00001_0102.txt

AA00026465_00001_0088.txt

AA00026465_00001_0040.txt

AA00026465_00001_0026.txt

AA00026465_00001_0079.txt

growthresponseof00medi_0010.txt

growthresponseof00medi_0022.txt

AA00026465_00001_pdf.txt

AA00026465_00001_0033.txt

growthresponseof00medi_0091.txt

growthresponseof00medi_0086.txt

growthresponseof00medi_0007.txt

AA00026465_00001_0096.txt

growthresponseof00medi_0014.txt

AA00026465_00001_0073.txt

AA00026465_00001_0032.txt

growthresponseof00medi_0058.txt

growthresponseof00medi_0019.txt

AA00026465_00001_0074.txt

growthresponseof00medi_0015.txt

growthresponseof00medi_0096.txt

AA00026465_00001_0065.txt

AA00026465_00001_0066.txt

AA00026465_00001_0062.txt

AA00026465_00001_0023.txt

growthresponseof00medi_0045.txt

growthresponseof00medi_0057.txt

AA00026465_00001_0042.txt

AA00026465_00001_0012.txt

growthresponseof00medi_0083.txt

AA00026465_00001_0030.txt

growthresponseof00medi_0012.txt

AA00026465_00001_0029.txt

growthresponseof00medi_0073.txt

growthresponseof00medi_0018.txt

AA00026465_00001_0045.txt

growthresponseof00medi_0033.txt

growthresponseof00medi_0028.txt

growthresponseof00medi_0035.txt

growthresponseof00medi_0050.txt

growthresponseof00medi_0051.txt

growthresponseof00medi_0087.txt

growthresponseof00medi_0044.txt

growthresponseof00medi_0004.txt

AA00026465_00001_0020.txt

AA00026465_00001_0044.txt

growthresponseof00medi_0000.txt

growthresponseof00medi_0076.txt

AA00026465_00001_0001.txt

growthresponseof00medi_0005.txt

AA00026465_00001_0049.txt

AA00026465_00001_0025.txt

AA00026465_00001_0104.txt

growthresponseof00medi_0067.txt

growthresponseof00medi_0023.txt

AA00026465_00001_0057.txt

growthresponseof00medi_0089.txt

AA00026465_00001_0095.txt

AA00026465_00001_0059.txt

AA00026465_00001_0068.txt

AA00026465_00001_0019.txt

AA00026465_00001_0097.txt

growthresponseof00medi_0101.txt

growthresponseof00medi_0097.txt

growthresponseof00medi_0026.txt

growthresponseof00medi_0103.txt

AA00026465_00001_0094.txt

AA00026465_00001_0008.txt

AA00026465_00001_0014.txt

growthresponseof00medi_0027.txt

EWSUT5ZQB_TQAXHH_xml.txt

AA00026465_00001_0016.txt

AA00026465_00001_0046.txt

AA00026465_00001_0078.txt

growthresponseof00medi_0008.txt

AA00026465_00001_0028.txt

AA00026465_00001_0067.txt

AA00026465_00001_0061.txt

growthresponseof00medi_0072.txt

AA00026465_00001_0041.txt

AA00026465_00001_0071.txt

growthresponseof00medi_0062.txt

growthresponseof00medi_0074.txt

growthresponseof00medi_0059.txt

AA00026465_00001_0013.txt

AA00026465_00001_0003.txt

AA00026465_00001_0089.txt

growthresponseof00medi_0030.txt

growthresponseof00medi_0085.txt

growthresponseof00medi_0053.txt

AA00026465_00001_0048.txt

AA00026465_00001_0054.txt

AA00026465_00001_0080.txt

growthresponseof00medi_0009.txt

AA00026465_00001_0077.txt

growthresponseof00medi_0054.txt

growthresponseof00medi_0082.txt

growthresponseof00medi_0043.txt

AA00026465_00001_0060.txt

AA00026465_00001_0056.txt

AA00026465_00001_0034.txt

growthresponseof00medi_0025.txt

AA00026465_00001_0087.txt

AA00026465_00001_0039.txt

AA00026465_00001_0101.txt

growthresponseof00medi_0056.txt

growthresponseof00medi_0068.txt

growthresponseof00medi_0064.txt

AA00026465_00001_0005.txt

growthresponseof00medi_0020.txt

AA00026465_00001_0009.txt

growthresponseof00medi_0102.txt

growthresponseof00medi_0065.txt

AA00026465_00001_0100.txt

AA00026465_00001_0018.txt

AA00026465_00001_0007.txt

growthresponseof00medi_0088.txt

growthresponseof00medi_0042.txt

AA00026465_00001_0098.txt

AA00026465_00001_0038.txt


Full Text























GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH
VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION



By



ONESIMO A. MEDINA




















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



UNIVERSITY OF FLORIDA 1987






























To the memory of my father Efrain, who influenced me in a very special way. I am very heartbroken that he died before this work was completed.

To my mother, Guillermina, for her never ending sacrifices, her love, and prayers.
















ACKNOWLEDGMENTS



It is with deep gratitude that I express thanks to the chairman and cochairman of my committee, Dr. David M. Sylvia and Dr. Albert E. Kretschmer, Jr., respectively, for their support, constant encouragement, guidance, and friendship. I also thank the other members of my committee, Dr. G. H. Snyder, Dr. G. Kidder, Dr. N. C. Schenck, and Dr. J. B. Sartain, for their suggestions, support,. and editorial comments.

Gratitude is extended to Mr. Tom Wilson for his friendship and valuable assistance rendered in the field portion of this project.

The moral support, love, and motivation of my brothers, Oquendo, Miosotis, and Gagarin were essential to the completion of my graduate program.

This research was funded in part through USDA ARS Tropical

Agricultural Development Grants 83-CRSR-2-2134 and 86-CRSR-2-2846. This support is greatly appreciated.

Most of all, warmest thanks go to my wife, Griselda, for her

understanding and encouragement during my graduate studies. Her many hours of assistance in typing the manuscript will always be remembered. I also thank my daughter Michelle for making her Mommy and Daddy very happy.






iii
















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS .... .... . ... . ..iii

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

LIST OF FIGURES. .................. . viii

ABSTRACT . . . . . . . xi

CHAPTER I

,INTRODUCTION ... .. ...... . ... . .. 1

CHAPTER II

THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL
FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA. . 3

Introduction . . . . . 3

Materials and Methods ................. 4

Results and Discussion ................. 7

CHAPTER III

GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH GLOMUS INTRARADICES . . . 14

Introduction . . . . . 14

Materials and Methods . . . . 15

Results and Discussion ,,.. .. -, .. . ... .17

CHAPTER IV

I GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I.
SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI
UNDER AMENDED SOIL CONDITIONS.. . . . . 23

Introduction . . . . . 23

iv









Materials and Methods . . . . 24

Results and Discussion . . . . 26

CHAPTER V

GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II.
EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT
DIFFERENT P LEVELS. . . . . . 34

Introduction . . . . . 34

Materials and Methods ................. 36

Results and Discussion ................. 37

CHAPTER VI

EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE
GROWTH AND UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM,
STYLOSANTHES GUIANENSIS, AND AESCHYNOMENE AMERICANA . 45

Introduction . . . . . 45

Materials-and Methods ........ ............ 46

Results and Discussion ................. 48

CHAPTER VII

GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND
AESCHYNOMENE AMERICANA TO INOCULATION WITH SELECTED VA
1MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS . 55

Introduction . . . . . 55

Materials and Methods ................. 57

Results and Discussion ................ 59

CHAPTER VIII

CONCLUSIONS . . . . . . 79

LITERATURE CITED . . . . . . 83

BIOGRAPHICAL SKETCH ...... . .... 91





v
















LIST OF TABLES


Page


Table 2-1. Chemical characteristics of the soils sampled
for VAM fungi associated with four tropical
forage legumes at four locations in south
Florida. ................ . ..... 8

Table 2-2. Analysis of variance for percentage of
mycorrhizal root colonization and total spore
density per 100 g of air-dried soil. ....... 9

Table 2-3. Mean percentage of mycorrhizal root colonization
and total spore density of VAM fungi among forage
legumes at four locations in south Florida. .... 10

Table 2-4. .Mean spore numbers of VAM fungal species
associated with four forage legumes at four
locations in south Florida. ............ 12

Table 3-1. Percentage of mycorrhizal root colonization of
tropical forage legumes in nonpasteurized (UP)
or pasteurized (P) soil in the greenhouse after
45 days. . . . .......... 19

Table 5-1. Regression equations and coefficients of
determination (r2) showing the relationship of P level to shoot dry weights, root fresh
weights, percentage and total root length
colonized. ................ ... .. 40

Table 5-2. Analysis of variance for shoot dry weights, root
fresh weights, percentage and total root
length colonized. ................. 41

Table 6-1. Mean squares and levels of significance from
the analysis of variance for shoot dry weight,
root fresh weight, P concentration, and total P
.* and.N uptake of forage legumes. ............ 49

Table 7-1. Shoot dry weights, P concentrations, and percentage
of roots colonized of Siratro and Aeschynomene
americana seedlings at transplanting. ........ 60

vi









Table 7-2. Analysis of variance for shoot dry weights of
Aeschynomene americana harvested 2 October 1986. 61

Table 7-3. Analysis of variance for shoot dry weights
from four Siratro harvests. .......... 63

Table 7-4. Regression equations and coefficients of
determination (r2) showing the relationship
of applied P level to shoot dry weight, percentage
of root colonization, P and N concentrations, and
total P and N uptake for the first harvest
of Siratro. .................. .. 66

Table 7-5. Regression equations and coefficients of
determination (r2) showing the relationship of applied P level to shoot dry weights for
the first, third, and fourth harvest and
percentage of root colonization for the
fourth harvestof Siratro. ... . ....... 67

Table 7-6. Regression equations and coefficients of
determination (r2) showing the relationship
of applied P level to shoot dry weight,
percentage of root colonization, P concentration,
and total P and N uptake of Aeschynomene
americana . ....... ... . 68

Table 7-7. Analysis of variance table for percentage root
colonized of Siratro by VAM fungi. ........ 71

Table 7-8. Analysis of variance table for percentage root
colonized, P concentration, total P, and total
Nof Aeschynomene americana. ... ...... 72

Table 7-9. Analysis of variance table for P concentration,
total P, N concentration, and total N of
Siratro for the first harvest. ......... ...75











vii
















LIST OF FIGURES

Page


Fig. 2-1. Collection sites for VAM fungi associated with
four tropical forage legumes in south Florida. 5

Fig. 3-1. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in nonpasteurized soil in the
greenhouse after 45 days. Legume species were
Aeschynomene americana (AA), Aeschynomene
villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL),
Stylosanthes hamata (SH), Stylosanthes guianensis
(SG), and ViYna adenantha (VA). Bars represent the mean of 3 replicates. Means with the same
letter within a species are not different
(P < 0.05) ................ .... .. 18

Fig. 3-2. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in pasteurized soil in the
greenhouse after 45 days. The legume species were Aeschynomene americana (AA), Arachis sp.
(AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA).
Bars represent the mean of 3 replications.
Means with the same letter within a species are
not different (P < 0.05). ............. 21

Fig. 4-1. Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum (ETU), or the control (CON) on the shoot dry weight
and root fresh weight of Siratro at two harvests.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05). ............. 27







viii










Fig. 4-2 Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum
(ETU), or the control (CON) on the percentage of root colonization and root length colonized
of Siratro at three and two harvests, respectively.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05). ........... 28

Fig. 4-3 Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments. ........ 30

Fig. 5-1. Effect of P application on shoot-dry weight,
root fresh weight, percentage of root colonized
by VAM fungi, and total root length colonized of Siratro grown in limed nonpasteurized soil
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). 38

Fig. 5-2. Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments in
nonpasteurized soil. ................ 42

Fig. 6-1. Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Stvlosanthes guianensis. ........ 50

Fig. 6-2. Effect of fungal inoculation and P applications on
the shoot- dry weight, r6otlfresh weight, rootcolonization, P concentration, total P, and
total N of Macroptilium atropurpureum. ....... 52

Fig. 6-3. Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Aeschynomene americana. ...... 53

Fig. 7-1. Effect of P application on shoot dry weights
for the first (A), second (B), third (C), and
fourth (D) harvest of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON).
Data points are means of five replicates. ...... 64






ix









Fig. 7-2. Effect of P application on shoot dry weight and
percentage of root colonized of Aeschynomene americana grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are
means of five replicates. ............. 65

Fig. 7-3. Effect of P application on percentage of root
colonized for the first (A) and fourth (B) harvest of Siratro grown in the field and
inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON).
Data points are means of five replicates. ...... 73

Fig. 7-4. Effect of P application on P concentration,
total P, N concentration, and total N of
Siratro grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are
means of five replicates. ............. 76

Fig. 7-5. Effect of P application on P concentration,
total P, and total N of Aeschynomene americana
grown in the field and inoculated with Glomus
etunicatum (ETU), Glomus intraradices (INT),
or not inoculated (CON). Data points are
means of five replicates. ............. 77




























x
















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



GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION



By


Onesimo A. Medina

December 1987


Chairman: Dr. D. M. Sylvia
Cochairman: Dr. A. E. Kretschmer, Jr. Major Department: Soil Science

Greenhouse and field studies were conducted to determine the growth response of tropical forage legumes to inoculation with vesiculararbuscular mycorrhizal (VAM) fungi and P applications and to evaluate the effectiveness of the indigenous VAM population versus introduced species.

Root and rhizosphere soil samples of four tropical forage legume species were collected at four locations in south Florida before initiating the greenhouse experiments. Six species of VAM fungi were isolated in this survey. The occurrrence of VAM fungal species, as determined by spore numbers, was affected by legume species and location.

Shoot dry and root dry weights of 'Siratro' (Macroptilium atropurpureum Urb.), aeschynomene (Aeschynomene americana L.), xi









Aeschynomene villosa Poir., Stylo (Stylosanthes guianensis SW.), and Stylosanthes hamata Taub. were increased in pasteurized and nonpasteurized limed soil in the greenhouse after inoculation with Glomus intraradices Schenck & Smith.

Inoculation with Glomus etunicatum Becker & Gerdemann and G.

intraradices also increased the growth of Siratro as compared to other VAM fungi and the noninoculated control in limed, nonpasteurized soil
-1
fertilized with 20 mg kg-1 of P. In other greenhouse experiments, G. etunicatum and G. intraradices were effective growth enhancers of Siratro
-1
over a practical range of 2.5 to 40 mg kg-1 of applied P in a limed, nonpasteurized soil. For both fungi, increasing P above 2.5 mg kg-1 increased the percentage and total root length colonized by VAM fungi. A positive correlation was found between mycorrhizal root colonization and shoot dry weight. In a limed, pasteurized soil, inoculation with G. etunicatum increased total P and N of Siratro at 12.5 and 25 mg kg-1
-1
of applied P, but not at 50 mg kg

The effectiveness of G. etunicatum and G. intraradices with Siratro and aeschynomene was corroborated in a field trial. These fungi increased the growth and uptake of P and N of both legumes over a range
-1
of applied P from 10 to 80 kg ha-1

Inoculation of forage legumes with effective VAM fungi enhanced

their growth. Growth enhancement occurred at P and lime levels used in commercial pasture production and in soils that had a large, but apparently ineffective indigenous VAM population.






xii
















CHAPTER I
INTRODUCTION


'-.-Forage legumes-are an important component of improved grass pastures and must be established rapidly and without excessive cost. The legumes serve both to increase forage quality and decrease the need for N fertilizer through N2-fixation.

Newly cleared lands incorporated into pasture production in south Florida are generally acidic and very low in total and available P throughout the soil profiles. While improvements to the productivity of these pastures may be obtained by the introduction of suitable legumes, effective N2-fixation and establishmentof legumes is frequently limited by the low levels of available P in these soils. Snyder et al. (1978) reported that large applications of P fertilizer are normally required for legume establishment and optimun growth in these soils. However, with the increasing cost of P fertilizer, alternative strategies for minimum P fertilizer input and efficient use of P must be adopted. One of these strategies may be via the management of vesicular-arbuscular mycorrhizal-(VAM) symbioses.

:' Vesichlar-arbuscular mycorrhizal fungi can improve the growth of legumes by increasing P uptake (Bethlenfalvay et al., 1985; Hayman, 1983). Phosphorus is often a growth-limiting factor since many legumes have P requirements and are poor scavengers of P. The VAM fungi may also increase nodulation and N2-fixation of legumes, primarily as an indirect


1










2

effect of improved P nutrition (Daft and l-Giahmi, 1976; Habte and Aziz, 1985).

Information concerning the association of VAM fungi with tropical forage legumes is sparse. Most of the growth response studies reported were done in either pasteurized soil or in small volumes of nonpasteurized soil. Except for the work of Saif (1987), little information is available on growth response of tropical forage legumes to inoculation with VAM fungi in nonpasteurized soil, especially under field conditions where introduced species of VAM fungi must compete with the indigenous VAM population.

Therefore, greenhouse studies were conducted in limed, pasteurized and nonpasteurized soil to improve the growth of tropical forage legumes through inoculation with effective VAM fungi and reduced P fertilization. In addition, the effect of inoculation with selected VAM isolates on growth and nutrient uptake of two tropical forage legumes under natural field conditions was investigated at different levels of applied P.
















CHAPTER II
THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA.



Introduction


There is widespread interest in the use of tropical forage legumes to increase production of tropical grasses in Florida's beef-cattle industry (Snyder et al., 1985). These legumes respond to inoculation with VAM fungi (Lynd et al., 1985; Saif, 1987)). However, before initiating fungal inoculation experiments with these forage legumes in south Florida, a survey was needed of the native populations of VAM fungi associated with several commercial forage legumes growing on a variety of soils.

Vesicular-arbuscular mycorrhizal associations have been observed in a wide variety of natural and agricultural ecosystems (Abbott and Robson, 1977a; Currah and Van Dyk, 1986; Harley and Harley, 1987). In Florida, the occurrence and distribution of VAM fungi in agronomic crops, including some tropical legumes (Schenck and Kinloch, 1980; Schenck and Smith, 1981), and sand-dune vegetation (Sylvia, 1986), has been reported. However, there is no information on the degree of native VAM colonization of tropical forage legumes in Florida or on the susceptibility of different species of legumes to various genera and species of VAM fungi.

The objective of this survey was to obtain quantitative data on the amount of root colonization and the species distribution of VAM fungi

3









4

associated with four cultivated tropical forage legumes from four different locations in south Florida.



Materials and Methods



Root and rhizosphere soil samples of four tropical forage legumes were collected from 11 to 17 October 1984, at four locations in south Florida: Deseret Ranches, Deer park; Fort Pierce, Agricultural Research and Education Center (AREC); Ona, AREC; and Basinger Ranch, 109 Ranch (Fig. 2-1). Most of the soils of the studied area belong to the order Spodosols. They are dominated by nearly level, somewhat poorly to poorly drained sandy soils with dark sandy subsoil layers. These soils are used primarily for pastures, vegetables, flowers, forestry, and citrus.

The forage legumes sampled were: 'Siratro' (Macroptilium atropurpureum Urb.), (except at Deseret Ranches), aeschynomene (Aeschynomene americana L.), Vigna adenantha Marechal, Mascherpa and Stainier, and carpon desmodium (Desmodium heterocarpon DC.). The legumes were mixed with pasture grasses at the time of sampling. Three rhizosphere soil samples were collected to a depth of 15 cm for each legume at each location. Samples, consisting of three subsamples of approximately 1.5 kg, were placed in plastic bags and transported to the laboratory on the same day.

Samples were sieved through a 4-mm screen, and 100 g subsamples were removed and stored at 50C for spore extraction. Legume roots were carefully separated manually from grass roots. A portion (0.5 g) of each














Deseret Fort Pierce Basinger Ona





















Fig. 2-1. Collection sites for VAM fungi associated with four
tropical forage legumes in south Florida.









6

root sample was cleared in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik and McGraw, 1982). Root colonization by VAM fungi was estimated by the gridline-intersect method of Giovannetti and Mosse (1980).

Chemical content of a composite soil sample from each location was determined by the Soil Testing Laboratory, University of Florida (Rhue and Kidder, 1984). Mehlich-I solution (0.05 M HC1 + 0.0125 M H2SO4) was used to extract Al, Ca, K, Mg, and P. All elements were analyzed in the filtrate by atomic absorption spectrophotometry, except P which was determined using the ammonium molybdate/ascorbic acid colorimetric method. Soil pH was determined using a 1:2 (v/v) soil:water ratio. Organic matter was estimated by oxidation with 1 N K2Cr207 in the presence of H2SO4.

Spores of VAM fungi were removed from soil by the wet sieving method of Daniels and Skipper (1982) using sieves with 425, 90, and 45 um openings. Fractions retained on 90 and 45 um sieves were centrifuged (1000 x g) for 3 min in water. The pellet was resuspended in 40% sucrose solution and centrifuged for 1.5 min. Spore species were identified where possible (Schenck and Smith, 1982; Trappe, 1982). In addition, spores or washed roots were placed in pasteurized Arredondo loamy sand surface soil (siliceous hyperthermic Grossarenic Paleudult) in 15-cm-diam plastic pots in the greenhouse and planted with bahiagrass (Paspalum notatum Flugge), carpon desmodium, or Siratro in an attempt to isolate VAM fungi in a manner similar to that described by Gerdemann and Trappe (1974) as the "inoculated pot culture" method.









7

Results and Discussion :



Results of soil pH and chemical analysis of soil samples reflect the different management regimes (including lime and fertilizer) (Table 2-1). ....

Differences in percentage of mycorrhizal root colonization and total spore density of air dry soil were significant among locations, legume species, and location x legume species interactions (Table 2-2). Total spore density at the four locations ranged from 5 to 679 per 100 g of air-dried soil, and the percentage of mycorrhizal root colonization varied from 3 to 41%. Miller et al. (1979) observed variable degree of mycorrhizal root colonization (4 to 74%) in forage grasses and legumes in Brazil. Except for carpon desmodium, legume species differed in percentage root colonization and total spore density among locations (Table 2-3). Fort Pierce had the highest total spore density for each legume species sampled except for Siratro.

Attempts were made to relate percentage root colonization and total spore density to soil P or the other soil chemical characteristics presented in Table 2-1, but no clear relationships were apparent. Abbott and Robson (1977a) and Hayman (1978) also reported that spore numbers were not correlated with soil P or soil pH in cultivated soils.

There was a positive correlation (P < 0.05) between root

colonization and total spore density for all legume species at Basinger (r-= 0,70) and Deseret (r =-0.76), but not at Fort Pierce and Ona. Giovannetti (1985) and Miller et al. (1979) reported a correlation









8

Table 2-1. Chemical characteristics of the soils sampled for VAM
fungi associated with four tropical forage legumes at four
locations in south Florida.



Location Legume O.M. pH Al Ca Mg K P
speciesz


-1
% mg kg soil---------AA 1.4 6.0 23 314 93 8 4 Fort Pierce DH 1.2 5.3 22 241 15 16 5
VA 1.3 5.5 25 242 21 20 4 MA 1.3 5.2 62 270 25 13 16

AA 3.4 6.1 44 1320 143 64 23 Ona DH 3.1 5.4 36 920 120 29 8
VA 2.5 4.9 22 480 70 46 8 MA 5.7 4.7 55 '800 95 43 5

AA 2.5 6.1 66 780 67 40 6 Deseret DH 2.3 6.0 27 1040 94 27 4
VA 2.8 7.2 30 1600 141 55 9

AA 4.7 5.3 28 960 32 28 4 Basinger DH 3.5 5.2 26 460 100 46 7
VA 3.6 5.1 28 540 111 58 8 MA 4.2 5.2 36 640 129 94 11


ZAA= Aeschynomene americana; DH= Desmodium heterocarpon; VA= Vigna adenantha; MA= Macroptilium atropurpureum.









9

Table 2-2. Analysis of variance for percentage of mycorrhizal root
colonization and total spore density per 100 g of airdried soil.



Mean Squares
Source of Degree of Root total spore variation freedom colonization density


% no. 100 g-1 soil

Location 3 447 325871 Legumes 3 321 73895 Interaction 8 299** 88933** Error 30 13 1320


**Significant at P < 0.01









10

Table 2-3. Mean percentage of mycorrhizal-root colonization and total
spore density of VAM fungi among forage legumes at four
locations in south Florida.




Location- Root Total spore colonization densityz


% no. 100 g-l soil

Aeschynomene americana

Ft. Pierce 7b 302a Ona 6b 160b Basinger 5b 8c Deseret 30a 146b

Desmodium heterocarpon

Ft. Pierce 12a 679a Ona 15a 376b Basinger 12a 19c Deseret 16a 36c

Vigna adenantha

Ft. Pierce 5c 535a Ona 41a 77b Basinger 20b 25c Deseret 25b 36c

Macroptilium atropurpureum

Ft. Pierce 15a 23b Ona 8b 294a Basinger 3b 5b



ZMeans within a column for each legume species followed by the same letter are not different (P < 0.05) according to Duncan's multiple range test.









11

between root colonization and spore density, while Hayman and Stovold (1979) and Giovannetti and Nicolson (1983) found no correlation. This apparent discrepancy may be due to different sampling methods. Giovannetti (1985) collected samples within the same plant species and sites, while the other researchers collected samples from many different plant species and sites.

Spore production and root colonization are influenced by seasonal variations (Giovannetti, 1985; Sylvia, 1986), host plant, stage of development (Saif and Khan, 1975; Schenck and Kinloch, 1980), and soil type (Lopes et al., 1983). In this survey there was only one sampling, so it was not possible to separate seasonal or host developmental effects on root colonization and total spore density.

The 6 species of VAM fungi collected in this survey were: Gigaspora heterogama (GH) Gerdemann & Trappe, Gigaspora margarita (GM) Becker & Hall, Glomus etunicatum (ETU) Becker & Gerdemann, Glomus intraradices (INT) Schenck & Smith, Glomus sp. (GS), and Acaulospora spinosa (AS) Walker & Trappe. The unidentified Glomus sp. was dark brown to black, 200-250 ipm in diam, and had 1 wall of 8-14 pm thickness.

The occurrence of fungal species, as determined by spore numbers, was affected by the legume host and location (Table 2-4). Iabal et al. (1975) and Schenck and Kinloch (1980) also recorded differences in spore numbers among plant species. The maximum number of spores of G. margarita occurred at Fort Pierce associated with aeschynomene. Spores of G. margarita were not found associated with Siratro at any of the four locations. Spores of G. heterogama, G. etunicatum, and G. intraradices









12

Table 2-4. Mean spore numbers of VAM fungal species associated with
four forage legumes at four locations in south Florida.




Species of VAM fungiE

Location GM GH ETU INT AS GS


Aeschynomene americana

Ft. Pierce 141a 66b 21a 42b 0 16 Deseret 9b 136a 0 0 0 0 Basinger 0 8c 0 0 0 0 Ona 0 16c 29a 114a 0 0

Desmodium heterocarpon

Ft. Pierce' 2a 114a 255b- 307a 0 0 Deseret 2a 9b 0 0 0 0 Basinger 0 0 19c 0 0 0 Ona 0 0 325a 51b 0 0

Macroptilium atropurpureum

Ft. Pierce 0 10a 0 4b 0 8 Basinger 0 5b 0 0 0 0 Ona 0 0 40 254a 0 0

Vigna adenantha

Ft. Pierce ib 39a 252a 241a 0 0 Deseret 20a 15b 0 0 0 0 Basinger 0 0 25b 0 0 0 Ona 28a 0 28b ib 9 0


ZMeans within a column for each legume species followed by the same letter are not different (P < 0.05) accordingly to Duncan's multiple range test.









13

were found associated with all legumes, in at least one of the locations. Glomus heterogama occurred in greatest numbers at Deseret and Fort Pierce associated with aeschynomene and carpon desmodium, respectively. The maximum number of spores of G. etunicatum occurred at Ona and Fort Pierce associated with carpon desmodium. A high number of spores of G. etunicatum was also found at Fort Pierce associated with Vigna adenantha. Glomus intraradices was recovered in greater numbers from carpon desmodium and Vigna adenantha at Fort Pierce as well as from Siratro at Ona. The unidentified Glomus sp. occurred in lower numbers than the other two species of Glomus; it was only found at Fort Pierce, associated with aeschynomene and Siratro. Acaulospora spinosa was only recovered from Vigna adenantha at Ona.

Overall root colonization by VAM fungi was low (most values below 20%) which indicates that (1) the native population of VAM fungi is not very infective and (2) field inoculation may be effective. Attempts to establish pot cultures of VAM fungi recovered in this survey were only successful with G. etunicatum and G. intraradices. These two fungi were shown to be effective in increasing the growth of several forage legumes and were chosen for further evaluations.
















CHAPTER III
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH GLOMUS INTRARADICES



Introduction



The use of forage legumes as companion crops to increase production of grasses is becoming an established practice in order to reduce the requirement for N fertilization (Rotar, 1983). In soils where P is a limiting factor, large applications of P fertilizer are required for legume establishment and optimum growth. However, with the increasing cost of P fertilizer, alternative strategies for minimum input and efficient use of P must be adopted. It is pertinent, therefore, to evaluate whether mycorrhizal associations with forage legumes can be manipulated in order to improve establishment, P nutrition, N2-fixation, and consequently yield.

Vesicular-arbuscular mycorrhizal fungi can improve the growth of

legumes by increasing P uptake (Bethlenfalvay et al., 1985; Chulan and Ragin, 1986; Harley and Smith, 1983; Hayman, 1983; Jensen, 1984). Phosphorus is often a growth-limiting factor since many legumes have high P requirements and are poor scavengers of P. The VAM fungi may also increase nodulation and N2-fixation of legumes, primarily as an indirect effect of improved P nutrition (Daft and El-Giahmi, 1976; Habte and Aziz, 1985; Newbould and Rangeley, 1984).


14









15

Most of the literature concerning the association of VAM fungi with forage legumes is on temperate species; e.g. alfalfa (Medicago sativa L.) (Kucey and Diab, 1984; Nielsen and Jensen, 1983; Satterlee et al., 1983), white clover (Trifolium repens L.) (Newbould and Rangeley, 1984; Powell, 1979; Rangeley et al., 1982), and subterranean clover (Trifolium subterraneum L.) (Abbott and Robson, 1978). Studies on tropical forage legumes have been limited to a few species such as tropical kudzu (Pueraria phaseoloides Benth) (Salinas et al., 1985; Waidyanatha et al., 1979), leucaena (Leucaena leucocephala Dewit) (Huang et al., 1985) and Stylo (Stylosanthes guianensis SW.) (Mosse, 1977; Waidyanatha et al., 1979).

The purpose of this investigation was to evaluate the effect of a VAM fungus, G. intraradices, on the growth of several tropical forage legumes in pasteurized and nonpasteurized soil under greenhouse conditions.



Materials and Methods



The top 15 cm of a virgin Oldsmar fine sand (sandy, siliceous,

hyperthermic Alfic Haplaquods) was collected from a newly cleared area at the Agricultural Research and Educational Center, Fort Pierce, FL. The low-P soil was air-dried and sieved through a 4-mm screen. The soil had an initial pH of 4.5 (1:2 soil:water suspension) and P, Ca, Mg, and K concentrations (extracted with 0.05 M HCl + 0.0125 M H2S04) of 1, 63, 21 and 12 mg kg soil, respectively. Lime, as high calcitic limestone, was thoroughly incorporated at a rate of 1500 mg kg- soil (equivalent










16

to 3,000 kg ha-1 assuming a 15-cm depth of soil ha-1 with a bulk density of 1.3 g cm-3) and allowed to equilibrate for 30 days before initiating the study. Solutions of P, K, Mg, Cu, Mn, Zn, B, and Mo also were thoroughly mixed with the-soil to supply rates of 10, 30, 12, 1.5, 1.0,

1.0, 0.50 and 0.10 mg kg-1, respectively. A portion of the soil was pasteurized at 600C for 4 h to eliminate the indigenous mycorrhizal fungi. Then 3 kg of soil was placed into 15-cm-diam plastic pots. The pH of the soil was 6.2 at the end of the experiment.

The legumes used in the experiment were: Siratro, aeschynomene,

Aeschynomene villosa Poir., Stylo, leucaena, Stylosanthes hamata Taub., cv. 'Verano', Vigna adenantha, and Arachis sp. Seeds were scarified with sandpaper, wetted, and sprinkled with type EL "cowpea" inoculum (Nitragin Co., Milwaukee, WI) prior to planting. Five seeds of the corresponding legumes were planted per pot, and plants were thinned to one per pot 10 d after emergence.

Glomus intraradices (isolate S311), used in this study, was isolated from cultivated Vigna adenantha at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-4). Fungal inoculum was produced in pot culture in pasteurized soil containing carpon desmodium as the host plant. Pot cultures were 10-weeks old when they were harvested, mixed and used to inoculate experimental pots. Ten grams per pot of the soil-root-fungus inoculum containing approximately 200 spores was spread in a 1-cm-thick layer, at a depth of 3 to 5 cm below the soil surface.' Noninoculated control treatments received 10 g of a soil-root mixture from noninoculated pot cultures that were free of VAM fungi.










17

The experimental treatments consisted of pasteurized or

nonpasteurized soil, with or without addition of G. intraradices inoculum. The pots were arranged on greenhouse benches in a completely randomized design with three replications per treatment. The average maximum and minimum greenhouse temperatures were 37 and 260C, respectively. Pots were watered as needed to maintain soil moisture near field capacity and were re-randomized every two weeks.

Plants were harvested after 45 d. Shoot and roots were dried at 700C for 48 h and weighed. The percentage of mycorrhizal root colonization was determined as described in Chapter II.

Significant treatment effects on shoot and root dry weights within legume species were analyzed by the T TEST procedure of the Statistical Analysis Systems (SAS Institute, 1982).





Results and Discussion



Inoculation with G. intraradices in nonpasteurized soil resulted in greater shoot dry weights (P < 0.05) for five of the seven legumes tested (Fig. 3-1). Greater shoot dry weights of these legumes were positively related to increased levels of mycorrhizal colonization following inoculation (Table 3-1).

Root dry weight results were similar to those for shoot dry weight, except for Stylo, where there was no increase in root dry weight as a result of mycorrhizal inoculation (Fig. 3-1).








18



I-- a a 5 INGC
S a 0 NO-INOC Ujl b .5 a

.0
_aa






oo
I

,- 0.8

aa INOC
1 0.6 O NON-INOC


5 0.4 a aa aa r b
0 .2 ba
O

0.0 .
PR RV RS MR LL SH SS VR LEGUME SPECIES

Fig. 3-1. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
nonpasteurized soil in the greenhouse after 45 days.
Legume species were Aeschynomene americana (AA),
Aeschynomene villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL), Stylosanthes
hamata (SH), Stylosanthes guianensis (SG), and Vigna
adenantha (VA). Bars represent the mean of 3 replicates.
Means with the same letter within a species are not
different (P < 0.05).









19

Table 3-1. Percentage of mycorrhizal root colonization of tropical
forage legumes in nonpasteurized (UP) or pasteurized (P)
soil in the greenhouse after 45 days.




Legume species Mycorrhizal Root colonizationE inoculation UP P





Aeschynomene americana + 22 3 5 0

Macroptilium atropurpureum + 49 35 20 0

Aeschynomene villosa + 10
6

Stylosanthes hamata + 12
6

Stylosanthes guianensis + 28
8

Vigna adenantha + 59 53 40 0

Leucaena leucocephala + 4 3 1 0

Arachis sp. + 31 12 35 2



ZBased on a composite of three samples for each legume.

*Treatment lost to glasshouse accident.










20

There was no increase in plant growth as a result of inoculation with G. intraradices in nonpasteurized soil for Arachis sp., leucaena, and Vigna adenantha. With the exception of leucaena, this may be attributed to effective colonization by the indigenous mycorrhizal fungi in the noninoculated soil (Table 3-1).

Only five legume species (Fig. 3-2) were evaluated in pasteurized soil; three were lost in a greenhouse accident. Siratro, Arachis sp., and Vigna adenantha had greater shoot and root dry weights after G. intraradices inoculation. This increase again was related to effective root colonization by G. intraradices (Table 3-1). In another study, Siratro was shown to respond to inoculation with several VAM fungi in pasteurized soil in the greenhouse (Lopes and De Olivera, 1980).

Inoculation with G. intraradices did not increase either the shoot or root dry weights of aeschynomene or leucaena in pasteurized soil. However, mycorrhizal colonization on both legumes was very low (3%).

Leucaena has been reported to be very mycorrrhizal dependent because it has few root hairs (Huang et al., 1985; Yost and Fox, 1979). The failure of VAM fungi to colonize it in this study in both pasteurized and nonpasteurized soil may be due to incompatibility between the plant and G. intraradices as well as native VAM fungi in the experimental soil, to inhibitory soil factors on the host-VAM symbiosis, or to the relatively slow development of the root system. It has been shown that some mycorrhizal fungi may be less effective on certain plant hosts. For example, Schroder et al. (1977) reported that Glomus macrocarpum Tul & Tul increased growth of onions but not of Stylosanthes sp.







21



a a a INOC 12.0 a
0 0 NON- INOC




1.0
a a

0 0.5

o 0 b





CD0.
LW a

aa aa




baa
0.2



RA RV RAS MR LL SH SG VP
LEGUME SPECIES
Fig. 3-2. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
pasteurized soil in the greenhouse after 45 days. The
legume species were Aeschynomene americana (AA), Arachis
sp. (AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA). Bars
represent the mean of 3 replications. Means with the same
letter within a species are not different (P < 0.05).










22

The failure to obtain good colonization of aeschynomene in

pasteurized soil was unexpected, since this legume was successfully colonized in nonpasteurized soil. Unfortunately, soil chemical properties were not determined after pasteurization. Legumes are sensitive (and hence VAM fungi) to elevated Mn, which is a common occurrence in heat treated soils.

The results obtained in this study clearly demonstrate that G. intraradices can successfully compete with some of the indigenous mycorrhizal fungi present in the experimental soil and promote growth of several legumes in nonpasteurized soil. This result agrees with earlier work by Abbott and Robson (1981), Mosse (1977), and Rangeley et al. (1982), which suggests a potential for successful field-scale inoculation with effective VAM fungi.
















CHAPTER IV
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I. SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI UNDER AMENDED SOIL CONDITIONS.



Introduction



Siratro, a cultivar developed by E. M. Hutton (1962) from two Mexican accessions of Macroptilium atropurpureum Urb., is a persistent, perennial forage legume adaptable to a wide range of soil and climatic conditions. It has become widespread and is among the most versatile forage legume grown throughout tropical regions of the world (Lynd et al., 1985). In pasteurized and nonpasteurized soils, increased growth of Siratro was attained after inoculation with Glomus fasciculatum Gerdemann & Trappe (Lopes and De Olivera, 1980; Lynd et al., 1985) and G. intraradices (Chapter III).

Hayman (1982) stated that VAM fungi are probably capable of symbiosis with most plants, at least to some degree. However, there is wide variation in the ability of VAM fungi to stimulate plant growth (Miller et al., 1985; Powell, 1982; Schubert and Hayman, 1986). Lopes and De Olivera (1980), using a gamma-irradiated soil of low P-content, studied the effect of inoculation with nine species of VAM fungi on the growth of Siratro. Only inoculation with G. fasciculatum and G. macrocarpum enhanced plant growth. Abbott and Robson (1981) defined the relative ability of a VAM fungus to stimulate plant growth as 'effectiveness' and

23









24

this defined term will be used in this paper. Wilson (1984) indicated that an evaluation of the effectiveness of the indigenous mycorrhizal population under amended soil conditions, as well as studies to select effective VAM fungi, are prerequisites for successful field inoculation.

Thus, the objective of the present study was to determine the effectiveness of several VAM fungi with Siratro in a limed, nonpasteurized soil with low P content under greenhouse conditions.



Materials and Methods



The soil used in this study, liming, and fertilizer amendments are described previously in Chapter III.

Plants were inoculated with the following VAM fungi: G. etunicatum (isolate S312) obtained from carpon desmodium at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-4); G. deserticola Trappe, Bloss & Menge (isolate S305) obtained from sea oats (Uniola paniculata L.) in a coastal dune, Anastasia, FL.; G. versiforme Berch & Fortin (isolate #231) obtained from N.C. Schenck, University of Florida, Gainesville, FL.; G. intraradices (isolate S311) obtained from Vigna adenantha at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-4); G. margarita (isolate #215) obtained from N.C. Schenck, University of Florida, Gainesville, FL. Isolates were maintained in pot cultures in pasteurized soil containing bahiagrass. Soils from 12-week-old pot cultures were used to inoculate experimental pots. The propagule densities of the native soil and inocula at the beginning of the experiment were determined by the most-probable-number









25

(MPN) technique using bahiagrass as the host plant and pasteurized Oldsmar fine sand as the diluent (Daniels and Skipper, 1982). The amount of inoculum used was adjusted to give equal inoculum densities among isolates. Each pot received approximately 240 propagules. Details on the fungal inoculation technique, planting, and watering were reported previously (Chapter III).

There were six inoculation treatments, five species of VAM fungi, and a control inoculated with non-VAM pot culture material. The pots were arranged on the greenhouse bench in a completely randomized block design with 15 replications per treatment.

The average maximum and minimum greenhouse temperatures during the experimental period were 32 and 190C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1200 u mol m s

Five randomly selected samples were harvested from each treatment after 20, 40, and 70 d of growth. At first harvest, shoot dry weight, percentage of root colonized by VAM fungi, plant height, and number of leaves were determined. In addition, root fresh weight and total root length colonized were determined at the second and third harvests. Shoot dry weight was determined by drying the material at 700C for 24 h. Percentage and total root length colonized were estimated by the gridline intersect method (Giovannetti and Mosse, 1980) after roots were cleared in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik and McGraw, 1982). Data were analyzed by Analysis of Variance Procedure, Statistical Analysis Systems (SAS Institute Inc., 1982). Duncan's multiple range test was used to separate treatment means when the F-test was significant (P < 0.05).









26

Results and Discussion



Plants inoculated with G. etunicatum and G. intraradices had higher shoot dry and root fresh weights than plants inoculated with the other VAM fungi or control plants, at 40 and 70 d after planting (Fig. 4-1). At both harvests, plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices. At the final harvest, plants inoculated with G. intraradices had higher root fresh weights than plants inoculated with G. etunicatum.

In contrast, plants inoculated with G. versiforme, G. margarita, and G. deserticola had shoot dry and root fresh weights that were not different from the noninoculated plants, except at the final harvest when plants inoculated with G. deserticola had higher shoot dry and root fresh weights than the control (Fig. 4-1). At 20 d, shoot dry weights were not different among treatments (mean = 0.70 g).

Percentage and total root length colonized by VAM fungi increased

with time (Fig. 4-2). Inoculation with G. etunicatum and G. intraradices resulted in the highest root colonization at all harvests. At 70 d, plants inoculated with G. etunicatum had the highest root colonization, followed by G. intraradices and then G. deserticola. There were no differences in root colonization among G. versiforme, G. margarita and control treatments. For the six treatments, total root length colonized by VAM fungi and percentage of mycorrhizal root colonization followed the same trend (Fig. 4-2).

Shoot dry weight of Siratro was correlated with total root length colonized by VAM fungi (r2 = 0.95^*) and percentage of mycorrhizal root









27







S40 ORYS C

E6 70 RYS



d d d

O













U) 4 d d
0b

~-



0,







CON MAR VER DES INT ETU








VAM FUNGI



Fig. 4-1. Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus
intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the shoot dry weight and root fresh weight of
Siratro at two harvests. Bars represent the means of five
replicates. Means with the same letter within a harvest
are not different (P < 0.05).
are not different (P C 0.05).








28


50

40 20 DAYS
Z 40
o 40 DAYS

< o S0 70 DAYS b

O
20
0
I- d

o 10



E

9 40 ORYS
Z 70 DAYS
0 b
0N

S4


Cr0 z



0 0_b b

CON MAR VER DES INT ETU

VAM FUNGI




Fig. 4-2 Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the percentage of root colonization and root length colonized of Siratro at three and two harvests,
respectively. Bars represent the means of five replicates.
Means with the same letter within a harvest are not
different (P < 0.05).









29

colonization (r2 = 0.83**). There was a quadratic relationship between shoot dry weight and length of Siratro root colonized by VAM fungi for all inoculated treatments (Fig. 4-3).

Plant height and number of leaves per plant were not different among treatments at any harvest. After 70 d, the mean plant height and number of leaves for all treatments were 90 and 15 cm, respectively.

There were 2 propagules per gram of soil in the native soil as determined by the MPN test at the beginning of the experiment.

Inoculum density is known to influence plant growth response to VAM fungal inoculation (Hass and Krikum, 1985; Wilson, 1984). Thus, one of the problems in comparing the efficacy of VAM fungi is ensuring uniform inoculum densities (Daniels et al., 1981). In this study, I used the MPN technique to provide a measure of the inoculum densities of the VAM fungi, and I adjusted the inoculum densities so that they were uniform for all inoculated treatments.

There were striking differences in the effectiveness of VAM fungi on Siratro. The results are consistent with the findings of others (Miller et al., 1985; Schubert and Hayman, 1986) indicating that different species and strains of VAM fungi vary considerably in the benefits they confer to the host plant. This experiment also confirmed previous work (Chapter III) which demonstrated that the native population of VAM fungi in this soil was less able to stimulate the growth of Siratro than effective, introduced species. It is possible that the decreasing soil acidity obtained by liming changed the native population of VAM fungi from effective to ineffective as compared to G. etunicatum and G. intraradices (Hayman and Tavares, 1985). Since it is necessary to lime







30










I
4

Aa
r: Y= 0.38 + 1.08X 0.07X
2 z
-r



0 2 4 6 8 10
ROOT LENGTH COLONIZED (m)










Fig. 4-3 Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments.










31

this soil for satisfactory establishment and growth of legumes (Snyder et al., 1985), VAM fungi must be selected for their effectiveness under amended soil conditions.

Powell (1980a) reported a relationship between the level of native inoculum density in the soil and plant growth response to mycorrhizal
-1
inoculation. When inoculum density was low (0.01-0.09 propagules g soil), there was a significant plant growth response to inoculation with
-i
VAM fungi. When inoculum density was higher (0.15-0.30 propagules g soil), there was little plant growth response to fungal inoculation. Likewise, good plant growth responses to inoculation with VAM fungi in soils with few indigenous endophytes have been reported by Mosse (1977) and Hall (1979). Thus it would seem that the most promising sites for inoculation with VAM fungi are those where indigenous populations of VAM fungi are very low. However, in this study, where the native inoculum
-1
density was relatively high (2 propagules g-1 soil), Siratro responded to inoculation with two of the four VAM fungi tested. In addition to the abundance of the indigenous VAM fungi, information about their infectivity and effectiveness is needed to assess potential sites for responsiveness to inoculation with effective VAM fungi.

The ineffectiveness of G. versiforme and G. margarita could be due to an innate symbiotic inefficiency, incompatibility, lack of competitiveness, or to inhibitory edaphic (e.g. soil pH or P level) or environmental factors (e.g. light and temperature). Hayman and Tavares (1985) showed clearly that different endophytes vary in their symbiotic effectiveness at different soil acidities. In addition, some endophytes may be less effective on certain plant hosts. For example,









32

Schroder et al. (1977) reported that G. macrocarpum increased growth of onions but decreased growth of Stylosanthes sp.

The most effective fungi in this study were those that colonized the root most rapidly. Sanders et al. (1977) and Abbott and Robson (1978) reported that VAM fungi differ in their rates of root colonization. Abbott and Robson (1981) stated that differences in the effectiveness of VAM fungi could be due to differences in their ability to (1) colonize the roots rapidly (infectivity), (2) produce external hyphae, and (3) to take up and transport P (efficiency). In this study, we measured only root colonization over time (infectivity) and found, under the conditions of this experiment, G. etunicatum was the most infective fungus.

Mycorrhizal root colonization, expressed as either percentage or

total root length colonized, was positively correlated with the shoot dry weight of Siratro. Abbott and Robson (1981) and Plenchette et al. (1982) also showed positive correlations between the magnitude of mycorrhizal root colonization and shoot dry weights of plants grown on P-deficient soils. These results suggest that differences in endophyte effectiveness may be evaluated on the basis of rates of root colonization. However, Hayman and Tavares (1985) demonstrated that final root colonization by VAM fungi may give little indication of the ability of an endophyte to stimulate plant growth. Abbott and Robson (1978) reported that VAM fungi which differ in effectiveness and rate of root colonization may have similar plateau levels of colonization at a late harvest. Therefore it is not surprising, when colonization is assessed at a relatively advanced stage of plant growth, that there is often little correlation between mycorrhizal root colonization and effectiveness.










33

Forage legumes are an important component of improved pastures and must be established rapidly and without excessive cost. When P is a major factor limiting the productivity of legumes, large applications of P fertilizer are normally required for legume establishment. However, with the increasing cost of P fertilizer, alternative strategies for minimum fertilizer input and efficient use must be adopted. These data demonstrate that by careful selection of effective VAM fungi, the growth of Siratro can be enhanced in a P-deficient native soil containing a less effective native VAM population than the introduced VAM fungi.
















CHAPTER V
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II. EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT DIFFERENT P LEVELS.



Introduction



Forage legumes are an important component of improved grass pastures. The legumes serve both to increase forage quality and decrease the need for fertilizer N through N2 fixation. Snyder et al. (1985) studied the responsiveness of several tropical legumes, including Siratro, to P and lime in a typical Florida Spodosol. They reported that lime and P rates of approximately 3000 and 75 kg ha-1 produced maximum yield and maximum economic return.

In an experiment with red clover in a field containing 10 mg NaHCO3-1
soluble P kg-1 soil, plants showed an early response to superphosphate, but by the end of the second year yields were high in all plots, equivalent to around 15 t ha-1 dry matter (Hayman et al., 1981). This result was attributed to one of the introduced endophytes, G. caledonium, which had spread and sporulated profusely throughout all the plots (including those inoculated with two other endophytes) and had previously enhanced growth of lucerne at this site. In upland pastures in Wales, Hayman and Mosse (1979) found that inoculation of white clover seedlings with a combination of G. Mosseae and G. fasciculatum "E3" in field plots given the standard dressing of 90 kg P ha-1 as basic slag doubled plant

34










35

growth and greatly enhanced tissue P content and nodulation. Growth responses at other sites varied from large to slightly negative, probably governed in part by the effectiveness of the indigenous VAM population (Hayman and Hampson, 1979).

Species (Miller et al., 1985; Schubert and Hayman, 1986; Thompson et al., 1986), and isolates within a species (Cooper, 1978), of VAM fungi can colonize plants at different rates. If the mycorrhizal growth response is related to the amount of early root colonization (Abbott and Robson, 1981; Chapter IV), then isolates of VAM fungi that colonize roots rapidly, at P levels found in established agricultural soils, may be most suitable for pasture inoculation.

Schubert and Hayman (1986) indicated that, in order to achieve a rational and effective use of inoculants, precise information on the performance of endophytes in soil amended with P was necessary. It is evident that the effect of soil P on symbiosis varies with the specific host and endophyte. Therefore, more research is needed to develop uniform and predictable endophyte-host responses.

In another study described in Chapter IV, G. etunicatum and G.

intraradices were found to be the most effective growth enhancers (out of 5 isolates) of Siratro in a soil amended with a moderate level of P and lime. In the present study, the objective was to evaluate the infectivity and effectiveness of these two fungi over a practical range of applied P.









36

Materials and Methods



The chemical properties of the soil used, liming, and fertilizer amendments are described previously (Chapter III).

Siratro was used as the host plant in this experiment. The VAM fungi tested were G. etunicatum (isolate S312) and G. intraradices (isolate S313). These isolates were maintained in pasteurized soil in pots with bahiagrass as the host. Soils from 10-week-old pot cultures were used to inoculate experimental pots. The propagule densities of the inocula were determined by the MPN technique (Daniels and Skipper, 1982) and approximately 240 propagules were added to each 15-cm-diam plastic pot. Details on the origin of the two fungal isolates were reported previously (Chapter II, Table 2-4). Fungal inoculation technique, planting, and watering were described in Chapter III.

The experiment was designed as a 3 x 4 factorial consisting of 3 inoculation treatments; G. intraradices, G. etunicatum, and noninoculated control, and four P treatments; 2.5, 10, 20, and 40 mg P kg-1 as Ca(H2PO4)2.H20 (equivalent to 5, 20, 40, and 80 kg P ha-1 assuming a 15-cm depth of soil ha-I with a bulk density of 1.3 g cm-3) Phosphorus was applied in solution one week before planting. Soil samples from each treatment were analyzed for extractable P at the beginning and end of the experiment using the Mehlich-I method (0.05 M HC1 + 0.0125 M H2S04). The twelve treatments were replicated five times and arranged on a greenhouse bench in a randomized complete block design.









37

The average maximum and minimum greenhouse temperatures during the experimental period were 34 and 260C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1800 p mol m s

Shoot dry and root fresh weights, and percentage and total root lengths colonized by VAM fungi, were determined after 60 d using procedures described previously (Chapter IV). Data for all variables were subjected to ANOVA procedures and regression analysis (SAS Institute Inc., 1982).



Results and Discussion



Phosphorus applications of 2.5, 10, 20, and 40 mg kg-I resulted in Mehlich-I extractable P in the soil of 5.6, 12.8, 21.6, and 38.0 mg kg-1 at the beginning of the experiment, respectively. These differences were still reflected at the end of the experiment when P concentrations were 4.4, 7.8, 15.2, and 24.6 mg kg-I

Shoot dry and root fresh weights of Siratro were increased by P

fertilization and fungal inoculation (Fig. 5-1). At 2.5 mg P kg-1, there was no difference in the shoot dry and root fresh weights among inoculated and control plants. At all other levels of P, inoculated plants had higher shoot dry and root fresh weights than control plants. Shoot dry weights were greatest for plants inoculated with G. etunicatum. There were no differences in root fresh weights between plants inoculated with G. etunicatum or G. intraradices. Inoculated plants had a quadratic relationship for shoot dry and root fresh weights and P application









38












O, o o
ETL
1 0 5 0 Z 30





APa PLIED_(mg/kg) _P APPLED_(mg/kg) CON"


i 5 on


t euri si
G




LL Z 3
0 -

0 0 o
0 10 20 30 40 0 0 t0 20 30 40

P APPLIED (mg/kg) P APPLIED (mg/kg)










Fig. 5-1. Effect of P application on shoot dry weight, root fresh
weight, percentage of root colonized by VAM fungi, and total root length colonized of Siratro grown in limed
nonpasteurized soil and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON).









39

(Table 5-1). Maximum yield of inoculated plants was achieved between 28 and 30 mg kg-1 of applied P. Control plants had a linear relationship for shoot dry weights and a quadratic relationship for root fresh weights. There were fungus x P interactions for shoot dry and root fresh weights (Table 5-2).

Percentage and total root length colonized by VAM fungi for the

inoculated treatments increased with P additions (Fig. 5-1). This effect was greater for G. etunicatum than for G. intraradices. Phosphate application did not alter the percentage and total root length colonized in the control plants. Inoculated plants had a quadratic relationship for percentage and total root length colonized and applied P (Table 5-1). Maximum colonization of inoculated treatments, expressed as either percentage or total length of colonized root, was attained between 32 and 35 mg kg-1 of applied P. There were fungus x P interactions for percentage and total root length colonized by VAM fungi (Table 5-2).

Shoot dry weight of Siratro over the range of applied P was highly correlated with percentage (r2 = 0.95"*) and total root length colonized by VAM fungi (r2 = 0.97**) (Fig. 5-2) for both inoculated treatments. Percentage of the root colonized by VAM fungi was very closely correlated (r2 = 0.98**) with the total root length colonized.

Growth enhancement from VAM inoculation at different levels of P has been reported to vary with VAM fungi (Hayman and Hampson, 1979; Hayman and Mosse, 1979; Schubert and Hayman, 1986; Thompson et al., 1986). For example, Schubert and Hayman (1986) indicated that, when large amounts of P were added (more than 100 mg kg-1 ), G. mosseae, G. versiforme, G.









40

Table 5-1. Regression equations and coefficients of determination (r2)
showing the relationship of P level to shoot dry weights,
root fresh weights, percentage and total root length
colonized.




Variable Regression equations r2


Glomus etunicatum

Shoot dry weight (g) = -0.12+0.34P-0.006P2 0.97* Root fresh weight (g) = 0.50+0.40P-0.006P2 0.96* Root colonization (%) = 4.47+2.62P-0.04P2 0.97"* Root length colonized (m) = -1.43+0.64P-0.009P2 0.96

Glomus intraradices

Shoot dry weight (g) = 0.16+0.24P-0.004P2 0.95 Root fresh weight (g) = 0.55+0.38P-0.006P2 0.94 Root colonization (%) = 5.42+1.86P-0.03P2 0.95" Root length colonized (m) = -1.08+0.49P-0.006P2 0.91"

Control

Shoot dry weight (g) = 0.72+0.06P 0.93** Root fresh weight (g) = 0.89+0.17P-0.002P2 0.92*


P = phosphorus level **significant at P < 0.01









41

Table 5-2. Analysis of variance for shoot dry weights, root fresh
weights, and percentage and total root length colonized.



Source of DF Shoot MS root MS root colonized MS variation percentage length


Block 4 0.07"" 0.04 7.39 0.069 Fungi (F) 2 11.07 18.01 2650.42 114.13 P rates (P) 3 34.19 59.37 1583.51 118.66 linear (Pl) 1 85.40 146.43 4156.00 305.28 quadratic (Pq) 1 13.97 30.10 577.33 32.23 Cubic (Pc) 1 3.20 1.57 17.14 2.48 F x P 6 1.65"* 2.55** 378.46"" 22.50"" F x P1 2 2.58"" 4.02" 965.54"" 55.69"* F x Pq 2 2.29*" 2.08** 159.17"" 7.820* F x Pc 2 0.08 1.55 10.67 3.99 Error 44 0.02 0.13 6.27 0.24


Significant at P < 0.01 MS = mean square








42










o o o





0
4 00



Coo 0~ Y=O.83+0.44X H 2 8o
O r=0.95 "
C()



0 3 1 g 12

ROOT LENGTH COLONIZED (m)











Fig. 5-2. Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments
in nonpasteurized soil.









43

macrocarpum and G. margarita were ineffective in stimulating growth of onion; however, G. caledonium and Glomus sp. 'E3' were generally effective at all P levels. In our study, G. etunicatum was more effective than G. intraradices at all but the lowest applied P level. Hence, there appears to be good potential for the selection of VAM fungi to enhance plant growth under amended soil conditions such as P fertilization and liming.

Phosphorus has been reported to increase, decrease or not affect root colonization by VAM fungi. However, it is difficult to compare results concerning the effect of P fertilization on mycorrhizal root colonization, because of differences in the range of added P, as well as other factors such as host plant and soil type. In this study, we used a range of 2.5 to 40 mg P kg-I because it represented P levels used in the production of tropical forage legumes in Florida on a similar soil (Snyder et al., 1985). At the lowest P level, G. etunicatum and G. intraradices did not colonize the root or improve growth of Siratro above that of the control plants. Barber and Lougham (1967) reported that, at a very low P level, competition for P occurs between plants and microflora. Habte and Manjunath (1987) and Same et al. (1983) indicated that the growth of VAM fungi is limited by P at very low levels. Between 10 to 40 mg P kg-1, percentage and total root length colonized by VAM fungi increased with P additions. These results agree with those of Abbott and Robson (1977b), Schubert and Hayman (1986), and Thompson et al. (1986), who reported an increase in the percentage and total root length colonized between 18 to 55 mg P kg-1. At high P levels (more than 100 mg kg- ), which are not feasible for field production of









44

forage legumes, percentage and total root length colonized by VAM fungi may be suppressed (Abbott and Robson, 1977b; Schubert and Hayman, 1986; Thompson et al., 1986).

The result that root colonization by VAM fungi, expressed as either percentage or total root length colonized, is positively correlated to shoot dry weight is consistent with the findings of Abbott and Robson (1981) and a previous study where its implications are discussed (Chapter IV).

I conclude that, in amended soils where the indigenous population of VAM fungi is less effective than some of the introduced species of VAM fungi, inoculation with effective VAM fungi can increase the plant growth. Furthermore, with highly mycorrhizal dependent crops such as tropical legumes, growth enhancement may occur at P levels actually used in commercial pasture production.
















CHAPTER VI
EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE GROWTH AND
UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM, STYLOSANTHES GUIANENSIS, AND AESCHYNOMENE AMERICANA



Introduction



Mycorrhizal colonization is important for legumes because it

increases their P uptake (Abbott and Robson, 1977b; Saif, 1987), and therefore nodulation and N2-fixation (Asimi et al., 1980; Bergersen, 1971; Gates and Wilson, 1974; Gibson, 1976).

Crush (1974) found that VAM fungi increased the growth and

nodulation of Centrosoma pubescens Benth, Stylo, and Trifolium repens L. Mosse et al. (1976) showed that effective nodulation of Centrosema, Stylosanthes, and Trifolium plants in a P-deficient Brazilian Cerrado soil could be achieved only by introducing both VAM fungi and P. Mycorrhizal fungi also have been shown to increase nodulation, N2fixation, plant growth, plant N and P content in Vigna unguiculata (Islam et al., 1980; Sanni, 1976), Medicago sativa (Barea et al., 1980), Pueraria phaseoloides and Stylo (Waidyanatha et al., 1979), Stylosanthes scabra (Purcino and Lynd, 1985), leucaena (Munns and Mosse, 1980; Purcino et al., 1986), and Siratro (Lynd et al., 1985). However, effective tripartite symbiosis (legume-rhizobium-VAM fungus) is influenced by soil and climatic conditions (Waidyanatha et al., 1979),



45









46

which are apparently species-specific (Burt and Miller, 1975; Mosse, 1972).

In a previous study (Chapter V), G. etunicatum was an effective growth enhancer of Siratro in a soil similar to the one used in the
-1
present study, with a moderate level of applied P (20-40 mg P kg-l).

The objective of this investigation was to determine the effect of inoculation with a VAM fungus, G. etunicatum, on the growth and plant uptake of P and N of three forage legumes at different P levels in pasteurized soil under greenhouse conditions.



Materials and Methods



The soil used in this investigation, liming, and basic fertilization are described previously (Chapter III). The soil chemical characteristics before soil fertility treatments and after pasteurization (700C for 4 h) were: pH 4.4 (soil:H20=l1:2); 1.4% organic matter; 2, 65, 11, and 12 mg kg-1 (Mehlich-I extractable) of P, Ca, Mg and K, respectively.

At planting three phosphorus levels were established by application in solution of 12.5, 25, and 50 mg P kg-1 as Ca(H2P04)2.H20 which is equivalent to 25, 50, and 100 kg P ha-1, assuming a 15-cm depth of soil ha-1 with a bulk density of 1.3 g cm-3

Glomus etunicatum (isolate S312) was isolated from carpon desmodium at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-4). Fungal inoculum was produced in pot culture in pasteurized










47

soil containing bahiagrass. The fungal inoculation technique was reported in Chapter III.

The legume species used in this experiment were: Siratro,

aeschynomene, and Stylo. Seeds were scarified with sandpaper, wetted, and sprinkled with type El "cowpea" inoculum (Nitragin Co., Milwaukee, WI) prior to planting. Three germinated seeds were planted per 620 ml "Deepots" (J.M. McConkey & Co, Inc, Summer, WA). After emergence, seedlings were thinned to one per pot.

The experiment was conducted as a 2 x 3 x 3 factorial consisting of 2 inoculation treatments, G. etunicatum and noninoculated control; three P levels, 12.5, 25, and 50 mg kg- ; the 3 legume species; and 3 replications. The 18 treatments were arranged on a nonshaded greenhouse bench in a completely randomized design. The average maximum and minimum greenhouse temperatures during the experimental period were 28 and 200C, respectively, and the average maximum photosynthetic photon flux density was 1793 u mol m-2 s-.

After 65 d plants were harvested, shoots dried (700C for 24 h),

weighed, and ground in a Wiley mill using a 20-mesh screen. Shoots were digested by the sealed-chamber procedure of Anderson (1986) and analyzed for P on a Jarrel-Ash 955 inductively-coupled argon plasma spectrometer (ICAP). For nitrogen analysis, samples were digested using a modification of the aluminum block digestion procedure of Gallaher et al. (1975) and ammonia in the digestate was determined by semiautomated colorimetry (Hambleton, 1977). Roots were washed from the soil, air-dried, and weighed. In addition, percentage and total root









48

length colonized were estimated as described in Chapter IV. A 0-4 scale

-was used to estimate the number of nodules per plant, with 1 = 20-50,

2 = 50-100, 3 = 100-150, and 4 = > 150 nodules. Data were subjected to ANOVA procedures and regression analysis (SAS institute Inc., 1982).



Results and Discussion



At harvest, both fungal inoculation and P applications increased shoot dry weight, plant P concentration, and total plant P and N of the three legumes (Table 6-1). Root fresh weight was increased for Stylo and Siratro but not for aeschynomene. There were fungus x P interactions for shoot dry and root fresh weights, and total plant N of Stylo. Siratro had fungus x P interactions for shoot dry weight and total plant P and N, whereas aeschynomene only had fungus x P interaction for shoot dry weight.

At all levels of applied P, shoot dry and root fresh weights and total N of Stylo were greater for mycorrhizal plants than nonmycorrhizal plants (Fig. 6-1). Differences between mycorrhizal and nonmycorrhizal plants were most pronounced at intermediate levels of applied P and diminished with further P addition.

Phosphorus concentration and total P of Stylo were not affected by fungus x P interactions (Table 6-1), but there was an overall effect of fungal inoculation (Fig. 6-1). Saif (1987), Mosse et al. (1976), and Waidyanantha et al. (1979) also reported an increase in plant growth, P concentration, and total P and N of Stylo in a pasteurized low P soil following inoculation with VAM fungi and P applications.









49

Table 6-1. Mean squares and levels of significance from the analysis
of variance for shoot dry weight, root fresh weight, P
concentration, and total P and N uptake of forage legumes.




Source of DF Shoot Root P Total Total Variation dry wt fresh wt conc P N



Stylosanthes guianensis

Fungus 1 0.45 1.13 0.011"* 6.93"" 500.97 P rates 2 0.64 0.21 0.016"" 9.13"* 790.32 Lineal 1 1.20 0.36 0.032"" 17.69"^ 1511.78 Quadratic 1 0.09 0.05 0.0004 0.57 68.86 Fungus x P 2 0.046** 0.15" 0.00007 0.65 47.14" Error 12 0.0057 0.026 0.012 0.18 8.48

Macroptilium atropurpureum

Fungus 1 1.40 0.58** 0.0076" 33.78 2360.30 P rates 2 7.28 1.81"" 0.028"" 163.11 10469.14 Lineal 1 14.54 3.55 0.056"" 325.94 20925.10 Quadratic 1 0.0072 0.05 0.0001 0.28 13.18 Fungus x P 2 0.54* 0.10 0.0014 10.37" 625.52* Error 12 0.029 0.027 0.00082 2.47 43.27

Aeschynomene americana

Fungus 1 0.13 0.042 0.0068 6.06 129.34" P rates 2 0.56 1.07 0.034"" 29.75 524.16 Lineal 1 1.09 2.01"* 0.064"" 58.70** 1035.46"* Quadratic 1 0.027 0.12" 0.0022 0.80 12.85 Fungus x P 2 0.037^ 0.030 0.00009 0.10 43.35 Error 12 0.0076 0.015 0.00052 0.25 21.75


*Significance at P < 0.05 Significant at P < 0.01











50











1.A 0.5 0oM s eM ANM
23 ~- -0.0 + 0.0EP 0.001P r 0.i m 0.4 Y + 0.02P t 0.70o
0 .2




0.4 0.32 2 ,_ Y 0C3 Y 0.0 + 0. pr 00 0
Y0 -0.0 0.019 tr 0

0.0 0. .2 5.0
o M A N o H N o o a .0.80
o- o a t; 3.0 0 Y Y -2.00 + 0.34 0.rP S- 2.0 r 0.90 Y -0.78 C.OBEP 0.0lP 4 r- 0.6M A

/.Y -0.-05 0.2P0 0. __0
1 Y -0.32 + 0.0P r* 0.8




5 50








0 0



.5 25.0 37.5 50.0 115 25.0 37.5 50.0 P APPLIED (mg/kg)








Fig. 6-1. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Stylosanthes
guianensis.









51

Mycorrhizal plants of Siratro had greater shoot dry weight, total P and total N than nonmycorrhizal plants at low and intermediate levels of applied P, but not at the highest level (Fig. 6-2). Overall, mycorrhizal plants had greater root fresh weight and plant P concentration than nonmycorrhizal plants. Other investigators (Lynd et al., 1985; Saif, 1987) have shown similar responses of Siratro to inoculation with VAM fungi and P additions. Differences in shoot dry weight between mycorrhizal and nonmycorrhizal plants of aeschynomene were only at the intermediate level of applied P (Fig. 6-3). Fungal inoculation did not affect the root fresh weight. Overall, mycorrhizal plants had greater P concentration, total P, and total N than nonmycorrhizal plants (Fig. 6-3).

Percentage and total root length colonized for mycorrhizal plants of Stylo (Fig. 6-1), Siratro (Fig. 6-2), and aeschynomene (Fig. 6-3) increased with the first addition of P. However, maximum colonization, expressed as either percentage or total length of colonized root of the three legumes, was attained at the intermediate levels of applied P.

The number of nodules in the three legumes increased with fungal inoculation and P applications. Mycorrhizal plants of Stylo had more nodules than nonmycorrhizal plants at all levels of applied P. However, mycorrhizal plants of Siratro and aeschynomene had more nodules than nonmycorrhizal plants only at 25 mg kg-1 of applied P.

Mycorrhizal plants required between 38-40 mg P kg-1 to achieve maximum shoot dry weight, whereas nonmycorrhizal plants required
-1
50 mg P kg-1 to produce approximately the same shoot dry weight, except for nonmycorrhizal Stylo (Fig 6-1) which even with 50 mg P kg-1 did not











52









7.0 0.4
o M oaM A

Y 0.47 + 0.2P 0.OP 1 r O." Y 0.17 + O.00P r 0.07S- ,



l Y m + 0.04P g r 0.@. .,


1.0 0.1 7.5 2.0
a M NM o M

.CY- 3.1O + 0.tIP 0.l1P r 0.84 4.14 + 0.24P Ir D.75W


.5 10.0



3.0 5.0




2
Y -1.7. + .P .0.mM


Yi -- r-0. a M .0A


Y 5.11 + Y.-P 1..7 3 O




-0.8Y 41.8 4 .4P r0. .gm

15
125 2i.0 07.5 .BU .0 37.5 P APPLIED (mg/kg)






Fig. 6-2. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Macroptilium
atropurpureum.











53



















1.5 a2+.
0.5









0 ma t" M
ST .07P P I7. 0.0+ O re- 0. CR
Y 1. L_+ 0. P r 0. P



0
2- M" --- I O -- .L 2 *0






115 2 t A. 1 2.



















shoot dry weight, root fresh weight, root colonization, P
2 0
-5 _7.5______2&0_7._M.


concentrio, t P, an totaNof





















americana.










54

reach the same shoot dry weight as myrc- rhizal plants. Fungal inoculation with G. etunicatum resulted in a 20% decrease in the amount of P required for maximum yield. This demonstrates the importance of VAM fungi in the P nutrition of tropical legumes and would represent an important savings to the farmer.

Growth responses of Stylo, Siratro, and aeschynomene associated with inoculation with VAM fungi are closely related to improved P uptake and N2-fixation. These results clearly support the findings of earlier studies that inoculation with VAM fungi not only stimulates plant growth and P uptake of legumes (Abbott and Robson, 1977b; Habte and Manjunath, 1987; Menge, 1983) but also nodulation and N2-fixation (Asimi et al., 1980; Lynd et al., 1985; Purcino et al., 1986) which was measured indirectly in this study by the total plant uptake of N.

Inoculation with effective species of VAM fungi and additions of P
-1
between 25 and 50 mg kg- were shown to improve the growth and uptake of P and N of Stylo, Siratro, and aeschynomene.
















CHAPTER VII
GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND AESCHYNOMENE AMERICANA
TO INOCULATION WITH SELECTED VA MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS



Introduction



The practical goal of studies on plant growth responses to

inoculation with VAM fungi is to obtain increased yield of plants growing under field conditions. Significant plant growth responses to inoculation with VAM fungi have been demonstrated in pot experiments using pasteurized and nonpasteurized soil for several tropical forage legumes such as: Pueraria phaseoloides (Salinas et al., 1985; Waidyanatha et al., 1979), leucaena (Habte and Manjunath, 1987; Huang et al., 1985), Siratro (Lynd et al., 1985), Stylo (Mosse, 1977) and aeschynomene (Chapter II and VI).

It has been pointed out that inoculation experiments with VAM fungi should include testing a series of P levels (Abbott and Robson, 1977b; Hall, 1978; Powell, 1980b) in order to select the optimum P level for a mycorrhizal response. Except for the work by Saif (1987), little information is available on plant growth response of tropical forage legumes to inoculation with VAM fungi in nonpasteurized soil under field conditions at different P levels. However, there is more data for temperate legumes. In general, the field sites where plants are most likely to respond to inoculation with VAM fungi are those containing 55









56

little soluble phosphate and a small or ineffective native population of VAM fungi. The experimental site selected for this study contains very little available P (i mg kg-1) and a reasonably high indigenous population of VAM fungi (2 propagules g-1 soil). The indigenous population, however, was less effective than two of the five introduced VAM fungi under amended soil conditions (Chapter IV).

Aeschynomene is used extensively in Florida as a forage legume to supply fixed nitrogen as protein and minerals to grazing animals (Hodges et al., 1982) Siratro has been used sparingly in Florida (Kretschmer, 1972), but is one of the most widely used legumes throughout tropical regions of the world (Lynd et al.,1985). However, for the satisfactory establishment and growth of aeschynomene and Siratro in highly acidic and P-deficient soils, lime and P fertilizer must be applied (Snyder et al., 1985). Recent greenhouse experiments have shown that better growth of forage legumes in these soils can be achieved by inoculation with selected species of VAM fungi (Chapter IV and V). Glomus etunicatum and G. intraradices were found to be effective growth enhancers of Siratro and aeschynomene in a nonpasteurized, limed (3000 kg ha-1) soil similar to the one used in the present study, over a range of 20-80 kg ha-1 of applied P. It is therefore of considerable interest to determine whether inoculation with selected isolates of VAM fungi can improve the establishment, growth, and nutrient uptake of Siratro and aeschynomene, under field conditions where soil was limed and fertilized with different P levels.










57

Materials and Methods



The soil used was a native Oldsmar fine sand (sandy, siliceous, hyperthermic Alfic Arenic Haplaquods) with a pH of 4.6 (soil:H20=1:2),

1.5% organic matter, and the following Mehlich-I extractable elements in mg kg-l: P 1.0, Ca 65, Mg 12, and K 15.

The experiment was designed as a 2 x 3 x 4 factorial consisting of two legume species; Siratro and A. americana; three inoculation treatments, G. etunicatum, G. intraradices, and the control; and four P treatments; 10, 30, 60, and 120 kg ha-1 as triple superphosphate. The 24 treatments were arranged in a randomized complete block design with ten replications per treatment. The P treatments were surface applied by hand on 3 July 1986, along with a basal application of lime (high calcitic limestone), Mg, nutritional spray (Diamond Fertilizer Co., Ft. Pierce, FL.), and Mo at 3000, 25, 22, and 0.2 kg ha-1, respectively, and incorporated using a rake to a depth of approximately 15 cm. Potassium was broadcasted on each plot at a rate of 60 kg ha-1 as KCL on 5 August 1986.

Seeds of Siratro and aeschynomene inoculated with rhizobium type El "Cowpea" inoculum (Nitragin Co., Milwaukee, WI) were sown in pasteurized Oldsmar fine sand, amended with high calcitic limestone at 1500 mg kg-1 and P at 12.5 mg kg- in cells of "speedling" styrofoam trays (72 cells per tray) on 20 June 1986.

Seedlings were inoculated or not inoculated with G. etunicatum

(isolate S312), G. intraradices (isolate S311). The amount of soil-root inoculum used for each VAM fungi was adjusted to give equal inoculum










58

densities as determined by the MPN technique (Daniels and Skipper, 1982). Approximately 180 propagules were added mid-way down each cell before seeding. Control seedlings received 15 g of a soil-root mixture from nonmycorrhizal pot cultures and the equivalent of 20 kg P ha-1, which was applied in solution 10 d after germination in an attempt to make the P status of the mycorrhizal and nonmycorrhizal seedlings similar at the time of transplanting.

Seeded trays were placed in a glasshouse for 6 wk, after which the whole cell content (each with one seedling) was transplanted to the field. Siratro seedlings were cut back to three nodes each before transplanting. Seedlings were transplanted on 4 August 1986.

One seedling was planted by hand in the middle of each 1.0 by 1.0 m plot which were surrounded by alleyways of 1.0 m. Extra seedlings were weighed, and P concentration and root colonization were determined. Harvests of Siratro and aeschynomene were made on 2 October 1986. Siratro, a perennial, also was harvested on 27 November 1986, 5 May 1987, and 29 June 1987. Herbage was dried at 750C for 24 h and weighed. Five subsamples per treatment from the first harvest of Siratro and aeschynomene foliage were analyzed for N and P content by automated colorimetry (Technicon Industrial Systems Method No. 334-74 W/B, Technicon Instruments Corp., Tarrytown, NY). Five root samples per treatment, consisting each of four subsamples, were used to assess mycorrhizal root colonization. Percentage of root colonized by VAM fungi of aeschynomene and Siratro (ist and 4th harvests), was estimated as described in Chapter III.










59

Soil samples (0-15 cm) were taken at the 4th harvest and analyzed for pH, P, Ca, Mg, and K using the Mehlich-I extractant method. All elements were determined by inductively coupled argon plasma (ICAP) spectrometry. Data for all variables were subjected to ANOVA procedures and regression analysis (SAS Institute Inc., 1982).



Results and Discussion



Pre-inoculated seedlings were used in this study to ensure the legume roots were well colonized with the selected VAM fungi, because this. was thought to be the most certain way to ensure establishment of the inoculum. Once a plant response is ascertained, methods of inoculation more applicable on a large scale can be tested.

As a result of lime application, soil pH increased from 4.6 to about

6.2, and extractable Ca increased up to about 650 mg kg-1. Phosphorus applications of 10, 30, 60, and 120 kg ha- resulted in extractable P in,
-i
the soil of 5.1, 8.6, 17.8, and 34.3 mg kg-1 at the end of the experiment, respectively. Thus the legume species and VAM fungi in the present study were exposed to a considerable range of soil P.

Seedlings inoculated with VAM fungi were similar in shoot dry weight and P concentration to control seedlings at transplanting (Table 7-1). Seedlings inoculated with VAM fungi were also well colonized, whereas no VAM colonization was detected on control seedlings (Table 7-1).

Phosphorus amendments and fungal inoculation increased shoot dry weights of aeschynomene (Table 7-2) and all harvests of Siratro










60

Table 7-1. Shoot dry weights, P concentrations, and percentage of
roots colonized of Siratro and aeschynomene seedlings at
transplanting.




VAM SiratroE aeschynomeneE
Inoculation Shoot dry Shoot Root shoot dry Shoot Root wt. P colon. wt. P colon.


mg --- % --- mg --- % --G. etunicatum 302 .18 52 304 .17 57 G. intraradices 305 .17 60 299 .20 55 Control 298 .16 0 302 .18 0


ZData are means of five replicates.










61

Table 7-2. Analysis of variance for shoot dry weights of Aeschynomene
americana harvested 2 October 1986.




Source of DF Mean squares variation


Block 9 52.57 Fungi (F) 2 1654.72"* Phosphorus (P) 3 6986.26*
Linear (Pl) 1 18700.24* Quadratic (Pq) 1 2250.32*
Cubic (Pc) 1 8.21 Fx P 6 91.79
F x Pl 2 53.72 F x Pq 2 215.53 F x Pc 2 6.11 Error 99 45.83


**Significant at P < 0.01










62

(Table 7-3). There were no fungi x P interactions for shoot dry weights of Siratro or aeschynomene.

At all levels of applied P and at all harvests, shoot dry weights of Siratro were greater for fungal inoculated plants than control plants (Fig. 7-1). Except for the third harvest, where the effect of fungal inoculation was less pronounced at all levels of applied P, differences between fungal inoculated and noninoculated plants were most marked at intermediate levels of applied P (30-90 kg ha-1) and diminished at the highest level (120 kg ha-1). The effect of inoculation with VAM fungi on the shoot dry weights of aeschynomene, at all levels of applied P, followed the same trend as that of Siratro although the response to mycorrhizal inoculation was greater (Fig. 7-2).

Inoculated plants of Siratro (Table 7-4 and 5) and aeschynomene

(Table 7-6) had a quadratic relationship between shoot dry weight and P application. Maximum shoot dry weight of Siratro was achieved between 75-85 kg ha-1 of P and for aeschynomene at 85 kg ha-1 of P, whereas control plants of both legumes, even with 120 kg ha-1 of P, did not reach the same shoot dry weight as fungal inoculated plants. Thus inoculation with VAM fungi resulted in at least a 30% savings (40 kg ha-1) in the amount of P fertilizer required for maximum yield. In a previous greenhouse experiment (Chapter VI), I found that inoculation with G. etunicatum resulted in a 20% decrease in the amount of P required for maximum yield of Siratro. Some previous reports of VAM field experiments with legumes in nonpasteurized soil (Black and Tinker, 1977; Khan, 1975) show responses to VAM inoculation only in the absence of P fertilizer.










63

Table 7-3. Analysis of variance for shoot dry weights from f,'ir
Siratro harvests.




Source of Mean squares variation DF Harvest 1 Harvest 2 Harvest 3 Harvest 4 2 Oct. 86 27 Nov. 86 5 May 87 29 June 87


Block 9(6)z 6.40 4.14 11.47 18.70" Fungi (F) 2 332.20 247.79* 69.20** 177.21"* Phosphorus (P) 3 807.52 1577.91"* 1598.20"" 1806.27'
Linear (Pl) 1 2107.47 4021.98" 4223.03"* 4630.88"* Quadratic (Pq) 1 305.04 711.49"" 559.16"" 773.57""
Cubic (Pc) 1 10.07 0.23 12.38 14.32 F x P 6 6.35"" 3.21 1.67 10.12
F x Pl 2 4.61"* 3.89 1.10 7.69 F x Pq 2 12.82"" 5.38 2.76 15.83 F x Pc 2 1.63 0.38 1.14 6.83 Error 99(66) 8.89 13.43 9.20 7.98

**Significant at P < 0.01 "Significant at P < 0.05

ZValues in parentheses are the degrees of freedom for harvests 2, 3, and
4 for which only 7 replicates were used.










64








40 40 TU
o CN A ETU a IN o CCN A ETU = 1NT

e 30 30


20 20









40 40
a C4\1 E INT 4 o AN ET'J I NT




SU 20


10 10


0 0
0 30 60 90 120 0 30 60 00 !2
PFFFUED (4ba) P ffRM (h1)












Fig. 7-1. Effect of P application on shoot dry weights for the first
(A), second (B), third (C), and fourth (D) harvest of
Siratro grown under field conditions and inoculated with
Glomus etunicatum (ETU), Glomus intraradices (INT), or not
inoculated (CON). Data points are means of five
replicates.










65


110
o CON A ETU 0 INT








50
-o
0
c3
r


20


o CON A ETU 0 INT
,- 60


3 45
-


0 30 60 90 120










C-,
CD 15



0 30 60 go 120

P APPLIED (kg/ha)









Fig. 7-2. Effect of P application on shoot dry weight and percentage
of root colonized of Aeschynomene americana grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.









66

Table 7-4. Regression equations and coefficients of determination
(r ) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P and N
concentrations, and total P and N uptake for the first
harvest of Siratro.




Variable Regression equations r2


Glomus etunicatum

Shoot dry wt. (g) = 12.36+0.33P-0.0022P2 0.80" Root coloniz. (%) = 5.31+0.77P-0.0043P2 0.91"* Plant P conc. (%) = 0.22+0.00072P 0.66* Total Plant P (mg) = 17.73+1.36P-0.0069P2 0.98* Plant N conc. (%) = 2.26+0.035P-0.00025P2 0.52 Total Plant N (mg) = 170.94+20.09P-0.13P2 0.91"*

Glomus intraradices

Shoot dry wt. (g) = 12.03+0.24P-0.0014P2 0.67* Root coloniz. (%) = 10.04+0.59P-0.0034P2 0.86"* Plant P conc. (%) = 0.16+0.0037P-0.000018P2 0.76^* Total plant P (mg) = 16.08+1.31P-0.0068P2 0.94"* Plant N conc. (%) = 2.12+0.041P-0.00026P2 0.56** Total plant N (mg) = 226.21+15.50P-0.095P2 0.91"*

Control

Shoot dry wt. (g) = 10.65+0.11P 0.65"" Plant P conc. (%) = 0.18+0.0011P 0.81 Total plant P (mg) = 12.36+0.48P 0.92'* Total plant N (mg) = 173.09+3.57P 0.89'*


P = phosphorus level ** Significant at P < 0.01









67

Table 7-5. Regression equations and coefficients of determination
(r ) showing the relationship of applied P level to shoot dry weights for the second, third, and fourth harvest and percentage of root colonization for the fourth harvest of
Siratro.




Variable Regression equation r2



Glomus etunicatum

Shoot dry wt (g), harvest 2 = 5.74+0.52P-0.0029P2 0.87^* Shoot dry wt (g), harvest 3 = 0.99+0.48P-0.0029P2 0.89 Shoot dry wt (g), harvest 4 = 5.98+0.55P-0.0032P2 0.87'* Root colon. (%), harvest 4 = 9.85+0.84P-0.0046P2 0.86 x

Glomus intraradices

Shoot dry wt (g), harvest 2 = 6.58+0.46P-0.0025P2 0.84** Shoot dry wt (g), harvest 3 = 2.05+0.43P-0.0025P2 0.86* Shoot dry wt (g), harvest 4 = 6.14+0.52P-0.0027P2 0.87 Root colon. (%), harvest 4 = 15.03+0.69P-0.0039P2 0.89"*

Control

Shoot dry wt (g), harvest 2 = 1.56+0.44P-0.0023P2 0.81"* Shoot dry wt (g), harvest 3 = -0.23+0.42P-0.0021P2 0.88** Shoot dry wt (g), harvest 4 = 3.38+0.42P-0.0017P2 0.91"*



P = phosphorus level 'Significant at P < 0.01









68

Table 7-6. Regression equations and coefficients of determination
(r ) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P
concentration, and total P and N uptake of Aeschynomene
americana.




Variable Regression equations r2



Glomus etunicatum

Shoot dry wt. (g) = 46.59+0.83P-0.0049P2 0.78*" Root coloniz. (%) = 7.89+0.88P-0.0047P2 0.89** Plant P Conc. (%) = 0.21+0.0037P-0.000019P2 0.80"* Total plant P (mg) = 67.54+4.96P-0.025P2 0.96** Total plant N (g) = 1.35+0.039P-0.00019P2 0.93""

Glomus intraradices

Shoot dry wt. (g) = 40.86+0.97P-0.0056P2 0.86* Root coloniz. (%) = 9.26+0.98P-0.0058P2 0.93"* Plant P Conc. (%) = 0.16+0.0053P-0.000029P2 0.76** Total plant P (mg) = 43.24+5.85P-0.031P2 0.95* Total plant N (g) = 1.18+0.043P-0.00022P2 0.91

Control

Shoot dry wt. (g) = 43.50+0.31P 0.76* Plant P Conc. (%) = 0.11+0.0045P-0.000021P2 0.95"" Total plant P (mg) = 28.08+3.63P-0.012P2 0.99"* Total plant N (g) = 1.44+0.015P 0.81"*


P = Phosphorus level Significant at P < 0.01










69

Hayman and Mosse (1979), however, reported improved growth of white clover in the field after inoculation with VAM fungi and the addition of 90 kg ha-I of P. They also indicated that responses to fungal
-1
inoculation were smaller with 23 kg ha-1 of P and were absent where no P was added. Similarly, Hall (1984) reported that inoculation with selected VAM fungi increased yield of white clover in the field only if 50 kg ha- of P was also applied.

Several greenhouse experiments on the phosphate response curves of fungal inoculated and noninoculated forage legumes have been carried out, mostly using clovers (Abbott and Robson, 1977b; Sparling and Tinker, 1978; Powell, 1980b) and Siratro (Lynd et al., 1985; Medina et al., 1987d) as the test plants. These authors applied soluble P fertilizers
-1
at rates ranging from 0 to 250 kg ha-1 and have reached the general conclusion that inoculation with VAM fungi markedly increases legume growth at low and intermediate rates of applied P. From the practical point of view, however, the interactions between phosphate additions and VAM on legumes are not always predictable and generalizable, because the responses are modulated by the incidence of several factors. These include the physical and chemical characteristics of the soil, plant species, VAM fungi, and the complex interactions between these factors.

At the first harvest of Siratro (Fig. 7-1A), plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices at all levels of applied P. However, in subsequent harvests of Siratro (Fig. 7-1BC and D) and for aeschynomene (Fig. 7-2) the response of shoot dry weight to inoculation with the two VAM fungi was not different.










70

Phosphorus additions and fungal inoculation increased percentage of root colonized by VAM fungi of Siratro (Table 7-7) and aeschynomene (Table 7-8). There were fungi x P interactions for percentage of root colonized by VAM fungi for for both legumes. Inoculated treatments had greater percentage of root colonized than control treatments at all levels of applied P. Percentage of root colonized by VAM fungi for the inoculated plants of Siratro (Fig. 7-3A and B) and aeschynomene
-1
(Fig. 7-2) increased linearly with P additions up to 60 kg ha
-i
Phosphorus application of 120 kg ha-1 did not affect the percentage of root colonized by VAM fungi. In a previous greenhouse study (Chapter V), I found that percentage of Siratro root length colonized by VAM fungi
-1
increased with P additions up to 40 mg kg- which is equivalent to 80
-1
kg ha-1 of P. Abbott and Robson (1977b) and Schubert and Hayman (1986), also in pot experiments, reported an increase in the percentage of root length colonized up to 55 mg kg-1 of P (110 kg ha-1).

Phosphorus applications did not alter the percentage of root

colonized in the control plants. The degree of root colonization of the control plants of Siratro increased from 9% in the first harvest to about 19% by the fourth harvest (Fig. 7-3A and B), but still failed to increase the shoot dry weight of the control plants compared to the inoculated plants. Fungal inoculated plants of Siratro (Table 7-4 and 5) had a quadratic relationship between percentage of root colonized and applied P. Maximum root colonization was attained between 85-90 kg ha-I of P for both legumes. At lower P additions, Siratro plants inoculated with G. intraradices had greater percentage of root colonized than plants inoculated with G. etunicatum (Fig. 7-3). However, there were no









71

Table 7-7. Analysis of variance for percentage root colonized of
Siratro by VAM fungi.




Source of DF Mean squares variation Harvest I Harvest 4 2 Oct. 86 29 June 87


Block 4 9.27 14.61 Fungi (F) 2 2394.82 1206.47 Phosphorus (P) 3 935.53 1230.09
Linear (Pl) 1 2195.71 2538.78 Quadratic (Pq) 1 552.25 1083.74 Cubic (Pc) 1 58.62 67.74 F x P 6 230.66** 86.820*
F x P1 2 510.93"" 174.04"^ F x Pq 2 112.49" 77.26**
F x Pc 2 68.55"" 9.16 Error 44 7.98 11.91


**Significant at P < 0.01 "Significant at P < 0.05









72

Table 7-8. Analysis of variance for percentage root colonized, P
concentration, and total P and N uptake of Aeschynomene
americana harvested 2 October 1986.




Mean Squares
Source of DF Root P conc. total P total N variation colon.


Block 4 10.11 0.00065 384.02 0.051 Fungi (F) 2 5276.62 0.0098** 9052.18 0.39 Phosphorus (P) 3 1844.64 0.051 75570.74 4.64**
Linear (Pl) 1 4408.21 0.12 196270.22 12.04* Quadratic (Pq) 1 1001.65 0.031 29179.92 1.89""
Cubic (Pc) 1 124.08 0.0012 1262.07 0.0021 F x P 6 378.66"' 0.0010 1190.91 0.026 F x P1 2 833.18"* 0.0021 1275.75"* 0.013 F x Pq 2 220.16"* 0.00052 1815.29"" 0.043 F x Pc 2 82.64** 0.00021 481.71 0.022 Error 44 11.60 0.0011 206.45 0.062


Significant at P < 0.01 "Significant at P < 0.05








73



50
o CON s ETU a INT

S 40

Cu
-r


20

10




60
o CON ETU INT









SB



0 30 60 90 120 P APPLIED (kg/ha)




Fig. 7-3. Effect of P application on percentage of root colonized
for the first (A) and fourth (B) harvest of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.










74

differences in root colonization between G. etunicatum and G. intraradices at any level of applied P. For aeschynomene, differences between G. intraradices and G. etunicatum were only at 30 kg ha-1 of applied P (Fig. 7-2).

At the first harvest, P amendments and fungal inoculation also had an effect on P concentration, total P uptake, N concentration, and total N uptake of Siratro (Table 7-9). These same parameters were increased for aeschynomene, except for N concentration (Table 7-8). There were fungi x P interactions for P concentration, total P uptake, N concentration, and total N uptake of Siratro. By the contrast, aeschynomene only had fungi x P interaction for total P uptake.

Inoculated plants of Siratro had greater P concentration,

total P uptake, N concentration, and total N uptake than control plants at low and intermediate levels of applied P (Fig. 7-4). At the highest level of applied P, total P uptake of Siratro plants inoculated with G. intraradices and P and N concentrations of plants inoculated with G. etunicatum did not differ from those of control plants. Fungal inoculated plants of aeschynomene also had greater P concentration, total P uptake, and total N uptake than control plants at low and intermediate levels of applied P, but not at the highest level (Fig. 7-5). The positive effect of inoculation with VAM fungi on P and N uptake have been shown for other forage legumes, for example, leucaena (Habte and Manjunath, 1987), Pueraria phaseoloides (Sanchez and Salinas, 1981), and field experiments with white clover (Hall, 1984; Hayman and Mosse, 1979) and Medicago sativa (Azcon-Aguilar and Barea, 1981). Potassium,









75

Table 7-9. Analysis of variance for P and N concentrations, and total
P and N uptake of Siratro for the first harvest.




Source of Mean Squares variation DF P Conc. total P N Conc. total N



Block 4 0.00033 26.06 0.42' 10491.95** Fungi (F) 2 0.0059 2556.03 3.03 540732.85 Phosphorus (P) 3 0.039 7248.85 1.47 544117.23
Linear (Pl) 1 0.11 19691.07 1.53 1006530.96 Quadratic (Pq) 1 0.0049 2055.48 2.87 625338.95 Cubic (Pc) 1 0.000077 0.010 0.010 487.84 F x P 6 0.0023 175.50 0.33 33638.14"*
F x Pl 2 0.0031" 53.34 0.21 7291.34
F x Pq 2 0.0039" 465.13"* 0.75" 92589.06""
F x Pc 2 0.00011 8.02 0.040 1034.01 Error 44 0.00087 26.84 0.14 3606.65


""Significant at P < 0.01 "Significant at P < 0.05










76





0.4
4

c 0.30.2



0. .I





100 a CN A ETJ INT [ CETMU INT











0 s 0 00 120 0 30 00 00 12
P FFWJR (kIhf) P FUE1 (4ha)
















Fig. 7-4. Effect of P application on P concentration, total P, N concentration, and total N of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.









77


0.5
o CCN a ETU D INT 0.4


0.3

0.2
z:
cr-J
a- 0.1

0.0 400
o CON a ETU 0 INT














0 z




1-













Fig. 7-5. Effect of P application on P concentration, total P, and
total N of Aeschynomene americana grown in the field
conditions and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.
points are means of five replicates.










78

Ca, Mg, and Zn concentrations of Siratro and aeschynomene were not related to P applications or fungal inoculation (data not presented).

In contrast to greenhouse pot experiments, field experiments on

inoculation with VAM fungi often have been unsuccessful. It is possibly that the main cause for the satisfactory response to fungal inoculation obtained in the present study was due to a largely ineffective indigenous VAM population as compared to G. etunicatum and G. intraradices under the amended soil conditions. In previous greenhouse experiments (Chapter IV and V) using a similar nonpasteurized soil than the one used in the present study, I found that G. etunicatum and G. intraradices were effective growth enhancers of Siratro, even with a reasonably high soil native VAM population. Similarly, Powell et al. (1980b) reported that indigenous VAM fungi were ineffective in many soils and that inoculation by more effective VAM fungi would result in positive responses, even in nonpasteurized soils containing a high indigenous VAM population.

In conclusion, the present study shows that effective inoculation with selected VAM fungi can have an important effect on growth of forage legumes in the field in soils that contain ineffective native VAM populations under amended soil conditions, even at moderate levels of applied P.
















CHAPTER VIII
CONCLUSIONS



The objective of this chapter is to summarize the work of the preceding six chapters.

The overall goal of this research project was to improve the establishment phases and growth of tropical forage legumes in newly cleared land at reduced P fertilization through inoculation with effective VAM fungi. In order to accomplish this goal, greenhouse studies were carried out in limed, pasteurized and nonpasteurized Oldsmar fine sand, collected from a newly cleared area at the Agricultural Research and Education Center, Fort Pierce, FL. A field experiment was conducted in the same nonpasteurized soil.

In Chapter II, quantitative data on the amount of root colonization and the species distribution of VAM fungi associated with four cultivated tropical forage legumes from four different locations in south Florida are reported. Differences in percentage of root colonization and total spore density were significant among locations, legume species, and location x legume species interactions. Legume species differed in percentage root colonization and total spore density among locations except for carpon desmodium, which showed no differences among locations in percentage root colonization. The six species of VAM fungi collected in this survey were: G. heterogama, G. margarita, G. etunicatum, G. intraradices, Glomus sp., and A. spinosa.

79










80

In Chapter III, the effect of inoculation with G. intraradices on the growth of several tropical forage legumes in P-deficient, nonpasteurized and pasteurized soil under greenhouse conditions is reported. Shoot and root dry weights were increased after inoculation in nonpasteurized soil for Siratro, aeschynomene, Aeschynnomene villosa, Stylosanthes hamata, and Stylo, but not for Arachis sp. and Vigna adenantha, which only responded to inoculation in pasteurized soil.

In Chapter IV, the effect of five species of VAM fungi, G.

etunicatum, G. deserticola, G. versiforme, G. intraradices, and G. margarita, on the growth of Siratro in a limed, nonpasteurized soil with an applied P level of 20 mg kg-I under greenhouse conditions was determined. Shoot dry and root fresh weights of plants inoculated with G. etunicatum and G. intraradices were higher than the other VAM fungal treatments and noninoculated plants. In addition, plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices. The indigenous population of VAM fungi was reasonably high (MPN = 2 propagules g-1 soil); however, plant yields were less than the best VAM treated plants. A positive correlation was found between mycorrhizal root colonization, expressed as either percentage or total root length colonized, and shoot dry weight. Glomus etunicatum colonized roots more rapidly than the other VAM fungi tested.

In Chapter V, the effect of G. etunicatum and G. intraradices, on the growth of Siratro in a limed, nonpasteurized soil, with applied P levels of 2.5, 10, 20, and 40 mg kg-1 under greenhouse conditions is
-1
reported. At 2.5 mg kg of applied P, there was no yield response to

inoculation. Above 2.5 mg kg-1 of applied P, plants inoculated with inoculation. Above 2.5 mg kg of applied P, plants inoculated with










81

either G. etunicatum or G. intraradices weighed more than control plants.
-1
Inoculated plants required between 28 and 30 mg P kg-1 to achieve maximum
-1
shoot dry weight, whereas control plants, even with 40 mg P kg- did not achieve maximum growth. Shoot dry weight response was better with G. etunicatum than with G. intraradices. For both fungi, increasing P
-1
above 2.5 mg kg-1 increased the percentage and total root length colonized by VAM fungi.

In Chapter VI, the effect of inoculation with G. etunicatum, on the growth and uptake of P and N of three forage legumes at applied P levels
-1
between 12.5 and 50 mg kg-1 in a limed, pasteurized soil under greenhouse conditions is reported. At all levels of applied P, shoot dry and root fresh weights, and total N of Stylo were greater for mycorrhizal plants than nonmycorrhizal plants. The differences were most pronounced at intermediate levels of applied P and diminished at the higher P levels. Mycorrhizal plants of Siratro had greater shoot dry weight, total P and N than nonmycorrhizal plants at low and intermediate P levels, but not at the highest level. Differences in shoot dry weight between mycorrhizal and nonmycorrhizal plants of aeschynomene were significant only at the intermediate level of applied P. Overall, inoculated plants of the three legumes studied had greater P concentration than control plants. Maximum root colonization, expressed as either percentage or total length of colonized root of the three legumes, was attained at intermediate levels of applied P.

In Chapter VII, the effect of inoculation with selected VAM isolates on growth and nutrient uptake of Siratro and aeschynomene under natural field conditions at applied P levels of 10, 30, 60, and 120 kg ha-1 is










82

reported. At all levels of applied P and for all harvests, shoot dry weights of Siratro were greater for fungal inoculated plants than noninoculated plants. Differences between fungal inoculated and noninoculated plants were most marked at 30 to 90 kg ha-1 of applied P
-1
and diminished at 120 kg ha-1. The effect of fungal inoculation on the shoot dry weights of aeschynomene, at all levels of applied P, was similar (but more pronounced) as that of Siratro. At the first harvest of Siratro, plants inoculated with G. etunicatum had higher shoot dry weights than G. intraradices plants at all levels of applied P. However, in subsequent harvests of Siratro and for aeschynomene the response of shoot dry weight to inoculation with the two VAM fungi was similar. Fungal inoculation resulted in at least a 30% savings (40 kg ha-1) in the amount of P fertilizer required for maximum yield. Inoculated treatments had greater percentage of root colonized than noninoculated treatments at all levels of applied P. Percentage of root colonized by VAM fungi for the inoculated plants of the two legumes increased linearly with P additions up to 60 kg ha-1. There were no differences in root colonization between G. etunicatum and G. intraradices at any level of applied P.

The results of these studies clearly demonstrate that inoculation

with effective VAM fungi can increase the growth of legumes in soils that may have a high, but largely ineffective native VAM population than introduced VAM fungi under amended soil conditions. Furthermore, with highly mycorrhizal dependent crops such as tropical forage legumes, a mycorrhizal growth response may occur at P levels normally used in commercial pasture production.
















LITERATURE CITED



Abbott, L.K., and A.D. Robson. 1977a. The distribution and abundance of
vesicular-arbuscular endophytes in some Western Australian soils.
Aust. J. Bot. 25: 515-522.

Abbott, L. K., and A. D. Robson. 1977b. Growth stimulation of
subterranean clover with vesicular-arbuscular mycorrhizas. Aust. J.
Agric. Res. 28:639-649.

Abbott, L. K., and A. D. Robson. 1978. Growth of subterranean clover
in relation to formation of endomycorrhizas by introduced and
indigenous fungi in a field soil. New Phytol. 81:575-587.

Abbott, L. K., and A. D. Robson. 1981. Infectivity and effectiveness
of vesicular-arbuscular mycorrhizal fungi: Effect of inoculum type.
Aust. J. Agric. Res. 32:631-639.

Anderson, D. L., and L. J. Henderson. 1986. Sealed chamber digestion
for plant nutrient analyses. Agron. J. 78:937-939.

Asimi, S., V. Gianinazzi-Pearson, and S. Gianinazzi. 1980. Influence of
increasing soil phosphorus levels on interactions between vesiculararbuscular mycorrhizae and Rhizobium in soybeans. Can. J. Bot.
58:2200-2206.

Azcon-Aguilar, C., and J. M. Barea. 1981. Field inoculation of Medicago
with VA mycorrhiza and rhizobium in phosphate-fixing agricultural
soil. Soil Biol. Biochem. 13:19-22.

Barber, D. A., and B. C. Lougham. 1967. The effect of microorganisms
in the absorption of inorganic plant nutrients. II. Uptake and utilization of phosphate by barley plants grown under sterile and
non-sterile conditions. J. Exp. Bot. 18:170-176.

Barea, J. M., J. L. Escudero, and C. Azcon-G. de Aguilar. 1980. Effects
of introduced and indigenous VA mycorrhizal fungi on nodulation and
nutrition of Medicago sativa in phosphate fixing soils as affected
by P fertilizers. Plant Soil 54:283-296.

Bergersen, F. J. 1971. Biochemistry of nitrogen fixation in legumes.
Annu. Rev. Plant Physiol. 22:124-140.



83










84

Bethlenfalvay, G.J., J.M. Ulrich, and M.S. Brown. 1985. Plant response
to mycorrhizal fungi: host, endophyte, and soil effects. Soil Sci.
Soc. Am. J. 49: 1164-1168.

Black, R. L. B., and P. B. Tinker. 1977. Interactions between effects
of vesicular-arbuscular mycorrhiza and fertilizer phosphorus on
yields of potatoes in the field. Nature 267:510-511.

Burt, R. L., and C. P. Miller. 1975. Stylosanthes--A source of pasture
legumes. Trop. Grassl. 9:117-123.

Chulan, A., and P. Ragu. 1986. Growth response of Theobroma cacao L.
seedlings to inoculation with vesicular-arbuscular mycorrhizal
fungi. Plant Soil 96:279-285.

Cooper, K. M. 1978. Adaption of mycorrhizal fungi to phosphate
fertilizers. p. 107. In A. R. Ferguson, R. L. Bieleski, and I. B.
Ferguson (eds.) Plant nutrition 1978. New Zealand DSIR Information
Series No. 134, Government Printer, Wellington.

Crush, J. R. 1974. Plant growth responses to vesicular-arbuscular
mycorrhiza. VII. Growth and nodulation of some herbage legumes.
New Phytol. 73:745-750.

Currah, R.S., and M. Van Dyk. 1986. A survey of some perennial vascular
plant species native to Alberta for occurrence of mycorrhizal fungi.
Canadian Field-Naturalist 100: 330-342.

Daft, M.I., and A.A. El-Giahmi. 1976. Studies on nodulated and
mycorrhizal peanuts. Ann. Appl. Biol. 83:273.

Daniels, B. A., P. M. McCool, and J. A. Menge. 1981. Comparative
inoculum potential of spores of six vesicular-arbuscular mycorrhizal
fungi. New Phytol. 89:385-391.

Daniels, B. A., and H. D. Skipper. 1982. Methods for recovery and
quantitative estimation of propagules from soil. p. 29-35. In N. C.
Schenck (ed.) Methods and principles of mycorrhizal research. Amer.
Phytopath. Soc., St. Paul, MN.

Gallaher, R. N., C. 0. Weldon, and J. G. Futral. 1975. An aluminum
block digester for plant and soil analysis. Soil Sci. Soc. Amer.
Proc. 39:803-806.

Gates, C. T., and J. R. Wilson. 1974. The interaction of nitrogen and
phosphorus on the growth, nutrient status and nodulation of
Stylosanthes humilis H.B.K. (Townsville stylo). Plant Soil 41:325333.

Gerdemann, J.W., and J.M. Trappe. 1974. The endogonaceae in the Pacific
Northwest. Mycol. Mem. 5: 1-76.










85

Gibson, A. H. 1976. Limitations to nitrogen fixation in legumes. In
W. E., Newton, and C. J. Nyman (eds.) Proc. ist Int. Symp. Nitrogen
Fixation, Vol.2. Wash. State Univ. Press, Pullman.

Giovannetti, M. 1985. Seasonal variations of vesicular-arbuscular
mycorrhizas and endogonaceous spores in a maritime sand dune.
Trans. Br. Mycol. Soc. 84: 679-684.

Giovannetti, M., and B. Mosse. 1980. Evaluation of techniques for
measuring vesicular-arbuscular mycorrhizal infection in roots. New
Phytol. 84:489-500.

Giovannetti, M., and T.H. Nicolson. 1983. Vesicular-arbuscular
mycorrhizas in Italian sand dunes. Trans. Br. Mycol. Soc. 80: 552557.

Habte, M., and T. Aziz. 1985. Response of Sesbania grandiflora to
inoculation of soil with vesicular-arbuscular mycorrhizal fungi.
Appl. Environ. Microbiol. 50: 701-703.

Habte, M., and A. Manjunath. 1987. Soil solution phosphorus status and
mycorrhizal dependency in Leucaena leucocephala. Appl. Environ.
Microbiol. 53:797-801.

Hall, I. R. 1978. Effects of endomycorrhizas on the competitive ability
of white clover. N. Z. J. Agr. Res. 21:509-515.

Hall, I. R. 1979. Soil pellets to introduce vesicular-arbuscular
mycorrhizal fungi into soil. Soil Biol. Biochem. 11:85-86.

Hall, I. R. 1984. Field trials assessing the effect of inoculating
agricultural soils with endomycorrhizal fungi. J. Agric. Sci.
102:725-731.

Hambleton, L. G. 1977. Semiautomated method for simultaneous
determination of phosphorus, calcium and crude protein in animal
feeds. J. Assoc. Offic. Agr. Chem. 60:845-852.

Harley, J.L., and E.L. Harley. 1987. A check-list of mycorrhiza in the
British flora. New Phytol. 105:1-102.

Harley, J. L., and S. E. Smith. 1983. Mycorrhizal symbiosis. Academic
Press, London.

Hass, J. H., and J. Krikum. 1985. Efficacy of endomycorrhizal fungus
isolates and inoculum quantities required for growth response. New
Phytol. 100:613-621.

Hayman, D.S. 1978. Mycorrhizal populations of sown pastures and native
vegetation in Otago, New Zealand. N. Z. J. Agr. Res. 21: 271-275.










86

Hayman, D. S. 1983. The physiology of vesicular-arbuiscular
endomycorrhizal symbiosis. Can. J. of Botany 61:944-963.

Hayman, D. S., and K. A. Hampson. 1978. VA mycorrhiza. Field
inoculation trial (white clover in Welsh upland soil). Rothamsted
Rep. Part 1:238-239.

Hayman, D. S., and B. Mosse. 1979. Improved growth of white clover in
hill grasslands by mycorrhizal inoculation. Ann. Appl. Biol.
93:141-148.

Hayman, D. S., R. J. Page, and C. A. Clarke. 1980. Vesiculararbuscular mycorrhiza: field inoculation studies with Red clover,
Sawyers I. Rothamsted Rep. Part 1:201-202.

Hayman, D.S., and G.E. Stovold. 1979. Spore populations and infectivity
of vesicular-arbuscular mycorrhizal fungi in New South Wales. Aust.
J. Bot. 27: 227-233.

Hayman, D. S., and M. Tavares. 1985. Plant growth responses to
vesicular-arbuscular mycorrhiza. XV. Influence of soil pH on the
symbiotic efficiency of different endophytes. New Phytol. 100:367377.

Hodges, E. M., A. E. Kretschmer, Jr., P. Mislevy, R. D. Roush, 0. C.
Ruelke, and G. H. Snyder. 1982. Production and utilization of the
tropical legume aeschynomene. Fla. Agric. Exp. Stn. Circ. S-290.

Huang, R., W.K. Smith, and R.S. Yost. 1985. Influence of vesiculararbuscular mycorrhiza on growth, water relations, and leaf
orientation in Leucaena leucocephala (Lam.) Dewit. New Phytol. 99:
229-243.

Hutton, E. M. 1962. Siratro-a tropical pasture legume bred from
Phaseolus atropurpureum. Aust. J. Agric. Anim. Hus. 2:117-125.

Iabal, S. H., K. Sultana, and B. Perveen. 1975. Endogone spore numbers
in the rhizosphere and the occurrence of vesicular-arbuscular
mycorrhizas in plants of economic importance. Biologia 21:227-237.

Islam, R., A. Ayanaba, and F. E. Sanders. 1980. Response of cowpea
(Vigna unguiculata) to inoculation with VA mycorrhizal fungi and rock phosphate fertilization in some unsterilized Nigerian soils.
Plant Soil 54:107-117.

Jensen, A. 1984. Responses of barley, pea, and maize to inoculation
with different vesicular-arbuscular mycorrhizal fungi in irradiated
soil. Plant Soil 78: 315-323.

Khan, A. G. 1975. Growth effects of VA mycorrhiza on crops in the
field. P. 149-435. In F. E. Sanders, B. Mosse, and P. B. Tinker
(eds.) Endomycorrhizas. Academic Press, London, New York.










87

Kormanik, P. P., and A. C. McGraw. 1982. Quantification of vesiculararbuscular mycorrhizae in plant roots. p. 37-45. In N. C. Schenck
(ed.) Methods and principles of mycorrhizal research. American
Phytopathological Society, St. Paul, MN.

Kretschmer, A. E., Jr. 1972. Siratro (Phaseolus atropurpureus DC.) A
summer-growing perennial pasture legume for central and south
Florida. Florida Agric. Exp. Stn. Circ. S-214.

Kucey, R.M., and G.E. Diab. 1984. Effects of lime, phosphorus, and
addition of vesicular-arbuscular (VA) mycorrhizal fungi on
indigenous VA fungi and on growth of alfalfa in a moderately acidic
soil. New Phytol. 98: 481-486.

Lopes, E. S., and E. De Olivera. 1980. Efeito de species de micorrizas
vesicular-arbuscular em Siratro (Macroptilium atropurpureum).
Bragantia 39:241-245.

Lopes, E.S., E. Oliveira, R. Dias, and N.C. Schenck. 1983. Occurrence
and distribution of vesicular-arbuscular mycorrhizal fungi in coffee
(Coffea arabica L.) plantations in central Sao Paulo state, Brazil.
Turrialba 33:417-422.

Lynd, J. Q., R. J. Tyrl, and A. A. C. Purcino. 1985. Mycorrhiza soil
fertility effects on regrowth, nodulation and nitrogenase activity
of Siratro (Macroptilium atropurpureum (DC) Urb.). J. Plant Nutr.
8:1047-1059.

Menge, J. A. 1983. the physiology of vesicular-arbuscular mycorrhizal
fungi in agriculture. Can. J. Bot. 61:1015-1024.

Miller, R.H., E.J.B.N. Cardoso, and C.O.N. Cardoso. 1979. Some
observations on mycorrhizal infection of tropical forage legumes and
grasses in Brazil. Summa Phytopathologica 5: 168-172.

Miller, D. D., P. A. Domoto, and C. Walker. 1985. Colonization and
efficacy of different endomycorrhizal fungi with apple seedlings at
two phosphorus levels. New Phytol. 100:393-402.

Mosse, B. 1972. The influence of soil type and endogone strain on the
growth of mycorrhizal plants in phosphate deficient soil. Rev.
Ecol. Biol. Sol. T. 9:529-537.

Mosse, B. 1977. Plant growth responses to vesicular-arbuscular
mycorrhiza. X. Responses of stylosanthes and maize to inoculation
in unsterile soils. New Phytol. 78:277-288.

Mosse, B., C. L. Powell, and D. S. Hayman. 1976. Plant growth responses
to vesicular-arbuscular mycorrhiza. IX. Interactions between VA
mycorrhiza, rock phosphate and symbiotic nitrogen fixation. New
Phytol. 76:331-342.










88

Munns, D. N., and B. Mosse. 1980. Mineral nutrition of legume crops.
In R. J. Summerfield, and A. H. Bunting (eds.) Advances in legume
science. Univ. Reading Press, England.

Newbould, P., and A. Rangeley. 1984. Effect of lime, phosphorus and
mycorrhizal fungi on growth, nodulation, and nitrogen fixation by
white clover ( Trifolium reopens) grown in UK hill soils. Plant Soil
76: 105-114.

Nielsen, J.P., and A. Jensen. 1983. Influence of vesicular-arbuscular
mycorrhiza fungi on growth and uptake of various nutrients as well
as uptake ratio of fertilizer P for lucerne (Medicago sativa).
Plant Soil 70: 165-172.

Pairunan, A. K., A. D. Robson, and L. K. Abbott. 1980. The
effectiveness of vesicular-arbuscular mycorrhizas in increasing
growth and phosphorus uptake of subterranean clover from phosphorus
sources of different solubilities. New Phytol. 84:327-338.

Plenchette, C., V. Furlan, and J. A. Fortin. 1982. Effects of different
endomycorrhizal fungi on five host plants grown on calcined montmorillonite clay. J. Amer. Soc. Hort. Sci. 107:535-538.

Powell, C.L1. 1979. Inoculation of white clover and ryegrass seed with
mycorrhizal fungi. New Phytol. 83: 81-85.

Powell, C. Ll. 1980a. Mycorrhizal infectivity of eroded soils. Soil
Biol. Biochem. 12:247-250.

Powell, C. Ll. 1980b. Phosphate response curves of mycorrhizal and nonmycorrhizal plants. I. Responses to superphosphate. N. Z. J. Agr.
Res. 23:225-231.

Powell, C. Ll. 1982. Selection of efficient VA mycorrhizal fungi.
Plant Soil 68:3-9.

Powell, C. Ll., M. Groters, and D. Metcalfe. 1980. Mycorrhizal
inoculation of a barley crop in the field. N. Z. J. Agr. Res.
23:107-109.

Purcino, A. A. C., C. Lurlarp, and J. Q. Lynd. 1986. Mycorrhiza and
soil fertility effects with growth, nodulation and nitrogen fixation
of Leucaena grown on a Typic Eutrustox. Commun. in Soil Sci. Plant
Anal. 17:473-489.

Purcino, A. A. C., and J. Q. Lynd. 1985. Tripartite symbiosis of
Stylosanthes scabra Vog. influenced by soil fertility treatments of
a Typic Eutrustox. Agron. J. 77:455-458.

Rangeley, A., M.J. Daft, and P. Newbould. 1982. The inoculatation of
white clover with mycorrhizal fungi in unsterile hill soils. New
Phytol. 92: 89-102.




Full Text
72
Table 7-8. Analysis of variance for percentage root colonized, P
concentration, and total P and N uptake of Aeschynomene
americana harvested 2 October 1986.
Source of
variation
DF
Mean Squares
Root
colon.
P cone.
total P
total N
Block
4
10.11
0.00065^
384.02
0.051
Fungi (F)
2
5276.62
0.009T*
9052.18
0.39**
Phosphorus (P)
3
1844.64
0.051^*
75570.74
4.64^
Linear (PI)
1
4408.21
0.12**
196270.22
12.04**
Quadratic (Pq) 1
1001.65
0.031**
29179.92
1.89**
Cubic (Pc)
1
124.08,
0.0012
1262.07
0.0021
F x P
6
378.66'^
0.0010
1190.91**
0.026
F x PI
2
833.18'{*
0.0021
1275.75x*
0.013
F x Pq
2
220.16^
0.00052
1815.29**
0.043
F x Pc
2
82.64**
0.00021
481.71
0.022
Error
44
11.60
0.0011
206.45
0.062
Significant at P < 0.01
''Significant at P < 0.05


LIST OF TABLES
, Page
Table 2-1. Chemical characteristics of the soils sampled
for VAM fungi associated with four tropical
forage legumes at four locations in south
Florida. . 8
Table 2-2. Analysis of variance for percentage of
mycorrhizal root colonization and total spore
density per 100 g of air-dried soil 9
Table 2-3. Mean percentage of mycorrhizal root colonization
and total spore density of VAM fungi among forage
legumes at four locations in south Florida 10
Table 2-4. Mean spore numbers of VAM fungal species
associated with four forage legumes at four
locations in south Florida 12
Table 3-1. Percentage of mycorrhizal root colonization of
tropical forage legumes in nonpasteurized (UP)
or pasteurized (P) soil in the greenhouse after
45 days. .! 19
i-fvv T'5.- ~r' '** "y. h rr*i' ~ f ;>
Table 5-1. Regression equations and coefficients of
determination (r^) showing the relationship
of P level to shoot dry weights, root fresh
weights, percentage and total root length
colonized 40
Table 5-2. Analysis of variance for shoot dry weights, root
1-*' fresh weights, percentage and total root
length colonized 41
Table 6-1. Mean squares and levels of significance from
the analysis of variance for shoot dry weight,
root fresh weight, P concentration, and total P
- !and.N uptake of forage legumes. ........... 49
Table 7-1. Shoot dry weights, P concentrations, and percentage
of roots colonized of Siratro and Aeschynomene
americana seedlings at transplanting 60
vi



PAGE 1

GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION By ONESIMO A. MEDINA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1987

PAGE 2

To the memory of my father Efrain, who influenced me in a very special way. I am very heartbroken that he died before this work wa completed. To my mother, Guillermina, for her never ending sacrifices, her love, and prayers.

PAGE 3

ACKNOWLEDGMENTS It is with deep gratitude that I express thanks to the chairman and cochairman of my committee. Dr. David M. Sylvia and Dr. Albert E. Kretschmer, Jr., respectively, for their support, constant encouragement, guidance, and friendship. I also thank the other members of my committee. Dr. G. H. Snyder, Dr. G. Kidder, Dr. N. C. Schenck, and Dr. J. B. Sartain, for their suggestions, support, and editorial comments. Gratitude is extended to Mr. Tom Wilson for his friendship and valuable assistance rendered in the field portion of this project. The moral support, love, and motivation of my brothers, Oquendo, Miosotis, and Gagarin were essential to the completion of my graduate program. This research was funded in part through USDA ARS Tropical Agricultural Development Grants 83-CRSR-2-2134 and 86-CRSR-2-2846 This support is greatly appreciated. Most of all, warmest thanks go to my wife, Griselda, for her understanding and encouragement during my graduate studies. Her many hours of assistance in typing the manuscript will always be remembered. I also thank my daughter Michelle for making her Mommy and Daddy very happy. Ill

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGMENTS ^. m LIST OF TABLES vi LIST OF FIGURES • • • •, viii ABSTRACT xi CHAPTER I INTRODUCTION 1 CHAPTER II THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA 3 Introduction 3 Materials and Methods 4 Results and Discussion 7 CHAPTER III > GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH GLOMUS INTRARADICES 14 Introduction s Materials and Methods 15 Mours Vi Result^ and. Discussion .j^ .....>.,..-..! .-.clj CHAPTER IV hrvi'-GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I. SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI UNDER AMENDED SOIL CONDITIONS 23 Introduction 23 iv

PAGE 5

Materials and Methods 24 Results and Discussion 26 CHAPTER V GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II. EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT DIFFERENT P LEVELS 34 Introduction 34 Materials and Methods 36 Results and Discussion 37 CHAPTER VI EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE GROWTH AND UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM STYLOSANTHES GUIANENSIS AND AESCHYNOMENE AMERICANA 45 Introduction 45 r Materials -and Methods .. 46 Results and Discussion 48 CHAPTER VII GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND AESCHYNOMENE AMERICANA TO INOCULATION WITH SELECTED VA ""MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS 55 Introduction 55 Materials and Methods 57 Results and Discussion 59 CHAPTER VIII CONCLUSIONS ............. 79 LITERATURE CITED 83 BIOGRAPHICAL SKETCH 91 SL1.VCVT;.)!-. or EFFKCTl x'^-: ''a ';^CO::,Wj..,AL h'Miri V

PAGE 6

LIST OF TABLES lj~"'S^r. li Page Table 2-1, Chemical characteristics of the soils sampled for VAM fungi associated with four tropical forage legumes at four locations in south Florida. 8 Table 2-2. Analysis of variance for percentage of mycorrhizal root colonization and total spore density per 100 g of air-dried soil 9 Table 2-3. Mean percentage of mycorrhizal root colonization and total spore density of VAM fungi among forage legumes at four locations in south Florida 10 Table 2-A. Mean spore numbers of VAM fungal species 1 associated with four forage legumes at four locations in south Florida 12 Table 3-1. Percentage of mycorrhizal root colonization of tropical forage legumes in nonpasteurized (UP) .. or pasteurized (P) soil in the greenhouse after ^c: 1^5 days. Table 5-1. Regression equations and coefficients of determination (r^) showing the relationship ^ of P level to shoot dry weights, root fresh weights, percentage and total root length colonized /^q Table 5-2. Analysis of variance for shoot dry weights, root fresh weights, percentage and total root length colonized } Table 6-1. Mean squares and levels of significance from C the analysis of variance for shoot dry weight, < root fresh weight, P concentration, and total P i:! A! ., SLand.N uptake of forage legumes 49 Table 7-1. Shoot dry weights, P concentrations, and percentage of roots colonized of Siratro and Aeschynomene americana seedlings at transplanting 60 vi

PAGE 7

Table 7-2. Table 7-3. Analysis of variance for shoot dry weights of Aeschynomene americana harvested 2 October 1986. Analysis of variance for shoot dry weights from four Siratro harvests. 61 53 Table 7-4. Table 7-5. Table 7-6. Table 7-7, Table 7Table 7-9. Regression equations and coefficients of determination (r^) showing the relationship of applied P level to shoot dry weight, percentage of root colonization, P and N concentrations, and total P and N uptake for the first harvest of Siratro Regression equations and coefficients of determination (r^) showing the relationship of applied P level to shoot dry weights for the first, third, and fourth harvest and percentage of root colonization for the fourth harvestof Siratro 1 p.pp 66 67 Regression equations and coefficients of determination (r^) showing the relationship of applied P level to shoot dry weight, percentage of root colonization, P concentration, and total P and N uptake of Aeschynomene americana 68 Analysis of variance table for percentage root colonized of Siratro by VAM fungi 71 Analysis of variance table for percentage root colonized, P concentration, total P, and total N.of Aeschynomene americana 72 Analysis of variance table for P concentration, total P, N concentration, and total N of Siratro for the first harvest 75

PAGE 8

LIST OF FIGURES Fig. 2-1. Collection sites for VAM fungi associated with four tropical forage legumes in south Florida. ... 5 Fig. 3-1. Effect of inoculation with Glomus intraradices on the shoot and root dry weights of tropical forage legumes in nonpasteurized soil in the greenhouse after 45 days. Legume species were Aeschynomene americana (AA), Aeschynomene villosa (AV), Arachis sp. (AS), Macroptilium atropurpureum (MA), Leucaena leucoephala (LL), Stylosanthes hamata (SH), Stylosanthes guianensis (SG), and Vigna adenantha (VA). Bars represent the mean of 3 replicates. Means with the same letter within a species are not different (P < 0.05) 18 Page Fig. 3-2. Effect of inoculation with Glomus intraradices on the shoot and root dry weights of tropical forage legumes in pasteurized soil in the greenhouse after A5 days. The legume species were Aeschynomene americana (AA), Arachis sp. (A.S), Macroptilium atropurpureum (MA), Leucaena leucocephala (LL), and Vigna adenantha (VA). Bars represent the mean of 3 replications. Means with the same letter within a species are not different (P < 0.05) 21 Fig. 4-1. Effect of inoculation with Gigaspora margarita (MAR), Glomus versiforme (VER), Glomus deserticola (DES), Glomus intraradices (INT), Glomus etunicatum (ETU), or the control (CON) on the shoot dry weight and root fresh weight of Siratro at two harvests. Bars represent the means of five replicates. Means with the same letter within a harvest are not different (P < 0.05) 27 viii

PAGE 9

Fig. 4-2 Effect of inoculation with Gigaspora margarita (MAR), Glomus versiforme (VER), Glomus deserticola (DES), Glomus intraradices (INT), Glomus etunicatum (ETU), or the control (CON) on the percentage of root colonization and root length colonized of Siratro at three and two harvests, respectively. Bars represent the means of five replicates. Means with the same letter within a harvest are not different (P < 0.05) 28 Fig. 4-3 Relationship between shoot dry weight and length of Siratro roots colonized by VAM fungi for all inoculated treatments 30 Fig. 5-1. Effect of P application on shootdry weight, root fresh weight, percentage of root colonized by VAM fungi, and total root length colonized of Siratro grown in limed nonpasteurized soil and inoculated with Glomus etunicattim (ETU), Glomus intraradices (INT), or not inoculated (CON). 38 Fig. 5-2. Relationship between shoot dry weight and length of Siratro roots colonized by VAM fungi for all inoculated treatments in nonpasteurized soil 42 Fig. 6-1. Effect of fungal inoculation and P applications on the shoot dry weight, root fresh weight, root colonization, P concentration, total P, and total N of Stylosanthes guianensis 50 Fig. 6-2. Effect of fungal inoculation and P applications on the shoot 'dry weight, root fresh weight, root colonization, P concentration, total P, and total N of Macroptilium atropurpurexim 52 Fig. 6-3. Effect of fungal inoculation and P applications on the shoot dry weight, root fresh weight, root colonization, P concentration, total P, and total N of Aeschynomene americana 53 Fig. 7-1. Effect of P application on shoot dry weights for the first (A), second (E), third (C), and fourth (D) harvest of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), intraradices (INT), or not inoculated (CON). Data points are means of five replicates 64 ix

PAGE 10

Fig. 7-2. Effect of P application on shoot dry weight and percentage of root colonized of Aeschynomene americana grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates 65 Fig. 7-3. Effect of P application on percentage of root colonized for the first (A) and fourth (B) harvest of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates 73 Fig. 7-4. Effect of P application on P concentration, total P, N concentration, and total N of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates 76 Fig. 7-5. Effect of P application on P concentration, total P, and total N of Aeschynomene americana grown in the field and inoculatsd, with, Gloimis etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates 77 X

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION By Onesimo A. Medina December 1987 Chairman: Dr. D. M. Sylvia ". .. ^. Cochairman: Dr. A. E. Kretschmer, Jr. Major Department: Soil Science Greenhouse and field studies were conducted to determine the growth response of tropical forage legumes to inoculation with vesiculararbuscular mycorrhizal (VAM) fungi and P applications and to evaluate the effectiveness of the indigenous VAM population versus introduced species Root and rhizosphere soil samples of four tropical forage legume species were collected at four locations in south Florida before initiating the greenhouse experiments. Six species of VAM fungi were isolated in this survey. The occurrrence of VAM fungal species, as determined by spore numbers, was affected by legume species and location. Shoot dry and root dry weights of 'Siratro' ( Macroptilium atropurpureum Urb.), aeschynomene ( Aeschynomene americana L.), xi

PAGE 12

Aeschynomene villosa Poir., Stylo ( Stylosanthes guianensis SW.) and Stylosanthes hamata Taub. were increased in pasteurized and nonpasteurized limed soil in the greenhouse after inoculation with Glomus intraradices Schenck & Smith. Inoculation with Glomus etunicatum Becker & Gerdemann and G. intraradices also increased the growth of Siratro as compared to other VAM fungi and the noninoculated control in limed, nonpasteurized soil fertilized with 20 mg kg ^ of P. In other greenhouse experiments, G. etunicatum and G. intraradices were effective growth enhancers of Siratro over a practical range of 2.5 to 40 mg kg ^ of applied P in a limed, nonpasteurized soil. For both fungi, increasing P above 2.5 mg kg ^ increased the percentage and total root length colonized by VAM fungi. A positive correlation was found between mycorrhizal root colonization and shoot dry weight. In a limed, pasteurized soil, inoculation with G. etunicatum increased total P and N of Siratro at 12.5 and 25 mg kg ^ of applied P, but not at 50 mg kg ^. The effectiveness of G. etunicatum and G. intraradices with Siratro and aeschynomene was corroborated in a field trial. These fungi increased the growth and uptake of P and N of both legvunes over a range of applied P from 10 to 80 kg ha'''". Inoculation of forage legximes with effective VAM fungi enhanced their growth. Growth enhancement occurred at P and lime levels used in commercial pasture production and in soils that had a large, but apparently ineffective indigenous VAM population. xii

PAGE 13

• CHAPTER I INTRODUCTION ;.riI.t.LForage legumes are an important component of improved grass pastures and must be established rapidly and without excessive cost. The legumes serve both to increase forage quality and decrease the need for N fertilizer through N2-fixation. Newly cleared lands incorporated into pasture production in south Florida are generally acidic and very low in total and available P throughout the soil profiles. While improvements to the productivity of these pastures may be obtained by the introduction of suitable legumes, effective N2-fixation and establishment' of legumes is frequently limited by the. low levels of available P in these soils. Snyder et al. (1978) reported that large applications of P fertilizer are normally required for legume establishment and optimun growth in these soils. However, with the increasing cost of P fertilizer, alternative strategies for minimum P fertilizer input and efficient use of P must be adopted. One of these strategies may be via the management of vesicular-arbuscular mycorrhizal (VAM) symbioses. .. tnsi: Vesiciilar-arbuscular mycorrhizal fungi can improve the growth of legumes by increasing P uptake (Bethlenfalvay et al., 1985; Hayman, 1983). Phosphorus is often a growth-limiting factor since many legumes have P requirements and are poor scavengers of P. The VAM fungi may also increase nodulation and N2-fixation of legumes, primarily as an indirect 1

PAGE 14

2 effect of improved P nutrition (Daft and "1-Giahmi, 1976; Habte and Aziz, 1985). Information concerning the association of VAM fungi with tropical forage legumes is sparse. Most of the growth response studies reported were done in either pasteurized soil or in small volumes of ~ ^ '-^'t.o nonpasteurized soil. Except for the work of Saif (1987), little information is available on growth response of tropical forage legumes to inoculation with VAM fungi in nonpasteurized soil, especially under field conditions where introduced species of VAM fungi must compete with the indigenous VAM population. Therefore, greenhouse studies were conducted in limed, pasteurized and nonpasteurized soil to improve the growth of tropical forage legumes through inoculation with effective VAM fungi and reduced P fertilization. In addition, the effect of inoculation with selected VAM isolates on growth and nutrient uptake of two tropical forage legumes under natural field conditions was investigated at different levels of applied P.

PAGE 15

CHAPTER II THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA. Introduction There is widespread interest in the use of tropical forage legumes to increase production of tropical grasses in Florida's beef -cattle industry (Snyder et al., 1985). These legumes respond to inoculation with VAM fungi (Lynd et al., 1985; Saif, 1987)). However, before initiating fungal inoculation experiments with these forage legximes in south Florida, a survey was needed of the native populations of VAM fungi associated with several commercial forage legumes growing on a variety of soils. Vesicular-arbuscular mycorrhizal associations have been observed in a wide variety of natural and agricultural ecosystems (Abbott and Robson, 1977a; Currah and Van Dyk, 1986; Harley and Harley, 1987). In Florida, the occurrence and distribution of VAM fungi in agronomic crops, including some tropical legumes (Schenck and Kinloch, 1980; Schenck and Smith, 1981), and sand-dune vegetation (Sylvia, 1986), has been reported. However, there is no information on the degree of native VAM colonization of tropical forage legumes in Florida or on the susceptibility of different species of legumes to various genera and species of VAM fungi. The objective of this survey was to obtain quantitative data on the amount of root colonization and the species distribution of VAM fungi

PAGE 16

4 associated with four cultivated tropical forage legumes from four different locations in south Florida. Materials and Methods Root and rhizosphere soil samples of four tropical forage legumes were collected from 11 to 17 October 198A, at four locations in south Florida: Deseret Ranches, Deer park; Fort Pierce, Agricultural Research and Education Center (AREC); Ona, AREC; and Basinger Ranch, 109 Ranch (Fig. 2-1). Most of the soils of the studied area belong to the order Spodosols. They are dominated by nearly level, somewhat poorly to poorly drained sandy soils with dark sandy subsoil layers. These soils are used primarily for pastures, vegetables, flowers, forestry, and citrus. The forage legumes sampled were: 'Siratro' ( Macroptilium atropurpureum Urb.), (except at Deseret Ranches), aeschynomene ( Aeschynomene americana L.), Vigna adenantha Marechal, Mascherpa and Stainier, and carpon desmodium ( Desmodium heterocarpon DC). The legiimes were mixed with pasture grasses at the time of sampling. Three rhizosphere soil samples were collected to a depth of 15 cm for each legiime at each location. Samples, consisting of three subsamples of approximately 1.5 kg, were placed in plastic bags and transported to the laboratory on the same day. Samples were sieved through a 4-mm screen, and 100 g subsamples were removed and stored at 5C for spore extraction. Legxime roots were carefully separated manually from grass roots. A portion (0.5 g) of each

PAGE 17

Fig. 2-1. Collection sites for VAM fungi associated with four tropical forage legumes in south Florida.

PAGE 18

6 root sample was cleared in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik and McGraw, 1982). Root colonization by VAM fxingi was estimated by the gridlineintersect method of Giovannetti and Mosse (1980). Chemical content of a composite soil sample from each location was determined by the Soil Testing Laboratory, University of Florida (Rhue and Kidder, 1984). Mehlich-I solution (0.05 M HCl + 0.0125 M H2SO4) was used to extract Al, Ca, K, Mg, and P. All elements were analyzed in the filtrate by atomic absorption spectrophotometry, except P which was determined using the ammonium molybdate/ascorbic acid colorimetric method. Soil pH was determined using a 1:2 (v/v) soil: water ratio. Organic matter was estimated by oxidation with 1 N K2Cr207 in the presence of H2SO4. Spores of VAM fungi were removed from soil by the wet sieving method of Daniels and Skipper (1982) using sieves with 425, 90, and 45 um openings. Fractions retained on 90 and 45 um sieves were centrifuged (1000 X g) for 3 min in water. The pellet was resuspended in 40% sucrose solution and centrifuged for 1.5 min. Spore species were identified where possible (Schenck and Smith, 1982; Trappe, 1982). In addition, spores or washed roots were placed in pasteurized Arredondo loamy sand surface soil (siliceous hyperthermic Grossarenic Paleudult) in 15-cm-diam plastic pots in the greenhouse and planted with bahiagrass ( Paspalum notatum Flugge), carpon desmodium, or Siratro in an attempt to isolate VAM fungi in a manner similar to that described by Gerdemann and Trappe (1974) as the "inoculated pot culture" method.

PAGE 19

7 Results and Discussion .<-.. j -i la Results of soil pH and chemical analysis of soil samples reflect the different management regimes (including lime and fertilizer) (Table 2-1). -"r-f -< -i: ^51-":,.i:;.nn ^ Differences in percentage of mycorrhizal root colonization and total spore density of air dry soil were significant among locations, legume species, and location x legume species interactions (Table 2-2). Total spore density at the four locations ranged from 5 to 679 per 100 g of air-dried soil, and the percentage of mycorrhizal root colonization varied from 3 to 41%. Miller et al. (1979) observed variable degree of mycorrhizal root colonization (4 to 74%) in forage grasses and legumes in Brazil. Except for carpon desmodium, legume species differed in percentage root colonization and total spore density among locations (Table 2-3). Fort Pierce had the highest total spore density for each legume species sampled except for Siratro. Attempts were made to relate percentage root colonization and total spore density to soil P or the other soil chemical characteristics presented in Table 2-1, but no clear relationships were apparent. Abbott and Robson (1977a) and Hayman (1978) also reported that spore numbers were not correlated with soil P or soil pH in cultivated soils. There was a positive correlation (P < 0.05) between root colonization and total spore density for all legume species at Basinger (r:= 0:70) and Deseret (r =0.76), but not at Fort Pierce and Ona. .p Giovannetti (1985) and Miller et al. (1979) reported a correlation

PAGE 20

8 Table 2-1. Chemical characteristics of the soils sampled for VAM fungi associated with four tropical forage leg\imes at four locations in south Florida. Location Leg^ime species^ O.M. pH Al Ca Mg K P ; fly ---mg kg 1 soilAA 1.4 6.0 23 314 93 8 4 rort rierce uti 1.2 5.3 22 241 15 16 5 VA 1.3 5 5 25 242 21 20 4 IT 1 5 2 62 270 25 13 16 AA 3.4 6.1 44 1320 143 64 23 Ona DH 3.1 5.4 36 920 120 29 8 VA 2.5 4.9 22 480 70 46 8 MA 5.7 4:7 55 800 '95 43 5' AA 2.5 6.1 66 780 67 40 6 Deseret DH 2.3 6.0 27 1040 94 27 4 VA • '2.8 7.2 30 1600 141 55 9 AA 4.7 5.3 28 960 32 28 4 Basinger DH 3.5 5.2 26 460 100 46 7 VA 3.6 5.1 28 540 111 58 8 MA 4.2 5.2 36 640 129 94 11 ^AAAeschynomene americana ; DH= Desmodium heterocarpon ; VA= Vigna adenantha ; MA= Macroptilium atropurpureum

PAGE 21

9 Table 2-2. Analysis of variance for percentage of mycorrhizal root colonization and total spore density per 100 g of air.. dried soil. Mean Squares Source of variation Degree of freedom Root colonization total spore density % no. 100 g"-'soil Location 3 A47 325871 Legumes 3 321. 73895., Interaction 8 299"* 88933** Error 30 13 1320 Significant at P < 0.01

PAGE 22

10 Table 2-3. Mean percentage of mycorrhizal root colonization and total spore density of VAM fungi among forage legumes at four locations in south Florida. Location Root colonization^ Total spore density^ Ft. Pierce Ona Basinger Deseret Aeschynomene americana 6b 30^ no. 100 g"-*soil 302' 160^ 146' Ft. Pierce Ona Basinger Deseret Ft. Pierce Ona Basinger Deseret Ft. Pierce Ona Basinger Desmodium heterocarpon 12^ 15^ 12^ 16^ Vigna adenantha 5^ 4ia 20^ 25b Macroptilium atropurpureum 15^ 679^ 376^ 19c 36^ 5353 77b 25c 36^ 23' 294' 2^Means within a column for each legume species followed by the same letter are not different (P < 0.05) according to Duncan's multiple range test. ^

PAGE 23

11 between root colonization and spore density, while Hayman and Stovnld -\ (1979) and Giovannetti and Nicolson (1983) found no correlation. This apparent discrepancy may be due to different sampling methods. Giovannetti (1985) collected samples within the same plant species and sites, while the other researchers collected samples from many different plant species and sites. Spore production and root colonization are influenced by seasonal variations (Giovannetti, 1985; Sylvia, 1986), host plant, stage of development (Saif and Khan, 1975; Schenck and Kinloch, 1980), and soil type (Lopes et al., 1983). In this survey there was only one sampling, so it was not possible to separate seasonal or host developmental effects on root colonization and total spore density. iThe 6 species of VAM fungi collected in this survey were: Gigaspora heterogama (GH) Gerdemann & Trappe, Gigaspora margarita (GM) Becker & Hall, Glomus etunicatum (ETU) Becker & Gerdemann, Glomus intraradices (INT) Schenck & Smith, Glomus sp. (GS), and Acaulospora spinosa (AS) Walker & Trappe. The unidentified Glomus sp. was dark brown to black, 200-250 pm in diam, and had 1 wall of 8-14 ym. thickness. : The occurrence of fungal species, as determined by spore nximbers, was affected by the legume host and location (Table 2-4). labal et al. and Schenck and Kinloch (1980) also recorded differences in spore numbers among plant species. The maximum number of spores of G. margarita occurred at Fort Pierce associated with aeschynomene. Spores of G. margarita were not found associated with Siratro at any of the four locations. Spores of G. heterogama G. etunicatum and G. intraradices

PAGE 24

12 Table 2-4. Mean spore numbers of VAM fungal species associated with four forage legumes at four locations in south Florida. Species of VAM fungi— Location GM GH ETU INT AS GS Aes chynomene americana Ft. Pierce 14ia 66^ 21^ 42^ 0 16 Deseret gb 136^ 0 0 0 0 Basinger 0 0 0 0 0 Ona 0 16^ 29a 114^ 0 0 Desmodium heterocarpon Ft. Pierce • 114^ 255b 307^ 0 0 Deseret 2^ gb 0 0 0 0 Basinger 0 0 igc 0 0 0 Ona 0 0 325^ 5lb 0 0 Macroptilium atropurpureum Ft. Pierce 0 10^ 0 4b 0 8 Basinger 0 5b 0 0 0 0 Ona 0 0 40 254^ 0 0 Vigna adenantha Ft. Pierce lb 252^ 241^ 0 0 Deseret 20^ 15b 0 0 0 0 Basinger 0 0 25b 0 0 0 Ona 28^ 0 28b lib 9 0 Means within a column for each legume species followed by the same letter are not different (P < 0.05) accordingly to Duncan's multiple range test.

PAGE 25

13 were found associated with all legumes, in at least one of the locations. Glomus heteroRama occurred in greatest numbers at Deseret and Fort Pierce associated with aeschynomene and carpon desmodium, respectively. The maximum number of spores of G. etunicatum occurred at Ona and Fort Pierce associated with carpon desmodium. A high number of spores of G. etunicatum was also found at Fort Pierce associated with Vigna adenantha Glomus intraradices was recovered in greater numbers from carpon desmodium and Vigna adenantha at Fort Pierce as well as from Siratro at Ona The unidentified Glomus sp. occurred in lower numbers than the other two species of Glomus; it was only found at Fort Pierce, associated with aeschynomene and Siratro. Acaulospora spinosa was only recovered from Vigna adenantha at Ona. Overall root colonization by VAM fungi was low (most values below 20%) which indicates that (1) the native population of VAM fungi is not very infective and (2) field inoculation may be effective. Attempts to establish pot cultures of VAM fungi recovered in this survey were only successful with G. etunicatum and G. intraradices These two fungi were shown to be effective in increasing the growth of several forage legumes and were chosen for further evaluations.

PAGE 26

CHAPTER III GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH GLOMUS INTRARADICES .-• .. Introduction ^ The use of forage legumes as companion crops to increase production of grasses is becoming an established practice in order to reduce the requirement for N fertilization (Rotar, 1983). In soils where P is a limiting factor, large applications of P fertilizer are required for legume establishment and optimum growth. However, with the increasing cost of P fertilizer, alternative strategies for minimum input and efficient use of P must be adopted. It is pertinent, therefore, to evaluate whether mycorrhizal associations with forage legumes can be manipulated in order to improve establishment, P nutrition, N2-fixation, and consequently yield. Vesicular-arbuscular mycorrhizal fungi can improve the growth of legumes by increasing P uptake (Bethlenf alvay et al., 1985; Chulan and Ragin, 1986; Harley and Smith, 1983; Hayman, 1983; Jensen, 1984). Phosphorus is often a growthlimiting factor since many legumes have high F requirements and are poor scavengers of P. The VAM fungi may also increase nodulation and N2-fixation of legumes, primarily as an indirect effect of improved P nutrition (Daft and El-Giahmi, 1976; Habte and Aziz, 1985; Newbould and Rangeley, 1984). 14

PAGE 27

15 Most of the literature concerning the association of VAM fungi with forage legumes is on temperate species; e.g. alfalfa (Medicago sativa L.) (Kucey and Diab, 198A; Nielsen and Jensen, 1983; Satterlee et al., 1983), white clover (Trifolium repens L.) (Newbould and Rangeley, 1984; Powell, 1979; Rangeley et al., 1982), and subterranean clover (Trifolium subterraneiMi L.) (Abbott and Robson, 1978). Studies on tropical forage legumes have been limited to a few species such as tropical kudzu ( Pueraria phaseoloides Benth) (Salinas et al., 1985; Waidyanatha et al., 1979), leucaena (Leucaena leucocephala Dewit) (Huang et al., 1985) and Stylo ( Stylosanthes guianensis SW.) (Mosse, 1977; Waidyanatha et al., 1979). ......... ., The purpose of this investigation was to evaluate the effect of a VAM fungus, G. intraradices on the growth of several tropical forage legumes in pasteurized and nonpasteurized soil under greenhouse conditions. • Materials and Methods The top 15 cm of a virgin Oldsmar fine sand (sandy, siliceous, hyperthermic Alfic Haplaquods) was collected from a newly cleared area at the Agricultural Research and Educational Center, Fort Pierce, FL. The low-P soil was air-dried and sieved through a A-mm screen. The soil had an initial pH of 4.5 (1:2 soil:water suspension) and P, Ca, Mg, and K concentrations (extracted with 0.05 M HCl +0.0125 M H2SO4) of 1, 63, 21 and 12 mg kg ^ soil, respectively. Lime, as high calcitic limestone, was thoroughly incorporated at a rate of 1500 mg kg"-^ soil (equivalent

PAGE 28

16 to 3,000 kg ha"^ assuming a 15-cm depth of soil ha"^ with a bulk density of 1.3 g cm"-^) and allowed to equilibrate for 30 days before initiating the study. Solutions of P, K, Mg, Cu, Mn, Zn, B, and Mo also were thoroughly mixed with tliesoil to supply rates of 10, 30, 12, 1,5, I'^'O;'"' 1.0, 0.50 and 0.10 mg kg \ respectively. A portion of the soil was pasteurized at 60C for A h to eliminate the indigenous mycorrhizal fungi. Then 3 kg of soil was placed into 15-cm-diam plastic pots. The pH of the soil was 6.2 at the end of the experiment. The legumes used in the experiment were: Siratro, aeschynomene, Aeschynomene villosa Poir., Stylo, leucaena, Stylosanthes hamata Taub., cv. 'Verano', Vigna adenantha, and Arachis sp. Seeds were scarified with sandpaper, wetted, and sprinkled with type EL "cowpea" inoculum (Nitragin Co., Milwaukee, WI) prior to planting. Five seeds of the corresponding legumes were planted per pot, and plants were thinned to one per pot 10 d after emergence. Glomus intraradices (isolate S311), used in this study, was isolated from cultivated Vigna adenantha at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-A). Fungal inoculum was produced in pot culture in pasteurized soil containing carpon desmodium as the host plant. Pot cultures were 10-weeks old when they were harvested, mixed and used to inoculate experimental pots. Ten grams per pot of the soil-rootfungus inoculum containing approximately 200 spores was spread in a 1-cm-thick layer, at a depth of 3 to 5 cm below the soil surface. Noninoculated control treatments received 10 g of a soil-root mixture from noninoculated pot cultures that were free of VAM fungi.

PAGE 29

17 The experimental treatments consisted of pasteurized or nonpasteurized soil, with or without addition of G. intraradices inoculum. The pots were arranged on greenhouse benches in a completely randomized design with three replications per treatment. The average maximum and minimum greenhouse temperatures were 37 and 26C, respectively. Pots were watered as needed to maintain soil moisture near field capacity and were re-randomized every two weeks. Plants were harvested after 45 d. Shoot and roots were dried at 70C for A8 h and weighed. The percentage of mycorrhizal root colonization was determined as described in Chapter II. Significant treatment effects on shoot and root dry weights within legume species were analyzed by the T TEST procedure of the Statistical Analysis Systems (SAS Institute, 1982). Results and Disc ussion ~ Inoculation with G. intraradices in nonpasteurized soil resulted in greater shoot dry weights (P < 0.05) for five of the seven legumes tested (Fig. 3-1). Greater shoot dry weights of these legumes were positively related to increased levels of mycorrhizal colonization following inoculation (Table 3-1). Root dry weight results were similar to those for shoot dry weight, except for Stylo, where there was no increase in root dry weight as a result of mycorrhizal inoculation (Fig. 3-1).

PAGE 30

18 INOC NON-INOC i INOC NON-INOC aa ^ RR RV PS MR LL SH SG VR LEGUME SPECIES Fig. 3-1. Effect of inoculation with Glomus intraradices on the shoot and root dry weights of tropical forage legumes in nonpasteurized soil in the greenhouse after 45 days. Legume species were Aeschynomene americana (AA), Aeschynomene villosa (AV), Arachis sp. (AS), Macroptilium atropurpureum (MA), Leucaena leucoephala (LL), Stvlos^es "^"'^t^ ^SH), Stylosanthes guianensis (SG), and Vigna adenantha (VA) Bars represent the mean of 3 ri^feates Means with the same letter within a species are not different (P < 0.05).

PAGE 31

19 Percentage of raycorrhizal root colonization of tropical forage legumes in nonpasteurized (UP) or pasteurized (P) soil in the greenhouse after 45 days. Legume species Mycorrhizal Root colonization^ inoculation UP P Aeschynomene americana + 22 J 5 0 Macropt ilium atropurpureum X i 35 20 0 Aeschynomene villosa + 10 6 Stylosanthes hamata + 12 •J* 6 Stylosanthes guianensis + 28 -k 8 Vigna adenantha + 59 53 40 0 Leucaena leucocephala + 4 3 1 0 Arachis sp. + 31 12 35 2 ^Based on a composite of three samples for each legume Treatment lost to glasshouse accident.

PAGE 32

20 There was no increase in plant growth as a result of inoculation with G. intraradices in nonpasteurized soil for Arachis sp., leucaena, and Vig"a adenantha With the exception of leucaena, this may be attributed to effective colonization by the indigenous mycorrhizal fungi in the noninoculated soil (Table 3-1). Only five legume species (Fig. 3-2) were evaluated in pasteurized soil; three were lost in a greenhouse accident. Siratro, Arachis sp., and Vigna adenantha had greater shoot and root dry weights after G. intraradices inoculation. This increase again was related to effective root colonization by G. intraradices (Table 3-1). In another study, Siratro was shown to respond to inoculation with several VAM fungi in pasteurized soil in the greenhouse (Lopes and De Olivera, 1980). Inoculation with G. intraradices did not increase either the shoot or root dry weights of aeschynomene or leucaena in pasteurized soil. However, mycorrhizal colonization on both legumes was very low (3%). Leucaena has been reported to be very mycorrrhizal dependent because it has few root hairs (Huang et al., 1985; Yost and Fox, 1979). The failure of VAM fungi to colonize it in this study in both pasteurized and nonpasteurized soil may be due to incompatibility between the plant and G. intraradices as well as native VAM fungi in the experimental soil, to inhibitory soil factors on the host-VAM symbiosis, or to the relatively slow development of the root system. It has been shown that some mycorrhizal fungi may be less effective on certain plant hosts. For example, Schroder et al. (1977) reported that Glomus macrocarpum Tul & Tul increased growth of onions but not of Stylosanthes sp.

PAGE 33

21 m RV RS MR LL SH SG VR LEGUME SPECIES Fig. 3-2. Effect of inoculation with Glomus intraradices on the shoot and root dry weights of tropical forage legumes in pasteurized soil in the greenhouse after 45 days. The legume species were Aeschynomene americana (AA), Arachis sp. (AS), Macroptilium atropurpureum (MA), Leucaena leucocephala (LL), and Vigna adenantha (VA). Bars represent the mean of 3 replications. Means with the same letter within a species are not different (P < 0.05).

PAGE 34

22 The fail are to obtain good colonization of aeschynomene in pasteurized soil was unexpected, since this legume was successfully colonized in nonpasteurized soil. Unfortunately, soil chemical properties were not determined after pasteurization. Legumes are sensitive (and hence VAM fungi) to elevated Mn, which is a common occurrence in heat treated soils. The results obtained in this study clearly demonstrate that G. intraradices can successfully compete with some of the indigenous mycorrhizal fungi present in the experimental soil and promote growth of several legumes in nonpasteurized soil. This result agrees with earlier work by Abbott and Robson (1981), Mosse (1977), and Rangeley et al. (1982), which suggests a potential for successful field-scale inoculatioi with effective VAM fungi.

PAGE 35

CHAPTER IV GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I. SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI UNDER AMENDED SOIL CONDITIONS. Introduction Siratro, a cultivar developed by E. M. Hutton (1962) from two Mexican accessions of Macroptilium atropurpureum Urb., is a persistent, perennial forage legume adaptable to a wide range of soil and climatic conditions. It has become widespread and is among the most versatile forage legume grown throughout tropical regions of the world (Lynd et al., 1985). In pasteurized and nonpasteurized soils, increased growth of Siratro was attained after inoculation with Glomus f asciculatxim Gerdemann & Trappe (Lopes and De Olivera, 1980; Lynd et al., 1985) and G. intraradices (Chapter III). Hayman (1982) stated that VAM fungi are probably capable of symbiosis with most plants, at least to some degree. However, there is wide variation in the ability of VAM fungi to stimulate plant growth (Miller et al., 1985; Powell, 1982; Schubert and Hayman, 1986). Lopes and De Olivera (1980), using a gammairradiated soil of low P-content, studied the effect of inoculation with nine species of VAM fungi on the growth of Siratro. Only inoculation with G. fasciculatum and G. macrocarpum enhanced plant growth. Abbott and Robson (1981) defined the relative ability of a VAM fungus to stimulate plant growth as 'effectiveness' and 23

PAGE 36

24 this defined term will be used in this paper. Wilson (1984) indicated that an evaluation of the effectiveness of the indigenous mycorrhizal population under amended soil conditions, as well as studies to select effective VAM fungi, are prerequisites for successful field inoculation. Thus, the objective of the present study was to determine the effectiveness of several VAM fungi with Siratro in a limed, nonpasteurized soil with low P content under greenhouse conditions. Materials and Methods The soil used in this study, liming, and fertilizer amendments are described previously in Chapter III. Plants were inoculated with the following VAM fungi: G. etunicatum (isolate S312) obtained from carpon desmodium at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-4); G. deserticola Trappe, Bloss & Menge (isolate S305) obtained from sea oats (Uniola paniculata L.) in a coastal dune, Anastasia, FL. ; G. versiforme Berch & Fortin (isolate #231) obtained from N.C. Schenck, University of Florida. Gainesville, FL.; G. intraradices (isolate S311) obtained from Vi£na adenantha at the Agricultural Research and Education Center, Ona. FL. (Chapter II. Table 2-4); G. margarita (isolate #215) obtained from N.C. Schenck. University of Florida. Gainesville. FL. Isolates were maintained in pot cultures in pasteurized soil containing bahiagrass. Soils from 12-week-old pot cultures were used to inoculate experimental pots. The propagule densities of the native soil and inocula at the beginning of the experiment were determined by the most-probable-number

PAGE 37

25 (MPN) technique using bahiagrass as the host plant and pasteurized Oldsmar fine sand as the diluent (Daniels and Skipper, 1982). The amount of inoculum used was adjusted to give equal inoculiim densities among isolates. Each pot received approximately 240 propagules. Details on the fungal inoculation technique, planting, and watering were reported previously (Chapter III). There were six inoculation treatments, five species of VAM fungi, and a control inoculated with non-VAM pot culture material. The pots were arranged on the greenhouse bench in a completely randomized block design with 15 replications per treatment. The average maximum and minimum greenhouse temperatures during the experimental period were 32 and 19C, respectively. Maximum photosynthetic photon flux density was 1200 )j mol m'^ s"'''. : Five randomly selected samples were harvested from each treatment after 20, AO, and 70 d of growth. At first harvest, shoot dry weight, percentage of root colonized by VAM fungi, plant height, and number of leaves were determined. In addition, root fresh weight and total root length colonized were determined at the second and third harvests. Shoot dry weight was determined by drying the material at 70C for 2A h. Percentage and total root length colonized were estimated by the gridline intersect method (Giovannetti and Mosse, 1980) after roots were cleared in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik and McGraw, 1982). Data were analyzed by Analysis of Variance Procedure, Statistical Analysis Systems (SAS Institute Inc., 1982). Duncan's multiple range test was used to separate treatment means when the F-test was significant (P < 0.05).

PAGE 38

26 Results and Discussion • Plants inoculated with G. etunicatum and G. intraradices had higher shoot dry and root fresh weights than plants inoculated with the other VAM fungi or control plants, at 40 and 70 d after planting (Fig. 4-1). At both harvests, plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices At the final harvest, plants inoculated with G. intraradices had higher root fresh weights than plants inoculated with G. etunicatum In contrast, plants inoculated with G. versiforme, G. margarita and G. deserticola had shoot dry and root fresh weights that were not different from the noninoculated plants, except at the final harvest when plants inoculated with G. deserticola had higher shoot dry and root fresh weights than the control (Fig. 4-1). At 20 d, shoot dry weights were not different among treatments (mean = 0.70 g). Percentage and total root length colonized by VAM fungi increased, with time (Fig. 4-2). Inoculation with G. etunicatum and G. intraradices resulted in the highest root colonization at all harvests. At 70 d, plants inoculated with G. etunicatum had the highest root colonization, followed by G. intraradices and then G. deserticola There were no differences in root colonization among G. versiforme G. margarita and control treatments. For the six treatments, total root length colonized by VAM fungi and percentage of raycorrhizal root colonization followed the same trend (Fig. 4-2). Shoot dry weight of Siratro was correlated with total root length colonized by VAM fungi (r^ = 0.95"") and percentage of mycorrhizal root

PAGE 39

27 8 GON MAR VER DES INT ETU VAM FUNGI Effect of inoculation with Gigaspora margarita (MAR), Glomus versiforme (VER), Glomus deserticola (DES), Glomus intraradices (INT), Glomus etunicatum (ETU), or the control (CON) on the shoot dry weight and root fresh weight of Siratro at two harvests. Bars represent the means of five replicates. Means with the same letter within a harvest are not different (P < 0.05),

PAGE 40

28 50 CON MAR VER DES 'NT ETU VAM FUNGI Fig. 4-2 Effect of inoculation with Gigaspora margarita (MAR), Glomus versiforme (VER), Glomus deserticola (DES), Glomus intraradices (INT), Glomus etunicatiim (ETU), or the control (con) on the percentage of root colonization and root length colonized of Siratro at three and two harvests, respectively. Bars represent the means of five replicates. Means with the same letter within a harvest are not different (P < 0,05).

PAGE 41

29 colonization (r^ = 0.83''"). There was a quadratic relationship between shoot dry weight and length of Siratro root colonized by VAM fungi for all inoculated treatments (Fig. 4-3). Plant height and number of leaves per plant were not different among treatments at any harvest. After 70 d, the mean plant height and number of leaves for all treatments were 90 and 15 cm, respectively. There were 2 propagules per gram of soil in the native soil as determined by the MPN test at the beginning of the experiment. Inoculum density is known to influence plant growth response to VAM fungal inoculation (Hass.and Krikum, 1985 ;. Wilson, 1984). Thus, one of the problems in comparing the efficacy of VAM fungi is ensuring uniform inoculum densities (Daniels et al., 1981). In this study, I used the MPN technique to provide a measure of the inoculum densities of the VAM fungi, and I adjusted the inoculum densities so that they were uniform for all inoculated treatments. There were striking differences in the effectiveness of VAM fungi on Siratro. The results are consistent with the findings of others (Miller et al., 1985; Schubert and Hayman, 1986) indicating that different species and strains of VAM fungi vary considerably in the benefits they confer to the host plant. This experiment also confirmed previous work (Chapter III) which demonstrated that the native population of VAM fungi in this soil was less able to stimulate the growth of Siratro than effective, introduced species. It is possible that the decreasing soil acidity obtained, by liming changed the native population of VAM fungi from effective to ineffective as compared to G. etunicatum and G. intraradices (Hayman and Tavares, 1985). Since it is necessary to lime

PAGE 42

30 Fig. A-3 Relationship between shoot dry weight and length of Siratro roots colonized by VAM fungi for all inoculated treatments.

PAGE 43

31 this soil for satisfactory establishment and growth of legumes (Snyder at al., 1985), VAM fungi must be selected for their effectiveness under amended soil conditions. Powell (1980a) reported a relationship between the level of native inoculum density in the soil and plant growth response to mycorrhizai inoculation. When inoculum density was low (0.01-0.09 propagules g'^ soil), there was a significant plant growth response to inoculation with VAM fungi. When inoculum density was higher (0.15-0.30 propagules g"^ soil), there was little plant growth response to fungal inoculation. Likewise, good plant growth responses to inoculation with VAM fungi in soils with few indigenous endophytes have been reported by Mosse (1977) and Hall (1979). Thus it would seem that the most promising sites for inoculation with VAM fungi are those where indigenous populations of VAM fungi are very low. However, in this study, where the native inoculum density was relatively high (2 propagules g"^ soil). Siratro responded to inoculation with two of the four VAM fungi tested. In addition to the abundance of the indigenous VAM fungi, information about their infectivity and effectiveness is needed to assess potential sites for responsiveness to inoculation with effective VAM fungi. The ineffectiveness of G. versiforme and G. margarita could be due to an innate symbiotic inefficiency, incompatibility, lack of competitiveness, or to inhibitory edaphic (e.g. soil pH or P level) or environmental factors (e.g. light and temperature). Hayman and Tavares (1985) showed clearly that different endophytes vary in their symbiotic effectiveness at different soil acidities. In addition, some endophytes may be less effective on certain plant hosts. For example.

PAGE 44

32 Schroder et al. (1977) reported that G. macrocarpijjn increased growth of onions but decreased growth of Stylosanthes sp. The most effective fungi in this study were those that colonized the root most rapidly. Sanders et al. (1977) and Abbott and Robson (1978) reported that VAM fungi differ in their rates of root colonization. Abbott and Robson (1981) stated that differences in the effectiveness of VAM fungi could be due to differences in their ability to (1) colonize the roots rapidly ( inf ectivity) (2) produce external hyphae, and (3) to take up and transport P (efficiency). In this study, we measured only root colonization over time ( inf ectivity) and found, under the conditions of this experiment, G. etunicatum was the most infective fungus. Mycorrhizal root colonization, expressed as either percentage or total root length colonized, was positively correlated with the shoot dry weight of Siratro. Abbott and Robson (1981) and Plenchette et al. (1982) also showed positive correlations between the magnitude of mycorrhizal root colonization and shoot dry weights of plants grown on P-deficient ^ soils. These results suggest that differences in endophyte effectiveness may be evaluated on the basis of rates of root colonization. However, Hayman and Tavares (1985) demonstrated that final root colonization by VAM fungi may give little indication of the ability of an endophyte to stimulate plant growth. Abbott and Robson (1978) reported that VAM fungi which differ in effectiveness and rate of root colonization may have similar plateau levels of colonization at a late harvest. Therefore it is not surprising, when colonization is assessed at a relatively advanced stage of plant growth, that there is often little correlation between mycorrhizal root colonization and effectiveness.

PAGE 45

33 Forage legumes are an important component of improved pastures and must be established rapidly and without excessive cost. When P is a major factor limiting the productivity of legumes, large applications of P fertilizer are normally required for legume establishment. However, with the increasing cost of P fertilizer, alternative strategies for minimum fertilizer input and efficient use must be adopted. These data demonstrate that by careful selection of effective VAM fungi, the growth of Siratro can be enhanced in a P-deficient native soil containing a less effective native VAM population than the introduced VAM fungi.

PAGE 46

CHAPTER V GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II. EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT DIFFERENT P LEVELS. Introduction Forage legiames are an important component of improved grass pastures. The legumes serve both to increase forage quality and decrease the need for fertilizer N through N2 fixation. Snyder et al. (1985) studied the responsiveness of several tropical legvimes, including Siratro, to P and lime in a typical Florida Spodosol. They reported that lime and P rates of approximately 3000 and 75 kg ha ^ produced maximum yield and maximum economic return. In an experiment with red clover in a field containing 10 mg NaHC03soluble P kg ^ soil, plants showed an early response to superphosphate, but by the end of the second year yields were high in all plots, equivalent to around 15 t ha"-^ dry matter (Hayman et al., 1981). This result was attributed to one of the introduced endophytes, G. caledonium, which had spread and sporulated profusely throughout all the plots (including those inoculated with two other endophytes) and had previously enhanced growth of lucerne at this site. In upland pastures in Wales, Hayman and Mosse (1979) found that inoculation of white clover seedlings with a combination of G. Mosseae and G. fasciculatum "E3" in field plots given the standard dressing of 90 kg P ha"^ as basic slag doubled plant 34

PAGE 47

35 growth and greatly enhanced tissue P content and nodulation. Growth responses at other sites varied from large to slightly negative, probably governed in part by the effectiveness of the indigenous VAM population (Hayman and Hampson, 1979), Species (Miller et al., 1985; Schubert and Hayman, 1986; Thompson et al., 1986), and isolates within a species (Cooper, 1978), of VAM fungi can colonize plants at different rates. If the mycorrhizal growth response is related to the amount of early root colonization (Abbott and Robson, 1981; Chapter IV), then isolates of VAM fungi that colonize roots rapidly, at P levels found in established agricultural soils, may be most suitable for pasture inoculation. Schubert and Hayman (1986) indicated that, in order to achieve a rational and effective use of inoculants, precise information on the performance of endophytes in soil amended with P was necessary. It is evident that the effect of soil P on symbiosis varies with the specific host and endophyte. Therefore, more research is needed to develop uniform and predictable endophyte-host responses. In another study described in Chapter IV, G. etunicatum and G. intraradices were found to be the most effective growth enhancers (out of 5 isolates) of Siratro in a soil amended with a moderate level of P and lime. In the present study, the objective was to evaluate the infectivity and effectiveness of these two fungi over a practical range of applied P.

PAGE 48

36 Materials and Methods The chemical properties of the soil used, liming, and fertilizer amendments are described previously (Chapter III). Siratro was used as the host plant in this experiment. The VAM fungi tested were G. etunicatum (isolate S312) and G. intraradices (isolate S313). These isolates were maintained in pasteurized soil in pots with bahiagrass as the host. Soils from 10-week-old pot cultures were used to inoculate experimental pots. The propagule densities of the inocula were determined by the MPN technique (Daniels and Skipper, .1982) and approximately 240 propagules were added to each 15-cm-diam plastic pot. Details on the origin of the two fungal isolates were reported previously (Chapter II, Table 2-A). Fungal inoculation technique, planting, and watering were described in Chapter III. The experiment was designed as a 3 x 4 factorial consisting of 3 inoculation treatments; G. intraradices G. etunicatum and noninoculated control, and four P treatments; 2.5, 10, 20, and 40 mg P kg ^ as Ca(H2P04)2.H20 (equivalent to 5, 20, 40, and 80 kg P ha'^ 1 T "I assuming a 15-cm depth of soil ha with a bulk density of 1.3 g cm Phosphorus was applied in solution one week before planting. Soil samples from each treatment were analyzed for extractable P at the beginning and end of the experiment using the Mehlich-I method (0.05 M HCl + 0.0125 M H2SO4). The twelve treatments were replicated five times and arranged on a greenhouse bench in a randomized complete block design.

PAGE 49

37 The average maximxim and minimum greenhouse temperatures during the experimental period were 34 and 26C, respectively. Maximum -2 -1 photosynthetic photon flux density was 1800 }i mol m s Shoot dry and root fresh weights, and percentage and total root lengths colonized by VAM fungi, were determined after 60 d using procedures described previously (Chapter IV). Data for all variables were subjected to ANOVA procedures and regression analysis (SAS Institute Inc., 1982). Results and Discussion Phosphorus applications of 2.5, 10, 20, and 40 mg kg"''' resulted in Mehlich-I extractable P in the soil of 5.6, 12.8, 21.6, and 38.0 mg kg"^ at the beginning of the experiment, respectively. These differences were still reflected at the end of the experiment when P concentrations were 4.4, 7.8, 15.2, and 24.6 mg kg'^.. Shoot dry and root fresh weights of Siratro were increased by P fertilization and fungal inoculation (Fig. 5-1). At 2.5 mg P kg"^, there was no difference in the shoot dry and root fresh weights among inoculated and control plants. At all other levels of P, inoculated plants had higher shoot dry and root fresh weights than control plants. Shoot dry weights were greatest for plants inoculated with G. etunicatum There were no differences in root fresh weights between plants inoculated with G. etunicatum or G. intraradices Inoculated plants had a quadratic relationship for shoot dry and root fresh weights and P application

PAGE 50

38 ETLI "JU r CON J/ ^ 10 20 30 40 P APPLIED (mg/kg) E Q UJ M O _1 o o X t— o UJ o o 1 2 3 s A /a A i A /I e 0 CCN 9 10 20 30 40 P APPLIED (mg/kg) Fig. 5-1. Effect of P application on shoot dry weight, root fresh weight, percentage of root colonized by VAM fungi, and total root length colonized of Siratro grown in limed nonpasteurized soil and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON).

PAGE 51

39 (Table 5-1). Maximum yield of inoculated plants was achieved between 28 and 30 rag kg ^ of applied P. Control plants had a linear relationship for shoot dry weights and a quadratic relationship for root fresh weights. There were fungus x P interactions for shoot dry and root fresh weights (Table 5-2). Percentage and total root length colonized by VAM fungi for the inoculated treatments increased with P additions (Fig. 5-1). This effect was greater for G. etunicatum than for G. intraradices Phosphate application did not alter the percentage and total root length colonized in the control plants. Inoculated plants had a quadratic relationship for percentage and total root length colonized and applied P (Table 5-1). Maximum colonization of inoculated treatments, expressed as either percentage or total length of colonized root, was attained between 32 and 35 mg kg"^ of applied P. There were fungus x P interactions for percentage and total root length colonized by VAM fungi (Table 5-2). Shoot dry weight of Siratro over the range of applied P was highly correlated with percentage (r^ = 0.95"*) and total root length colonized by VAM fungi (r^ = 0.97"'^) (Fig. 5-2) for both inoculated treatments. Percentage of the root colonized by VAM fungi was very closely correlated (r^ = 0.98^*) with the total root length colonized. Growth enhancement from VAM inoculation at different levels of P has been reported to vary with VAM fungi (Hayman and Hampson, 1979; Hayman and Mosse, 1979; Schubert and Hayman, 1986; Thompson et al., 1986). For example, Schubert and Hayman (1986) indicated that, when large amounts of P were added (more than 100 mg kg"^), G. mosseae. G. versiforme. G.

PAGE 52

40 Table 5-1. Regression equations and coefficients of determination (r^) showing the relationship of P level to shoot dry weights, • root fresh weights, percentage and total root length colonized. Variable Regression equations Shoot dry weight (g) Root fresh weight (g) Root colonization (%) Root length colonized (m) Glomus etunicatum = -0.12+0.34P-0.005p2 = 0.50+0.40P-0.006p2 = 4.47+2.62P-0.0Ap2 = -1.43+0.6AP-0.009p2 0.97' 0.96'' 0.97' 0.96' Glomus intraradices Shoot dry weight (g) Root fresh weight (g) Root colonization (%) Root length colonized (m) = 0,16+0.2AP-0.004P2 = 0.55+0.38P-0.006p2 = 5.42+1.86P-0.03p2 = -1.08+0.49P-0.006p2 0.95',"' 0 94"' 0.95'"'' 0.91"' Control Shoot dry weight (g) Root fresh weight (g) = 0.72+0.06P = 0.89+0. 17P-0.002p2 0.93; 0.92' P = phosphorus level **signif icant at P < 0.01

PAGE 53

41 Table 5-2. Analysis of variance for shoot dry weights, root fresh weights, ana percentage and total root length colonized. Source of variation DF Shoot MS root MS root colonized MS percentage length Block 4 Fungi (F) 2 P rates (P) 3 linear (PI) 1 quadratic (Pq) 1 Cubic (Pc) 1 F X P 6 F X PI 2 F X Pq 2 F X Pc 2 Error 44 0.07"" 11.07 34.19 85.40 13.97 3.20 1.65 2.58^'* 2.29''* 0.08 0.02 0.04 18.01 59.37 146.43 30.10 1.57 2.557' 4.02"' 2.08"' 1.55 0.13 7.39 2650.42 1583.51 4156.00 577.33 17.14, 378.46' 965.54' 159.17"'' 10.67 6.27 0.069 114.13 118.66 305.28 32.23 2.48 22.50"' 55 69""' .82"' .99 0.24 '"'significant at P < 0.01 MS = mean square

PAGE 54

42 — 003^0 Y=0.83+0.44X r=0 95 1 O 3 6 9 12 ROOT LENGTH COLONIZED Cm) Relationship between shoot dry weight and length of Siratro roots colonized by VAM fungi for all inoculated treatments in nonpasteurized soil.

PAGE 55

43 macrocarpum and G. margarita were ineffective in stimulating growth of onion; however, G. caledonium and Glomus sp. 'E3' were generally effective at all P levels. In our study, G. etunicatum was more effective than G. intraradices at all but the lowest applied P level. Hence, there appears to be good potential for the selection of VAM fungi to enhance plant growth under amended soil conditions such as P fertilization and liming. Phosphorus has been reported to increase, decrease or not affect root colonization by VAM fungi. However, it is difficult to compare results concerning the effect of P fertilization on mycorrhizal root colonization, because of differences in the range of added P, as well as other factors such as host plant and soil type. In this study, we used a range of 2.5 to 40 mg P kg ^ because it represented P levels used in the production of tropical forage legumes in Florida on a similar soil (Snyder et al., 1985). At the lowest P level, G. etunicatum and G. intraradices did not colonize the root or improve growth of Siratro above that of the control plants. Barber and Lougham (1967) reported that, at a very low P level, competition for P occurs between plants and microflora. Habte and Manjunath (1987) and Same et al. (1983) indicated that the growth of VAM fungi is limited by P at very low levels. Between 10 to 40 mg P kg ^, percentage and total root length colonized by VAM fungi increased with P additions. These results agree with those of Abbott and Robson (1977b), Schubert and Hayman (1986), and Thompson et al. (1986), who reported an increase in the percentage and total root length colonized between 18 to 55 mg P kg'^. At high P levels (more than 100 mg kg'^, which are not feasible for field production of

PAGE 56

44 forage legumes, percentage and total root length colonized by VAM fungi may be suppressed (Abbott and Robson 1977b; Schubert and Hayman, 1986; Thompson et al., 1986). The result that root colonization by VAM fungi, expressed as either percentage or total root length colonized, is positively correlated to shoot dry weight is consistent with the findings of Abbott and Robson (1981) and a previous study where its implications are discussed (Chapter IV). I conclude that, in amended soils where the indigenous population of VAM fungi is less effective than some of the introduced species of VAM fungi, inoculation with effective VAM fungi can increase the plant growth. Furthermore, with highly mycorrhizal dependent crops such as tropical legumes, growth enhancement may occur at P levels actually used in commercial pasture production.

PAGE 57

CHAPTER VI EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE GROWTH AND UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM STYLOSANTHES GUIANENSIS AND AESCHYNOMENE AMERICANA Introduction Mycorrhizal colonization is important for legumes because it increases their P uptake (Abbott and Robson, 1977b; Saif, 1987), and therefore nodulation and N2-fixation (Asimi et al., 1980; Bergersen, 1971; Gates and Wilson, 1974; Gibson, 1976). Crush (1974) found that VAM fungi increased the growth and nodulation of Centrosoma pubescens Benth, Stylo, and Trifoliiun repens L. Mos se et al. (1976) showed that effective nodulation of Centrosema Stylosanthes and Trifolixim plants in a P-deficient Brazilian Cerrado soil could be achieved only by introducing both VAM fungi and P. Mycorrhizal fungi also have been shown to increase nodulation, fixation, plant growth, plant N and P content in Vigna unguiculata (Islam et al., 1980; Sanni, 1976), Medicago sativa (Barea et al., 1980), Pueraria phaseoloides and Stylo (Waidyanatha et al., 1979), Stylosanthes scabra (Purcino and Lynd, 1985), leucaena (Munns and Mosse, 1980; Purcino et al,, 1986), and Siratro (Lynd et al., 1985). However, effective tripartite symbiosis ( legume-rhizobium-VAM fungus) is influenced by soil and climatic conditions (Waidyanatha et al., 1979), 45

PAGE 58

46 which are apparently species-specific (Burt and Miller, 1975; Mosse, 1972). In a previous study (Chapter V), G. etunicatum was an effective growth enhancer of Siratro in a soil similar to the one used in the present study, with a moderate level of applied P (20-40 mg P kg'^). The objective of this investigation was to determine the effect of inoculation with a VAM fungus, G. etunicatum on the growth and plant uptake of P and N of three forage legumes at different P levels in pasteurized soil under greenhouse conditions. Materials and Methods The soil used in this investigation, liming, and basic fertilization are described previously (Chapter III). The soil chemical characteristics before soil fertility treatments and after pasteurization (70C for A h) were: pH A.4 (soil:H20=l : 2) ; l.A% organic matter; 2, 65, 11, and 12 mg kg ^ (Mehlich-I extractable) of P, Ca, Mg and K, respectively. At planting three phosphorus levels were established by application in solution of 12.5, 25, and 50 mg P kg"-^ as Ca(H2P04)2.H2O which is equivalent to 25, 50, and 100 kg P ha"\ assuming a 15-cm depth of soil ha'^ with a bulk density of 1.3 g cm"^. Glomus etunicatum (isolate S312) was isolated from carpon desmodium at the Agricultural Research and Education Center, Ona, FL. (Chapter II, Table 2-A). Fungal inoculum was produced in pot culture in pasteurized

PAGE 59

47 soil containing bahiagrass. The fungal inoculation technique was reported in Chapter III. The legume species used in this experiment were: Siratro, aeschynomene, and Stylo. Seeds were scarified with sandpaper, wetted, and sprinkled with type El "cowpea" inoculum (Nitragin Co., Milwaukee, WI) prior to planting. Three germinated seeds were planted per 620 ml "Deepots" (J.M. McConkey & Co, Inc, Summer, WA). After emergence, seedlings were thinned to one per pot. The experiment was conducted as a 2 x 3 x 3 factorial consisting of 2 inoculation treatments, etunicatum and noninoculated control; three P levels, 12.5, 25, and 50 mg kg ^ ; the 3 legume species; and 3 replications. The 18 treatments were arranged on a nonshaded greenhouse bench in a completely randomized design. The average maximum and minimum greenhouse temperatures during the experimental period were 28 and 20^, respectively, and the average maximum photosynthetic photon flux density was 1793 u mol m~^ s"-*-. After 65 d plants were harvested, shoots dried (70C for 24 h), weighed, and ground in a Wiley mill using a 20-mesh screen. Shoots were digested by the sealed-chamber procedure of Anderson (1986) and analyzed for P on a Jarrel-Ash 955 inductively-coupled argon plasma spectrometer (ICAP). For nitrogen analysis, samples were digested using a modification of the aluminum block digestion procedure of Gallaher et al. (1975) and ammonia in the digestate was determined by semiautomated colorimetry (Hambleton, 1977). Roots were washed from the soil, air-dried, and weighed. In addition, percentage and total root

PAGE 60

48 length colonized were estimated as described in Chapter IV. A 0-A scale •was used to estimate the number of nodules per plant, with 1 = 20-50, 2 = 50-100, 3 = 100-150, and A = > 150 nodules. Data were subjected to ANOVA procedures and regression analysis (SAS institute Inc., 1982). Results and Discussion At harvest, both fungal inoculation and P applications increased shoot dry weight, plant P concentration, and total plant P and N of the three legiimes (Table 6-1). Root fresh weight was increased for Stylo and Siratro but not for aeschynomene. There were ftingus x P interactions for shoot dry and root fresh weights, and total plant N of Stylo. Siratro had fungus x P interactions for shoot dry weight and total plant P and N, whereas aeschynomene only had fungus x P interaction for shoot dry weight. At all levels of applied P, shoot dry and root fresh weights and total N of Stylo were greater for mycorrhizal plants than nonmycorrhizal plants (Fig. 6-1). Differences between mycorrhizal and nonmycorrhizal plants were most pronounced at intermediate levels of applied P and diminished with further P addition. ,, Phosphorus concentration and total P of Stylo were not affected by fungus X P interactions (Table 6-1), but there was an overall effect of fungal inoculation (Fig. 6-1). Saif (1987), Mosse et al. (1976), and Waidyanantha et al. (1979) also reported an increase in plant growth, P concentration, and total P and N of Stylo in a pasteurized low P soil following inoculation with VAM fungi and P applications.

PAGE 61

49 Table 6-1. Mean squares and levels of significance from the analysis of variance for shoot dry weight, root fresh weight, P -.. -. concentration, and total P and N uptake of forage legumes. Source of DF Shoot Root P Total Total Variation dry wt fresh wt cone P N Stylosanthes guianensis Fungus 1 0.A5 1.13 0.011** 6.93 7 500.97 P rates 2 0.6A 0.21 0.016** 9.13** 790.32 Lineal 1 1.20 0.36 0 032** 17.69** 1511.78 Quadratic 1 0.09 0.05 0.0004 0.57 68.86 Fungus X P 2 0.046** 0.15** 0.00007 0.65 ^ 47.14* Error 12 0.0057 0.026 0.012 0. 18 8.48 Macropt ilium atropurpureum Fungus 1 1.40 0.58** 0.0076* 33.78 2360.30 P rates 2 7.28 1.81** 0.028** 163.11 10469.14 1 14.54 3.55** 0.056** zuy z J iu Quadratic 1 0.0072 0.05 0.0001 0.28. 13.18 Fungus X P 2 0 54** 0.10 0.0014 10.37" 625.52** Error 12 0.029 0.027 0.00082 2.47 43.27 Aeschynomene americana Fungus 1 0.13 0.042. 0.0068** 6.06" 129.34* P rates 2 0.56 1.07** 0.034** 29.75"" 524.16** Lineal 1 1.09 2.01** 0.064** 58.70** 1035.46** Quadratic 1 0.027. 0.12* 0.0022 0.80 12.85 Fungus X P 2 0.037* 0.030 0.00009 0.10 43.35 Error 12 0.0076 0.015 0.00052 0.25 21.75 *Significance at P < 0.05 **Signif icant at P < 0.01

PAGE 62

50 12,5 25.0 07.5 5D.0 12.5 25.0 37.5 BD.O P APPLIED
PAGE 63

51 Mycorrhizal plants of Siratro had greater shoot dry weight, total P and total N than nonmycorrhizal plants at low and intermediate levels of applied P, but not at the highest level (Fig. 6-2). Overall, mycorrhizal plants had greater root fresh weight and plant P concentration than nonmycorrhizal plants. Other investigators (Lynd et al., 1985; Saif, 1987) have shown similar responses of Siratro to inoculation with VAM fungi and P additions. Differences in shoot dry weight between mycorrhizal and nonmycorrhizal plants of aeschynomene were only at the intermediate level of applied P (Fig. 6-3). Fungal inoculation did not affect the root fresh weight. Overall, mycorrhizal plants had greater P concentration, total P, and total N than nonmycorrhizal plants (Fig. 6-3). Percentage and total root length colonized for mycorrhizal plants of Stylo (Fig. 6-1), Siratro (Fig. 6-2), and aeschynomene (Fig. 6-3) increased with the first addition of P. However, maximum colonization, expressed as either percentage or total length of colonized root of the three legumes, was attained at the intermediate levels of applied P. The number of nodules in the three legumes increased with fungal inoculation and P applications. Mycorrhizal plants of Stylo had more nodules than nonmycorrhizal plants at all levels of applied P. However, mycorrhizal plants of Siratro and aeschynomene had more nodules than nonmycorrhizal plants only at 25 mg kg~^ of applied P. Mycorrhizal plants required between 38-AO mg P kg'^ to achieve maximum shoot dry weight, whereas nonmycorrhizal plants required 50 mg P kg ^ to produce approximately the same shoot dry weight, except for nonmycorrhizal Stylo (Fig 6-1) which even with 50 mg P kg"^ did not

PAGE 64

52 P APPLIED Cmg/kg) Fig. 6-2. Effect of fungal inoculation and P applications on the shoot dry weight, root fresh weight, root colonization, P concentration, total P, and total N of Macroptilium atropurpureum

PAGE 65

53 T .aO + O.BB r 0.88' I2J S.0 P. APPLIED (mg/kg) ;J7.5 SD.0 Fig. 6-3. Effect of fungal inoculation and P applications on the shoot dry weight, root fresh weight, root colonization, P concentration, total P, and total N of Aeschynomene americana.

PAGE 66

5A reach the same shoot dry weight as myr-^rhizal plants. Fungal inoculation with G. etunicatum resulted in a 20% decrease in the amount of P required for maximum yield. This demonstrates the importance of VAM fungi in the P nutrition of tropical legumes and would represent an important savings to the farmer. Growth responses of Stylo, Siratro, and aeschynomene associated with inoculation with VAM fungi are closely related to improved P uptake and N2-fixation. These results clearly support the findings of earlier studies that inoculation with VAM fungi not only stimulates plant growth and P uptake of legumes (Abbott and Robson, 1977b; Habte and Manjunath, 1987; Menge, 1983) but also nodulation and N2-fixation (Asimi et al., 1980; Lynd et al., 1985; Purcino et al., 1986) which was measured indirectly in this study by the total plant uptake of N. Inoculation with effective species of VAM fungi and additions of P between 25 and 50 mg kg ^, were shown to improve the growth and uptake of P and N of Stylo, Siratro, and aeschynomene.

PAGE 67

CHAPTER VII GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND AESCHYNOMENE AMERICANA TO INOCULATION WITH SELECTED VA MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS Introduction The practical goal of studies on plant growth responses to inoculation with VAM fungi is to obtain increased yield of plants growing under field conditions. Significant plant growth responses to inoculation with VAM fungi have been demonstrated in pot experiments using pasteurized and nonpasteurized soil for several tropical forage legumes such as: Pueraria phaseoloides (Salinas et al., 1985; Waidyanatha et al., 1979), leucaena (Habte and Manjunath, 1987; Huang et al., 1985), Siratro (Lynd et al., 1985), Stylo (Mosse, 1977) and aeschynomene (Chapter II and VI). It has been pointed out that inoculation experiments with VAM fungi should include testing a series of P levels (Abbott and Robson, 1977b; Hall, 1978; Powell, 1980b) in order to select the optimiim P level for a mycorrhizal response. Except for the work by Saif (1987), little information is available on plant growth response of tropical forage legumes to inoculation with VAM fungi in nonpasteurized soil under field conditions at different P levels. However, there is more data for temperate legumes. In general, the field sites where plants are most likely to respond to inoculation with VAM fungi are those containing 55

PAGE 68

56 little soluble phosphate and a small or ineffective native population of VAM fungi. The experimental site selected for this study contains very little available P (1 mg kg ^) and a reasonably high indigenous population of VAM fungi (2 propagules g ^ soil). The indigenous population, however, was less effective than two of the five introduced VAM fungi under amended soil conditions (Chapter IV). Aeschynomene is used extensively in Florida as a forage legume to supply fixed nitrogen as protein and minerals to grazing animals (Hodges et al., 1982) Siratro has been used sparingly in Florida (Kretschmer, 1972), but is one of the most widely used legumes throughout tropical regions of the world (Lynd et al.,1985). However, for the satisfactory establishment and growth of aeschynomene and Siratro in highly acidic and P-deficient soils, lime and P fertilizer must be applied (Snyder et al., 1985). Recent greenhouse experiments have shown that better growth of forage legumes in these soils can be achieved by inoculation with selected species of VAM fungi (Chapter IV and V). Glomus etunicatum and G. intraradices were found to be effective growth enhancers of Siratro and aeschynomene in a nonpasteurized, limed (3000 kg ha"''') soil similar to the one used in the present study, over a range of 20-80 kg ha"''" of applied P. It is therefore of considerable interest to determine whether inoculation with selected isolates of VAM fungi can improve the establishment, growth, and nutrient uptake of Siratro and aeschynomene, under field conditions where soil was limed and fertilized with different P levels.

PAGE 69

57 Materials and Methods The soil used was a native Oldsmar fine sand (sandy, siliceous, hyperthermic Alfic Arenic Haplaquods) with a pH of 4.6 (soil:H20=l : 2) 1.5% organic matter, and the following Mehlich-I extractabie elements in mg kg"-^: P 1.0, Ca 65, Mg 12, and K 15. The experiment was designed as a 2 x 3 x 4 factorial consisting of two legume species; Siratro and A. americana ; three inoculation treatments, G. etunicatum, G. intraradices and the control; and four P treatments; 10, 30, 60, and 120 kg ha ^ as triple superphosphate. The 24 treatments were arranged in a randomized complete block design with ten replications per treatment. The P treatments were surface applied by hand on 3 July 1986, along with a basal application of lime (high calcitic limestone), Mg, nutritional spray (Diamond Fertilizer Co., Ft. Pierce, FL.), and Mo at 3000, 25, 22, and 0.2 kg ha"\ respectively, and incorporated using a rake to a depth of approximately 15 cm. Potassium was broadcasted on each plot at a rate of 60 kg ha'-"" as KCL on 5 August 1986. Seeds of Siratro and aeschynomene inoculated with rhizobium type El "Cowpea" inoculum (Nitragin Co., Milwaukee, WI) were sown in pasteurized Oldsmar fine sand, amended with high calcitic limestone at 1500 mg kg'-^ and P at 12.5 mg kg \ in cells of "speedling" styrofoam trays (72 cells per tray) on 20 June 1986. • • ;i .. Seedlings were inoculated or not inoculated with G. etunicatum (isolate S312), G. intraradices (isolate S311). The amount of soil-root inoculum used for each VAM fungi was adjusted to give equal inoculum

PAGE 70

58 densities as determined by the MPN technique (Daniels and Skipper, 1982). Approximately 180 propagules were added mid-way down each cell before seeding. Control seedlings received 15 g of a soil-root mixture from nonmycorrhizal pot cultures and the equivalent of 20 kg P ha~^, which was applied in solution 10 d after germination in an attempt to make the P status of the mycorrhizal and nonmycorrhizal seedlings similar at the time of transplanting. Seeded trays were placed in a glasshouse for 6 wk, after which the whole cell content (each with one seedling) was transplanted to the field. Siratro seedlings were cut back to three nodes each before transplanting. Seedlings were transplanted on 4 August 1986. One seedling was planted by hand in the middle of each 1.0 by 1.0 ra plot which were surrounded by alleyways of 1.0 m. Extra seedlings were weighed, and P concentration and root colonization were determined. Harvests of Siratro and aeschynomene were made on 2 October 1986. Siratro, a perennial, also was harvested on 27 November 1986, 5 May 1987, and 29 June 1987. Herbage was dried at 75C for 24 h and weighed. Five subsamples per treatment from the first harvest of Siratro and aeschynomene foliage were analyzed for N and P content by automated colorimetry (Technicon Industrial Systems Method No. 334-74 W/B, Technicon Instruments Corp., Tarrytown, NY). Five root samples per treatment, consisting each of four subsamples, were used to assess mycorrhizal root colonization. Percentage of root colonized by VAM fungi of aeschynomene and Siratro (1st and 4th harvests), was estimated as described in Chapter III.

PAGE 71

59 Soil samples (0-15 cm) were taken at the 4th harvest and analyzed for pH, P, Ca, Mg, and K using the Mehlich-I extractant method. All elements were determined by inductively coupled argon plasma (ICAP) spectrometry. Data for all variables were subjected to ANOVA procedures and regression analysis (SAS Institute Inc., 1982). > Results and Discussion Preinoculated seedlings were used in this study to ensure the legume roots were well colonized with the selected VAM fungi, because this was thought to be the most certain way to ensure establishment of the inoculum. Once a plant response is ascertained, methods of inoculation more applicable on a large scale can be tested. ... As a result of lime application, soil pH increased from 4.6 to about 6.2, and extractable Ca increased up to about 650 mg kg"'''. Phosphorus applications of 10, 30, 60, and 120 kg ha"-^ resulted in extractable P in. the soil of 5.1, 8.6, 17.8, and 34.3 mg kg"-^ at the end of the experiment, respectively. Thus the legume species and VAM fungi in the present study were exposed to a considerable range of soil P. Seedlings inoculated with VAM fungi were similar in shoot dry weight and P concentration to control seedlings at transplanting (Table 7-1). Seedlings inoculated with VAM fungi were also well colonized, whereas no VAM colonization was detected on control seedlings (Table 7-1). Phosphorus amendments and fungal inoculation increased shoot dry weights of aeschynomene (Table 7-2) and all harvests of Siratro

PAGE 72

60 Shoot dry weights, P concentrations, and percentage of roots colonized of Siratro and aeschynomene seedlings at transplanting. ; VAM Siratro— aes chynomene— Inoculation Shoot dry Shoot Root shoot dry Shoot Root wt. P colon. wt. P colon. rag — % rag — % — G. etunicatum 302 .18 52 30A .17 57 G. intraradices 305 .17 60 299 .20 55 Control 298 .16 0 302 .18 0 ^Data are means of five replicates

PAGE 73

61 Table 7-2. Analysis of variance for shoot dry weights of Aeschynomene americana harvested 2 October 1986. Source of variation DF — — — Mean squares • Block 9 52.57, Fungi (F) 2 165A.72"" Phosphorus (P) 3 6986.26"* Linear (PI) 1 18700.24"* Quadratic (Pq) 1 2250.32"'* Cubic (Pc) 1 8.21 F X P 6 91.79 F X PI 2 53.72 F X Pq 2 215.53 F X Pc 2 6.11 Error 99 45.83 ""Significant at P < 0.01

PAGE 74

62 (Table 7-3). There were no fxingi x P interactions for shoot dry weights of Siratro or aeschynomene. At all levels of applied P and at all harvests, shoot dry weights of Siratro were greater for fungal inoculated plants than control plants (Fig. 7-1). Except for the third harvest, where the effect of fungal inoculation was less pronounced at all levels of applied P, differences between fungal inoculated and noninoculated plants were most marked at intermediate levels of applied P (30-90 kg ha"-*-) and diminished at the highest level (120 kg ha"b. The effect of inoculation with VAM fungi on the shoot dry weights of aeschynomene, at all levels of applied P, followed the same trend as that of Siratro although the response to mycorrhizal inoculation was greater (Fig. 7-2). Inoculated plants of Siratro (Table 7-4 and 5) and aeschynomene (Table 7-6) had a quadratic relationship between shoot dry weight and P application. Maximum shoot dry weight of Siratro was achieved between 75-85 kg ha ^ of P and for aeschynomene at 85 kg ha"-*of P, whereas control plants of both legumes, even with 120 kg ha'^ of P, did not reach the same shoot dry weight as fungal inoculated plants. Thus inoculation with VAM fungi resulted in at least a 30% savings (40 kg ha"^ in the amount of P fertilizer required for maximum yield. In a previous greenhouse experiment (Chapter VI), I found that inoculation with G. etunicatum resulted in a 20% decrease in the amount of P required for maximum yield of Siratro. Some previous reports of VAM field experiments with legumes in nonpasteurized soil (Black and Tinker, 1977; Khan, 1975) show responses to VAM inoculation only in the absence of P fertilizer.

PAGE 75

63 Table 7-3. Analysis of variance for shoot dry weights from four Siratro harvests. Source of Mean squares variation DF Harvest 1 2 Oct. 86 Harvest 2 27 Nov. 86 Harvest 3 5 May 87 Harvest 4 29 June 87 Block 9(6)z 6, ,A0 4. 14 11. .47 18, .70* Fungi (F) 2 332. ,20 247. 79** 69, 20** 177, .21** Phosphorus (P) 3 807. .52 1577. 91" 1598, .20** 1806, .27** Linear (PI) 1 2107. .hi 4021. 98** 4223, .03** 4630, .88"" Quadratic (Pq) 1 305. .04 711. 49** 559, .16** 773, .57** Cubic (Pc) 1 10. ,07 0. 23 12. .38 14. .32 F X P 6 6. 35** 3. 21 1, ,67 10. .12 F X PI 2 k. ,61** 3. 89 1. ,10 7. ,69 F X Pq 2 12. 82"* 5. 38 2. ,76 15. ,83 F X Pc 2 1. ,63 0. 38 1. ,14 6. ,83 Error 99(66) 8. 89 13. 43 9. ,20 7. ,98 ''''significant at P < 0.01 *Signif leant at P < 0.05 ^Values in parentheses are the degrees of freedom for harvests 2, 3, and 4 for which only 7 replicates were used.

PAGE 76

64 Effect of P application on shoot dry weights for the first (A), second (B), third (C), and fourth (D) harvest of Siratro grown under field conditions and inoculated with Gloi""s etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates.

PAGE 77

65 Effect of P application on shoot dry weight and percentage of root colonized of Aeschynomene americana grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates.

PAGE 78

66 Table 7-A. Regression equations and coefficients of determination (r'^) showing the relationship of applied P level to shoot dry weight, percentage of root colonization, P and N concentrations, and total P and N uptake for the first harvest of Siratro. Variable Regression equations -pGlomus etunicatiim Shoot dry wt. (g) = 12 36+0 33P-0 0022p2 0.80** Root coloniz. (%) = 5 31+0 77P-0 0043p2 0.91** Plant P cone. (%) = 0 22+0 00072P 0.66** Total Plant P (mg) = 17 73+1 36P-0 0069P^ 0.98** Plant N cone. (%) = 2 26+0 035P-0 00025p2 0.52** Total Plant N (mg) = 170. 94+20. 09P-0. 13p2 0.91** Glomus intraradices Shoot dry wt. (g) = 12 03+0 24P-0 0014p2 0.67** Root coloniz. (%) = 10 04+0 59P-0 0034P^ 0.86** Plant P cone. (%) = 0 16+0 0037P-0 OOOOISP^ 0.76** Total plant P (mg) = 16 08+1 31P-0 0068P^ 0.94** Plant N cone. (%) = 2 12+0 041P-0 00026p2 0.56** Total plant N (mg) = 226. 21+15. 50P-0.095p2 0.91** Control Shoot dry wt. (g) = 10.65+0. IIP 0.65** Plant P cone. (%) = 0.18+0. 001 IP 0.81** Total plant P (mg) = 12.36+0.48P 0.92** Total plant N (mg) = 173. 09+3. 57P 0.89** P = phosphorus level Significant at P < 0.01

PAGE 79

67 Table 7-5. Regression equaticas and coefficients of determination (r"^) showing the relationship of applied P level to shoot dry weights for the second, third, and fourth harvest and percentage of root colonization for the fourth harvest of Siratro. Variable Regression equation Shoot dry wt (g), harvest 2 Shoot dry wt (g), harvest 3 Shoot dry wt (g), harvest 4 Root colon. (%), harvest 4 Glomus etunicatum = 5.74+0.52P-0.0029p2 = 0.99+0.48P-0.0029p2 = 5.98+0.55P-0.0032P^ = 9.85+0.84P-0.0046p2 0.87', 0.89"; 0.87' 0.86 Shoot dry wt (g), harvest 2 Shoot dry wt (g), harvest 3 Shoot dry wt (g), harvest 4 Root colon. (%), harvest 4 Glomus intraradices = 6.58+0.46P-0.0025p2 = 2.05+0.43P-0.0025p2 = 6.14+0.52P-0.0027p2 = 15.03+0.69P-0.0039p2 0.84'"'^ 0.86"* 0.87"* 0.89** Shoot dry wt (g), harvest 2 Shoot dry wt (g), harvest 3 Shoot dry wt (g), harvest 4 Control = 1.56+0.44P-0.0023p2 = -0.23+0.42P-0.0021p2 = 3.38+0.42P-0.0017p2 0.81 0.88' 0.91 P = phosphorus level 'Significant at P < 0.01

PAGE 80

68 Regression equations and coefficients of determination (r ) showing the relationship of applied P level to shoot dry weight, percentage of root colonization, P concentration, and total P and N uptake of Aeschynomene americana. Variable Regression equations r2 Glomus etunicatum Shoot dry wt. (g) = 46.59+0.83P-0.0049p2 0.78** T") 1 T • / Of \ Root coloniz. (%) = 7.89+0.88P-0.00A7P'^ 0.89" riant P Cone. (%) = 0.21+0.0037P-0.000019p2 0.80** Total plant P (mg) = 67.54+4.96P-0.025p2 0.96** Total plant N (g) = 1.35+0.039P-0.00019p2 Glomus intraradices 0.93** Shoot dry wt. (g) = 40.86+0.97P-0.0056P^ 0.86*f Root coloniz. (%) = 9.26+0.98P-0.0058p2 0 93** Plant P Cone. (%) = 0.16+0.0053P-0.000029p2 0.76** Total plant P (mg) = 43.24+5.85P-0.031p2 0.95"* 0.91** Total plant N (g) = 1.18+0.043P-0.00022p2 Control Shoot dry wt. (g) = 43.50+0.31P 0.76** Plant P Cone. (%) = 0.11+0.0045P-0.000021p2 0.95** 0.99 0.81** Total plant P (mg) = 28.08+3.63P-0.012p2 Total plant N (g) = 1.44+0.015P = Phosphorus level "* Significant at P < 0.01

PAGE 81

69 Hayman and Mosse (1979), however, reported improved growth of white clover in the field after inoculation with VAM fungi and the addition of 90 kg ha ^ of P. They also indicated that responses to fungal inoculation were smaller with 23 kg ha"^ of P and were absent where no P was added. Similarly, Hall (1984) reported that inoculation with selected VAM fungi increased yield of white clover in the field only if 50 kg ha ^ of P was also applied. ~ ^" Several greenhouse experiments on the phosphate response curves of fungal inoculated and noninoculated forage legumes have been carried out, mostly using clovers (Abbott and Robson, 1977b; Sparling and Tinker, 1978; Powell, 1980b) and Siratro (Lynd et al., 1985; Medina et al., 1987d) as the test plants. These authors applied soluble P fertilizers at rates ranging from 0 to 250 kg ha"^ and have reached the general conclusion that inoculation with VAM fungi markedly increases legume growth at low and intermediate rates of applied P. From the practical point of view, however, the interactions between phosphate additions and VAM on legumes are not always predictable and generalizable, because the responses are modulated by the incidence of several factors. These include the physical and chemical characteristics of the soil, plant species, VAM fungi, and the complex interactions between these factors. At the first harvest of Siratro (Fig. 7-lA). plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices at all levels of applied P. However, in subsequent harvests of Siratro (Fig. 7-lBC and D) and for aeschynomene (Fig. 7-2) the response of shoot dry weight to inoculation with the two VAM fungi was not different.

PAGE 82

70 Phosphorus additions and fungal inoculation increased percentage of root colonized by VAM fungi of Siratro (Table 7-7) and aeschynomene (Table 7-8). There were fungi x P interactions for percentage of root colonized by VAM fungi for for both legumes. Inoculated treatments had greater percentage of root colonized than control treatments at all levels of applied P. Percentage of root colonized by VAM fungi for the inoculated plants of Siratro (Fig. 7-3A and B) and aeschynomene (Fig. 7-2) increased linearly with P additions up to 60 kg ha"''". Phosphorus application of 120 kg ha'-"did not affect the percentage of root colonized by VAM fungi. In a previous greenhouse study (Chapter V), I found that percentage of Siratro root length colonized by VAM fungi increased with P additions up to 40 mg kg"-"-, which is equivalent to 80 kg ha ^ of P. Abbott and Robson (1977b) and Schubert and Hayman (1986), also in pot experiments, reported an increase in the percentage of root length colonized up to 55 mg kg"-^ of P (110 kg ha'"*"). Phosphorus applications did not alter the percentage of root colonized in the control plants. The degree of root colonization of the control plants of Siratro increased from 9% in the first harvest to about 19% by the fourth harvest (Fig. 7-3A and B), but still failed to increase the shoot dry weight of the control plants compared to the inoculated plants. Fungal inoculated plants of Siratro (Table 7-4 and 5) had a quadratic relationship between percentage of root colonized and applied P. Maximum root colonization was attained between 85-90 kg ha"^ of P for both legumes. At lower P additions, Siratro plants inoculated with G. intraradices had greater percentage of root colonized than plants inoculated with G. etunicatum (Fig. 7-3). However, there were no

PAGE 83

71 Table 7-7. Analysis of variance for percentage root colonized of Siratro by VAM fungi. Source of variation DF Harvest 1 2 Oct. 86 Mean squares Harvest 4 29 June 87 Block 4 Fungi (F) .2 Phosphorus (P) 3 Linear (PI) 1 Quadratic (Pq) 1 Cubic (Pc) 1 F X P 6 F X PI 2 F X Pq 2 F X Pc 2 Error 44 9.27 2394.82 935.53 2195.71 552.25 58.62. 230.66'' 510.93* 112.49'' 68.55"'" 7.98 14.61 1206.47 1230.09 2538.78 1083.74 67.74 86.82'; 174.04'' 77.26'' 9.16 11.91 Significant at P < 0.01 'Significant at P < 0.05

PAGE 84

72 Table 7-8. Analysis of variance for percentage root colonized, P concentration, and total P and N uptake of Aeschynomene americana harvested 2 October 1986. Mean Squares Source of DF Root P cone. total P total N variation colon. Block 4 10. 11 Fungi (F) 2 5276. 62 Phosphorus (P) 3 1844. 64 Linear (PI) 1 4408. 21 Quadratic (Pq) 1 1001. 65 Cubic (Pc) 1 124. 08 F X P 6 378. 66"* F X PI 2 833. 18** F X Pq 2 220. 16** F X Pc 2 82. 64"""' Error 4A 11. 60 0.00065 384.02 0. ,051 0.0098** 9052.18 0. 39** 0.051** 75570.74 4. 64';'; 0.12**^. 196270.22 12. 04** 0.031** 29179.92 1. 89** 0.0012 1262.07, 0. 0021 0.0010 1190.91** 0. 026 0.0021 1275.75*; 0. 013 0.00052 1815.29** 0. 043 0.00021 481.71 0. 022 0.0011 206.45 0. 062 ^Significant at P < 0.01 *Signif leant at P < 0.05

PAGE 85

73 Effect of P application on percentage of root colonized for the first (A) and fourth (B) harvest of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), ^^""^^s intraradices (INT), or not inoculated (CON). Data points are means of five replicates.

PAGE 86

74 differences in root colonization between G. etunicattim and G. intraradices at any level of applied P. For aeschynomene, differences between G. intraradices and G. etunicatum were only at 30 kg ha~''' of applied P (Fig. 7-2). At the first harvest, P amendments and fiingal inoculation also had an effect on P concentration, total P uptake, N concentration, and total N uptake of Siratro (Table 7-9). These same parameters were increased for aeschynomene, except for N concentration (Table 7-8). There were fungi X P interactions for P concentration, total P uptake, N concentration, and total N uptake of Siratro. By the contrast, aeschynomene only had fungi x P interaction for total P uptake. Inoculated plants of Siratro had greater P concentration, total P uptake, N concentration, and total N uptake than control plants at low and intermediate levels of applied P (Fig. 7-4). At the highest level of applied P, total P uptake of Siratro plants inoculated with G. intraradices and P and N concentrations of plants inoculated with G. etunicatum did not differ from those of control plants. Fungal inoculated plants of aeschynomene also had greater P concentration, total P uptake, and total N uptake than control plants at low and intermediate levels of applied P, but not at the highest level (Fig. 7-5). The positive effect of inoculation with VAM fungi on P and N uptake have been shown for other forage legumes, for example, leucaena (Habte and Manjunath, 1987), Pueraria phaseoloides (Sanchez and Salinas, 1981), and field experiments with white clover (Hall, 1984; Hayman and Mosse, r 1979) and Medicago sativa (Azcon-Aguilar and Barea, 1981). Potassium,

PAGE 87

75 Analysis of variance for P and N concentrations, and total P and N uptake of Siratro for the first harvest. Source of Mean Squares variation DF P Cone. total P N Cone. total N Block 4 0. 00033 26 .06 0.42" 10491. 95'"' Fungi (F) 2 0. 0059 2556 .03 3.03 540732. 85 Phosphorus (P) 3 0. 039 7248 .85 1.47 544117. 23 Linear (PI) 1 0. 11 19691 .07 1.53 1006530. 96 Quadratic (Pq) 1 0. 0049 2055 .48 2.87 625338. 95 Cubic (Pc) 1 0. 000077 0 .010 o.oip 487. 84 F X P 6 0. 0023'' 175 .50"* 0.33" 33638. 14'" F X PI 2 0. 003lf 53 .34 0.21. 7291. 34 F X Pq 2 0. 0039" 465 .13"* 0 75* 92589. 06""" F X Pc 2 0. 00011 8 .02 0.040 1034. 01 Error 44 0. 00087 26 .84 0.14 3606. 65 "'Significant at P < 0.01 'Significant at P < 0.05

PAGE 88

76 Fig. 7-4. Effect of P application on P concentration, total P, N concentration, and total N of Siratro grown in the field and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Data points are means of five replicates.

PAGE 89

77 P fiPPLIED (kg/ha) Effect of P application on P concentration, total P, and total N of Aeschynomene americana grown in the field conditions and inoculated with Glomus etunicatum (ETU), Glomus intraradices (INT), or not inoculated (CON). Dat points are means of five replicates.

PAGE 90

78 Ca, Mg, and Zn concentrations of Siratro and aeschynomene were not related to P applications or fungal inoculation (data not presented). In contrast to greenhouse pot experiments, field experiments on inoculation with VAM fungi often have been unsuccessful. It is possibly that the main cause for the satisfactory response to fungal inoculation obtained in the present study was due to a largely ineffective indigenous VAM population as compared to G. etunicatum and G. intraradices under the amended soil conditions. In previous greenhouse experiments (Chapter IV and V) using a similar nonpasteurized soil than the one used in the present study, I found that G. etunicatum and G. intraradices were effective growth enhancers of Siratro, even with a reasonably high soil native VAM population. Similarly, Powell et al. (1980b) reported that indigenous VAM fungi were ineffective in many soils and that inoculation by more effective VAM fungi would result in positive responses, even in nonpasteurized soils containing a high indigenous VAM population. In conclusion, the present study shows that effective inoculation with selected VAM fungi can have an important effect on growth of forage legumes in the field in soils that contain ineffective native VAM populations under amended soil conditions, even at moderate levels of applied P.

PAGE 91

CHAPTER VIII CONCLUSIONS The objective of this chapter is to summarize the work of the preceding six chapters. The overall goal of this research project was to improve the establishment phases and growth of tropical forage legumes in newly cleared land at reduced P fertilization through inoculation with effective VAM fungi. In order to accomplish this goal, greenhouse studies were carried out in limed, pasteurized and nonpasteurized Oldsmar fine sand, collected from a newly cleared area at the Agricultural Research and Education Center, Fort Pierce, FL. A field experiment was conducted in the same nonpasteurized soil. In Chapter II, quantitative data on the amount of root colonization and the species distribution of VAM fungi associated with four cultivated tropical forage legumes from four different locations in south Florida are reported. Differences in percentage of root colonization and total spore density were significant among locations, legume species, and location x legiune species interactions. Legume species differed in percentage root colonization and total spore density among locations except for carpon desmodium, which showed no differences among locations in percentage root colonization. The six species of VAM fungi collected in this survey were: G. heterogama G. margarita G. etunicatum, G. intraradices Glomus sp., and A. spinosa 79

PAGE 92

80 In Chapter III, the effect of inoculation with G. intraradices on the growth of several tropical forage legumes in P-def icient, nonpasteurized and pasteurized soil under greenhouse conditions is reported. Shoot and root dry weights were increased after inoculation in nonpasteurized soil for Siratro, aeschynomene, Aeschynnomene villosa, Stylosanthes hamata and Stylo, but not for Arachis sp. and Vigna adenantha, which only responded to inoculation in pasteurized soil. In Chapter IV, the effect of five species of VAM fungi, G. etunicatum G. deserticola, G. versiforme G. intraradices and G. margarita. on the growth of Siratro in a limed, nonpasteurized soil with an applied P level of 20 mg kg"-"" under greenhouse conditions was determined. Shoot dry and root fresh weights of plants inoculated with G. etunicatum and G. intraradices were higher than the other VAM fungal treatments and noninoculated plants. In addition, plants inoculated with G. etunicatum had higher shoot dry weights than plants inoculated with G. intraradices. The indigenous population of VAM fungi was reasonably high (MPN = 2 propagules g ^ soil); however, plant yields were less than the best VAM treated plants. A positive correlation was found between mycorrhizal root colonization, expressed as either percentage or total root length colonized, and shoot dry weight. Glomus etunicatum colonized roots more rapidly than the other VAM fungi tested. In Chapter V, the effect of G. etunicatum and G. intraradices on the growth of Siratro in a limed, nonpasteurized soil, with applied P levels of 2.5. 10, 20. and 40 mg kg'^ under greenhouse conditions is reported. At 2.5 mg kg"^ of applied P. there was no yield response to inoculation. Above 2.5 mg kg"^ of applied P, plants inoculated with

PAGE 93

81 either G. etunicatiim or G. intraradices weighed more than control plants. Inoculated plants required between 28 and 30 mg P kg to achieve maximum shoot dry weight, whereas control plants, even with AO mg P kg ^, did not achieve maximum growth. Shoot dry weight response was better with G. etunicatum than with G. intraradices For both fungi, increasing P above 2.5 mg kg ^ increased the percentage and total root length colonized by VAM fungi. In Chapter VI, the effect of inoculation with G. etunicatum on the growth and uptake of P and N of three forage legumes at applied P levels between 12.5 and 50 mg kg ^ in a limed, pasteurized soil under greenhouse conditions is reported. At all levels of applied P, shoot dry and root fresh weights, and total N of Stylo were greater for mycorrhizal plants than nonmycorrhizal plants. The differences were most pronounced at intermediate levels of applied P and diminished at the higher P levels. Mycorrhizal plants of Siratro had greater shoot dry weight, total P and N than nonmycorrhizal plants at low and intermediate P levels, but not at the highest level. Differences in shoot dry weight between mycorrhizal and nonmycorrhizal plants of aeschynomene were significant only at the intermediate level of applied P. Overall, inoculated plants of the three legumes studied had greater P concentration than control plants. Maximum root colonization, expressed as either percentage or total length of colonized root of the three legumes, was attained at intermediate levels of applied P. In Chapter VII, the effect of inoculation with selected VAM isolates on growth and nutrient uptake of Siratro and aeschynomene under natural field conditions at applied P levels of 10, 30, 60, and 120 kg ha"'" is

PAGE 94

82 reported. At all levels of applied P and for all harvests, shoot dry weights of Siratro were greater for fungal inoculated plants than -. noninoculated plants. Differences between fungal inoculated and noninoculated plants were most marked at 30 to 90 kg ha ^ of applied P and diminished at 120 kg ha ^. The effect of fungal inoculation on the shoot dry weights of aeschynomene, at all levels of applied P, was similar (but more pronounced) as that of Siratro. At the first harvest of Siratro, plants inoculated with G. etunicatum had higher shoot dry weights than G. intraradices plants at all levels of applied P. However, in subsequent harvests of Siratro and for aeschynomene the response of shoot dry weight to inoculation with the two VAM fungi was similar. Fungal inoculation resulted in at least a 30% savings (40 kg ha ''') in the amount of P fertilizer required for maximum yield. Inoculated treatments had greater percentage of root colonized than noninoculated treatments at all levels of applied P. Percentage of root colonized by VAM fungi for the inoculated plants of the two legumes increased linearly with P additions up to 60 kg ha ^. There were no differences in root colonization between G. etunicatum and G. intraradices at any level of applied P. The results of these studies clearly demonstrate that inoculation with effective VAM fungi can increase the growth of legumes in soils that may have a high, but largely ineffective native VAM population than introduced VAM fungi under amended soil conditions. Furthermore, with highly mycorrhizal dependent crops such as tropical forage legumes, a mycorrhizal growth response may occur at P levels normally used in commercial pasture production.

PAGE 95

LITERATURE CITED Abbott, L.K., and A.D. Robson. 1977a. The distribution and abundance of vesicular-arbuscular endophytes in some Western Australian soils. Aust. J. Bot. 25: 515-522. Abbott, L. K. and A. D. Robson. 1977b. Growth stimulation of subterranean clover with vesicular-arbuscular mycorrhizas. Aust. J Agric. Res. 28:639-649. Abbott, L. K., and A. D. Robson. 1978. Growth of subterranean clover in relation to formation of endomycorrhizas by introduced and indigenous fungi in a field soil. New Phytol. 81:575-587. Abbott, L. K., and A. D. Robs on. 1981. Infectivity and effectiveness of vesicular-arbuscular mycorrhizal fungi: Effect of inoculum type Aust. J. Agric. Res. 32:631-639. Anderson, D. L. and L. J. Henderson. 1986. Sealed chamber digestion for plant nutrient analyses. Agron. J. 78:937-939. Asimi, S., V. Gianinazzi-Pearson, and S. Gianinazzi. 1980. Influence o increasing soil phosphorus levels on interactions between vesicular arbuscular mycorrhizae and Rhizobium in soybeans. Can. J. Bot. 58:2200-2206. Azcon-Aguilar, C, and J. M. Barea. 1981. Field inoculation of Medicag with VA mycorrhiza and rhizobium in phosphatefixing agricultural soil. Soil Biol. Biochem. 13:19-22. Barber, D. A., and B. C. Lougham. 1967. The effect of microorganisms in the absorption of inorganic plant nutrients. II. Uptake and utilization of phosphate by barley plants grown under sterile and non-sterile conditions. J. Exp. Bot. 18:170-176. Barea, J. M., J. L. Escudero, and C. Azcon-G. de Aguilar. 1980. Effect of introduced and indigenous VA mycorrhizal fungi on nodulation and nutrition of Medicago sativa in phosphate fixing soils as affected !.. by P fertilizers. Plant Soil 54:283-296. Bergersen, F. J. 1971. Biochemistry of nitrogen fixation in legumes. Annu. Rev. Plant Physiol. 22:124-140. 83

PAGE 96

84 Bethlenfalvay, G.J., J.M. Ulrich, and M.S. Brown. 1985. Plant response to mycorrhizal fungi: host, endophyte, and soil effects. Soil Sci. Soc. Am. J. 49: 1164-1168. Black, R. L. B. and P. B. Tinker. 1977. Interactions between effects of vesicular-arbuscular mycorrhiza and fertilizer phosphorus on yields of potatoes in the field. Nature 267:510-511. Burt, R. L., and C. P. Miller. 1975. Stylosanthes--A source of pasture legumes. Trop. Grassl. 9:117-123. Chulan, A., and P. Ragu. 1986. Growth response of Theobroma cacao L. seedlings to inoculation with vesicular-arbuscular mycorrhizal fungi. Plant Soil 96:279-285. Cooper, K. M. 1978. Adaption of mycorrhizal fungi to phosphate fertilizers, p. 107. In A. R. Ferguson, R. L. Bieleski, and I. B. Ferguson (eds.) Plant nutrition 1978. New Zealand DSIR Information Series No. 134, Government Printer, Wellington. Crush, J. R. 1974. Plant growth responses to vesicular-arbuscular mycorrhiza. VII. Growth and nodulation of some herbage legumes. New Phytol. 73:745-750. Currah, R.S., and M. Van Dyk. 1986. A survey of some perennial vascular plant species native to Alberta for occurrence of mycorrhizal fungi. Canadian Field-Naturalist 100: 330-342. Daft, M.I., and A. A. El-Giahmi. 1976. Studies on nodulated and mycorrhizal peanuts. Ann. Appl. Biol. 83:273. Daniels, B. A., P. M. McCool, and J. A. Menge. 1981. Comparative inoculum potential of spores of six vesicular-arbuscular mycorrhizal fungi. New Phytol. 89:385-391. Daniels, B. A., and H. D. Skipper. 1982. Methods for recovery and quantitative estimation of propagules from soil. p. 29-35. In N. C. Schenck (ed.) Methods and principles of mycorrhizal research. Amer. Phytopath. Soc, St. Paul, MN. Gallaher, R. N. C. 0. Weldon, and J. G. Futral. 1975. An aluminum • block digester for plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39:803-806. Gates, C. T., and J. R. Wilson. 1974. The interaction of nitrogen and phosphorus on the growth, nutrient status and nodulation of Stylosanthes humilis H.B.K. (Townsville stylo). Plant Soil 41:325333. Gerdemann, J.W. and J.M. Trappe. 1974. The endogonaceae in the Pacific J Northwest. Mycol. Mem. 5: 1-76. j

PAGE 97

85 Gibson, A. H. 1976. Limitations to nitrogen fixation in legumes. In W. E. Newton, and C. J. Nyman (eds.) Proc. 1st Int. Symp. Nitrogen Fixation, Vol.2. Wash. State Univ. Press, Pullman. Giovannetti, M. 1985. Seasonal variations of vesicular-arbuscular mycorrhizas and endogonaceous spores in a maritime sand dune. Trans. Br. Mycol. Soc. 84: 679-684. Giovannetti, M. and B. Mosse. 1980. Evaluation of techniques for measuring vesicular-arbuscular mycorrhizal infection in roots. New Phytol. 84:489-500. Giovannetti, M., and T.H. Nicolson. 1983. Vesicular-arbuscular mycorrhizas in Italian sand dunes. Trans. Br. Mycol. Soc. 80: 552557. Habte, M. and T. Aziz. 1985. Response of Sesbania grandif lora to inoculation of soil with vesicular-arbuscular mycorrhizal fungi. Appl. Environ. Microbiol. 50: 701-703. Habte, M., and A. Manjunath. 1987. Soil solution phosphorus status and mycorrhizal dependency in Leucaena leucocephala Appl. Environ. Microbiol. 53:797-801. Hall, I. R. 1978. Effects of endomycorrhizas on the competitive ability of white clover. N. Z. J. Agr. Res. 21:509-515. Hall, I. R. 1979. Soil pellets to introduce vesicular-arbuscular mycorrhizal fungi into soil. Soil Biol. Biochem. 11:85-86. Hall, I. R. 1984. Field trials assessing the effect of inoculating agricultural soils with endomycorrhizal fungi. J. Agric. Sci. 102:725-731. Hambleton, L. G. 1977. Semiautoraated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. J. Assoc. Offic. Agr. Chem. 60:845-852. Harley, J.L., and E.L. Harley. 1987. A check-list of mycorrhiza in the British flora. New Phytol. 105:1-102. Harley, J. L. and S. E. Smith. 1983. Mycorrhizal symbiosis. Academic Press, London. Hass, J. H., and J. Krikum. 1985. Efficacy of endomycorrhizal fungus isolates and inoculum quantities required for growth response. New Phytol. 100:613-621. Hayman, D.S. 1978. Mycorrhizal populations of sown pastures and native vegetation in Otago, New Zealand. N. Z. J. Agr. Res. 21: 271-275.

PAGE 98

86 Hayman, D. S. 1983. The physiology of vesicular-arbuscular endomycorrhizal symbiosis. Can. J. of Botany 61:94A-963. Hayman, D. S., and K. A. Hampson. 1978. VA mycorrhiza. Field inoculation trial (white clover in Welsh upland soil). Rothamsted Rep. Part 1:238-239. Hayman, D. S., and B. Mosse. 1979. Improved growth of white clover in hill grasslands by mycorrhizal inoculation. Ann. Appl. Biol. 93:1A1-1A8. Hayman, D. S., R. J. Page, and C. A. Clarke. 1980. Vesiculararbuscular mycorrhiza: field inoculation studies with Red clover. Sawyers I. Rothamsted Rep. Part 1:201-202. Hayman, D.S., and G.E. Stovold. 1979. Spore populations and infectivity of vesicular-arbuscular mycorrhizal fungi in New South Wales. Aust. J. Bot. 27: 227-233. Hayman, D. S., and M. Tavares. 1985. Plant growth responses to vesicular-arbuscular mycorrhiza. XV. Influence of soil pH on the symbiotic efficiency of different endophytes. New Phvtol. 100:367377. Hodges, E. M., A. E. Kretschmer, Jr., P. Mislevy, R. D. Roush, 0. C. Ruelke, and G. H. Snyder. 1982. Production and utilization of the tropical legxime aeschynomene. Fla. Agric. Exp. Stn. Circ. S-290. Huang, R., W.K. Smith, and R.S. Yost. 1985. Influence of vesiculararbuscular mycorrhiza on growth, water relations, and leaf orientation in Leucaena leucocephala (Lam.) Dewit. New Phytol. 99: 229-243. Hutton, E. M. 1962. Siratro-a tropical pasture legvime bred from Phaseolus atropurpureum Aust. J. Agric. Anim. Hus. 2:117-125. labal, S. H., K. Sultana, and B. Perveen. 1975. Endogone spore numbers in the rhizosphere and the occurrence of vesicular-arbuscular mycorrhizas in plants of economic importance. Biologia 21:227-237. Islam, R., A. Ayanaba, and F. E. Sanders. 1980. Response of cowpea ( Vigna unguiculata ) to inoculation with VA mycorrhizal fungi and rock phosphate fertilization in some unsterilized Nigerian soils. Plant Soil 54:107-117. Jensen, A. 1984. Responses of barley, pea, and maize to inoculation with different vesicular-arbuscular mycorrhizal fungi in irradiated soil. Plant Soil 78: 315-323. Khan, A. G. 1975. Growth effects of VA mycorrhiza on crops in the field. P. 149-435. In F. E. Sanders, B. Mosse, and P. B. Tinker (eds.) Endomycorrhizas. Academic Press, London, New York.

PAGE 99

87 Kormanik, P. P., and A. C. McGraw. 1982. Quantification of vesiculararbuscular mycorrhizae in plant roots, p. 37-45. In N. C. Schenck (ed.) Methods and principles of mycorrhizal research. American Phytopathological Society, St. Paul, MN. Kretschmer, A. E., Jr. 1972. Siratro ( Phaseolus atropurpureus DC.) A summer -growing perennial pasture legume for central and south Florida. Florida Agric. Exp. Stn. Circ. S-214. Kucey, R.M. and G.E. Diab. 1984. Effects of lime, phosphorus, and addition of vesicular-arbuscular (VA) mycorrhizal fungi on indigenous VA fungi and on growth of alfalfa in a moderately acidic soil. New Phytol. 98: 481-486. Lopes, E. S., and E. De Olivera. 1980. Efeito de species de micorrizas vesicular-arbuscular em Siratro ( Macroptilium atropurpureum ) Bragantia 39:241-245. Lopes, E.S., E. Oliveira, R. Dias, and N.C. Schenck. 1983. Occurrence and distribution of vesicular-arbuscular mycorrhizal fungi in coffee ( Coffea arabica L.) plantations in central Sao Paulo state, Brazil. Turrialba 33:417-422. Lynd, J. Q., R. J. Tyrl, and A. A. C. Purcino. 1985. Mycorrhiza soil fertility effects on regrowth, nodulation and nitrogenase activity of Siratro ( Macroptilixim atropurpureum (DC) Urb.). J. Plant Nutr. 8:1047-1059. Menge, J. A. 1983. the physiology of vesicular-arbuscular mycorrhizal fungi in agriculture. Can. J. Bot. 61:1015-1024. Miller, R.H., E.J. B.N. Cardoso, and C.O.N. Cardoso. 1979. Some observations on mycorrhizal infection of tropical forage legumes and grasses in Brazil. Summa Phytopathologica 5: 168-172. Miller, D. D., P. A. Domoto, and C. Walker. 1985. Colonization and efficacy of different endomycorrhizal fungi with apple seedlings at two phosphorus levels. New Phytol. 100:393-402. Mosse, B. 1972. The influence of soil type and endogone strain on the growth of mycorrhizal plants in phosphate deficient soil. Rev. Ecol. Biol. Sol. T. 9:529-537. Mosse, B. 1977. Plant growth responses to vesicular-arbuscular mycorrhiza. X. Responses of stylosanthes and maize to inoculation in unsterile soils. New Phytol. 78:277-288. Mosse, B., C. L. Powell, and D. S. Hayman. 1976. Plant growth responses to vesicular-arbuscular mycorrhiza. IX. Interactions between VA mycorrhiza, rock phosphate and symbiotic nitrogen fixation. New Phytol. 76:331-342.

PAGE 100

88 Munns, D. N., and B. Mosse. 1980. Mineral nutrition of legume crops. In R. J. Summerfield, and A. H. Bunting (eds.) Advances in legume science. Univ. Reading Press, England. Newbould, P., and A. Rangeley. 198A. Effect of lime, phosphorus and mycorrhizal fungi on growth, nodulation, and nitrogen fixation by white clover ( Trifolium repens ) grown in UK hill soils. Plant Soil 76: 105-llA. Nielsen, J. P., and A. Jensen. 1983. Influence of vesicular-arbuscular mycorrhiza fungi on growth and uptake of various nutrients as well as uptake ratio of fertilizer P for lucerne ( Medicago sativa ) Plant Soil 70: 165-172. Pairunan, A. K., A. D. Robson, and L. K. Abbott. 1980. The effectiveness of vesicular-arbuscular mycorrhizas in increasing growth and phosphorus uptake of subterranean clover from phosphorus sources of different solubilities. New Phytol. 84:327-338. Plenchette, C, V. Furlan, and J. A. Fortin. 1982. Effects of different endomycorrhizal fungi on five host plants grown on calcined montmorillonite clay. J. Amer. Soc. Hort. Sci. 107:535-538. Powell, C.Ll. 1979. Inoculation of white clover and ryegrass seed with mycorrhizal fungi. New Phytol. 83: 81-85. Powell, C. LI. 1980a. Mycorrhizal infectivity of eroded soils. Soil Biol. Biochem. 12:247-250. Powell, C. LI. 1980b. Phosphate response curves of mycorrhizal and nonraycorrhizal plants. I. Responses to superphosphate. N. Z. J. Agr. Res. 23:225-231. Powell, C. LI. 1982. Selection of efficient VA mycorrhizal fungi. Plant Soil 68:3-9. Powell, C. LI., M. Groters, and D. Metcalfe. 1980. Mycorrhizal inoculation of a barley crop in the field. N. Z. J. Agr. Res. 23:107-109. Purcino, A. A. C, C. Lurlarp, and J. Q. Lynd. 1986. Mycorrhiza and soil fertility effects with growth, nodulation and nitrogen fixation of Leucaena grown on a Typic Eutrustox. Commun. in Soil Sci. Plant Anal. 17:473-489. Purcino, A. A. C, and J. Q. Lynd. 1985. Tripartite symbiosis of Stylosanthes scabra Vog. influenced by soil fertility treatments of a Typic Eutrustox. Agron. J. 77:455-458. Rangeley, A., M.J. Daft, and P. Newbould. 1982. The inoculatation of white clover with mycorrhizal fungi in unsterile hill soils. New Phytol. 92: 89-102.

PAGE 101

89 Rhue, R.D., and G. Kidder. 198A. Procedures used by the IFAS extension soil testing laboratory and interpretation of results. Fla. Coop. Ext. Serv. Circ. 596. Rotar, P.P. 1983. Legumes in action. P. 21-22. In R.L. Burt et al. (eds.) The role of Centrosema, Desmodium, and Stylosanthes in improving tropical pastures. Westview Press, Colorado. Saif, S. R. 1987. Growth responses of tropical forage plant species to vesicular-arbuscular mycorrhizae. I. Growth, mineral uptake and mycorrhizal dependency. Plant Soil 97:23-35. Saif, S.R., and A.G. Khan. 1975. The influence of season and stage of development of plant on Endogone mycorrhiza of field-grown wheat. Can. J. Microbiology 21: 1021-1024. Salinas, J.G., J.I. Sanz, and E. Sieverding. 1985. Importance of VA mycorrhizae for phosphorus supply to pasture plants in tropical oxisols. Plant Soil 84: 347-360. Same, B. I., A. D. Robson, and L. K. Abbott. 1983. Phosphorus, soluble carbohydrates and endomycorrhizal infection. Soil Biol. Biochem. 15:593-597. Sanchez, P. A., and J. G. Salinas. 1981. Lowinput technology for managing oxisols and ultisols in tropical America. Adv. in Agronomy 34:279-406. Sanders, F. E., P. B. Tinker, R. L. B. Black, and S. M. Palmerley. 1977. The development of endomycorrhizal root systems. I. Spread of infection and growth-promoting effects with four species of vesicular-arbuscular endophyte. New Phytol. 78:257-268. Sanni, S. 0. 1976. Vesicular-arbuscular mycorrhiza in some Nigerian soils and their effect on the growth of cowpea ( Vigna unguiculata ) tomato ( Lycopersicon esculentum ) and maize (Zea mays). New Phytol 77:667-671. ^ SAS Institute Inc. 1982. SAS user 's guide: Statistics. SAS Institute Inc., Gary, N.C. Satterlee, L., B. Melton, B. McCaslin, and D. Miller. 1983. Mycorrhizal effects on plant growth, phosphorus uptake, and N2(C2Ha) fixation in two alfalfa populations. Agron. J. 75: 715-716. Schenck, N.C, and R.A. Kinloch. 1980. Incidence of mycorrhizal fungi on six field crops in monoculture on a newly cleared woodland site Mycologia 72: 445-456. Schenck, N.C, and G.S. Smith. 1981. Distribution and occurrence of vesicular-arbuscular mycorrhizal fungi on Florida agricultural crops. Proc. Soil and Crop Sci. Soc. of Fla. 40:171-175.

PAGE 102

90 Schenck, N.C., and G.S. Smith. 1982. Additional new and unreported species of mycorrhizal fungi (Endogonaceae) from Florida. Mycologia 7A: 77-92. Schroder, V., D. S. Hayman, and B. Mosse. 1977. Influence of VA mycorrhiza on plant growth. Rothamsted Report for 1976, part 1, p. 286 Schubert, A., and D. S. Hayman. 1986. Plant growth responses to vesicular-arbuscular mycorrhiza. XVI. Effectiveness of different endophytes at different levels of soil phosphate. New Phytol. 103:79-90. Snyder, G. H., A. E. Kretschmer, Jr., and J. Alvarez. 1985. Agronomic and economic response of three tropical legumes to lime and phosphorus in an acid infertile spodosol. Agron. J. 77:427-432. Sylvia, D.M. 1986. Spatial and temporal distribution of vesiculararbuscular mycorrhizal fungi associated with Uniola paniculata in Florida foredunes. Mycologia 78: 728-734. Thompson, B. D. A. D. Robson, and L. K. Abbott. 1986. Effects of phosphorus on the formation of mycorrhizas by Gigaspora calospora and Glomus fasciculatum in relation to root carbohydrates. New Phytol. 103:751-765. Trappe, J. M. 1982. Synoptic keys to the genera and species of zygomycetous (vesicular-arbuscular) mycorrhizal fungi. Phytopathology 72: 1102-1108. Waidyanatha, U. P. De S., N. Yogaratnan, and W. A. Ariyaratne. 1979. Mycorrhizal infection on growth and nitrogen fixation of Pueraria and Stylosanthes and uptake of phosphorus from two rock phosphates New Phytol. 82:147-152. Wilson, J. M. 1984. Competition of infection between vesiculararbuscular mycorrhizal fungi. New Phytol. 97:427-435. Yost, R.S., and R.L. Fox. 1979. Contribution of mycorrhizae to P nutrition of crops growing on an oxisol. Agron. J. 71: 903-908.

PAGE 103

BIOGRAPHICAL SKETCH Onesimo Adonis Medina was born in San Pedro de Macoris, Dominican Republic, on November 2, 1953. He received his B. Sc. degree in agronomy in April, 1977, from the Universidad Nacional Pedro Henriquez Urena, Santo Domingo, Dom. Rep. In June, 1977, he was hired by the Dominican Agrarian Institute (IAD) as an Assistant Investigator in the Soil Survey Division. He began graduate studies in August, 1979, and received a M. Sc. degree in soil science with a specialization in soil fertility from the University of Hawaii in 1981. On returning to the Dominican Republic, he was hired by the National Livestock Research Center (CENIP) as an Associate Investigator in the Soil Fertility and Plant Nutrition Division. He also served as an Assistant Professor in the Agronomy and Soil Science Department of the Universidad Nacional Pedro Henriquez Urena. In August, 1984, he enrolled to the graduate program of the University of Florida, Department of Soil Science, and received the degree of Doctor of Philosophy in December, 1987. He has been married to Griselda Terrero since 1979 and they have one daughter, Michelle. 91

PAGE 104

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. Dr. D. M. Sylvia, Chairman Assistant 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. Dr. A. E. Kretschmer, Cochairma: 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. Dr. G. KiddS? Associate Professor of Soil Science I certify that I have read this study and that in ray 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. P D^'. jTxb. Sartain Professor of Soil Science

PAGE 105

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. 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. Professor of Plant Pathology This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1987 Dean, ^o^legeofAgri^Tture J i^lti Dean, Graduate School


A3
macrocarpum and G. margarita were ineffective in stimulating growth of
onion; however, G. caledonium and Glomus sp. 'E3' were generally
effective at all P levels. In our study, G. etunicatum was more
effective than G. intraradices at all but the lowest applied P level.
Hence, there appears to be good potential for the selection of VAM fungi
to enhance plant growth under amended soil conditions such as P
fertilization and liming.
Phosphorus has been reported to increase, decrease or not affect root
colonization by VAM fungi. However, it is difficult to compare results
concerning the effect of P fertilization on mycorrhizal root
colonization, because of differences in the range of added P, as well as
other factors such as host plant and soil type. In this study, we used
a range of 2.5 to AO mg P kg ^ because it represented P levels used in
the production of tropical forage legumes in Florida on a similar soil
(Snyder et al., 1985). At the lowest P level, G. etunicatum and G.
intraradices did not colonize the root or improve growth of Siratro above
that of the control plants. Barber and Lougham (1967) reported that, at
a very low P level, competition for P occurs between plants and
microflora. Habte and Manjunath (1987) and Same et al. (1983) indicated
that the growth of VAM fungi is limited by P at very low levels. Between
10 to AO mg P kg ^, percentage and total root length colonized by VAM
fungi increased with P additions. These results agree with those of
Abbott and Robson (1977b), Schubert and Hayman (1986), and
Thompson et al. (1986), who reported an increase in the percentage and
total root length colonized between 18 to 55 mg P kg-1. At high P levels
(more than 100 mg kg ^), which are not feasible for field production of


g-DCTT CRY EI&T (g) SCOT CRY YEEHT (g)
64
0 30 00 00 120
P APPLIED (kgAn)
Fig. 7-1. Effect of P application on shoot dry weights for the first
(A), second (B), third (C), and fourth (D) harvest of
Siratro grown under field conditions and inoculated with
Glomus etunicatum (ETU), Glomus intraradices (INT), or not
inoculated (CON). Data points are means of five
replicates.


9
Table 2-2. Analysis of variance for percentage of mycorrhizal root
colonization and total spore density per 100 g of air-
dried soil.
Source of
variation
Degree of
freedom
Mean
Squares
Root
colonization
total spore
density
%
no. 100 g^ soil
Location
3
447
325871
Legumes
3
321
73895
Interaction
8
299**
88933**
Error
30
13
1320
Significant at P < 0.01


BIOGRAPHICAL SKETCH
Onesimo Adonis Medina was born in San Pedro de Macoris, Dominican
Republic, on November 2, 1953. He received his B. Sc. degree in
agronomy in April, 1977, from the Universidad Nacional Pedro Henriquez
Urena, Santo Domingo, Dom. Rep. In June, 1977, he was hired by the
Dominican Agrarian Institute (IAD) as an Assistant Investigator in the
Soil Survey Division.
He began graduate studies in August, 1979, and received a M. Sc.
degree in soil science with a specialization in soil fertility from the
University of Hawaii in 1981. On returning to the Dominican Republic,
he was hired by the National Livestock Research Center (CENIP) as an
Associate Investigator in the Soil Fertility and Plant Nutrition
Division. He also served as an Assistant Professor in the Agronomy and
V
Soil Science Department of the Universidad Nacional Pedro Henriquez
Urena.
In August, 1984, he enrolled to the graduate program of the
University of Florida, Department of Soil Science, and received the
degree of Doctor of Philosophy in December, 1987. He has been married
to Griselda Terrero since 1979 and they have one daughter, Michelle.
91


70
Phosphorus additions and fungal inoculation increased percentage of
root colonized by VAM fungi of Siratro (Table 7-7) and aeschynomene
(Table 7-8). There were fungi x P interactions for percentage of root
colonized by VAM fungi for for both legumes. Inoculated treatments had
greater percentage of root colonized than control treatments at all
levels of applied P. Percentage of root colonized by VAM fungi for the
inoculated plants of Siratro (Fig. 7-3A and B) and aeschynomene
(Fig. 7-2) increased linearly with P additions up to 60 kg ha ^.
Phosphorus application of 120 kg ha ^ did not affect the percentage of
root colonized by VAM fungi. In a previous greenhouse study (Chapter
V), I found that percentage of Siratro root length colonized by VAM fungi
increased with P additions up to 40 mg kg \ which is equivalent to 80
kg ha ^ of P. Abbott and Robson (1977b) and Schubert and Hayman (1986),
also in pot experiments, reported an increase in the percentage of root
length colonized up to 55 mg kg ^ of P (110 kg ha ^).
Phosphorus applications did not alter the percentage of root
colonized in the control plants. The degree of root colonization of the
control plants of Siratro increased from 9% in the first harvest to about
19% by the fourth harvest (Fig. 7-3A and B), but still failed to increase
the shoot dry weight of the control plants compared to the inoculated
plants. Fungal inoculated plants of Siratro (Table 7-4 and 5) had a
quadratic relationship between percentage of root colonized and applied
P. Maximum root colonization was attained between 85-90 kg ha1 of P
for both legumes. At lower P additions, Siratro plants inoculated with
G. intraradices had greater percentage of root colonized than plants
inoculated with G. etunicatum (Fig. 7-3). However, there were no


10
Table 2-3. Mean percentage of mycorrhizal root colonization and total
spore density of VAM fungi among forage legumes at four
locations in south Florida.
Location
Root
colonization2
i JOUd.'c
Total spore
density2
%
no. 100 g-^ soil
Aeschynomene americana
Ft. Pierce
7b
302a
Ona
6b
160b
Basinger
. 5b
. 8C
Deseret
30a
146b
Desmodium heterocarpon
Ft. Pierce
12a
679a
Ona
15a
376b
Basinger
12a
19c
Deseret
16a
36c
Vigna adenantha
Ft. Pierce
5C
535a
Ona
41a
77b
Basinger
20b
25
Deseret
25b
36c
Macroptilium atropurpureum
Ft. Pierce
15a
23b
Ona
8b
294a
Basinger
3b
5b
zMeans within a column for each legume species followed by the same
letter are not different (P < 0.05) according to Duncan's multiple range
test.


ROOT COLONIZATION (Z) ROOT COLONIZATION (23
73
Fig. 7-
Effect of P application on percentage of root colonized
for the first (A) and fourth (B) harvest of Siratro grown
in the field and inoculated with Glomus etunicatum (ETU).
Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.


80
In Chapter III, the effect of inoculation with G. intraradices on
the growth of several tropical forage legumes in P-deficient,
nonpasteurized and pasteurized soil under greenhouse conditions is
reported. Shoot and root dry weights were increased after inoculation
in nonpasteurized soil for Siratro, aeschynomene, Aeschynnomene villosa,
Stylosanthes hamata, and Stylo, but not for Arachis sp. and Vigna
adenantha, which only responded to inoculation in pasteurized soil.
In Chapter IV, the effect of five species of VAM fungi, G.
etunicatum, G. deserticola, G. versiforme, G. intraradices, and G.
margarita, on the growth of Siratro in a limed, nonpasteurized soil with
an applied P level of 20 mg kg ^ under greenhouse conditions was
determined. Shoot dry and root fresh weights of plants inoculated with
G. etunicatum and G. intraradices were higher than the other VAM fungal
treatments and noninoculated plants. In addition, plants inoculated with
G. etunicatum had higher shoot dry weights than plants inoculated with
G. intraradices. The indigenous population of VAM fungi was reasonably
high (MPN = 2 propagules g ^ soil); however, plant yields were less than
the best VAM treated plants. A positive correlation was found between
mycorrhizal root colonization, expressed as either percentage or total
root length colonized, and shoot dry weight. Glomus etunicatum colonized
roots more rapidly than the other VAM fungi tested.
In Chapter V, the effect of G. etunicatum and G. intraradices, on
the growth of Siratro in a limed, nonpasteurized soil, with applied P
levels of 2.5, 10, 20, and AO mg kg ^ under greenhouse conditions is
reported. At 2.5 mg kg ^ of applied P, there was no yield response to
inoculation. Above 2.5 mg kg ^ of applied P, plants inoculated with


12
Table 2-4. Mean spore numbers of VAM fungal species associated with
four forage legumes at four locations in south Florida.
Location
Species of VAM fungi-
-.} ; -.1
GM
GH
ETU
INT
AS
GS
Aeschynomene americana
Ft. Pierce
141a
66b
21a
42b
0
16
Deseret
9b
136a
0
0
0
0
Basinger
0
8C
0
0
0
0
On a
0
16c
29a
114a
0
0
Desmodium heterocarpon
Ft. Pierce
2a
114a
255b
307a
0
0
Deseret
2a
9b
0
0
0
0
Basinger
0
0
19c
0
0
0
Ona
0
0
325a
51b
0
0
Macroptilium atropurpureum
Ft. Pierce
0
10a
0
4b
0
8
Basinger
0
5b
0
0
0
0
Ona
0
0
40
254a
0
0
Vigna adenantha
Ft. Pierce
lb
39a
252a
241a
0
0
Deseret
20a
15b
0
0
0
0
Basinger
0
0
25b
0
0
0
Ona
28a
0
28b
llb
9
0
zMeans within
a column
for each
legume species followed by the same
letter are not different (P < 0.05) accordingly to Duncan's multiple
range test.


To the memory of my father Efrain, who influenced me in a very
special way. I am very heartbroken that he died before this work was
completed.
To my mother, Guillermina, for her never ending sacrifices, her
love, and prayers.


69
Hayman and Mosse (1979), however, reported improved growth of white
clover in the field after inoculation with VAM fungi and the addition of
90 kg ha of P. They also indicated that responses to fungal
inoculation were smaller with 23 kg ha ^ of P and were absent where no P
was added. Similarly, Hall (1984) reported that inoculation with
selected VAM fungi increased yield of white clover in the field only if
50 kg ha 1 of P was also applied.
Several greenhouse experiments on the phosphate response curves of
fungal inoculated and noninoculated forage legumes have been carried out,
mostly using clovers (Abbott and Robson, 1977b; Sparling and Tinker,
1978; Powell, 1980b) and Siratro (Lynd et al., 1985; Medina et al.,
1987d) as the test plants. These authors applied soluble P fertilizers
at rates ranging from 0 to 250 kg ha ^ and have reached the general
conclusion that inoculation with VAM fungi markedly increases legume
growth at low and intermediate rates of applied P. From the practical
point of view, however, the interactions between phosphate additions and
VAM on legumes are not always predictable and generalizable, because the
responses are modulated by the incidence of several factors. These
include the physical and chemical characteristics of the soil, plant
species, VAM fungi, and the complex interactions between these factors.
At the first harvest of Siratro (Fig. 7-1A), plants inoculated with
G. etunicatum had higher shoot dry weights than plants inoculated with
G. intraradices at all levels of applied P. However, in subsequent
harvests of Siratro (Fig. 7-1BC and D) and for aeschynomene (Fig. 7-2)
the response of shoot dry weight to inoculation with the two VAM fungi
was not different.


63
Table 7-3. Analysis of variance for shoot dry weights from four
Siratro harvests.
Source of
Mean squares
variation
DF
Harvest 1
2 Oct. 86
Harvest 2 Harvest 3
27 Nov. 86 5 May 87
Harvest A
29 June 87
Block
9(6)z
6. AO
A.1A
11.A7
18.70'
Fungi (F)
2
332.20
2A7.79**
69.20**
177.21**
Phosphorus (P)
3
807.52
1577.91**
1598.20**
1806.27**
Linear (PI)
1
2107.A7
A021.98**
A223.03*
A630.88**
Quadratic (Pq)
1
305.0A
711.A9**
559.16**
773.57**
Cubic (Pc)
1
10.07
6.35
0.23
12.38
1A.32
F x P
6
3.21
1.67
10.12
F x PI
2
A. 61.**
3.89
1.10
7.69
F x Pq
2
12.82
5.38
2.76
15.83
F x Pc
2
1.63
0.38
1.1A
6.83
Error
99(66)
8.89
13.A3
9.20
7.98
''Significant at
P < 0.01
Significant at
P < 0.05
zValues in parentheses are the degrees of freedom for harvests 2, 3, and
A for which only 7 replicates were used.


24
this defined term will be used in this paper. Wilson (1984) indicated
that an evaluation of the effectiveness of the indigenous mycorrhizal
population under amended soil conditions, as well as studies to select
effective VAM fungi, are prerequisites for successful field inoculation.
Thus, the objective of the present study was to determine the
effectiveness of several VAM fungi with Siratro in a limed,
nonpasteurized soil with low P content under greenhouse conditions.
Materials and Methods
The soil used in this study, liming, and fertilizer amendments are
described previously in Chapter III.
Plants were inoculated with the following VAM fungi: G. etunicatum
(isolate S312) obtained from carpon desmodium at the Agricultural
Research and Education Center, Ona, FL. (Chapter II, Table 2-4); G.
deserticola Trappe, Bloss & Menge (isolate S305) obtained from sea oats
(Unila paniculata L.) in a coastal dune, Anastasia, FL.; G. versiforme
Berch & Fortin (isolate #231) obtained from N.C. Schenck, University of
Florida, Gainesville, FL.; G. intraradices (isolate S311) obtained from
Vigna adenantha at the Agricultural Research and Education Center, Ona,
FL. (Chapter II, Table 2-4); G. margarita (isolate #215) obtained from
N.C. Schenck, University of Florida, Gainesville, FL. Isolates were
maintained in pot cultures in pasteurized soil containing bahiagrass.
Soils from 12-week-old pot cultures were used to inoculate experimental
pots. The propagule densities of the native soil and inocula at the
beginning of the experiment were determined by the most-probable-number


74
differences in root colonization between G. etunicatum and G.
intraradices at any level of applied P. For aeschynomene, differences
between G. intraradices and G. etunicatum were only at 30 kg ha ^ of
applied P (Fig. 7-2).
At the first harvest, P amendments and fungal inoculation also had
an effect on P concentration, total P uptake, N concentration, and total
N uptake of Siratro (Table 7-9). These same parameters were increased
for aeschynomene, except for N concentration (Table 7-8). There were
fungi x P interactions for P concentration, total P uptake, N
concentration, and total N uptake of Siratro. By the contrast,
aeschynomene only had fungi x P interaction for total P uptake.
Inoculated plants of Siratro had greater P concentration,
total P uptake, N concentration, and total N uptake than control plants
at low and intermediate levels of applied P (Fig. 7-4). At the highest
level of applied P, total P uptake of Siratro plants inoculated with G.
intraradices and P and N concentrations of plants inoculated with G.
etunicatum did not differ from those of control plants. Fungal
inoculated plants of aeschynomene also had greater P concentration, total
P uptake, and total N uptake than control plants at low and intermediate
levels of applied P, but not at the highest level (Fig. 7-5). The
positive effect of inoculation with VAM fungi on P and N uptake have been
shown for other forage legumes, for example, leucaena (Habte and
Manjunath, 1987), Pueraria phaseoloides (Sanchez and Salinas, 1981),
and field experiments with white clover (Hall, 1984; Hayman and Mosse,
1979) and Medicago sativa (Azcon-Aguilar and Barea, 1981). Potassium,


78
Ca, Mg, and Zn concentrations of Siratro and aeschynomene were not
related to P applications or fungal inoculation (data not presented).
In contrast to greenhouse pot experiments, field experiments on
inoculation with VAM fungi often have been unsuccessful. It is possibly
that the main cause for the satisfactory response to fungal inoculation
obtained in the present study was due to a largely ineffective indigenous
VAM population as compared to G. etunicatum and G. intraradices under
the amended soil conditions. In previous greenhouse experiments (Chapter
IV and V) using a similar nonpasteurized soil than the one used in the
present study, I found that G. etunicatum and G. intraradices were
effective growth enhancers of Siratro, even with a reasonably high soil
native VAM population. Similarly, Powell et al. (1980b) reported that
indigenous VAM fungi were ineffective in many soils and that inoculation
by more effective VAM fungi would result in positive responses, even in
nonpasteurized soils containing a high indigenous VAM population.
In conclusion, the present study shows that effective inoculation
with selected VAM fungi can have an important effect on growth of forage
legumes in the field in soils that contain ineffective native VAM
populations under amended soil conditions, even at moderate levels of
applied P.


46
which are apparently species-specific (Burt and Miller, 1975; Mosse,
1972).
In a previous study (Chapter V), G. etunicatum was an effective
growth enhancer of Siratro in a soil similar to the one used in the
present study, with a moderate level of applied P (20-40 mg P kg ^).
The objective of this investigation was to determine the effect of
inoculation with a VAM fungus, G. etunicatum, on the growth and plant
uptake of P and N of three forage legumes at different P levels in
pasteurized soil under greenhouse conditions.
Materials and Methods
The soil used in this investigation, liming, and basic fertilization
are described previously (Chapter III). The soil chemical
characteristics before soil fertility treatments and after pasteurization
(70C for 4 h) were: pH 4.4 (soil:H20=l:2); 1.4% organic matter; 2, 65,
11, and 12 mg kg ^ (Mehlich-I extractable) of P, Ca, Mg and K,
respectively.
At planting three phosphorus levels were established by application
in solution of 12.5, 25, and 50 mg P kg 1 as Cai^PO^^.^O which is
equivalent to 25, 50, and 100 kg P ha ^, assuming a 15-cm depth of soil
ha-l with a bulk density of 1.3 g cm-^.
Glomus etunicatum (isolate S312) was isolated from carpon desmodium
at the Agricultural Research and Education Center, Ona, FL. (Chapter II,
Table 2-4). Fungal inoculum was produced in pot culture in pasteurized


15
Most of the literature concerning the association of VAM fungi with
forage legumes is on temperate species; e.g. alfalfa (Medicago sativa L.)
(Kucey and Diab, 1984; Nielsen and Jensen, 1983; Satterlee et al., 1983),
white clover (Trifolium repens L.) (Newbould and Rangeley, 1984; Powell,
1979; Rangeley et al., 1982), and subterranean clover (Trifolium
subterraneum L.) (Abbott and Robson, 1978). Studies on tropical forage
legumes have been limited to a few species such as tropical kudzu
(Pueraria phaseoloides Benth) (Salinas et al., 1985; Waidyanatha et al.,
1979), leucaena (Leucaena leucocephala Dewit) (Huang et al., 1985) and
Stylo (Stylosanthes guianensis SW.) (Mosse, 1977; Waidyanatha et al.,
1979).
The purpose of this investigation was to evaluate the effect of a
VAM fungus, G. intraradices, on the growth of several tropical forage
legumes in pasteurized and nonpasteurized soil under greenhouse
conditions.
Materials and Methods
The top 15 cm of a virgin Oldsmar fine sand (sandy, siliceous,
hyperthermic Alfic Haplaquods) was collected from a newly cleared area
at the Agricultural Research and Educational Center, Fort Pierce, FL.
The low-P soil was air-dried and sieved through a 4-mm screen. The soil
had an initial pH of 4.5 (1:2 soil:water suspension) and P, Ca, Mg, and
K concentrations (extracted with 0.05 M HC1 + 0.0125 M H2SO4) of 1, 63,
21 and 12 mg kg 1 soil, respectively. Lime, as high calcitic limestone,
was thoroughly incorporated at a rate of 1500 mg kg1 soil (equivalent


51
Mycorrhizal plants of Siratro had greater shoot dry weight, total P
and total N than nonmycorrhizal plants at low and intermediate levels of
applied P, but not at the highest level (Fig. 6-2). Overall, mycorrhizal
plants had greater root fresh weight and plant P concentration than
nonmycorrhizal plants. Other investigators (Lynd et al., 1985; Saif,
1987) have shown similar responses of Siratro to inoculation with VAM
fungi and P additions. Differences in shoot dry weight between
mycorrhizal and nonmycorrhizal plants of aeschynomene were only at the
intermediate level of applied P (Fig. 6-3). Fungal inoculation did not
affect the root fresh weight. Overall, mycorrhizal plants had greater P
concentration, total P, and total N than nonmycorrhizal plants
(Fig. 6-3).
Percentage and total root length colonized for mycorrhizal plants
of Stylo (Fig. 6-1), Siratro (Fig. 6-2), and aeschynomene (Fig. 6-3)
increased with the first addition of P. However, maximum colonization,
expressed as either percentage or total length of colonized root of the
three legumes, was attained at the intermediate levels of applied P.
The number of nodules in the three legumes increased with fungal
inoculation and P applications. Mycorrhizal plants of Stylo had more
nodules than nonmycorrhizal plants at all levels of applied P. However,
mycorrhizal plants of Siratro and aeschynomene had more nodules than
nonmycorrhizal plants only at 25 mg kg1 of applied P.
Mycorrhizal plants required between 38-40 mg P kg-1 to achieve
maximum shoot dry weight, whereas nonmycorrhizal plants required
50 mg P kg to produce approximately the same shoot dry weight, except
for nonmycorrhizal Stylo (Fig 6-1) which even with 50 mg P kg1 did not


CHAPTER VIII
CONCLUSIONS
The objective of this chapter is to summarize the work of the
preceding six chapters.
The overall goal of this research project was to improve the
establishment phases and growth of tropical forage legumes in newly
cleared land at reduced P fertilization through inoculation with
effective VAM fungi. In order to accomplish this goal, greenhouse
studies were carried out in limed, pasteurized and nonpasteurized Oldsmar
fine sand, collected from a newly cleared area at the Agricultural
Research and Education Center, Fort Pierce, FL. A field experiment was
conducted in the same nonpasteurized soil.
In Chapter II, quantitative data on the amount of root colonization
and the species distribution of VAM fungi associated with four cultivated
tropical forage legumes from four different locations in south Florida
are reported. Differences in percentage of root colonization and total
spore density were significant among locations, legume species, and
location x legume species interactions. Legume species differed in
percentage root colonization and total spore density among locations
except for carpon desmodium, which showed no differences among locations
in percentage root colonization. The six species of VAM fungi collected
in this survey were: G. heterogama, G. margarita, G. etunicatum, G.
intraradices, Glomus sp., and A. spinosa.
79


ROOT FRESH WEIGHT (g) SHOOT DRY WEIGHT (g)
38
P APPLIED (mg/kg)
CC o
/*>
s
tr
P APPLIED (mg/kg)
Fig. 5-1. Effect of P application on shoot dry weight, root fresh
weight, percentage of root colonized by VAM fungi, and
total root length colonized of Siratro grown in limed
nonpasteurized soil and inoculated with Glomus etunicatum
(ETU), Glomus intraradices (INT), or not inoculated (CON).


58
densities as determined by the MPN technique (Daniels and Skipper, 1982).
Approximately 180 propagules were added mid-way down each cell before
seeding. Control seedlings received 15 g of a soil-root mixture from
nonmycorrhizal pot cultures and the equivalent of 20 kg P ha ^, which
was applied in solution 10 d after germination in an attempt to make the
P status of the mycorrhizal and nonmycorrhizal seedlings similar at the
time of transplanting.
Seeded trays were placed in a glasshouse for 6 wk, after which the
whole cell content (each with one seedling) was transplanted to the
field. Siratro seedlings were cut back to three nodes each before
transplanting. Seedlings were transplanted on 4 August 1986.
One seedling was planted by hand in the middle of each 1.0 by 1.0 m
plot which were surrounded by alleyways of 1.0m. Extra seedlings were
weighed, and P concentration and root colonization were determined.
Harvests of Siratro and aeschynomene were made on 2 October 1986.
Siratro, a perennial, also was harvested on 27 November 1986, 5 May 1987,
and 29 June 1987. Herbage was dried at 75C for 24 h and weighed. Five
subsamples per treatment from the first harvest of Siratro and
aeschynomene foliage were analyzed for N and P content by automated
colorimetry (Technicon Industrial Systems Method No. 334-74 W/B,
Technicon Instruments Corp., Tarrytown, NY). Five root samples per
treatment, consisting each of four subsamples, were used to assess
mycorrhizal root colonization. Percentage of root colonized by VAM fungi
of aeschynomene and Siratro (1st and 4th harvests), was estimated as
described in Chapter III.


Fig. 7-2
Fig. 7-3
Fig. 7-4
Fig. 7-5
Effect of P application on shoot dry weight and
percentage of root colonized of Aeschynomene
americana grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices
(INT), or not inoculated (CON). Data points are
means of five replicates 65
Effect of P application on percentage of root
colonized for the first (A) and fourth (B)
harvest of Siratro grown in the field and
inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON).
Data points are means of five replicates 73
Effect of P application on P concentration,
total P, N concentration, and total N of
Siratro grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices
(INT), or not inoculated (CON). Data points are
means of five replicates 76
Effect of P application on P concentration,
total P, and total N of Aeschynomene americana
grown in the field and inoculated with Glomus
etunicatum (ETU), Glomus intraradices (INT),
or not inoculated (CON). Data points are
means of five replicates 77
x


71
Table 7-7. Analysis of variance for percentage root colonized of
Siratro by VAM fungi.
Source of
variation
DF
Mean
Harvest 1
2 Oct. 86
squares
Harvest 4
29 June 87
Block
4
9.27
14.61
Fungi (F)
2
2394.82
1206.47
Phosphorus (P)
3
935.53
1230.09
Linear (PI)
1
2195.71
2538.78
Quadratic (Pq)
1
552.25
1083.74
Cubic (Pc)
1
58.62
67.74
F x P
6
230.66**
86.82'*
F x PI
2
510.937'7'
174.04**
F x Pq
2
112.49**
77.26**
F x Pc
2
68.55'
9.16
Error
44
7.98
11.91
Significant at P < 0.05
Significant at P < 0.01


ROOT m.CHIZRTION (Z) ROO FRESH VE1Q1T (g) 3IOOI GFY WEIGHT (g)
52
9 _
e £
P APPLIED Cmg/kg)
Fig. 6 2. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Macroptilium
atropurpureum.


5
Fig. 2-1.
Collection sites for VAM fungi associated with four
tropical forage legumes in south Florida.


CHAPTER II
THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGI ON TROPICAL
FORAGE LEGUMES IN SOUTH FLORIDA.
Introduction
t j.. :.viT.' 1. u.;.* Ci -o r. : ..* W.
There is widespread interest in the use of tropical forage legumes
to increase production of tropical grasses in Florida's beef-cattle
industry (Snyder et al., 1985). These legumes respond to inoculation
with VAM fungi (Lynd et al., 1985; Saif, 1987)). However, before
initiating fungal inoculation experiments with these forage legumes in
south Florida, a survey was needed of the native populations of VAM fungi
associated with several commercial forage legumes growing on a variety
of soils.
Vesicular-arbuscular mycorrhizal associations have been observed in
a wide variety of natural and agricultural ecosystems (Abbott and Robson,
1977a; Currah and Van Dyk, 1986; Harley and Harley, 1987). In Florida,
the occurrence and distribution of VAM fungi in agronomic crops,
including some tropical legumes (Schenck and Kinloch, 1980; Schenck and
Smith, 1981), and sand-dune vegetation (Sylvia, 1986), has been reported.
However, there is no information on the degree of native VAM colonization
of tropical forage legumes in Florida or on the susceptibility of
different species of legumes to various genera and species of VAM fungi.
The objective of this survey was to obtain quantitative data on the
amount of root colonization and the species distribution of VAM fungi
3


87
Kormanik, P. P., and A. C. McGraw. 1982. Quantification of vesicular-
arbuscular mycorrhizae in plant roots, p. 37-45. In N. C. Schenck
(ed.) Methods and principles of mycorrhizal research. American
Phytopathological Society, St. Paul, MN.
Kretschmer, A. E., Jr. 1972. Siratro (Phaseolus atropurpureus DC.) A
summer-growing perennial pasture legume for central and south
Florida. Florida Agrie. Exp. Stn. Circ. S-214.
Kucey, R.M., and G.E. Diab. 1984. Effects of lime, phosphorus, and
addition of vesicular-arbuscular (VA) mycorrhizal fungi on
indigenous VA fungi and on growth of alfalfa in a moderately acidic
soil. New Phytol. 98: 481-486.
Lopes, E. S., and E. De Olivera. 1980. Efeito de species de micorrizas
vesicular-arbuscular em Siratro (Macroptilium atropurpureum).
Bragantia 39:241-245.
Lopes, E.S., E. Oliveira, R. Dias, and N.C. Schenck. 1983. Occurrence
and distribution of vesicular-arbuscular mycorrhizal fungi in coffee
(Coffea arabica L.) plantations in central Sao Paulo state, Brazil.
Turrialba 33:417-422.
Lynd, J. Q., R. J. Tyrl, and A. A. C. Purcino. 1985. Mycorrhiza soil
fertility effects on regrowth, nodulation and nitrogenase activity
of Siratro (Macroptilium atropurpureum (DC) Urb.). J. Plant Nutr.
8:1047-1059.
Menge, J. A. 1983. the physiology of vesicular-arbuscular mycorrhizal
fungi in agriculture. Can. J. Bot. 61:1015-1024.
Miller, R.H., E.J.B.N. Cardoso, and C.O.N. Cardoso. 1979. Some
observations on mycorrhizal infection of tropical forage legumes and
grasses in Brazil. Summa Phytopathologica 5: 168-172.
Miller, D. D., P. A. Domoto, and C. Walker. 1985. Colonization and
efficacy of different endomycorrhizal fungi with apple seedlings at
two phosphorus levels. New Phytol. 100:393-402.
Mosse, B. 1972. The influence of soil type and endogone strain on the
growth of mycorrhizal plants in phosphate deficient soil. Rev.
Ecol. Biol. Sol. T. 9:529-537.
Mosse, B. 1977. Plant growth responses to vesicular-arbuscular
mycorrhiza. X. Responses of stylosanthes and maize to inoculation
in unsterile soils. New Phytol. 78:277-288.
Mosse, B., C. L. Powell, and D. S. Hayman. 1976. Plant growth responses
to vesicular-arbuscular mycorrhiza. IX. Interactions between VA
mycorrhiza, rock phosphate and symbiotic nitrogen fixation. New
Phytol. 76:331-342.


35
growth and greatly enhanced tissue P content and nodulation. Growth
responses at other sites varied from large to slightly negative, probably
governed in part by the effectiveness of the indigenous VAM population
(Hayman and Hampson, 1979).
Species (Miller et al., 1985; Schubert and Hayman, 1986; Thompson et
al., 1986), and isolates within a species (Cooper, 1978), of VAM fungi
can colonize plants at different rates. If the mycorrhizal growth
response is related to the amount of early root colonization (Abbott and
Robson, 1981; Chapter IV), then isolates of VAM fungi that colonize roots
rapidly, at P levels found in established agricultural soils, may be most
suitable for pasture inoculation.
Schubert and Hayman (1986) indicated that, in order to achieve a
rational and effective use of inoculants, precise information on the
performance of endophytes in soil amended with P was necessary. It is
evident that the effect of soil P on symbiosis varies with the specific
host and endophyte. Therefore, more research is needed to develop
uniform and predictable endophyte-host responses.
In another study described in Chapter IV, G. etunicatum and G.
intraradices were found to be the most effective growth enhancers (out
of 5 isolates) of Siratro in a soil amended with a moderate level of P
and lime. In the present study, the objective was to evaluate the
infectivity and effectiveness of these two fungi over a practical range
of applied P.


39
(Table 5-1). Maximum yield of inoculated plants was achieved between 28
and 30 mg kg of applied P. Control plants had a linear relationship
for shoot dry weights and a quadratic relationship for root fresh
weights. There were fungus x P interactions for shoot dry and root fresh
weights (Table 5-2).
Percentage and total root length colonized by VAM fungi for the
inoculated treatments increased with P additions (Fig. 5-1). This effect
was greater for G. etunicatum than for G. intraradices. Phosphate
application did not alter the percentage and total root length colonized
in the control plants. Inoculated plants had a quadratic relationship
for percentage and total root length colonized and applied P
(Table 5-1). Maximum colonization of inoculated treatments, expressed
as either percentage or total length of colonized root, was attained
between 32 and 35 mg kg-^ of applied P. There were fungus x P
interactions for percentage and total root length colonized by VAM fungi
(Table 5-2).
Shoot dry weight of Siratro over the range of applied P was highly
correlated with percentage (r2 = 0.95") and total root length colonized
O V*. J,
by VAM fungi (rz = 0.97"') (Fig. 5-2) for both inoculated treatments.
Percentage of the root colonized by VAM fungi was very closely correlated
(r2 = 0.98*>') with the total root length colonized.
Growth enhancement from VAM inoculation at different levels of P has
been reported to vary with VAM fungi (Hayman and Hampson, 1979; Hayman
and Mosse, 1979; Schubert and Hayman, 1986; Thompson et al., 1986). For
example, Schubert and Hayman (1986) indicated that, when large amounts
of P were added (more than 100 mg kg *), G. mosseae, G. versiforme, G.


41
Table 5-2. Analysis of variance for shoot dry weights, root fresh
weights, and percentage and total root length colonized.
Source of
variation
DF
Shoot MS
root MS
root colonized MS
percentage length
Block
4
A,'.
0.07
0.04
7.39
0.069
Fungi (F)
2
11.07
18.01
2650.42
114.13
P rates (P)
3
34.19
59.37
1583.51
118.66
linear (PI)
1
85.40
146.43
4156.00
305.28
quadratic (Pq)
1
13.97
30.10
577.33
32.23
Cubic (Pc)
1
3.20
1.57
17.14,
2.48
F x P
6
1.65**
2.55'"'*
378.46""
22.50**
F x PI
2
2.58"
4.02**
965.54**
55.69**
F x Pq
2
2.29**
2.08**
159.17**
7.82""
F x Pc
2
0.08
1.55
10.67
3.99
Error
44
0.02
0.13
6.27
0.24
**
Significant at P < 0.01
MS = mean square


81
either G. etunicatum or G. intraradices weighed more than control plants.
Inoculated plants required between 28 and 30 mg P kg ^ to achieve maximum
shoot dry weight, whereas control plants, even with 40 mg P kg 1, did
not achieve maximum growth. Shoot dry weight response was better with
G. etunicatum than with G. intraradices. For both fungi, increasing P
above 2.5 mg kg 1 increased the percentage and total root length
colonized by VAM fungi.
In Chapter VI, the effect of inoculation with G. etunicatum, on the
growth and uptake of P and N of three forage legumes at applied P levels
between 12.5 and 50 mg kg 1 in a limed, pasteurized soil under greenhouse
conditions is reported. At all levels of applied P, shoot dry and root
fresh weights, and total N of Stylo were greater for mycorrhizal plants
than nonmycorrhizal plants. The differences were most pronounced at
intermediate levels of applied P and diminished at the higher P levels.
Mycorrhizal plants of Siratro had greater shoot dry weight, total P and
N than nonmycorrhizal plants at low and intermediate P levels, but not
at the highest level. Differences in shoot dry weight between
mycorrhizal and nonmycorrhizal plants of aeschynomene were significant
only at the intermediate level of applied P. Overall, inoculated plants
of the three legumes studied had greater P concentration than control
plants. Maximum root colonization, expressed as either percentage or
total length of colonized root of the three legumes, was attained at
intermediate levels of applied P.
In Chapter VII, the effect of inoculation with selected VAM isolates
on growth and nutrient uptake of Siratro and aeschynomene under natural
field conditions at applied P levels of 10, 30, 60, and 120 kg ha'1 is


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS .' iii
LIST OF TABLES vi
LIST OF FIGURES viii
ABSTRACT xi
CHAPTER I
..INTRODUCTION 1
CHAPTER II
THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL
FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA 3
Introduction 3
Materials and Methods A
Results and Discussion 7
CHAPTER III
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH GLOMUS INTRARADICES 14
Introduction 1A
Materials and Methods 15
hours vi Results, and Discussion ,,:M.. ...... ^ ... .Jc. ... 1,7
CHAPTER IV
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I.
SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI
UNDER AMENDED SOIL CONDITIONS 23
Introduction 23
iv


31
this soil for satisfactory establishment and growth of legumes (Snyder
et al., 1985), VAM fungi must be selected for their effectiveness under
amended soil conditions.
Powell (1980a) reported a relationship between the level of native
inoculum density in the soil and plant growth response to mycorrhizal
inoculation. When inoculum density was low (0.01-0.09 propagules g
soil), there was a significant plant growth response to inoculation with
VAM fungi. When inoculum density was higher (0.15-0.30 propagules g ^
soil), there was little plant growth response to fungal inoculation.
Likewise, good plant growth responses to inoculation with VAM fungi in
soils with few indigenous endophytes have been reported by Mosse (1977)
and Hall (1979). Thus it would seem that the most promising sites for
inoculation with VAM fungi are those where indigenous populations of VAM
fungi are very low. However, in this study, where the native inoculum
density was relatively high (2 propagules g ^ soil), Siratro responded
to inoculation with two of the four VAM fungi tested. In addition to the
abundance of the indigenous VAM fungi, information about their
infectivity and effectiveness is needed to assess potential sites for
responsiveness to inoculation with effective VAM fungi.
The ineffectiveness of G. versiforme and G. margarita could be due
to an innate symbiotic inefficiency, incompatibility, lack of
competitiveness, or to inhibitory edaphic (e.g. soil pH or P level) or
environmental factors (e.g. light and temperature). Hayman and Tavares
(1985) showed clearly that different endophytes vary in their symbiotic
effectiveness at different soil acidities. In addition, some endophytes
may be less effective on certain plant hosts. For example,


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION
By
Onesimo A. Medina
December 1987
Chairman: Dr. D. M. Sylvia
Cochairman: Dr. A. E. Kretschmer, Jr.
Major Department: Soil Science
Greenhouse and field studies were conducted to determine the growth
response of tropical forage legumes to inoculation with vesicular-
arbuscular mycorrhizal (VAM) fungi and P applications and to evaluate
the effectiveness of the indigenous VAM population versus introduced
species.
Root and rhizosphere soil samples of four tropical forage legume
species were collected at four locations in south Florida before
initiating the greenhouse experiments. Six species of VAM fungi were
isolated in this survey. The occurrrence of VAM fungal species, as
determined by spore numbers, was affected by legume species and location.
Shoot dry and root dry weights of 'Siratro' (Macroptilium
atropurpureum Urb.), aeschynomene (Aeschynomene americana L.),
xi


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.
Dr^ G. H. Siiyder
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.
A
C'2^S^c/'^CyK~
i, v.
Dr. N. C. Schenck
Professor of Plant Pathology
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December 1987
Dean, Graduate School


40
Table 5-1. Regression equations and coefficients of determination (r2)
showing the relationship of P level to shoot dry weights,
root fresh weights, percentage and total root length
colonized.
Variable
Regression equations
r2
Shoot dry weight (g)
Glomus etunicatum
= -0.12+0.34P-0.006P2
0.97**
Root fresh weight (g)
= 0.50+0.40P-0.006P2
0.96';'*
Root colonization (%)
= 4.47+2.62P-0.04P2
0.97**
Root length colonized
(m)
= -1.43+0.64P-0.009P2
0.96**
Shoot dry weight (g)
Glomus intraradices
= 0.16+0.24P-0.004P2
0.95**
Root fresh weight (g)
= 0.55+0.38P-0.006P2
0.94**
Root colonization (%)
= 5.42+1.86P-0.03P2
0.95**
Root length colonized
(m)
= -1.08+0.49P-0.006P2
0.91
Shoot dry weight (g)
Control
= 0.72+0.06P
0.93"*
Root fresh weight (g)
= 0.89+0.17P-0.002P2
0.92**
**significant at P < 0.01
P = phosphorus level


' v r.
CHAPTER IV
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN
NONPASTEURIZED SOIL. I. SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI
UNDER AMENDED SOIL CONDITIONS.
Introduction
Siratro, a cultivar developed by E. M. Hutton (1962) from two Mexican
accessions of Macroptilium atropurpureum Urb., is a persistent, perennial
forage legume adaptable to a wide range of soil and climatic conditions.
It has become widespread and is among the most versatile forage legume
grown throughout tropical regions of the world (Lynd et al., 1985).
In pasteurized and nonpasteurized soils, increased growth of Siratro was
attained after inoculation with Glomus fasciculatum Gerdemann & Trappe
(Lopes and De Olivera, 1980; Lynd et al., 1985) and G. intraradices
(Chapter III).
Hayman (1982) stated that VAM fungi are probably capable of symbiosis
with most plants, at least to some degree. However, there is wide
variation in the ability of VAM fungi to stimulate plant growth (Miller
et al., 1985; Powell, 1982; Schubert and Hayman, 1986). Lopes and De
Olivera (1980), using a gamma-irradiated soil of low P-content, studied
the effect of inoculation with nine species of VAM fungi on the growth
of Siratro. Only inoculation with G. fasciculatum and G. macrocarpum
enhanced plant growth. Abbott and Robson (1981) defined the relative
ability of a VAM fungus to stimulate plant growth as 'effectiveness1 and
23


60
Table 7-1. Shoot dry weights, P concentrations, and percentage of
roots colonized of Siratro and aeschynomene seedlings at
transplanting.
VAM
Inoculation
Siratro
aeschynomene
Shoot dry
wt.
Shoot
P
Root
colon.
shoot dry
wt.
Shoot Root
P colon.
mg
--- %

mg
%
G. etunicatum
302
.18
52
304
.17 57
G. intraradices
305
.17
60
299
.20 55
Control
298
.16
0
302
.18 0
zData are means of five replicates.


19
Table 3-1. Percentage of mycorrhizal root colonization of tropical
forage legumes in nonpasteurized (UP) or pasteurized (P)
soil in the greenhouse after 45 days.
Legume species
Mycorrhizal
inoculation
Root colonization
UP P
%
Aeschynomene americana
+
22
3
5
0
Macroptilium atropurpureum
+
49
35
20
0
Aeschynomene villosa
+
10
*
6
Stylosanthes hamata
+
12
A
6
Stylosanthes guianensis
+
28
VC
8
Vigna adenantha
+
59
53
-
40
0
Leucaena leucocephala
+
4
3
1
0
Arachis sp.
+
31
12
- -
35
2
zBased on a composite of three samples for each legume.
Treatment lost to glasshouse accident.


25
(MPN) technique using bahiagrass as the host plant and pasteurized
Oldsmar fine sand as the diluent (Daniels and Skipper, 1982). The amount
of inoculum used was adjusted to give equal inoculum densities among
isolates. Each pot received approximately 240 propagules. Details on
the fungal inoculation technique, planting, and watering were reported
previously (Chapter III).
There were six inoculation treatments, five species of VAM fungi, and
a control inoculated with non-VAM pot culture material. The pots were
arranged on the greenhouse bench in a completely randomized block design
with 15 replications per treatment.
The average maximum and minimum greenhouse temperatures during the
experimental period were 32 and 19C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1200 p mol m s
Five randomly selected samples were harvested from each treatment
after 20, 40, and 70 d of growth. At first harvest, shoot dry weight,
percentage of root colonized by VAM fungi, plant height, and number of
leaves were determined. In addition, root fresh weight and total root
length colonized were determined at the second and third harvests. Shoot
dry weight was determined by drying the material at 70C for 24 h.
Percentage and total root length colonized were estimated by the gridline
intersect method (Giovannetti and Mosse, 1980) after roots were cleared
in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik
and McGraw, 1982). Data were analyzed by Analysis of Variance Procedure,
Statistical Analysis Systems (SAS Institute Inc., 1982). Duncan's
multiple range test was used to separate treatment means when the F-test
was significant (P < 0.05).


57
Materials and Methods
The soil used was a native Oldsmar fine sand (sandy, siliceous,
hyperthermic Alfic Arenic Haplaquods) with a pH of 4.6 (soil:^0=1:2),
1.5% organic matter, and the following Mehlich-I extractable elements in
mg kg : P 1.0, Ca 65, Mg 12, and K 15.
The experiment was designed as a 2 x 3 x 4 factorial consisting of
two legume species; Siratro and A. americana; three inoculation
treatments, G. etunicatum, G. intraradices, and the control; and four P
treatments; 10, 30, 60, and 120 kg ha 1 as triple superphosphate. The
24 treatments were arranged in a randomized complete block design with
ten replications per treatment. The P treatments were surface applied
by hand on 3 July 1986, along with a basal application of lime (high
calcitic limestone), Mg, nutritional spray (Diamond Fertilizer Co., Ft.
Pierce, FL.), and Mo at 3000, 25, 22, and 0.2 kg ha \ respectively, and
incorporated using a rake to a depth of approximately 15 cm. Potassium
was broadcasted on each plot at a rate of 60 kg ha 1 as KCL on 5 August
1986.
Seeds of Siratro and aeschynomene inoculated with rhizobium type El
"Cowpea" inoculum (Nitragin Co., Milwaukee, WI) were sown in pasteurized
Oldsmar fine sand, amended with high calcitic limestone at 1500 mg kg'1
and P at 12.5 mg kg \ in cells of "speedling" styrofoam trays (72 cells
per tray) on 20 June 1986.
Seedlings were inoculated or not inoculated with G. etunicatum
(isolate S312), G. intraradices (isolate S311). The amount of soil-root
inoculum used for each VAM fungi was adjusted to give equal inoculum


90
Schenck, N.C., and G.S. Smith. 1982. Additional new and unreported
species of mycorrhizal fungi (Endogonaceae) from Florida. Mycologia
74: 77-92.
Schroder, V., D. S. Hayman, and B. Mosse. 1977. Influence of VA
mycorrhiza on plant growth. Rothamsted Report for 1976, part 1, p.
286.
Schubert, A., and D. S. Hayman. 1986. Plant growth responses to
vesicular-arbuscular mycorrhiza. XVI. Effectiveness of different
endophytes at different levels of soil phosphate. New Phytol.
103:79-90.
Snyder, G. H., A. E. Kretschmer, Jr., and J. Alvarez. 1985. Agronomic
and economic response of three tropical legumes to lime and
phosphorus in an acid infertile spodosol. Agron. J. 77:427-432.
Sylvia, D.M. 1986. Spatial and temporal distribution of vesicular-
arbuscular mycorrhizal fungi associated with Unila paniculata in
Florida foredunes. Mycologia 78: 728-734.
Thompson, B. D., A. D. Robson, and L. K. Abbott. 1986. Effects of
phosphorus on the formation of mycorrhizas by Gigaspora calospora
and Glomus fasciculatum in relation to root carbohydrates. New
Phytol. 103:751-765.
Trappe, J. M. 1982. Synoptic keys to the genera and species of
zygomycetous (vesicular-arbuscular) mycorrhizal fungi.
Phytopathology 72: 1102-1108.
Waidyanatha, U. P. De S., N. Yogaratnan, and W. A. Ariyaratne. 1979.
Mycorrhizal infection on growth and nitrogen fixation of Pueraria
and Stylosanthes and uptake of phosphorus from two rock phosphates.
New Phytol. 82:147-152.
Wilson, J. M. 1984. Competition of infection between vesicular-
arbuscular mycorrhizal fungi. New Phytol. 97:427-435.
Yost, R.S., and R.L. Fox. 1979. Contribution of mycorrhizae to P
nutrition of crops growing on an oxisol. Agron. J. 71: 903-908.


TOTFL FLFNT P (mg) FLflNT P CCNC. C)
76
P PPPLIH3 (kg/ha)
P fffl-ia (Ig/ha)
Effect of P application on P concentration, total P, N
concentration, and total N of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.
Fig. 7-4.


ROOT LENGTH COLONIZED (m) ROOT COLONIZATION (%)
28
Fig. 4-2 Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus
intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the percentage of root colonization and root
length colonized of Siratro at three and two harvests,
respectively. Bars represent the means of five replicates.
Means with the same letter within a harvest are not
different (P < 0.05).


CHAPTER III
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH
GLOMUS INTRARADICES -- ---
Introduction
The use of forage legumes as companion crops to increase production
of grasses is becoming an established practice in order to reduce the
requirement for N fertilization (Rotar, 1983). In soils where P is a
limiting factor, large applications of P fertilizer are required for
legume establishment and optimum growth. However, with the increasing
cost of P fertilizer, alternative strategies for minimum input and
efficient use of P must be adopted. It is pertinent, therefore, to
evaluate whether mycorrhizal associations with forage legumes can be
manipulated in order to improve establishment, P nutrition, N2-fixation,
and consequently yield.
Vesicular-arbuscular mycorrhizal fungi can improve the growth of
legumes by increasing P uptake (Bethlenfalvay et al., 1985; Chulan and
Ragin, 1986; Harley and Smith, 1983; Hayman, 1983; Jensen, 1984).
Phosphorus is often a growth-limiting factor since many legumes have high
P requirements and are poor scavengers of P. The VAM fungi may also
increase nodulation and N2_fixation of legumes, primarily as an indirect
effect of improved P nutrition (Daft and El-Giahmi, 1976; Habte and Aziz,
1985; Newbould and Rangeley, 1984).
14


67
Table 7-5. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weights for the second, third, and fourth harvest and
percentage of root colonization for the fourth harvest of
Siratro.
Variable
Regression equation
r2
Glomus etunicatum
Shoot dry wt (g), harvest
2
=
5.7A+0.52P-0.0029P2
0.87**
Shoot dry wt (g), harvest
3
=
0.99+0.A8P-0.0029P2
0.89^
Shoot dry wt (g), harvest
A
=
5.98+0.55P-0.0032P2
0.87"*
Root colon. (%), harvest
A
=
9.85+0.8AP-0.00A6P2
0.86x*
Glomus intraradices
Shoot dry wt (g), harvest
2
=
6.58+0.A6P-0.0025P2
0.8A**
Shoot dry wt (g), harvest
3
=
2.05+0.A3P-0.0025P2
0.86**
Shoot dry wt (g), harvest
A
=
6.1A+0.52P-0.0027P2
0.87**
Root colon. (%), harvest
A
=
15.03+0.69P-0.0039P2
0.89**
Control
Shoot dry wt (g), harvest
2
=
1.56+0.AAP-0.0023P2
0.81**
Shoot dry wt (g), harvest
3
=
-0.23+0.A2P-0.0021P2
0.88**
Shoot dry wt (g), harvest
A
3.38+0.A2P-0.0017P2
0.91**
"''Significant at P < 0.01
P = phosphorus level


GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH
VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION
By
ONESIMO A. MEDINA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987

To the memory of my father Efrain, who influenced me in a very
special way. I am very heartbroken that he died before this work was
completed.
To my mother, Guillermina, for her never ending sacrifices, her
love, and prayers.

ACKNOWLEDGMENTS
It is with deep gratitude that I express thanks to the chairman and
cochairman of my committee, Dr. David M. Sylvia and Dr. Albert E.
Kretschmer, Jr., respectively, for their support, constant encouragement,
guidance, and friendship. I also thank the other members of my
committee, Dr. G. H. Snyder, Dr. G. Kidder, Dr. N. C. Schenck, and Dr.
J. B. Sartain, for their suggestions, support, and editorial comments.
Gratitude is extended to Mr. Tom Wilson for his friendship and
valuable assistance rendered in the field portion of this project.
The moral support, love, and motivation of my brothers, Oquendo,
Miosotis, and Gagarin were essential to the completion of my graduate
program.
This research was funded in part through USDA ARS Tropical
Agricultural Development Grants 83-CRSR-2-2134 and 86-CRSR-2-2846. This
support is greatly appreciated.
Most of all, warmest thanks go to my wife, Griselda, for her
understanding and encouragement during my graduate studies. Her many
hours of assistance in typing the manuscript will always be remembered.
I also thank my daughter Michelle for making her Mommy and Daddy very
happy.
iii

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS .' iii
LIST OF TABLES vi
LIST OF FIGURES viii
ABSTRACT xi
CHAPTER I
..INTRODUCTION 1
CHAPTER II
THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL
FUNGI ON TROPICAL FORAGE LEGUMES IN SOUTH FLORIDA 3
Introduction 3
Materials and Methods A
Results and Discussion 7
CHAPTER III
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH GLOMUS INTRARADICES 14
Introduction 1A
Materials and Methods 15
hours vi Results, and Discussion ,,:M.. ...... ^ ... .Jc. ... 1,7
CHAPTER IV
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. I.
SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI
UNDER AMENDED SOIL CONDITIONS 23
Introduction 23
iv

Materials and Methods
24
Results and Discussion
26
CHAPTER V
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II.
EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT
DIFFERENT P LEVELS
Introduction
Materials and Methods 36
Results and Discussion 37
34
: i
34
CHAPTER VI
EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE
GROWTH AND UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM,
STYLOSANTHES GUIANENSIS. AND AESCHYNOMENE AMERICANA 45
Introduction 45
Materials and Methods . 46
Results and Discussion 48
CHAPTER VII
GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND
AESCHYNOMENE AMERICANA TO INOCULATION WITH SELECTED VA
' MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS 55
Introduction 55
Materials and Methods 57
Results and Discussion 59
CHAPTER VIII
Result :.iid D- sou* w
CONCLUSIONS ,
LITERATURE CITED
BIOGRAPHICAL SKETCH -
SELECT!OF OF EFFECTIVE .'A VCO: KHJ..AL M.'rJfF.
79
83
91
v

LIST OF TABLES
, Page
Table 2-1. Chemical characteristics of the soils sampled
for VAM fungi associated with four tropical
forage legumes at four locations in south
Florida. . 8
Table 2-2. Analysis of variance for percentage of
mycorrhizal root colonization and total spore
density per 100 g of air-dried soil 9
Table 2-3. Mean percentage of mycorrhizal root colonization
and total spore density of VAM fungi among forage
legumes at four locations in south Florida 10
Table 2-4. Mean spore numbers of VAM fungal species
associated with four forage legumes at four
locations in south Florida 12
Table 3-1. Percentage of mycorrhizal root colonization of
tropical forage legumes in nonpasteurized (UP)
or pasteurized (P) soil in the greenhouse after
45 days. .! 19
i-fvv T'5.- ~r' '** "y. h rr*i' ~ f ;>
Table 5-1. Regression equations and coefficients of
determination (r^) showing the relationship
of P level to shoot dry weights, root fresh
weights, percentage and total root length
colonized 40
Table 5-2. Analysis of variance for shoot dry weights, root
1-*' fresh weights, percentage and total root
length colonized 41
Table 6-1. Mean squares and levels of significance from
the analysis of variance for shoot dry weight,
root fresh weight, P concentration, and total P
- !and.N uptake of forage legumes. ........... 49
Table 7-1. Shoot dry weights, P concentrations, and percentage
of roots colonized of Siratro and Aeschynomene
americana seedlings at transplanting 60
vi

Table 7-2. Analysis of variance for shoot dry weights of
Aeschynomene americana harvested 2 October 1986. . 61
Table 7-3. Analysis of variance for shoot dry weights
from four Siratro harvests. 63
Table 7-4. Regression equations and coefficients of
determination (r^) showing the relationship
of applied P level to shoot dry weight, percentage
of root colonization, P and N concentrations, and
total P and N uptake for the first harvest
of Siratro 66
Table 7-5. Regression equations and coefficients of
determination (r~) showing the relationship
of applied P level to shoot dry weights for
the first, third, and fourth harvest and
percentage of root colonization for the
fourth harvestof Siratro 67
Table 7-6. Regression equations and coefficients of
determination (r~) showing the relationship
of applied P level to shoot dry weight,
percentage of root colonization, P concentration,
and total P and N uptake of Aeschynomene
americana. . 68
Table 7-7. Analysis of variance table for percentage root
colonized of Siratro by VAM fungi 71
Table 7-8. Analysis of variance table for percentage root
colonized, P concentration, total P, and total
N of Aeschynomene americana. ...... 72
Table 7-9. Analysis of variance table for P concentration,
total P, N concentration, and total N of
Siratro for the first harvest 75
C

LIST OF FIGURES
Fig. 2-
Fig. 3-
Fig. 3-2
Fig. 4-1
t/.K \ t lac-. i : .3 c ~c"rT -r_;
1. Collection sites for VAM fungi associated with
four tropical forage legumes in south Florida. .
1. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in nonpasteurized soil in the
greenhouse after 45 days. Legume species were
Aeschynomene americana (AA), Aeschynomene
villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL),
Stylosanthes hamata (SH), Stylosanthes guianensis
(SG), and Vigna adenantha (VA). Bars represent
the mean of 3 replicates. Means with the same
letter within a species are not different
(P < 0.05)
. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in pasteurized soil in the
greenhouse after 45 days. The legume species
were Aeschynomene americana (AA), Arachis sp.
(AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA).
Bars represent the mean of 3 replications.
Means with the same letter within a species are
not different (P<0.05)
. Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum
(ETU), or the control (CON) on the shoot dry weight
and root fresh weight of Siratro at two harvests.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05)
Page
5
18
21
. 27
viii

Fig. 4-2
Fig. 4-3
Fig. 5-1.
Fig. 5-2.
Fig. 6-1.
Fig. 6-2.
Fig. 6-3.
Fig. 7-1.
Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum
(ETU), or the control (CON) on the percentage
of root colonization and root length colonized
of Siratro at three and two harvests, respectively.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05) 28
Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments 30
Effect of P application on shoot-dry weight,
root fresh weight, percentage of root colonized
by VAM fungi, and total root length colonized
of Siratro grown in limed nonpasteurized soil
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). 38
Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments in
nonpasteurized soil 42
Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Stylosanthes guianensis 50
Effect of fungal inoculation and P applications on
the shoot dry weight, root-fresh weight, root
colonization, P concentration, total P, and
total N of Macroptilium atropurpureum 52
Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Aeschynomene americana. 53
Effect of P application on shoot dry weights
for the first (A), second (B), third (C), and
fourth (D) harvest of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON).
Data points are means of five replicates 64
xx

Fig. 7-2
Fig. 7-3
Fig. 7-4
Fig. 7-5
Effect of P application on shoot dry weight and
percentage of root colonized of Aeschynomene
americana grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices
(INT), or not inoculated (CON). Data points are
means of five replicates 65
Effect of P application on percentage of root
colonized for the first (A) and fourth (B)
harvest of Siratro grown in the field and
inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON).
Data points are means of five replicates 73
Effect of P application on P concentration,
total P, N concentration, and total N of
Siratro grown in the field and inoculated
with Glomus etunicatum (ETU), Glomus intraradices
(INT), or not inoculated (CON). Data points are
means of five replicates 76
Effect of P application on P concentration,
total P, and total N of Aeschynomene americana
grown in the field and inoculated with Glomus
etunicatum (ETU), Glomus intraradices (INT),
or not inoculated (CON). Data points are
means of five replicates 77
x

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO
INOCULATION WITH VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION
By
Onesimo A. Medina
December 1987
Chairman: Dr. D. M. Sylvia
Cochairman: Dr. A. E. Kretschmer, Jr.
Major Department: Soil Science
Greenhouse and field studies were conducted to determine the growth
response of tropical forage legumes to inoculation with vesicular-
arbuscular mycorrhizal (VAM) fungi and P applications and to evaluate
the effectiveness of the indigenous VAM population versus introduced
species.
Root and rhizosphere soil samples of four tropical forage legume
species were collected at four locations in south Florida before
initiating the greenhouse experiments. Six species of VAM fungi were
isolated in this survey. The occurrrence of VAM fungal species, as
determined by spore numbers, was affected by legume species and location.
Shoot dry and root dry weights of 'Siratro' (Macroptilium
atropurpureum Urb.), aeschynomene (Aeschynomene americana L.),
xi

Aeschynomene villosa Poir., Stylo (Stylosanthes guianensis SW.), and
Stylosanthes hamata Taub. were increased in pasteurized and
nonpasteurized limed soil in the greenhouse after inoculation with Glomus
intraradices Schenck & Smith.
Inoculation with Glomus etunicatum Becker & Gerdemann and G.
intraradices also increased the growth of Siratro as compared to other
VAM fungi and the noninoculated control in limed, nonpasteurized soil
fertilized with 20 mg kg ^ of P. In other greenhouse experiments, G.
etunicatum and G. intraradices were effective growth enhancers of Siratro
over a practical range of 2.5 to 40 mg kg ^ of applied P in a limed,
nonpasteurized soil. For both fungi, increasing P above 2.5 mg kg ^
increased the percentage and total root length colonized by VAM fungi.
A positive correlation was found between mycorrhizal root colonization
and shoot dry weight. In a limed, pasteurized soil, inoculation with
G. etunicatum increased total P and N of Siratro at 12.5 and 25 mg kg ^
-1
of applied P, but not at 50 mg kg
' .. -s
The effectiveness of G. etunicatum and G. intraradices with Siratro
and aeschynomene was corroborated in a field trial. These fungi
increased the growth and uptake of P and N of both legumes over a range
of applied P from 10 to 80 kg ha
Inoculation of forage legumes with effective VAM fungi enhanced
their growth. Growth enhancement occurred at P and lime levels used in
commercial pasture production and in soils that had a large, but
apparently ineffective indigenous VAM population.
xii

CHAPTER I
INTRODUCTION
-Forage legumes are an important component of improved grass pastures
and must be established rapidly and without excessive cost. The legumes
serve both to increase forage quality and decrease the need for N
fertilizer through ^fixation.
Newly cleared lands incorporated into pasture production in south
Florida are generally acidic and very low in total and available P
throughout the soil profiles. While improvements to the productivity of
these pastures may be obtained by the introduction of suitable legumes,
effective N2~fixation and establishment:of legumes is frequently limited
by the_low levels of available P in these soils. Snyder et al. (1978)
reported that large applications of P fertilizer are normally required
for legume establishment and optimun growth in these soils. However, .
with the increasing cost of P fertilizer, alternative strategies for
minimum P fertilizer input and efficient use of P must be adopted. One
of these strategies may be via the management of vesicular-arbuscular
mycorrhizal (VAM) symbioses.
r.n -i VeSi ciliar arbuscular mycorthizal fungi can improve the growth of
legumes by increasing P uptake (Bethlenfalvay et al., 1985; Hayman,
1983). Phosphorus is often a growth-limiting factor since many legumes
have P requirements and are poor scavengers of P. The VAM fungi may also
increase nodulation and ^-fixation of legumes, primarily as an indirect
1

2
effect of improved P nutrition (Daft and ^i-Giahmi, 1976; Habte and Aziz,
1985).
Information concerning the association of VAM fungi with tropical
forage legumes is sparse. Most of the growth response studies reported
were done in either pasteurized soil or in small volumes of
nonpasteurized soil. Except for the work of Saif (1987), little
information is available on growth response of tropical forage legumes
to inoculation with VAM fungi in nonpasteurized soil, especially under
field conditions where introduced species of VAM fungi must compete with
the indigenous VAM population.
Therefore, greenhouse studies were conducted in limed, pasteurized
and nonpasteurized soil to improve the growth of tropical forage legumes
through inoculation with effective VAM fungi and reduced P fertilization.
In addition, the effect of inoculation with selected VAM isolates on
growth and nutrient uptake of two tropical forage legumes under natural
field conditions was investigated at different levels of applied P.
have
lh
L so

CHAPTER II
THE OCCURRENCE OF VESICULAR-ARBUSCULAR MYCORRHIZAL FUNGI ON TROPICAL
FORAGE LEGUMES IN SOUTH FLORIDA.
Introduction
t j.. :.viT.' 1. u.;.* Ci -o r. : ..* W.
There is widespread interest in the use of tropical forage legumes
to increase production of tropical grasses in Florida's beef-cattle
industry (Snyder et al., 1985). These legumes respond to inoculation
with VAM fungi (Lynd et al., 1985; Saif, 1987)). However, before
initiating fungal inoculation experiments with these forage legumes in
south Florida, a survey was needed of the native populations of VAM fungi
associated with several commercial forage legumes growing on a variety
of soils.
Vesicular-arbuscular mycorrhizal associations have been observed in
a wide variety of natural and agricultural ecosystems (Abbott and Robson,
1977a; Currah and Van Dyk, 1986; Harley and Harley, 1987). In Florida,
the occurrence and distribution of VAM fungi in agronomic crops,
including some tropical legumes (Schenck and Kinloch, 1980; Schenck and
Smith, 1981), and sand-dune vegetation (Sylvia, 1986), has been reported.
However, there is no information on the degree of native VAM colonization
of tropical forage legumes in Florida or on the susceptibility of
different species of legumes to various genera and species of VAM fungi.
The objective of this survey was to obtain quantitative data on the
amount of root colonization and the species distribution of VAM fungi
3

4
associated with four cultivated tropical forage legumes from four
different locations in south Florida.
Materials and Methods
Root and rhizosphere soil samples of four tropical forage legumes
were collected from 11 to 17 October 198A, at four locations in south
Florida: Deseret Ranches, Deer park; Fort Pierce, Agricultural Research
and Education Center (AREC); Ona, AREC; and Basinger Ranch, 109 Ranch
(Fig. 2-1). Most of the soils of the studied area belong to the order
Spodosols. They are dominated by nearly level, somewhat poorly to poorly
drained sandy soils with dark sandy subsoil layers. These soils are used
primarily for pastures, vegetables, flowers, forestry, and citrus.
The forage legumes sampled were: 'Siratro' (Macroptilium
atropurpureum Urb.), (except at Deseret Ranches), aeschynomene
(Aeschynomene americana L.), Vigna adenantha Marechal, Mascherpa and
Stainier, and carpon desmodium (Desmodium heterocarpon DC.). The legumes
were mixed with pasture grasses at the time of sampling. Three
rhizosphere soil samples were collected to a depth of 15 cm for each
legume at each location. Samples, consisting of three subsamples of
approximately 1.5 kg, were placed in plastic bags and transported to the
lu.vt- . i: v*.
laboratory on the same day.
Samples were sieved through a 4-mm screen, and 100 g subsamples were
removed and stored at 5C for spore extraction. Legume roots were
carefully separated manually from grass roots. A portion (0.5 g) of each

5
Fig. 2-1.
Collection sites for VAM fungi associated with four
tropical forage legumes in south Florida.

6
root sample was cleared in 10% KOH and stained with 0.05% trypan blue in
lactophenol (Kormanik and McGraw, 1982). Root colonization by VAM fungi
was estimated by the gridline-intersect method of Giovannetti and Mosse
(1980).
Chemical content of a composite soil sample from each location was
determined by the Soil Testing Laboratory, University of Florida (Rhue
and Kidder, 1984). Mehlich-I solution (0.05 M HC1 + 0.0125 M H2SO4) was
used to extract Al, Ca, K, Mg, and P. All elements were analyzed in the
filtrate by atomic absorption spectrophotometry, except P which was
determined using the ammonium molybdate/ascorbic acid colorimetric
method. Soil pH was determined using a 1:2 (v/v) soilrwater ratio.
Organic matter was estimated by oxidation with 1 N ^0^0-7 in the
presence of H2SO4.
Spores of VAM fungi were removed from soil by the wet sieving method
of Daniels and Skipper (1982) using sieves with 425, 90, and 45 um
openings. Fractions retained on 90 and 45 um sieves were centrifuged
(1000 x g) for 3 min in water. The pellet was resuspended in 40% sucrose
solution and centrifuged for 1.5 min. Spore species were identified
where possible (Schenck and Smith, 1982; Trappe, 1982). In addition,
spores or washed roots were placed in pasteurized Arredondo loamy sand
surface soil (siliceous hyperthermic Grossarenic Paleudult) in 15-cm-diam
plastic pots in the greenhouse and planted with bahiagrass (Paspalum
notatum Flugge), carpon desmodium, or Siratro in an attempt to isolate
VAM fungi in a manner similar to that described by Gerdemann and Trappe
(1974) as the "inoculated pot culture" method.

7
Results and Discussion j : n
Results of soil pH and chemical analysis of soil samples reflect
the different management regimes (including lime and fertilizer)
(Table 2-1). 'r .'.'.on war
Differences in percentage of mycorrhizal root colonization and total
spore density of air dry soil were significant among locations, legume
species, and location x legume species interactions (Table 2-2). Total
spore density at the four locations ranged from 5 to 679 per 100 g of
air-dried soil, and the percentage of mycorrhizal root colonization
varied from 3 to 41%. Miller et al. (1979) observed variable degree of
mycorrhizal root colonization (4 to 74%) in forage grasses and legumes
in Brazil. Except for carpon desmodium, legume species differed in
percentage root colonization and total spore density among locations
(Table 2-3). Fort Pierce had the highest total spore density for each
legume species sampled except for Siratro.
Attempts were made to relate percentage root colonization and total
spore density to soil P or the other soil chemical characteristics
presented in Table 2-1, but no clear relationships were apparent. Abbott
and Robson (1977a) and Hayman (1978) also reported that spore numbers
were not correlated with soil P or soil pH in cultivated soils.
There was a positive correlation (P < 0.05) between root
colonization and total spore density for all legume species at Basinger
(r = 0;70) and Deseret (r =0.76), but not at Fort Pierce and Ona.
Giovannetti (1985) and Miller et al. (1979) reported a correlation

8
Table 2-1. Chemical characteristics of the soils sampled for VAM
fungi associated with four tropical forage legumes at four
locations in south Florida.
Location
Legume
species2
O.M.
PH
A1
Ca
Mg
K
P
c > | 1
%
<
-mg kg
soil-
AA
1.4
6.0
23
314
93
8
4
Fort Pierce
DH
1.2
5.3
22
241
15
16
5
VA
1.3
5.5
25
242
21
20
4
MA
1.3
5.2
62
270
25
13
16
AA
3.4
6.1
44
1320
143
64
23
Ona
DH
3.1
5.4
36
920
120
29
8
VA
2.5
4.9
22
480
70
46
8
MA
5.7
4.7
55
' 800
~ 95
43
5
AA
2.5
6.1
66
780
67
40
6
Deseret
DH
2.3
6.0
27
1040
94
27
4
l.n f r {.'1_ 1
VA
2.8
7.2
30
1600
141
55
9
AA
4.7
5.3
28
960
32
28
4
Basinger
DH
3.5
5.2
26
460
100
46
7
VA
3.6
5.1
28
540
111
58
8
MA
4.2
5.2
36
640
129
94
11
ZAA- Aeschynomene americana; DH= Desmodium heterocarpon;
VA= Vigna adenantha; MA= Macroptilium atropurpureum.

9
Table 2-2. Analysis of variance for percentage of mycorrhizal root
colonization and total spore density per 100 g of air-
dried soil.
Source of
variation
Degree of
freedom
Mean
Squares
Root
colonization
total spore
density
%
no. 100 g^ soil
Location
3
447
325871
Legumes
3
321
73895
Interaction
8
299**
88933**
Error
30
13
1320
Significant at P < 0.01

10
Table 2-3. Mean percentage of mycorrhizal root colonization and total
spore density of VAM fungi among forage legumes at four
locations in south Florida.
Location
Root
colonization2
i JOUd.'c
Total spore
density2
%
no. 100 g-^ soil
Aeschynomene americana
Ft. Pierce
7b
302a
Ona
6b
160b
Basinger
. 5b
. 8C
Deseret
30a
146b
Desmodium heterocarpon
Ft. Pierce
12a
679a
Ona
15a
376b
Basinger
12a
19c
Deseret
16a
36c
Vigna adenantha
Ft. Pierce
5C
535a
Ona
41a
77b
Basinger
20b
25
Deseret
25b
36c
Macroptilium atropurpureum
Ft. Pierce
15a
23b
Ona
8b
294a
Basinger
3b
5b
zMeans within a column for each legume species followed by the same
letter are not different (P < 0.05) according to Duncan's multiple range
test.

11
between root colonization and spore density, while Hayman and Stovold
(1979) and Giovannetti and Nicolson (1983) found no correlation. This
apparent discrepancy may be due to different sampling methods.
Giovannetti (1985) collected samples within the same plant species and
sites, while the other researchers collected samples from many different
plant species and sites.
Spore production and root colonization are influenced by seasonal
variations (Giovannetti, 1985; Sylvia, 1986), host plant, stage of
development (Saif and Khan, 1975; Schenck and Kinloch, 1980), and soil
type (Lopes et al., 1983). In this survey there was only one sampling,
so it was not possible to separate seasonal or host developmental effects
on root colonization and total spore density.
The 6 species of VAM fungi collected in this survey were: Gigaspora
heterogama (GH) Gerdemann & Trappe, Gigaspora margarita (GM) Becker &
Hall, Glomus etunicatum (ETU) Becker & Gerdemann, Glomus intraradices
(INT) Schenck & Smith, Glomus sp. (GS), and Acaulospora spinosa (AS)
Walker & Trappe. The unidentified Glomus sp. was dark brown to black,
200-250 pm in diam, and had 1 wall of 8-14 pm thickness.
The occurrence of fungal species, as determined by spore numbers,
was affected by the legume host and location (Table 2-4). labal et al.
(1975) and Schenck and Kinloch (1980) also recorded differences in spore
numbers among plant species. The maximum number of spores of G.
margarita occurred at Fort Pierce associated with aeschynomene. Spores
of G. margarita were not found associated with Siratro at any of the four
locations. Spores of G. heterogama, G. etunicatum, and G. intraradices

12
Table 2-4. Mean spore numbers of VAM fungal species associated with
four forage legumes at four locations in south Florida.
Location
Species of VAM fungi-
-.} ; -.1
GM
GH
ETU
INT
AS
GS
Aeschynomene americana
Ft. Pierce
141a
66b
21a
42b
0
16
Deseret
9b
136a
0
0
0
0
Basinger
0
8C
0
0
0
0
On a
0
16c
29a
114a
0
0
Desmodium heterocarpon
Ft. Pierce
2a
114a
255b
307a
0
0
Deseret
2a
9b
0
0
0
0
Basinger
0
0
19c
0
0
0
Ona
0
0
325a
51b
0
0
Macroptilium atropurpureum
Ft. Pierce
0
10a
0
4b
0
8
Basinger
0
5b
0
0
0
0
Ona
0
0
40
254a
0
0
Vigna adenantha
Ft. Pierce
lb
39a
252a
241a
0
0
Deseret
20a
15b
0
0
0
0
Basinger
0
0
25b
0
0
0
Ona
28a
0
28b
llb
9
0
zMeans within
a column
for each
legume species followed by the same
letter are not different (P < 0.05) accordingly to Duncan's multiple
range test.

13
were found associated with all legumes, in at least one of the locations.
Glomus heterogama occurred in greatest numbers at Deseret and Fort Pierce
associated with aeschynomene and carpon desmodium, respectively. The
maximum number of spores of G. etunicatum occurred at Ona and Fort Pierce
associated with carpon desmodium. A high number of spores of G.
etunicatum was also found at Fort Pierce associated with Vigna adenantha.
Glomus intraradices was recovered in greater numbers from carpon
desmodium and Vigna adenantha at Fort Pierce as well as from Siratro at
Ona. The unidentified Glomus sp. occurred in lower numbers than the
other two species of Glomus; it was only found at Fort Pierce, associated
with aeschynomene and Siratro. Acaulospora spinosa was only recovered
from Vigna adenantha at Ona.
Overall root colonization by VAM fungi was low (most values below
20%) which indicates that (1) the native population of VAM fungi is not
very infective and (2) field inoculation may be effective. Attempts to
establish pot cultures of VAM fungi recovered in this survey were only
successful with G. etunicatum and G. intraradices. These two fungi were
shown to be effective in increasing the growth of several forage legumes
and were chosen for further evaluations.

CHAPTER III
GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH
GLOMUS INTRARADICES -- ---
Introduction
The use of forage legumes as companion crops to increase production
of grasses is becoming an established practice in order to reduce the
requirement for N fertilization (Rotar, 1983). In soils where P is a
limiting factor, large applications of P fertilizer are required for
legume establishment and optimum growth. However, with the increasing
cost of P fertilizer, alternative strategies for minimum input and
efficient use of P must be adopted. It is pertinent, therefore, to
evaluate whether mycorrhizal associations with forage legumes can be
manipulated in order to improve establishment, P nutrition, N2-fixation,
and consequently yield.
Vesicular-arbuscular mycorrhizal fungi can improve the growth of
legumes by increasing P uptake (Bethlenfalvay et al., 1985; Chulan and
Ragin, 1986; Harley and Smith, 1983; Hayman, 1983; Jensen, 1984).
Phosphorus is often a growth-limiting factor since many legumes have high
P requirements and are poor scavengers of P. The VAM fungi may also
increase nodulation and N2_fixation of legumes, primarily as an indirect
effect of improved P nutrition (Daft and El-Giahmi, 1976; Habte and Aziz,
1985; Newbould and Rangeley, 1984).
14

15
Most of the literature concerning the association of VAM fungi with
forage legumes is on temperate species; e.g. alfalfa (Medicago sativa L.)
(Kucey and Diab, 1984; Nielsen and Jensen, 1983; Satterlee et al., 1983),
white clover (Trifolium repens L.) (Newbould and Rangeley, 1984; Powell,
1979; Rangeley et al., 1982), and subterranean clover (Trifolium
subterraneum L.) (Abbott and Robson, 1978). Studies on tropical forage
legumes have been limited to a few species such as tropical kudzu
(Pueraria phaseoloides Benth) (Salinas et al., 1985; Waidyanatha et al.,
1979), leucaena (Leucaena leucocephala Dewit) (Huang et al., 1985) and
Stylo (Stylosanthes guianensis SW.) (Mosse, 1977; Waidyanatha et al.,
1979).
The purpose of this investigation was to evaluate the effect of a
VAM fungus, G. intraradices, on the growth of several tropical forage
legumes in pasteurized and nonpasteurized soil under greenhouse
conditions.
Materials and Methods
The top 15 cm of a virgin Oldsmar fine sand (sandy, siliceous,
hyperthermic Alfic Haplaquods) was collected from a newly cleared area
at the Agricultural Research and Educational Center, Fort Pierce, FL.
The low-P soil was air-dried and sieved through a 4-mm screen. The soil
had an initial pH of 4.5 (1:2 soil:water suspension) and P, Ca, Mg, and
K concentrations (extracted with 0.05 M HC1 + 0.0125 M H2SO4) of 1, 63,
21 and 12 mg kg 1 soil, respectively. Lime, as high calcitic limestone,
was thoroughly incorporated at a rate of 1500 mg kg1 soil (equivalent

16
to 3,000 kg ha"l assuming a 15-cm depth of soil ha-^ with a bulk density
of 1.3 g cra'^) and allowed to equilibrate for 30 days before initiating
the study. Solutions of P, K, Mg, Cu, Mn, Zn, B, and Mo also were
thoroughly mixed with the-soil to supply rates of 10, 30, 12, 1.5, 1.0,
1.0, 0.50 and 0.10 mg kg \ respectively. A portion of the soil was
pasteurized at 60C for A h to eliminate the indigenous mycorrhizal
fungi. Then 3 kg of soil was placed into 15-cm-diam plastic pots. The
; T*
pH of the soil was 6.2 at the end of the experiment.
The legumes used in the experiment were: Siratro, aeschynomene,
Aeschynomene villosa Poir., Stylo, leucaena, Stylosanthes hamata Taub.,
cv. 'Verano1, Vigna adenantha, and Arachis sp. Seeds were scarified with
sandpaper, wetted, and sprinkled with type EL "cowpea" inoculum (Nitragin
Co., Milwaukee, Wl) prior to planting. Five seeds of the corresponding
legumes were planted per pot, and plants were thinned to one per pot 10
d after emergence.
Glomus intraradices (isolate S311), used in this study, was isolated
from cultivated Vigna adenantha at the Agricultural Research and
Education Center, Ona, FL. (Chapter II, Table 2-4). Fungal inoculum was
produced in pot culture in pasteurized soil containing carpon desmodium
as the host plant. Pot cultures were 10-weeks old when they were
harvested, mixed and used to inoculate experimental pots. Ten grams per
pot of the soil-root-fungus inoculum containing approximately 200 spores
was spread in a 1-cm-thick layer, at a depth of 3 to 5 cm below the soil
surface. Noninoculated control treatments received 10 g of a soil-root
mixture from noninoculated pot cultures that were free of VAM fungi.

17
The experimental treatments consisted of pasteurized or
nonpasteurized soil, with or without addition of G. intraradices
inoculum. The pots were arranged on greenhouse benches in a completely
randomized design with three replications per treatment. The average
maximum and minimum greenhouse temperatures were 37 and 26C,
respectively. Pots were watered as needed to maintain soil moisture near
field capacity and were re-randomized every two weeks.
Plants were harvested after 45 d. Shoot and roots were dried at 70C
for 48 h and weighed. The percentage of mycorrhizal root colonization
was determined as described in Chapter II.
Significant treatment effects on shoot and root dry weights within
legume species were analyzed by the T TEST procedure of the Statistical
Analysis Systems (SAS Institute, 1982).
Results and Discussion
Inoculation with G. intraradices in nonpasteurized soil resulted in
greater shoot dry weights (P < 0.05) for five of the seven legumes tested
(Fig. 3-1). Greater shoot dry weights of these legumes were positively
related to increased levels of mycorrhizal colonization following
inoculation (Table 3-1).
Root dry weight results were similar to those for shoot dry weight,
except for Stylo, where there was no increase in root dry weight as a
result of mycorrhizal inoculation (Fig. 3-1).

ROOT DRY WEIGHT (g) SHOOT DRY WEIGHT (g)
18
? R
<
' C,* C.
2.0
1 .5
1 .0
0.5
0.0
0.8
0.6
0.4
0.2
0.0
LEGUME SPECIES
Fig. 3-1. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
nonpasteurized soil in the greenhouse after 45 days.
Legume species were Aeschynomene americana (AA),
Aeschynomene villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL), Stvlosanthes
hamata (SH), Stylosanthes guianensis (SG), and Vigna
adenantha (VA). Bars represent the mean of 3 replicates.
Means with the same letter within a species are not
different (P < 0.05).

19
Table 3-1. Percentage of mycorrhizal root colonization of tropical
forage legumes in nonpasteurized (UP) or pasteurized (P)
soil in the greenhouse after 45 days.
Legume species
Mycorrhizal
inoculation
Root colonization
UP P
%
Aeschynomene americana
+
22
3
5
0
Macroptilium atropurpureum
+
49
35
20
0
Aeschynomene villosa
+
10
*
6
Stylosanthes hamata
+
12
A
6
Stylosanthes guianensis
+
28
VC
8
Vigna adenantha
+
59
53
-
40
0
Leucaena leucocephala
+
4
3
1
0
Arachis sp.
+
31
12
- -
35
2
zBased on a composite of three samples for each legume.
Treatment lost to glasshouse accident.

20
There was no increase in plant growth as a result of inoculation with
G. intraradices in nonpasteurized soil for Arachis sp., leucaena, and
Vigna adenantha. With the exception of leucaena, this may be attributed
to effective colonization by the indigenous mycorrhizal fungi in the
noninoculated soil (Table 3-1).
Only five legume species (Fig. 3-2) were evaluated in pasteurized
soil; three were lost in a greenhouse accident. Siratro, Arachis sp.,
and Vigna adenantha had greater shoot and root dry weights after G.
intraradices inoculation. This increase again was related to effective
root colonization by G. intraradices (Table 3-1). In another study,
Siratro was shown to respond to inoculation with several VAM fungi in
pasteurized soil in the greenhouse (Lopes and De Olivera, 1980).
Inoculation with G. intraradices did not increase either the shoot
or root dry weights of aeschynomene or leucaena in pasteurized soil.
However, mycorrhizal colonization on both legumes was very low (3%).
Leucaena has been reported to be very mycorrrhizal dependent because
it has few root hairs (Huang et al., 1985; Yost and Fox, 1979). The
failure of VAM fungi to colonize it in this study in both pasteurized and
nonpasteurized soil may be due to incompatibility between the plant and
G. intraradices as well as native VAM fungi in the experimental soil,
to inhibitory soil factors on the host-VAM symbiosis, or to the
relatively slow development of the root system. It has been shown that
some mycorrhizal fungi may be less effective on certain plant hosts.
For example, Schroder et al. (1977) reported that Glomus macrocarpum Tul
& Tul increased growth of onions but not of Stylosanthes sp.

ROOT DRY WEIGHT (g) SHOOT DRY WEIGHT (g)
21
2.5
2.0
1 .5
1 .0
0.5
o:o
fi RV RS MR LL SH SG VR
LEGUME SPECIES
Fig. 3-2. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
pasteurized soil in the greenhouse after 45 days. The
legume species were Aeschynomene americana (AA), Arachis
sp. (AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA). Bars
represent the mean of 3 replications. Means with the same
letter within a species are not different (P < 0.05).

22
The failure to obtain good colonization of aeschynomene in
pasteurized soil was unexpected, since this legume was successfully
colonized in nonpasteurized soil. Unfortunately, soil chemical
properties were not determined after pasteurization. Legumes are
sensitive (and hence VAM fungi) to elevated Mn, which is a common
occurrence in heat treated soils.
The results obtained in this study clearly demonstrate that G.
intraradices can successfully compete with some of the indigenous
mycorrhizal fungi present in the experimental soil and promote growth of
several legumes in nonpasteurized soil. This result agrees with earlier
work by Abbott and Robson (1981), Mosse (1977), and Rangeley et al.
(1982), which suggests a potential for successful field-scale inoculation
with effective VAM fungi.

' v r.
CHAPTER IV
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN
NONPASTEURIZED SOIL. I. SELECTION OF EFFECTIVE VA MYCORRHIZAL FUNGI
UNDER AMENDED SOIL CONDITIONS.
Introduction
Siratro, a cultivar developed by E. M. Hutton (1962) from two Mexican
accessions of Macroptilium atropurpureum Urb., is a persistent, perennial
forage legume adaptable to a wide range of soil and climatic conditions.
It has become widespread and is among the most versatile forage legume
grown throughout tropical regions of the world (Lynd et al., 1985).
In pasteurized and nonpasteurized soils, increased growth of Siratro was
attained after inoculation with Glomus fasciculatum Gerdemann & Trappe
(Lopes and De Olivera, 1980; Lynd et al., 1985) and G. intraradices
(Chapter III).
Hayman (1982) stated that VAM fungi are probably capable of symbiosis
with most plants, at least to some degree. However, there is wide
variation in the ability of VAM fungi to stimulate plant growth (Miller
et al., 1985; Powell, 1982; Schubert and Hayman, 1986). Lopes and De
Olivera (1980), using a gamma-irradiated soil of low P-content, studied
the effect of inoculation with nine species of VAM fungi on the growth
of Siratro. Only inoculation with G. fasciculatum and G. macrocarpum
enhanced plant growth. Abbott and Robson (1981) defined the relative
ability of a VAM fungus to stimulate plant growth as 'effectiveness1 and
23

24
this defined term will be used in this paper. Wilson (1984) indicated
that an evaluation of the effectiveness of the indigenous mycorrhizal
population under amended soil conditions, as well as studies to select
effective VAM fungi, are prerequisites for successful field inoculation.
Thus, the objective of the present study was to determine the
effectiveness of several VAM fungi with Siratro in a limed,
nonpasteurized soil with low P content under greenhouse conditions.
Materials and Methods
The soil used in this study, liming, and fertilizer amendments are
described previously in Chapter III.
Plants were inoculated with the following VAM fungi: G. etunicatum
(isolate S312) obtained from carpon desmodium at the Agricultural
Research and Education Center, Ona, FL. (Chapter II, Table 2-4); G.
deserticola Trappe, Bloss & Menge (isolate S305) obtained from sea oats
(Unila paniculata L.) in a coastal dune, Anastasia, FL.; G. versiforme
Berch & Fortin (isolate #231) obtained from N.C. Schenck, University of
Florida, Gainesville, FL.; G. intraradices (isolate S311) obtained from
Vigna adenantha at the Agricultural Research and Education Center, Ona,
FL. (Chapter II, Table 2-4); G. margarita (isolate #215) obtained from
N.C. Schenck, University of Florida, Gainesville, FL. Isolates were
maintained in pot cultures in pasteurized soil containing bahiagrass.
Soils from 12-week-old pot cultures were used to inoculate experimental
pots. The propagule densities of the native soil and inocula at the
beginning of the experiment were determined by the most-probable-number

25
(MPN) technique using bahiagrass as the host plant and pasteurized
Oldsmar fine sand as the diluent (Daniels and Skipper, 1982). The amount
of inoculum used was adjusted to give equal inoculum densities among
isolates. Each pot received approximately 240 propagules. Details on
the fungal inoculation technique, planting, and watering were reported
previously (Chapter III).
There were six inoculation treatments, five species of VAM fungi, and
a control inoculated with non-VAM pot culture material. The pots were
arranged on the greenhouse bench in a completely randomized block design
with 15 replications per treatment.
The average maximum and minimum greenhouse temperatures during the
experimental period were 32 and 19C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1200 p mol m s
Five randomly selected samples were harvested from each treatment
after 20, 40, and 70 d of growth. At first harvest, shoot dry weight,
percentage of root colonized by VAM fungi, plant height, and number of
leaves were determined. In addition, root fresh weight and total root
length colonized were determined at the second and third harvests. Shoot
dry weight was determined by drying the material at 70C for 24 h.
Percentage and total root length colonized were estimated by the gridline
intersect method (Giovannetti and Mosse, 1980) after roots were cleared
in 10% KOH and stained with 0.05% trypan blue in lactophenol (Kormanik
and McGraw, 1982). Data were analyzed by Analysis of Variance Procedure,
Statistical Analysis Systems (SAS Institute Inc., 1982). Duncan's
multiple range test was used to separate treatment means when the F-test
was significant (P < 0.05).

26
Results and Discussion
Plants inoculated with G. etunicatum and G. intraradices had higher
shoot dry and root fresh weights than plants inoculated with the other
VAM fungi or control plants, at 40 and 70 d after planting (Fig. 4-1).
At both harvests, plants inoculated with G. etunicatum had higher shoot
dry weights than plants inoculated with G. intraradices. At the final
harvest, plants inoculated with G. intraradices had higher root fresh
weights than plants inoculated with G. etunicatum.
In contrast, plants inoculated with G. versiforme, G. margarita, and
G. deserticola had shoot dry and root fresh weights that were not
different from the noninoculated plants, except at the final harvest when
plants inoculated with G. deserticola had higher shoot dry and root fresh
weights than the control (Fig. 4-1). At 20 d, shoot dry weights were
not different among treatments (mean = 0.70 g).
rPercentage and total root length colonized by VAM fungi increased
with time (Fig. 4-2). Inoculation with G. etunicatum and G. intraradices
resulted in the highest root colonization at all harvests. At 70 d,
plants inoculated with G. etunicatum had the highest root colonization,
followed by G. intraradices and then G. deserticola. There were no
differences in root colonization among G. versiforme, G. margarita and
control treatments. For the six treatments, total root length colonized
by VAM fungi and percentage of mycorrhizal root colonization followed
the same trend (Fig. 4-2).
Shoot dry weight of Siratro was correlated with total root length
colonized by VAM fungi (r2 = 0.95""") and percentage of mycorrhizal root

ROOT FRESH WEIGHT (g) SHOOT DRY WEIGHT (g)
27
Fig. 4-1. Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus
intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the shoot dry weight and root fresh weight of
Siratro at two harvests. Bars represent the means of five
replicates. Means with the same letter within a harvest
are not different (P < 0.05).

ROOT LENGTH COLONIZED (m) ROOT COLONIZATION (%)
28
Fig. 4-2 Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus
intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the percentage of root colonization and root
length colonized of Siratro at three and two harvests,
respectively. Bars represent the means of five replicates.
Means with the same letter within a harvest are not
different (P < 0.05).

29
colonization (r^ = 0.83''"). There was a quadratic relationship between
shoot dry weight and length of Siratro root colonized by VAM fungi for
all inoculated treatments (Fig. 4-3).
Plant height and number of leaves per plant were not different among
treatments at any harvest. After 70 d, the mean plant height and number
of leaves for all treatments were 90 and 15 cm, respectively.
There were 2 propagules per gram of soil in the native soil as
determined by the MPN test at the beginning of the experiment.
Inoculum density is known to influence plant growth response to VAM
fungal inoculation (Hass and Krikum, 1985Wilson, 1984). Thus, one of
the problems in comparing the efficacy of VAM fungi is ensuring uniform
inoculum densities (Daniels et al., 1981). In this study, I used the
MPN technique to provide a measure of the inoculum densities of the VAM
fungi, and I adjusted the inoculum densities so that they were uniform
for all inoculated treatments.
There were striking differences in the effectiveness of VAM fungi on
Siratro. The results are consistent with the findings of others (Miller
et al., 1985; Schubert and Hayman, 1986) indicating that different
species and strains of VAM fungi vary considerably in the benefits they
confer to the host plant. This experiment also confirmed previous work
(Chapter III) which demonstrated that the native population of VAM fungi
in this soil was less able to stimulate the growth of Siratro than
effective, introduced species. It is possible that the decreasing soil
acidity obtained by liming changed the native population of VAM fungi
from effective to ineffective as compared to G. etunicatum and G.
intraradices (Hayman and Tavares, 1985). Since it is necessary to lime

SHOOT DRY WEIGHT (g)
30
Fig. 4-
ROOT LENGTH COLONIZED (m)
Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments.

31
this soil for satisfactory establishment and growth of legumes (Snyder
et al., 1985), VAM fungi must be selected for their effectiveness under
amended soil conditions.
Powell (1980a) reported a relationship between the level of native
inoculum density in the soil and plant growth response to mycorrhizal
inoculation. When inoculum density was low (0.01-0.09 propagules g
soil), there was a significant plant growth response to inoculation with
VAM fungi. When inoculum density was higher (0.15-0.30 propagules g ^
soil), there was little plant growth response to fungal inoculation.
Likewise, good plant growth responses to inoculation with VAM fungi in
soils with few indigenous endophytes have been reported by Mosse (1977)
and Hall (1979). Thus it would seem that the most promising sites for
inoculation with VAM fungi are those where indigenous populations of VAM
fungi are very low. However, in this study, where the native inoculum
density was relatively high (2 propagules g ^ soil), Siratro responded
to inoculation with two of the four VAM fungi tested. In addition to the
abundance of the indigenous VAM fungi, information about their
infectivity and effectiveness is needed to assess potential sites for
responsiveness to inoculation with effective VAM fungi.
The ineffectiveness of G. versiforme and G. margarita could be due
to an innate symbiotic inefficiency, incompatibility, lack of
competitiveness, or to inhibitory edaphic (e.g. soil pH or P level) or
environmental factors (e.g. light and temperature). Hayman and Tavares
(1985) showed clearly that different endophytes vary in their symbiotic
effectiveness at different soil acidities. In addition, some endophytes
may be less effective on certain plant hosts. For example,

32
Schroder et al. (1977) reported that G. macrocarpum increased growth of
onions but decreased growth of Stylosanthes sp.
The most effective fungi in this study were those that colonized the
root most rapidly. Sanders et al. (1977) and Abbott and Robson (1978)
reported that VAM fungi differ in their rates of root colonization.
Abbott and Robson (1981) stated that differences in the effectiveness of
VAM fungi could be due to differences in their ability to (1) colonize
the roots rapidly (infectivity), (2) produce external hyphae, and (3) to
take up and transport P (efficiency). In this study, we measured only
root colonization over time (infectivity) and found, under the conditions
of this experiment, G. etunicatum was the most infective fungus.
Mycorrhizal root colonization, expressed as either percentage or
total root length colonized, was positively correlated with the shoot dry
weight of Siratro. Abbott and Robson (1981) and Plenchette et al. (1982)
also showed positive correlations between the magnitude of mycorrhizal
root colonization and shoot dry weights of plants grown on P-deficient
soils. These results suggest that differences in endophyte effectiveness
may be evaluated on the basis of rates of root colonization. However,
Hayman and Tavares (1985) demonstrated that final root colonization by
VAM fungi may give little indication of the ability of an endophyte to
stimulate plant growth. Abbott and Robson (1978) reported that VAM fungi
which differ in effectiveness and rate of root colonization may have
similar plateau levels of colonization at a late harvest. Therefore it
is not surprising, when colonization is assessed at a relatively advanced
stage of plant growth, that there is often little correlation between
mycorrhizal root colonization and effectiveness.

33
Forage legumes are an important component of improved pastures and
must be established rapidly and without excessive cost. When P is a
major factor limiting the productivity of legumes, large applications of
P fertilizer are normally required for legume establishment. However,
with the increasing cost of P fertilizer, alternative strategies for
minimum fertilizer input and efficient use must be adopted. These data
demonstrate that by careful selection of effective VAM fungi, the growth
of Siratro can be enhanced in a P-deficient native soil containing a less
effective native VAM population than the introduced VAM fungi.

CHAPTER V
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN
NONPASTEURIZED SOIL. II. EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT
DIFFERENT P LEVELS.
Introduction
Forage legumes are an important component of improved grass pastures.
The legumes serve both to increase forage quality and decrease the need
for fertilizer N through N2 fixation. Snyder et al. (1985) studied the
responsiveness of several tropical legumes, including Siratro, to P and
lime in a typical Florida Spodosol. They reported that lime and P rates
of approximately 3000 and 75 kg ha ^ produced maximum yield and maximum
economic return.
In an experiment with red clover in a field containing 10 mg NaHCC^-
soluble P kg ^ soil, plants showed an early response to superphosphate,
but by the end of the second year yields were high in all plots,
equivalent to around 15 t ha ^ dry matter (Hayman et al., 1981). This
result was attributed to one of the introduced endophytes, G. caledonium,
which had spread and sporulated profusely throughout all the plots
(including those inoculated with two other endophytes) and had previously
enhanced growth of lucerne at this site. In upland pastures in Wales,
Hayman and Mosse (1979) found that inoculation of white clover seedlings
with a combination of G. Mosseae and G. fasciculatum "E3" in field plots
given the standard dressing of 90 kg P ha 1 as basic slag doubled plant
34

35
growth and greatly enhanced tissue P content and nodulation. Growth
responses at other sites varied from large to slightly negative, probably
governed in part by the effectiveness of the indigenous VAM population
(Hayman and Hampson, 1979).
Species (Miller et al., 1985; Schubert and Hayman, 1986; Thompson et
al., 1986), and isolates within a species (Cooper, 1978), of VAM fungi
can colonize plants at different rates. If the mycorrhizal growth
response is related to the amount of early root colonization (Abbott and
Robson, 1981; Chapter IV), then isolates of VAM fungi that colonize roots
rapidly, at P levels found in established agricultural soils, may be most
suitable for pasture inoculation.
Schubert and Hayman (1986) indicated that, in order to achieve a
rational and effective use of inoculants, precise information on the
performance of endophytes in soil amended with P was necessary. It is
evident that the effect of soil P on symbiosis varies with the specific
host and endophyte. Therefore, more research is needed to develop
uniform and predictable endophyte-host responses.
In another study described in Chapter IV, G. etunicatum and G.
intraradices were found to be the most effective growth enhancers (out
of 5 isolates) of Siratro in a soil amended with a moderate level of P
and lime. In the present study, the objective was to evaluate the
infectivity and effectiveness of these two fungi over a practical range
of applied P.

36
Materials and Methods
The chemical properties of the soil used, liming, and fertilizer
amendments are described previously (Chapter III).
Siratro was used as the host plant in this experiment. The VAM
fungi tested were G. etunicatum (isolate S312) and G. intraradices
(isolate S313). These isolates were maintained in pasteurized soil in
pots with bahiagrass as the host. Soils from 10-week-old pot cultures
were used to inoculate experimental pots. The propagule densities of
the inocula were determined by the MPN technique (Daniels and Skipper,
.1982) and approximately 240 propagules were added to each 15-cm-diam
plastic pot. Details on the origin of the two fungal isolates were
reported previously (Chapter II, Table 2-4). Fungal inoculation
technique, planting, and watering were described in Chapter III.
The experiment was designed as a 3 x 4 factorial consisting of 3
inoculation treatments; G. intraradices, G. etunicatum, and
noninoculated control, and four P treatments; 2.5, 10, 20, and 40 mg P
kg 1 as CaO^PO^^.I^O (equivalent to 5, 20, 40, and 80 kg P ha ^
-1 -3)
assuming a 15-cm depth of soil ha with a bulk density of 1.3 g cm .
Phosphorus was applied in solution one week before planting. Soil
samples from each treatment were analyzed for extractable P at the
beginning and end of the experiment using the Mehlich-I method (0.05 M
HC1 + 0.0125 M H2SO4). The twelve treatments were replicated five times
and arranged on a greenhouse bench in a randomized complete block design

37
The average maximum and minimum greenhouse temperatures during the
experimental period were 34 and 26C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1800 ju mol m s
Shoot dry and root fresh weights, and percentage and total root
lengths colonized by VAM fungi, were determined after 60 d using
procedures described previously (Chapter IV). Data for all variables
were subjected to ANOVA procedures and regression analysis (SAS Institute
Inc., 1982).
Results and Discussion
Phosphorus applications of 2.5, 10, 20, and 40 mg kg ^ resulted in
Mehlich-I extractable P in the soil of 5.6, 12.8, 21.6, and 38.0
mg kg~l at the beginning of the experiment, respectively. These
differences were still reflected at the end of the experiment when P
concentrations were 4.4, 7.8, 15.2, and 24.6 mg kg *.
Shoot dry and root fresh weights of Siratro were increased by P
fertilization and fungal inoculation (Fig. 5-1). At 2.5 mg P kg--*-, there
was no difference in the shoot dry and root fresh weights among
inoculated and control plants. At all other levels of P, inoculated
plants had higher shoot dry and root fresh weights than control plants.
Shoot dry weights were greatest for plants inoculated with G. etunicatum.
There were no differences in root fresh weights between plants inoculated
with G. etunicatum or G. intraradices. Inoculated plants had a quadratic
relationship for shoot dry and root fresh weights and P application

ROOT FRESH WEIGHT (g) SHOOT DRY WEIGHT (g)
38
P APPLIED (mg/kg)
CC o
/*>
s
tr
P APPLIED (mg/kg)
Fig. 5-1. Effect of P application on shoot dry weight, root fresh
weight, percentage of root colonized by VAM fungi, and
total root length colonized of Siratro grown in limed
nonpasteurized soil and inoculated with Glomus etunicatum
(ETU), Glomus intraradices (INT), or not inoculated (CON).

39
(Table 5-1). Maximum yield of inoculated plants was achieved between 28
and 30 mg kg of applied P. Control plants had a linear relationship
for shoot dry weights and a quadratic relationship for root fresh
weights. There were fungus x P interactions for shoot dry and root fresh
weights (Table 5-2).
Percentage and total root length colonized by VAM fungi for the
inoculated treatments increased with P additions (Fig. 5-1). This effect
was greater for G. etunicatum than for G. intraradices. Phosphate
application did not alter the percentage and total root length colonized
in the control plants. Inoculated plants had a quadratic relationship
for percentage and total root length colonized and applied P
(Table 5-1). Maximum colonization of inoculated treatments, expressed
as either percentage or total length of colonized root, was attained
between 32 and 35 mg kg-^ of applied P. There were fungus x P
interactions for percentage and total root length colonized by VAM fungi
(Table 5-2).
Shoot dry weight of Siratro over the range of applied P was highly
correlated with percentage (r2 = 0.95") and total root length colonized
O V*. J,
by VAM fungi (rz = 0.97"') (Fig. 5-2) for both inoculated treatments.
Percentage of the root colonized by VAM fungi was very closely correlated
(r2 = 0.98*>') with the total root length colonized.
Growth enhancement from VAM inoculation at different levels of P has
been reported to vary with VAM fungi (Hayman and Hampson, 1979; Hayman
and Mosse, 1979; Schubert and Hayman, 1986; Thompson et al., 1986). For
example, Schubert and Hayman (1986) indicated that, when large amounts
of P were added (more than 100 mg kg *), G. mosseae, G. versiforme, G.

40
Table 5-1. Regression equations and coefficients of determination (r2)
showing the relationship of P level to shoot dry weights,
root fresh weights, percentage and total root length
colonized.
Variable
Regression equations
r2
Shoot dry weight (g)
Glomus etunicatum
= -0.12+0.34P-0.006P2
0.97**
Root fresh weight (g)
= 0.50+0.40P-0.006P2
0.96';'*
Root colonization (%)
= 4.47+2.62P-0.04P2
0.97**
Root length colonized
(m)
= -1.43+0.64P-0.009P2
0.96**
Shoot dry weight (g)
Glomus intraradices
= 0.16+0.24P-0.004P2
0.95**
Root fresh weight (g)
= 0.55+0.38P-0.006P2
0.94**
Root colonization (%)
= 5.42+1.86P-0.03P2
0.95**
Root length colonized
(m)
= -1.08+0.49P-0.006P2
0.91
Shoot dry weight (g)
Control
= 0.72+0.06P
0.93"*
Root fresh weight (g)
= 0.89+0.17P-0.002P2
0.92**
**significant at P < 0.01
P = phosphorus level

41
Table 5-2. Analysis of variance for shoot dry weights, root fresh
weights, and percentage and total root length colonized.
Source of
variation
DF
Shoot MS
root MS
root colonized MS
percentage length
Block
4
A,'.
0.07
0.04
7.39
0.069
Fungi (F)
2
11.07
18.01
2650.42
114.13
P rates (P)
3
34.19
59.37
1583.51
118.66
linear (PI)
1
85.40
146.43
4156.00
305.28
quadratic (Pq)
1
13.97
30.10
577.33
32.23
Cubic (Pc)
1
3.20
1.57
17.14,
2.48
F x P
6
1.65**
2.55'"'*
378.46""
22.50**
F x PI
2
2.58"
4.02**
965.54**
55.69**
F x Pq
2
2.29**
2.08**
159.17**
7.82""
F x Pc
2
0.08
1.55
10.67
3.99
Error
44
0.02
0.13
6.27
0.24
**
Significant at P < 0.01
MS = mean square

SHOOT DRY WEIGHT (g)
42
ROOT LENGTH COLONIZED Cm)
Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments
in nonpasteurized soil.
Fig.
5-2.

A3
macrocarpum and G. margarita were ineffective in stimulating growth of
onion; however, G. caledonium and Glomus sp. 'E3' were generally
effective at all P levels. In our study, G. etunicatum was more
effective than G. intraradices at all but the lowest applied P level.
Hence, there appears to be good potential for the selection of VAM fungi
to enhance plant growth under amended soil conditions such as P
fertilization and liming.
Phosphorus has been reported to increase, decrease or not affect root
colonization by VAM fungi. However, it is difficult to compare results
concerning the effect of P fertilization on mycorrhizal root
colonization, because of differences in the range of added P, as well as
other factors such as host plant and soil type. In this study, we used
a range of 2.5 to AO mg P kg ^ because it represented P levels used in
the production of tropical forage legumes in Florida on a similar soil
(Snyder et al., 1985). At the lowest P level, G. etunicatum and G.
intraradices did not colonize the root or improve growth of Siratro above
that of the control plants. Barber and Lougham (1967) reported that, at
a very low P level, competition for P occurs between plants and
microflora. Habte and Manjunath (1987) and Same et al. (1983) indicated
that the growth of VAM fungi is limited by P at very low levels. Between
10 to AO mg P kg ^, percentage and total root length colonized by VAM
fungi increased with P additions. These results agree with those of
Abbott and Robson (1977b), Schubert and Hayman (1986), and
Thompson et al. (1986), who reported an increase in the percentage and
total root length colonized between 18 to 55 mg P kg-1. At high P levels
(more than 100 mg kg ^), which are not feasible for field production of

44
forage legumes, percentage and total root length colonized by VAM fungi
may be suppressed (Abbott and Robson, 1977b; Schubert and Hayman, 1986;
Thompson et al., 1986).
The result that root colonization by VAM fungi, expressed as either
percentage or total root length colonized, is positively correlated to
shoot dry weight is consistent with the findings of Abbott and Robson
(1981) and a previous study where its implications are discussed
(Chapter IV).
I conclude that, in amended soils where the indigenous population of
VAM fungi is less effective than some of the introduced species of VAM
fungi, inoculation with effective VAM fungi can increase the plant
growth. Furthermore, with highly mycorrhizal dependent crops such as
tropical legumes, growth enhancement may occur at P levels actually used
in commercial pasture production.

CHAPTER VI
EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE GROWTH AND
UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM, STYLOSANTHES GUIANENSIS,
AND AESCHYNOMENE AMERICANA
Introduction
Mycorrhizal colonization is important for legumes because it
increases their P uptake (Abbott and Robson, 1977b; Saif, 1987), and
therefore nodulation and N2_fixation (Asimi et al., 1980; Bergersen,
1971; Gates and Wilson, 1974; Gibson, 1976).
Crush (1974) found that VAM fungi increased the growth and
nodulation of Centrosoma pubescens Benth, Stylo, and Trifolium repens L.
Mosse et al. (1976) showed that effective nodulation of Centrosema,
Stylosanthes, and Trifolium plants in a P-deficient Brazilian Cerrado
soil could be achieved only by introducing both VAM fungi and P.
Mycorrhizal fungi also have been shown to increase nodulation, N2~
fixation, plant growth, plant N and P content in Vigna unguiculata (Islam
et al., 1980; Sanni, 1976), Medicago sativa (Barea et al., 1980),
Pueraria phaseoloides and Stylo (Waidyanatha et al., 1979), Stylosanthes
scabra (Purcino and Lynd, 1985), leucaena (Munns and Mosse, 1980;
Purcino et al., 1986), and Siratro (Lynd et al., 1985). However,
effective tripartite symbiosis (legume-rhizobium-VAM fungus) is
influenced by soil and climatic conditions (Waidyanatha et al., 1979),
45

46
which are apparently species-specific (Burt and Miller, 1975; Mosse,
1972).
In a previous study (Chapter V), G. etunicatum was an effective
growth enhancer of Siratro in a soil similar to the one used in the
present study, with a moderate level of applied P (20-40 mg P kg ^).
The objective of this investigation was to determine the effect of
inoculation with a VAM fungus, G. etunicatum, on the growth and plant
uptake of P and N of three forage legumes at different P levels in
pasteurized soil under greenhouse conditions.
Materials and Methods
The soil used in this investigation, liming, and basic fertilization
are described previously (Chapter III). The soil chemical
characteristics before soil fertility treatments and after pasteurization
(70C for 4 h) were: pH 4.4 (soil:H20=l:2); 1.4% organic matter; 2, 65,
11, and 12 mg kg ^ (Mehlich-I extractable) of P, Ca, Mg and K,
respectively.
At planting three phosphorus levels were established by application
in solution of 12.5, 25, and 50 mg P kg 1 as Cai^PO^^.^O which is
equivalent to 25, 50, and 100 kg P ha ^, assuming a 15-cm depth of soil
ha-l with a bulk density of 1.3 g cm-^.
Glomus etunicatum (isolate S312) was isolated from carpon desmodium
at the Agricultural Research and Education Center, Ona, FL. (Chapter II,
Table 2-4). Fungal inoculum was produced in pot culture in pasteurized

47
soil containing bahiagrass. The fungal inoculation technique was
reported in Chapter III.
The legume species used in this experiment were: Siratro,
aeschynomene, and Stylo. Seeds were scarified with sandpaper, wetted,
and sprinkled with type El "cowpea" inoculum (Nitragin Co., Milwaukee,
WI) prior to planting. Three germinated seeds were planted per 620 ml
"Deepots" (J.M. McConkey & Co, Inc, Summer, WA). After emergence,
seedlings were thinned to one per pot.
The experiment was conducted as a 2 x 3 x 3 factorial consisting of
2 inoculation treatments, etunicatum and noninoculated control; three
P levels, 12.5, 25, and 50 mg kg ^; the 3 legume species; and 3
replications. The 18 treatments were arranged on a nonshaded greenhouse
bench in a completely randomized design. The average maximum and minimum
greenhouse temperatures during the experimental period were 28 and 20^,
respectively, and the average maximum photosynthetic photon flux density
was 1793 u mol m-^ s-^.
After 65 d plants were harvested, shoots dried (70C for 24 h),
weighed, and ground in a Wiley mill using a 20-mesh screen. Shoots were
digested by the sealed-chamber procedure of Anderson (1986) and analyzed
for P on a Jarrel-Ash 955 inductively-coupled argon plasma spectrometer
(ICAP). For nitrogen analysis, samples were digested using a
modification of the aluminum block digestion procedure of Gallaher
et al. (1975) and ammonia in the digestate was determined by
semiautomated colorimetry (Hambleton, 1977). Roots were washed from the
soil, air-dried, and weighed. In addition, percentage and total root

48
length colonized were estimated as described in Chapter IV. A 0-4 scale
was used to estimate the number of nodules per plant, with 1 = 20-50,
2 = 50-100, 3 = 100-150, and 4 = > 150 nodules. Data were subjected to
ANOVA procedures and regression analysis (SAS institute Inc., 1982).
Results and Discussion
At harvest, both fungal inoculation and P applications increased
shoot dry weight, plant P concentration, and total plant P and N of the
three legumes (Table 6-1). Root fresh weight was increased for Stylo
and Siratro but not for aeschynomene. There were fungus x P interactions
for shoot dry and root fresh weights, and total plant N of Stylo.
Siratro had fungus x P interactions for shoot dry weight and total plant
P and N, whereas aeschynomene only had fungus x P interaction for shoot
dry weight.
At all levels of applied P, shoot dry and root fresh weights
and total N of Stylo were greater for mycorrhizal plants than
nonmycorrhizal plants (Fig. 6-1). Differences between mycorrhizal and
nonmycorrhizal plants were most pronounced at intermediate levels of
applied P and diminished with further P addition.
Phosphorus concentration and total P of Stylo were not affected by
fungus x P interactions (Table 6-1), but there was an overall effect of
fungal inoculation (Fig. 6-1). Saif (1987), Mosse et al. (1976), and
Waidyanantha et al. (1979) also reported an increase in plant growth, P
concentration, and total P and N of Stylo in a pasteurized low P soil
following inoculation with VAM fungi and P applications.

49
Table 6-1. Mean squares and levels of significance from the analysis
of variance for shoot dry weight, root fresh weight, P
concentration, and total P and N uptake of forage legumes.
Source of
Variation
DF
Shoot
dry wt
Root P
fresh wt cone
Total
P
Total
N
Stylosanthes guianensis
Fungus
1
0.45
1.13
0.011**
6.93^
500.97
P rates
2
0.64
0.21
0.016"
9.13*
790.32
Lineal
1
1.20
0.36
0.032*
17.69**
1511.78
Quadratic
1
0.09
0.05
0.0004
0.57
68.86
Fungus x P
2
0.046**
0.15**
0.00007
0.65
47.14*
Error
12
0.0057
0.026
0.012
0.18
8.48
Macroptilium atropurpureum
Fungus
1
1.40
0.58**
0.0076*
33.78
2360.30
P rates
2
7.28
1.81**
0.028""
163.11
10469.14
Lineal
1
14.54
3.55"
0.056'
325.94
20925.10
Quadratic
1
0.0072
0.05
0.0001
0.28
13.18
Fungus x P
2
0.54
0.10
0.0014
10.37
625.52**
Error
12
0.029
0.027
0.00082
2.47
43.27
Aeschynomene
americana
Fungus
1
0.13
0.042
0.0068**
6.06**
129.34*
P rates
2
0.56
1.07**
0.034""
29.75**
524.16**
Lineal
1
1.09
2.01*
0.064"
58.70*
1035.46**
Quadratic
1
0.027,
0.12
0.0022
0.80
12.85
Fungus x P
2
0.037"
0.030
0.00009
0.10
43.35
Error
12
0.0076
0.015
0.00052
0.25
21.75
Significance at P < 0.05 ''"'Significant at P < 0.01

root cfiLONiMim m Rnnr fresh veisii (g) snnor dry yeihit (g)
50
(mg/kg)
Fig. 6 1. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Stylosanthes
guianensis.

51
Mycorrhizal plants of Siratro had greater shoot dry weight, total P
and total N than nonmycorrhizal plants at low and intermediate levels of
applied P, but not at the highest level (Fig. 6-2). Overall, mycorrhizal
plants had greater root fresh weight and plant P concentration than
nonmycorrhizal plants. Other investigators (Lynd et al., 1985; Saif,
1987) have shown similar responses of Siratro to inoculation with VAM
fungi and P additions. Differences in shoot dry weight between
mycorrhizal and nonmycorrhizal plants of aeschynomene were only at the
intermediate level of applied P (Fig. 6-3). Fungal inoculation did not
affect the root fresh weight. Overall, mycorrhizal plants had greater P
concentration, total P, and total N than nonmycorrhizal plants
(Fig. 6-3).
Percentage and total root length colonized for mycorrhizal plants
of Stylo (Fig. 6-1), Siratro (Fig. 6-2), and aeschynomene (Fig. 6-3)
increased with the first addition of P. However, maximum colonization,
expressed as either percentage or total length of colonized root of the
three legumes, was attained at the intermediate levels of applied P.
The number of nodules in the three legumes increased with fungal
inoculation and P applications. Mycorrhizal plants of Stylo had more
nodules than nonmycorrhizal plants at all levels of applied P. However,
mycorrhizal plants of Siratro and aeschynomene had more nodules than
nonmycorrhizal plants only at 25 mg kg1 of applied P.
Mycorrhizal plants required between 38-40 mg P kg-1 to achieve
maximum shoot dry weight, whereas nonmycorrhizal plants required
50 mg P kg to produce approximately the same shoot dry weight, except
for nonmycorrhizal Stylo (Fig 6-1) which even with 50 mg P kg1 did not

ROOT m.CHIZRTION (Z) ROO FRESH VE1Q1T (g) 3IOOI GFY WEIGHT (g)
52
9 _
e £
P APPLIED Cmg/kg)
Fig. 6 2. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Macroptilium
atropurpureum.

53
P APPLIED (mg/kg)
Fig. 6-3. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Aeschynomene
americana.

54
reach the same shoot dry weight as myco-rrhizal plants. Fungal
inoculation with G. etunicatum resulted in a 20% decrease in the amount
of P required for maximum yield. This demonstrates the importance of VAM
fungi in the P nutrition of tropical legumes and would represent an
important savings to the farmer.
Growth responses of Stylo, Siratro, and aeschynomene associated with
inoculation with VAM fungi are closely related to improved P uptake and
^fixation. These results clearly support the findings of earlier
studies that inoculation with VAM fungi not only stimulates plant growth
and P uptake of legumes (Abbott and Robson, 1977b; Habte and Manjunath,
1987; Menge, 1983) but also nodulation and ^-fixation (Asimi et al.,
1980; Lynd et al., 1985; Purcino et al., 1986) which was measured
indirectly in this study by the total plant uptake of N.
Inoculation with effective species of VAM fungi and additions of P
between 25 and 50 mg kg ^, were shown to improve the growth and uptake
of P and N of Stylo, Siratro, and aeschynomene.

CHAPTER VII
GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND AESCHYNOMENE AMERICANA
TO INOCULATION WITH SELECTED VA MYCORRHIZAL FUNGI IN THE FIELD AT
DIFFERENT P LEVELS
Introduction
The practical goal of studies on plant growth responses to
inoculation with VAM fungi is to obtain increased yield of plants growing
under field conditions. Significant plant growth responses to
inoculation with VAM fungi have been demonstrated in pot experiments
using pasteurized and nonpasteurized soil for several tropical forage
legumes such as: Pueraria phaseoloides (Salinas et al., 1985;
Waidyanatha et al., 1979), leucaena (Habte and Manjunath, 1987; Huang et
al., 1985), Siratro (Lynd et al., 1985), Stylo (Mosse, 1977) and
aeschynomene (Chapter II and VI).
It has been pointed out that inoculation experiments with VAM fungi
should include testing a series of P levels (Abbott and Robson, 1977b;
Hall, 1978; Powell, 1980b) in order to select the optimum P level for a
mycorrhizal response. Except for the work by Saif (1987), little
information is available on plant growth response of tropical forage
legumes to inoculation with VAM fungi in nonpasteurized soil under field
conditions at different P levels. However, there is more data for
temperate legumes. In general, the field sites where plants are most
likely to respond to inoculation with VAM fungi are those containing
55

56
little soluble phosphate and a small or ineffective native population of
VAM fungi. The experimental site selected for this study contains very
little available P (1 mg kg and a reasonably high indigenous
population of VAM fungi (2 propagules g ^ soil). The indigenous
population, however, was less effective than two of the five introduced
VAM fungi under amended soil conditions (Chapter IV).
Aeschynomene is used extensively in Florida as a forage legume to
supply fixed nitrogen as protein and minerals to grazing animals (Hodges
et al., 1982) Siratro has been used sparingly in Florida (Kretschmer,
1972), but is one of the most widely used legumes throughout tropical
regions of the world (Lynd et al.,1985). However, for the satisfactory
establishment and growth of aeschynomene and Siratro in highly acidic
and P-deficient soils, lime and P fertilizer must be applied (Snyder et
al., 1985). Recent greenhouse experiments have shown that better growth
of forage legumes in these soils can be achieved by inoculation with
selected species of VAM fungi (Chapter IV and V). Glomus etunicatum and
G. intraradices were found to be effective growth enhancers of Siratro
and aeschynomene in a nonpasteurized, limed (3000 kg ha ^) soil similar
to the one used in the present study, over a range of 20-80 kg ha ^ of
applied P. It is therefore of considerable interest to determine whether
inoculation with selected isolates of VAM fungi can improve the
j.1 . r.- .. \ \ 7. '* T
establishment, growth, and nutrient uptake of Siratro and aeschynomene,
under field conditions where soil was limed and fertilized with different
P levels.

57
Materials and Methods
The soil used was a native Oldsmar fine sand (sandy, siliceous,
hyperthermic Alfic Arenic Haplaquods) with a pH of 4.6 (soil:^0=1:2),
1.5% organic matter, and the following Mehlich-I extractable elements in
mg kg : P 1.0, Ca 65, Mg 12, and K 15.
The experiment was designed as a 2 x 3 x 4 factorial consisting of
two legume species; Siratro and A. americana; three inoculation
treatments, G. etunicatum, G. intraradices, and the control; and four P
treatments; 10, 30, 60, and 120 kg ha 1 as triple superphosphate. The
24 treatments were arranged in a randomized complete block design with
ten replications per treatment. The P treatments were surface applied
by hand on 3 July 1986, along with a basal application of lime (high
calcitic limestone), Mg, nutritional spray (Diamond Fertilizer Co., Ft.
Pierce, FL.), and Mo at 3000, 25, 22, and 0.2 kg ha \ respectively, and
incorporated using a rake to a depth of approximately 15 cm. Potassium
was broadcasted on each plot at a rate of 60 kg ha 1 as KCL on 5 August
1986.
Seeds of Siratro and aeschynomene inoculated with rhizobium type El
"Cowpea" inoculum (Nitragin Co., Milwaukee, WI) were sown in pasteurized
Oldsmar fine sand, amended with high calcitic limestone at 1500 mg kg'1
and P at 12.5 mg kg \ in cells of "speedling" styrofoam trays (72 cells
per tray) on 20 June 1986.
Seedlings were inoculated or not inoculated with G. etunicatum
(isolate S312), G. intraradices (isolate S311). The amount of soil-root
inoculum used for each VAM fungi was adjusted to give equal inoculum

58
densities as determined by the MPN technique (Daniels and Skipper, 1982).
Approximately 180 propagules were added mid-way down each cell before
seeding. Control seedlings received 15 g of a soil-root mixture from
nonmycorrhizal pot cultures and the equivalent of 20 kg P ha ^, which
was applied in solution 10 d after germination in an attempt to make the
P status of the mycorrhizal and nonmycorrhizal seedlings similar at the
time of transplanting.
Seeded trays were placed in a glasshouse for 6 wk, after which the
whole cell content (each with one seedling) was transplanted to the
field. Siratro seedlings were cut back to three nodes each before
transplanting. Seedlings were transplanted on 4 August 1986.
One seedling was planted by hand in the middle of each 1.0 by 1.0 m
plot which were surrounded by alleyways of 1.0m. Extra seedlings were
weighed, and P concentration and root colonization were determined.
Harvests of Siratro and aeschynomene were made on 2 October 1986.
Siratro, a perennial, also was harvested on 27 November 1986, 5 May 1987,
and 29 June 1987. Herbage was dried at 75C for 24 h and weighed. Five
subsamples per treatment from the first harvest of Siratro and
aeschynomene foliage were analyzed for N and P content by automated
colorimetry (Technicon Industrial Systems Method No. 334-74 W/B,
Technicon Instruments Corp., Tarrytown, NY). Five root samples per
treatment, consisting each of four subsamples, were used to assess
mycorrhizal root colonization. Percentage of root colonized by VAM fungi
of aeschynomene and Siratro (1st and 4th harvests), was estimated as
described in Chapter III.

59
Soil samples (0-15 cm) were taken at the 4th harvest and analyzed
for pH, P, Ca, Mg, and K using the Mehlich-I extractant method. All
elements were determined by inductively coupled argon plasma (ICAP)
spectrometry. Data for all variables were subjected to ANOVA procedures
and regression analysis (SAS Institute Inc., 1982).
Results and Discussion
Pre-inoculated seedlings were used in this study to ensure the
legume roots were well colonized with the selected VAM fungi, because
this was thought to be the most certain way to ensure establishment of
the inoculum. Once a plant response is ascertained, methods of
inoculation more applicable on a large scale can be tested.
As a result of lime application, soil pH increased from 4.6 to about
6.2, and extractable Ca increased up to about 650 mg kg ^. Phosphorus
applications of 10, 30, 60, and 120 kg ha ^ resulted in extractable P in
the soil of 5.1, 8.6, 17.8, and 34.3 mg kg ^ at the end of the
experiment, respectively. Thus the legume species and VAM fungi in the
present study were exposed to a considerable range of soil P.
Seedlings inoculated with VAM fungi were similar in shoot dry weight
and P concentration to control seedlings at transplanting (Table 7-1).
Seedlings inoculated with VAM fungi were also well colonized, whereas no
VAM colonization was detected on control seedlings (Table 7-1).
Phosphorus amendments and fungal inoculation increased shoot dry
weights of aeschynomene (Table 7-2) and all harvests of Siratro

60
Table 7-1. Shoot dry weights, P concentrations, and percentage of
roots colonized of Siratro and aeschynomene seedlings at
transplanting.
VAM
Inoculation
Siratro
aeschynomene
Shoot dry
wt.
Shoot
P
Root
colon.
shoot dry
wt.
Shoot Root
P colon.
mg
--- %

mg
%
G. etunicatum
302
.18
52
304
.17 57
G. intraradices
305
.17
60
299
.20 55
Control
298
.16
0
302
.18 0
zData are means of five replicates.

61
Table 7-2. Analysis of variance for shoot dry weights of Aeschynomene
americana harvested 2 October 1986.
Source of
variation
DF
Mean squares
Block
9
52.57
Fungi (F)
2
1654.72**
Phosphorus (P)
3
6986.26**
Linear (PI)
1
18700.24**
Quadratic (Pq)
1
2250.32**
Cubic (Pc)
1
8.21
F x P
6
91.79
F x PI
2
53.72
F x Pq
2
215.53
F x Pc
2
6.11
Error
99
45.83
Significant at P < 0.01

62
(Table 7-3). There were no fungi x P interactions for shoot dry weights
of Siratro or aeschynomene.
At all levels of applied P and at all harvests, shoot dry weights
of Siratro were greater for fungal inoculated plants than control plants
(Fig. 7-1). Except for the third harvest, where the effect of fungal
inoculation was less pronounced at all levels of applied P, differences
between fungal inoculated and noninoculated plants were most marked at
intermediate levels of applied P (30-90 kg ha ^) and diminished at the
highest level (120 kg ha ^). The effect of inoculation with VAM fungi
on the shoot dry weights of aeschynomene, at all levels of applied P,
followed the same trend as that of Siratro although the response to
mycorrhizal inoculation was greater (Fig. 7-2).
Inoculated plants of Siratro (Table 7-4 and 5) and aeschynomene
(Table 7-6) had a quadratic relationship between shoot dry weight and P
application. Maximum shoot dry weight of Siratro was achieved between
75-85 kg ha ^ of P and for aeschynomene at 85 kg ha ^ of P, whereas
control plants of both legumes, even with 120 kg ha ^ of P, did not reach
the same shoot dry weight as fungal inoculated plants. Thus inoculation
with VAM fungi resulted in at least a 30% savings (40 kg ha ^) in the
amount of P fertilizer required for maximum yield. In a previous
greenhouse experiment (Chapter VI), I found that inoculation with G.
etunicatum resulted in a 20% decrease in the amount of P required for
maximum yield of Siratro. Some previous reports of VAM field experiments
with legumes in nonpasteurized soil (Black and Tinker, 1977; Khan, 1975)
show responses to VAM inoculation only in the absence of P fertilizer.

63
Table 7-3. Analysis of variance for shoot dry weights from four
Siratro harvests.
Source of
Mean squares
variation
DF
Harvest 1
2 Oct. 86
Harvest 2 Harvest 3
27 Nov. 86 5 May 87
Harvest A
29 June 87
Block
9(6)z
6. AO
A.1A
11.A7
18.70'
Fungi (F)
2
332.20
2A7.79**
69.20**
177.21**
Phosphorus (P)
3
807.52
1577.91**
1598.20**
1806.27**
Linear (PI)
1
2107.A7
A021.98**
A223.03*
A630.88**
Quadratic (Pq)
1
305.0A
711.A9**
559.16**
773.57**
Cubic (Pc)
1
10.07
6.35
0.23
12.38
1A.32
F x P
6
3.21
1.67
10.12
F x PI
2
A. 61.**
3.89
1.10
7.69
F x Pq
2
12.82
5.38
2.76
15.83
F x Pc
2
1.63
0.38
1.1A
6.83
Error
99(66)
8.89
13.A3
9.20
7.98
''Significant at
P < 0.01
Significant at
P < 0.05
zValues in parentheses are the degrees of freedom for harvests 2, 3, and
A for which only 7 replicates were used.

g-DCTT CRY EI&T (g) SCOT CRY YEEHT (g)
64
0 30 00 00 120
P APPLIED (kgAn)
Fig. 7-1. Effect of P application on shoot dry weights for the first
(A), second (B), third (C), and fourth (D) harvest of
Siratro grown under field conditions and inoculated with
Glomus etunicatum (ETU), Glomus intraradices (INT), or not
inoculated (CON). Data points are means of five
replicates.

RDOT COLONIZATION CD SHOOT DRY HEIGHT (g)
65
P RPPLIED (kg/ha)
Fig. 7-2. Effect of P application on shoot dry weight and percentage
of root colonized of Aeschynomene americana grown in the
field and inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.

66
Table 7-4. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P and N
concentrations, and total P and N uptake for the first
harvest of Siratro.
Variable
Regression equations
r2
Glomus etunicatum
Shoot dry wt.
(g)
=
12.36+0.33P-0.0022P2
0.80**
Root coloniz.
(%)
=
5.31+0.77P-0.0043P2
0.91**
Plant P cone.
(%)
=
0.22+0.00072P
0.66**
Total Plant P
(mg)
=
17.73+1.36P-0.0069P2
0.98**
Plant N cone.
(%)
=
2.26+0.035P-0.00025P2
0.52**
Total Plant N
(mg)
=
170.94+20.09P-0.13P2
0.91
Glomus intraradices
Shoot dry wt.
(g)
=
12.03+0.24P-0.0014P2
0.67**
Root coloniz.
(%)
=
10.04+0.59P-0.0034P2
0.86'c*
Plant P cone.
(%)
=
0.16+0.0037P-0.000018P2
0.76"*
Total plant P
(mg)
=
16.08+1.31P-0.0068P2
0.94**
Plant N cone.
(%)
=
2.12+0.041P-0.00026P2
0.56"*
Total plant N
(mg)
=
226.21+15.50P-0.095P2
0.91**
Control
Shoot dry wt.
(g)
=
10.65+0.IIP
0.65**
Plant P cone.
(%)
=
0.18+0.001IP
0.81**
Total plant P
(mg)
=
12.36+0.48P
0.92**
Total plant N
(mg)
=
173.09+3.57P
0.89"'*
P = phosphorus level
Significant at P < 0.01

67
Table 7-5. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weights for the second, third, and fourth harvest and
percentage of root colonization for the fourth harvest of
Siratro.
Variable
Regression equation
r2
Glomus etunicatum
Shoot dry wt (g), harvest
2
=
5.7A+0.52P-0.0029P2
0.87**
Shoot dry wt (g), harvest
3
=
0.99+0.A8P-0.0029P2
0.89^
Shoot dry wt (g), harvest
A
=
5.98+0.55P-0.0032P2
0.87"*
Root colon. (%), harvest
A
=
9.85+0.8AP-0.00A6P2
0.86x*
Glomus intraradices
Shoot dry wt (g), harvest
2
=
6.58+0.A6P-0.0025P2
0.8A**
Shoot dry wt (g), harvest
3
=
2.05+0.A3P-0.0025P2
0.86**
Shoot dry wt (g), harvest
A
=
6.1A+0.52P-0.0027P2
0.87**
Root colon. (%), harvest
A
=
15.03+0.69P-0.0039P2
0.89**
Control
Shoot dry wt (g), harvest
2
=
1.56+0.AAP-0.0023P2
0.81**
Shoot dry wt (g), harvest
3
=
-0.23+0.A2P-0.0021P2
0.88**
Shoot dry wt (g), harvest
A
3.38+0.A2P-0.0017P2
0.91**
"''Significant at P < 0.01
P = phosphorus level

68
Table 7-6. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P
concentration, and total P and N uptake of Aeschynomene
americana.
Variable
Regression equations
r2
Glomus etunicatum
Shoot dry wt.
(g)
=:
46.59+0.83P-0.0049P2
0.78**
Root coloniz.
(%)
=
7.89+0.88P-0.0047P2
0.89**
Plant P Cone.
(%)
=
0.21+0.0037P-0.000019P2
0.80**
Total plant P
(mg)
=
67.54+4.96P-0.025P2
0.96**
Total plant N
(g)
=
1.35+0.039P-0.00019P2
0.93'*
Glomus intraradices
Shoot dry wt.
(g)
=
40.86+0.97P-0.0056P2
0.86
Root coloniz.
(%)
=
9.26+0.98P-0.0058P2
0.93**
Plant P Cone.
(%)
=
0.16+0.0053P-0.000029P2
0.76
Total plant P
(mg)
=
43.24+5.85P-0.03IP2
0.95"'*
Total plant N
(g)
=
1.18+0.043P-0.00022P2
0.91"x
Control
Shoot dry wt.
(g)
-
43.50+0.31P
0.76**
Plant P Cone.
(%)
=
0.11+0.0045P-0.00002IP2
0.95**
Total plant P
(mg)
=
28.08+3.63P-0.012P2
0.99"*
Total plant N
(g)
=
1.44+0.015P
0.81**
P = Phosphorus level
Significant at P < 0.01

69
Hayman and Mosse (1979), however, reported improved growth of white
clover in the field after inoculation with VAM fungi and the addition of
90 kg ha of P. They also indicated that responses to fungal
inoculation were smaller with 23 kg ha ^ of P and were absent where no P
was added. Similarly, Hall (1984) reported that inoculation with
selected VAM fungi increased yield of white clover in the field only if
50 kg ha 1 of P was also applied.
Several greenhouse experiments on the phosphate response curves of
fungal inoculated and noninoculated forage legumes have been carried out,
mostly using clovers (Abbott and Robson, 1977b; Sparling and Tinker,
1978; Powell, 1980b) and Siratro (Lynd et al., 1985; Medina et al.,
1987d) as the test plants. These authors applied soluble P fertilizers
at rates ranging from 0 to 250 kg ha ^ and have reached the general
conclusion that inoculation with VAM fungi markedly increases legume
growth at low and intermediate rates of applied P. From the practical
point of view, however, the interactions between phosphate additions and
VAM on legumes are not always predictable and generalizable, because the
responses are modulated by the incidence of several factors. These
include the physical and chemical characteristics of the soil, plant
species, VAM fungi, and the complex interactions between these factors.
At the first harvest of Siratro (Fig. 7-1A), plants inoculated with
G. etunicatum had higher shoot dry weights than plants inoculated with
G. intraradices at all levels of applied P. However, in subsequent
harvests of Siratro (Fig. 7-1BC and D) and for aeschynomene (Fig. 7-2)
the response of shoot dry weight to inoculation with the two VAM fungi
was not different.

70
Phosphorus additions and fungal inoculation increased percentage of
root colonized by VAM fungi of Siratro (Table 7-7) and aeschynomene
(Table 7-8). There were fungi x P interactions for percentage of root
colonized by VAM fungi for for both legumes. Inoculated treatments had
greater percentage of root colonized than control treatments at all
levels of applied P. Percentage of root colonized by VAM fungi for the
inoculated plants of Siratro (Fig. 7-3A and B) and aeschynomene
(Fig. 7-2) increased linearly with P additions up to 60 kg ha ^.
Phosphorus application of 120 kg ha ^ did not affect the percentage of
root colonized by VAM fungi. In a previous greenhouse study (Chapter
V), I found that percentage of Siratro root length colonized by VAM fungi
increased with P additions up to 40 mg kg \ which is equivalent to 80
kg ha ^ of P. Abbott and Robson (1977b) and Schubert and Hayman (1986),
also in pot experiments, reported an increase in the percentage of root
length colonized up to 55 mg kg ^ of P (110 kg ha ^).
Phosphorus applications did not alter the percentage of root
colonized in the control plants. The degree of root colonization of the
control plants of Siratro increased from 9% in the first harvest to about
19% by the fourth harvest (Fig. 7-3A and B), but still failed to increase
the shoot dry weight of the control plants compared to the inoculated
plants. Fungal inoculated plants of Siratro (Table 7-4 and 5) had a
quadratic relationship between percentage of root colonized and applied
P. Maximum root colonization was attained between 85-90 kg ha1 of P
for both legumes. At lower P additions, Siratro plants inoculated with
G. intraradices had greater percentage of root colonized than plants
inoculated with G. etunicatum (Fig. 7-3). However, there were no

71
Table 7-7. Analysis of variance for percentage root colonized of
Siratro by VAM fungi.
Source of
variation
DF
Mean
Harvest 1
2 Oct. 86
squares
Harvest 4
29 June 87
Block
4
9.27
14.61
Fungi (F)
2
2394.82
1206.47
Phosphorus (P)
3
935.53
1230.09
Linear (PI)
1
2195.71
2538.78
Quadratic (Pq)
1
552.25
1083.74
Cubic (Pc)
1
58.62
67.74
F x P
6
230.66**
86.82'*
F x PI
2
510.937'7'
174.04**
F x Pq
2
112.49**
77.26**
F x Pc
2
68.55'
9.16
Error
44
7.98
11.91
Significant at P < 0.05
Significant at P < 0.01

72
Table 7-8. Analysis of variance for percentage root colonized, P
concentration, and total P and N uptake of Aeschynomene
americana harvested 2 October 1986.
Source of
variation
DF
Mean Squares
Root
colon.
P cone.
total P
total N
Block
4
10.11
0.00065^
384.02
0.051
Fungi (F)
2
5276.62
0.009T*
9052.18
0.39**
Phosphorus (P)
3
1844.64
0.051^*
75570.74
4.64^
Linear (PI)
1
4408.21
0.12**
196270.22
12.04**
Quadratic (Pq) 1
1001.65
0.031**
29179.92
1.89**
Cubic (Pc)
1
124.08,
0.0012
1262.07
0.0021
F x P
6
378.66'^
0.0010
1190.91**
0.026
F x PI
2
833.18'{*
0.0021
1275.75x*
0.013
F x Pq
2
220.16^
0.00052
1815.29**
0.043
F x Pc
2
82.64**
0.00021
481.71
0.022
Error
44
11.60
0.0011
206.45
0.062
Significant at P < 0.01
''Significant at P < 0.05

ROOT COLONIZATION (Z) ROOT COLONIZATION (23
73
Fig. 7-
Effect of P application on percentage of root colonized
for the first (A) and fourth (B) harvest of Siratro grown
in the field and inoculated with Glomus etunicatum (ETU).
Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.

74
differences in root colonization between G. etunicatum and G.
intraradices at any level of applied P. For aeschynomene, differences
between G. intraradices and G. etunicatum were only at 30 kg ha ^ of
applied P (Fig. 7-2).
At the first harvest, P amendments and fungal inoculation also had
an effect on P concentration, total P uptake, N concentration, and total
N uptake of Siratro (Table 7-9). These same parameters were increased
for aeschynomene, except for N concentration (Table 7-8). There were
fungi x P interactions for P concentration, total P uptake, N
concentration, and total N uptake of Siratro. By the contrast,
aeschynomene only had fungi x P interaction for total P uptake.
Inoculated plants of Siratro had greater P concentration,
total P uptake, N concentration, and total N uptake than control plants
at low and intermediate levels of applied P (Fig. 7-4). At the highest
level of applied P, total P uptake of Siratro plants inoculated with G.
intraradices and P and N concentrations of plants inoculated with G.
etunicatum did not differ from those of control plants. Fungal
inoculated plants of aeschynomene also had greater P concentration, total
P uptake, and total N uptake than control plants at low and intermediate
levels of applied P, but not at the highest level (Fig. 7-5). The
positive effect of inoculation with VAM fungi on P and N uptake have been
shown for other forage legumes, for example, leucaena (Habte and
Manjunath, 1987), Pueraria phaseoloides (Sanchez and Salinas, 1981),
and field experiments with white clover (Hall, 1984; Hayman and Mosse,
1979) and Medicago sativa (Azcon-Aguilar and Barea, 1981). Potassium,

75
Table 7-9. Analysis of variance for P and N concentrations, and total
P and
N uptake of
Siratro for
the first harvest.
Source of
Mean
Squares
variation
DF
P Cone.
total P
N Cone. total N
Block
4
0.00033
26.06
0.42
10491.95**
Fungi (F)
2
0.0059
2556.03
3.03
540732.85
Phosphorus (P)
3
0.039
7248.85
1.47
544117.23
Linear (PI)
1
0.11
19691.07
1.53
1006530.96
Quadratic (Pq)
1
0.0049
2055.48
2.87
625338.95
Cubic (Pc)
1
0.000077
0.010
0.010
487.84
F x P
6
0.00237
175.50*
0.33
33638.14**
F x PI
2
0.0031';
53.34,
0.21,
7291.34
F x Pq
2
0.0039
465.13**
0.75
92589.06x*
F x Pc
2
0.00011
8.02
0.040
1034.01
Error
44
0.00087
26.84
0.14
3606.65
"Significant at
P <
0.01
Significant at P
< 0.05

TOTFL FLFNT P (mg) FLflNT P CCNC. C)
76
P PPPLIH3 (kg/ha)
P fffl-ia (Ig/ha)
Effect of P application on P concentration, total P, N
concentration, and total N of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.
Fig. 7-4.

77
C 3
CD
C_J
Q_
I
cr
32
CD!
cd 2
I
CD
CON
ETU
INT
30 60 60 120
P PPLIED (kg/ho)
Fig. 7-5. Effect of P application on P concentration, total P, and
total N of Aeschynomene americana grown in the field
conditions and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.

78
Ca, Mg, and Zn concentrations of Siratro and aeschynomene were not
related to P applications or fungal inoculation (data not presented).
In contrast to greenhouse pot experiments, field experiments on
inoculation with VAM fungi often have been unsuccessful. It is possibly
that the main cause for the satisfactory response to fungal inoculation
obtained in the present study was due to a largely ineffective indigenous
VAM population as compared to G. etunicatum and G. intraradices under
the amended soil conditions. In previous greenhouse experiments (Chapter
IV and V) using a similar nonpasteurized soil than the one used in the
present study, I found that G. etunicatum and G. intraradices were
effective growth enhancers of Siratro, even with a reasonably high soil
native VAM population. Similarly, Powell et al. (1980b) reported that
indigenous VAM fungi were ineffective in many soils and that inoculation
by more effective VAM fungi would result in positive responses, even in
nonpasteurized soils containing a high indigenous VAM population.
In conclusion, the present study shows that effective inoculation
with selected VAM fungi can have an important effect on growth of forage
legumes in the field in soils that contain ineffective native VAM
populations under amended soil conditions, even at moderate levels of
applied P.

CHAPTER VIII
CONCLUSIONS
The objective of this chapter is to summarize the work of the
preceding six chapters.
The overall goal of this research project was to improve the
establishment phases and growth of tropical forage legumes in newly
cleared land at reduced P fertilization through inoculation with
effective VAM fungi. In order to accomplish this goal, greenhouse
studies were carried out in limed, pasteurized and nonpasteurized Oldsmar
fine sand, collected from a newly cleared area at the Agricultural
Research and Education Center, Fort Pierce, FL. A field experiment was
conducted in the same nonpasteurized soil.
In Chapter II, quantitative data on the amount of root colonization
and the species distribution of VAM fungi associated with four cultivated
tropical forage legumes from four different locations in south Florida
are reported. Differences in percentage of root colonization and total
spore density were significant among locations, legume species, and
location x legume species interactions. Legume species differed in
percentage root colonization and total spore density among locations
except for carpon desmodium, which showed no differences among locations
in percentage root colonization. The six species of VAM fungi collected
in this survey were: G. heterogama, G. margarita, G. etunicatum, G.
intraradices, Glomus sp., and A. spinosa.
79

80
In Chapter III, the effect of inoculation with G. intraradices on
the growth of several tropical forage legumes in P-deficient,
nonpasteurized and pasteurized soil under greenhouse conditions is
reported. Shoot and root dry weights were increased after inoculation
in nonpasteurized soil for Siratro, aeschynomene, Aeschynnomene villosa,
Stylosanthes hamata, and Stylo, but not for Arachis sp. and Vigna
adenantha, which only responded to inoculation in pasteurized soil.
In Chapter IV, the effect of five species of VAM fungi, G.
etunicatum, G. deserticola, G. versiforme, G. intraradices, and G.
margarita, on the growth of Siratro in a limed, nonpasteurized soil with
an applied P level of 20 mg kg ^ under greenhouse conditions was
determined. Shoot dry and root fresh weights of plants inoculated with
G. etunicatum and G. intraradices were higher than the other VAM fungal
treatments and noninoculated plants. In addition, plants inoculated with
G. etunicatum had higher shoot dry weights than plants inoculated with
G. intraradices. The indigenous population of VAM fungi was reasonably
high (MPN = 2 propagules g ^ soil); however, plant yields were less than
the best VAM treated plants. A positive correlation was found between
mycorrhizal root colonization, expressed as either percentage or total
root length colonized, and shoot dry weight. Glomus etunicatum colonized
roots more rapidly than the other VAM fungi tested.
In Chapter V, the effect of G. etunicatum and G. intraradices, on
the growth of Siratro in a limed, nonpasteurized soil, with applied P
levels of 2.5, 10, 20, and AO mg kg ^ under greenhouse conditions is
reported. At 2.5 mg kg ^ of applied P, there was no yield response to
inoculation. Above 2.5 mg kg ^ of applied P, plants inoculated with

81
either G. etunicatum or G. intraradices weighed more than control plants.
Inoculated plants required between 28 and 30 mg P kg ^ to achieve maximum
shoot dry weight, whereas control plants, even with 40 mg P kg 1, did
not achieve maximum growth. Shoot dry weight response was better with
G. etunicatum than with G. intraradices. For both fungi, increasing P
above 2.5 mg kg 1 increased the percentage and total root length
colonized by VAM fungi.
In Chapter VI, the effect of inoculation with G. etunicatum, on the
growth and uptake of P and N of three forage legumes at applied P levels
between 12.5 and 50 mg kg 1 in a limed, pasteurized soil under greenhouse
conditions is reported. At all levels of applied P, shoot dry and root
fresh weights, and total N of Stylo were greater for mycorrhizal plants
than nonmycorrhizal plants. The differences were most pronounced at
intermediate levels of applied P and diminished at the higher P levels.
Mycorrhizal plants of Siratro had greater shoot dry weight, total P and
N than nonmycorrhizal plants at low and intermediate P levels, but not
at the highest level. Differences in shoot dry weight between
mycorrhizal and nonmycorrhizal plants of aeschynomene were significant
only at the intermediate level of applied P. Overall, inoculated plants
of the three legumes studied had greater P concentration than control
plants. Maximum root colonization, expressed as either percentage or
total length of colonized root of the three legumes, was attained at
intermediate levels of applied P.
In Chapter VII, the effect of inoculation with selected VAM isolates
on growth and nutrient uptake of Siratro and aeschynomene under natural
field conditions at applied P levels of 10, 30, 60, and 120 kg ha'1 is

82
reported. At all levels of applied P and for all harvests, shoot dry
weights of Siratro were greater for fungal inoculated plants than
noninoculated plants. Differences between fungal inoculated and
noninoculated plants were most marked at 30 to 90 kg ha of applied P
and diminished at 120 kg ha ^. The effect of fungal inoculation on the
shoot dry weights of aeschynomene, at all levels of applied P, was
similar (but more pronounced) as that of Siratro. At the first harvest
of Siratro, plants inoculated with G. etunicatum had higher shoot dry
weights than G. intraradices plants at all levels of applied P. However,
in subsequent harvests of Siratro and for aeschynomene the response of
shoot dry weight to inoculation with the two VAM fungi was similar.
Fungal inoculation resulted in at least a 30% savings (AO kg ha ^) in
the amount of P fertilizer required for maximum yield. Inoculated
treatments had greater percentage of root colonized than noninoculated
treatments at all levels of applied P. Percentage of root colonized by
VAM fungi for the inoculated plants of the two legumes increased linearly
with P additions up to 60 kg ha *. There were no differences in root
colonization between G. etunicatum and G. intraradices at any level of
applied P.
The results of these studies clearly demonstrate that inoculation
with effective VAM fungi can increase the growth of legumes in soils that
may have a high, but largely ineffective native VAM population than
introduced VAM fungi under amended soil conditions. Furthermore, with
highly mycorrhizal dependent crops such as tropical forage legumes, a
mycorrhizal growth response may occur at P levels normally used in
commercial pasture production.

LITERATURE CITED
Abbott, L.K., and A.D. Robson. 1977a. The distribution and abundance of
vesicular-arbuscular endophytes in some Western Australian soils.
Aust. J. Bot. 25: 515-522.
Abbott, L. K., and A. D. Robson. 1977b. Growth stimulation of
subterranean clover with vesicular-arbuscular mycorrhizas. Aust. J.
Agrie. Res. 28:639-649.
Abbott, L. K., and A. D. Robson. 1978. Growth of subterranean clover
in relation to formation of endomycorrhizas by introduced and
indigenous fungi in a field soil. New Phytol. 81:575-587.
Abbott, L. K., and A. D. Robson. 1981. Infectivity and effectiveness
of vesicular-arbuscular mycorrhizal fungi: Effect of inoculum type.
Aust. J. Agrie. Res. 32:631-639.
Anderson, D. L., and L. J. Henderson. 1986. Sealed chamber digestion
for plant nutrient analyses. Agron. J. 78:937-939.
Asimi, S., V. Gianinazzi-Pearson, and S. Gianinazzi. 1980. Influence of
increasing soil phosphorus levels on interactions between vesicular-
arbuscular mycorrhizae and Rhizobium in soybeans. Can. J. Bot.
58:2200-2206.
Azcon-Aguilar, C., and J. M. Barea. 1981. Field inoculation of Medicago
with VA mycorrhiza and rhizobium in phosphate-fixing agricultural
soil. Soil Biol. Biochem. 13:19-22.
Barber, D. A., and B. C. Lougham. 1967. The effect of microorganisms
in the absorption of inorganic plant nutrients. II. Uptake and
utilization of phosphate by barley plants grown under sterile and
non-sterile conditions. J. Exp. Bot. 18:170-176.
Barea, J. M., J. L. Escudero, and C. Azcon-G. de Aguilar. 1980. Effects
of introduced and indigenous VA mycorrhizal fungi on nodulation and
nutrition of Medicago sativa in phosphate fixing soils as affected
by P fertilizers. Plant Soil 54:283-296.
Bergersen, F. J. 1971. Biochemistry of nitrogen fixation in legumes.
Annu. Rev. Plant Physiol. 22:124-140.
83

84
Bethlenfalvay, G.J., J.M. Ulrich, and M.S. Brown. 1985. Plant response
to mycorrhizal fungi: host, endophyte, and soil effects. Soil Sci.
Soc. Am. J. 49: 1164-1168.
Black, R. L. B., and P. B. Tinker. 1977. Interactions between effects
of vesicular-arbuscular mycorrhiza and fertilizer phosphorus on
yields of potatoes in the field. Nature 267:510-511.
Burt, R. L., and C. P. Miller. 1975. Stylosanthes--A source of pasture
legumes. Trop. Grassl. 9:117-123.
Chulan, A., and P. Ragu. 1986. Growth response of Theobroma cacao L.
seedlings to inoculation with vesicular-arbuscular mycorrhizal
fungi. Plant Soil 96:279-285.
Cooper, K. M. 1978. Adaption of mycorrhizal fungi to phosphate
fertilizers, p. 107. In A. R. Ferguson, R. L. Bieleski, and I. B.
Ferguson (eds.) Plant nutrition 1978. New Zealand DSIR Information
Series No. 134, Government Printer, Wellington.
Crush, J. R. 1974. Plant growth responses to vesicular-arbuscular
mycorrhiza. VII. Growth and nodulation of some herbage legumes.
New Phytol. 73:745-750.
Currah, R.S., and M. Van Dyk. 1986. A survey of some perennial vascular
plant species native to Alberta for occurrence of mycorrhizal fungi.
Canadian Field-Naturalist 100: 330-342.
Daft, M.I., and A.A. El-Giahmi. 1976. Studies on nodulated and
mycorrhizal peanuts. Ann. Appl. Biol. 83:273.
Daniels, B. A., P. M. McCool, and J. A. Menge. 1981. Comparative
inoculum potential of spores of six vesicular-arbuscular mycorrhizal
fungi. New Phytol. 89:385-391.
Daniels, B. A., and H. D. Skipper. 1982. Methods for recovery and
quantitative estimation of propagules from soil. p. 29-35. In N. C.
Schenck (ed.) Methods and principles of mycorrhizal research. Amer.
Phytopath. Soc., St. Paul, MN.
Gallaher, R. N., C. 0. Weldon, and J. G. Futral. 1975. An aluminum
block digester for plant and soil analysis. Soil Sci. Soc. Amer.
Proc. 39:803-806.
Gates, C. T., and J. R. Wilson. 1974. The interaction of nitrogen and
phosphorus on the growth, nutrient status and nodulation of
Stylosanthes humilis H.B.K. (Townsville stylo). Plant Soil 41:325-
333.
Gerdemann, J.W., and J.M. Trappe. 1974. The endogonaceae in the Pacific
Northwest. Mycol. Mem. 5: 1-76.

85
Gibson, A. H. 1976. Limitations to nitrogen fixation in legumes. In
W. E., Newton, and C. J. Nyman (eds.) Proc. 1st Int. Symp. Nitrogen
Fixation, Vol.2. Wash. State Univ. Press, Pullman.
Giovannetti, M. 1985. Seasonal variations of vesicular-arbuscular
mycorrhizas and endogonaceous spores in a maritime sand dune.
Trans. Br. Mycol. Soc. 8A: 679-684.
Giovannetti, M., and B. Mosse. 1980. Evaluation of techniques for
measuring vesicular-arbuscular mycorrhizal infection in roots. New
Phytol. 84:489-500.
Giovannetti, M., and T.H. Nicolson. 1983. Vesicular-arbuscular
mycorrhizas in Italian sand dunes. Trans. Br. Mycol. Soc. 80: 552-
557.
Habte, M., and T. Aziz. 1985. Response of Sesbania grandiflora to
inoculation of soil with vesicular-arbuscular mycorrhizal fungi.
Appl. Environ. Microbiol. 50: 701-703.
Habte, M., and A. Manjunath. 1987. Soil solution phosphorus status and
mycorrhizal dependency in Leucaena leucocephala. Appl. Environ.
Microbiol. 53:797-801.
Hall, I. R. 1978. Effects of endomycorrhizas on the competitive ability
of white clover. N. Z. J. Agr. Res. 21:509-515.
Hall, I. R. 1979. Soil pellets to introduce vesicular-arbuscular
mycorrhizal fungi into soil. Soil Biol. Biochem. 11:85-86.
Hall, I. R. 1984. Field trials assessing the effect of inoculating
agricultural soils with endomycorrhizal fungi. J. Agrie. Sci.
102:725-731.
Hambleton, L. G. 1977. Semiautomated method for simultaneous
determination of phosphorus, calcium and crude protein in animal
feeds. J. Assoc. Offic. Agr. Chem. 60:845-852.
Harley, J.L., and E.L. Harley. 1987. A check-list of mycorrhiza in the
British flora. New Phytol. 105:1-102.
Harley, J. L., and S. E. Smith. 1983. Mycorrhizal symbiosis. Academic
Press, London.
Hass, J. H., and J. Krikum. 1985. Efficacy of endomycorrhizal fungus
isolates and inoculum quantities required for growth response. New
Phytol. 100:613-621.
Hayman, D.S. 1978. Mycorrhizal populations of sown pastures and native
vegetation in Otago, New Zealand. N. Z. J. Agr. Res. 21: 271-275.

86
Hayman, D. S. 1983. The physiology of vesicular-arbuscular
endomycorrhizal symbiosis. Can. J. of Botany 61:944-963.
Hayman, D. S., and K. A. Hampson. 1978. VA mycorrhiza. Field
inoculation trial (white clover in Welsh upland soil). Rothamsted
Rep. Part 1:238-239.
Hayman, D. S., and B. Mosse. 1979. Improved growth of white clover in
hill grasslands by mycorrhizal inoculation. Ann. Appl. Biol.
93:141-148.
Hayman, D. S., R. J. Page, and C. A. Clarke. 1980. Vesicular-
arbuscular mycorrhiza: field inoculation studies with Red clover,
Sawyers I. Rothamsted Rep. Part 1:201-202.
Hayman, D.S., and G.E. Stovold. 1979. Spore populations and infectivity
of vesicular-arbuscular mycorrhizal fungi in New South Wales. Aust.
J. Bot. 27: 227-233.
Hayman, D. S., and M. Tavares. 1985. Plant growth responses to
vesicular-arbuscular mycorrhiza. XV. Influence of soil pH on the
symbiotic efficiency of different endophytes. New Phytol. 100:367-
377.
Hodges, E. M., A. E. Kretschmer, Jr., P. Mislevy, R. D. Roush, 0. C.
Ruelke, and G. H. Snyder. 1982. Production and utilization of the
tropical legume aeschynomene. Fla. Agrie. Exp. Stn. Circ. S-290.
Huang, R., W.K. Smith, and R.S. Yost. 1985. Influence of vesicular-
arbuscular mycorrhiza on growth, water relations, and leaf
orientation in Leucaena leucocephala (Lam.) Dewit. New Phytol. 99
229-243.
Hutton, E. M. 1962. Siratro-a tropical pasture legume bred from
Phaseolus atropurpureum. Aust. J. Agrie. Anim. Hus. 2:117-125.
labal, S. H., K. Sultana, and B. Perveen. 1975. Endogone spore numbers
in the rhizosphere and the occurrence of vesicular-arbuscular
mycorrhizas in plants of economic importance. Biologia 21:227-237.
Islam, R., A. Ayanaba, and F. E. Sanders. 1980. Response of cowpea
(Vigna unguiculata) to inoculation with VA mycorrhizal fungi and
rock phosphate fertilization in some unsterilized Nigerian soils.
Plant Soil 54:107-117.
i i j . ....
Jensen, A. 1984. Responses of barley, pea, and maize to inoculation
with different vesicular-arbuscular mycorrhizal fungi in irradiated
soil. Plant Soil 78: 315-323.
Khan, A. G. 1975. Growth effects of VA mycorrhiza on crops in the
field. P. 149-435. In F. E. Sanders, B. Mosse, and P. B. Tinker
(eds.) Endomycorrhizas. Academic Press, London, New York.

87
Kormanik, P. P., and A. C. McGraw. 1982. Quantification of vesicular-
arbuscular mycorrhizae in plant roots, p. 37-45. In N. C. Schenck
(ed.) Methods and principles of mycorrhizal research. American
Phytopathological Society, St. Paul, MN.
Kretschmer, A. E., Jr. 1972. Siratro (Phaseolus atropurpureus DC.) A
summer-growing perennial pasture legume for central and south
Florida. Florida Agrie. Exp. Stn. Circ. S-214.
Kucey, R.M., and G.E. Diab. 1984. Effects of lime, phosphorus, and
addition of vesicular-arbuscular (VA) mycorrhizal fungi on
indigenous VA fungi and on growth of alfalfa in a moderately acidic
soil. New Phytol. 98: 481-486.
Lopes, E. S., and E. De Olivera. 1980. Efeito de species de micorrizas
vesicular-arbuscular em Siratro (Macroptilium atropurpureum).
Bragantia 39:241-245.
Lopes, E.S., E. Oliveira, R. Dias, and N.C. Schenck. 1983. Occurrence
and distribution of vesicular-arbuscular mycorrhizal fungi in coffee
(Coffea arabica L.) plantations in central Sao Paulo state, Brazil.
Turrialba 33:417-422.
Lynd, J. Q., R. J. Tyrl, and A. A. C. Purcino. 1985. Mycorrhiza soil
fertility effects on regrowth, nodulation and nitrogenase activity
of Siratro (Macroptilium atropurpureum (DC) Urb.). J. Plant Nutr.
8:1047-1059.
Menge, J. A. 1983. the physiology of vesicular-arbuscular mycorrhizal
fungi in agriculture. Can. J. Bot. 61:1015-1024.
Miller, R.H., E.J.B.N. Cardoso, and C.O.N. Cardoso. 1979. Some
observations on mycorrhizal infection of tropical forage legumes and
grasses in Brazil. Summa Phytopathologica 5: 168-172.
Miller, D. D., P. A. Domoto, and C. Walker. 1985. Colonization and
efficacy of different endomycorrhizal fungi with apple seedlings at
two phosphorus levels. New Phytol. 100:393-402.
Mosse, B. 1972. The influence of soil type and endogone strain on the
growth of mycorrhizal plants in phosphate deficient soil. Rev.
Ecol. Biol. Sol. T. 9:529-537.
Mosse, B. 1977. Plant growth responses to vesicular-arbuscular
mycorrhiza. X. Responses of stylosanthes and maize to inoculation
in unsterile soils. New Phytol. 78:277-288.
Mosse, B., C. L. Powell, and D. S. Hayman. 1976. Plant growth responses
to vesicular-arbuscular mycorrhiza. IX. Interactions between VA
mycorrhiza, rock phosphate and symbiotic nitrogen fixation. New
Phytol. 76:331-342.

88
Munns, D. N., and B. Mosse. 1980. Mineral nutrition of legume crops.
In R. J. Summerfield, and A. H. Bunting (eds.) Advances in legume
science. Univ. Reading Press, England
Newbould, P., and A. Rangeley. 1984. Effect of lime, phosphorus and
mycorrhizal fungi on growth, nodulation, and nitrogen fixation by
white clover ( Trifolium repens) grown in UK hill soils. Plant Soil
76: 105-114.
Nielsen, J.P., and A. Jensen. 1983. Influence of vesicular-arbuscular
mycorrhiza fungi on growth and uptake of various nutrients as well
as uptake ratio of fertilizer P for lucerne (Medicago sativa).
Plant Soil 70: 165-172.
Pairunan, A. K., A. D. Robson, and L. K. Abbott. 1980. The
effectiveness of vesicular-arbuscular mycorrhizas in increasing
growth and phosphorus uptake of subterranean clover from phosphorus
sources of different solubilities. New Phytol. 84:327-338.
Plenchette, C., V. Furlan, and J. A. Fortin. 1982. Effects of different
endomycorrhizal fungi on five host plants grown on calcined
montmorillonite clay. J. Amer. Soc. Hort. Sci. 107:535-538.
Powell, C.L1. 1979. Inoculation of white clover and ryegrass seed with
mycorrhizal fungi. New Phytol. 83: 81-85.
Powell, C. LI. 1980a. Mycorrhizal infectivity of eroded soils. Soil
Biol. Biochem. 12:247-250.
Powell, C. LI. 1980b. Phosphate response curves of mycorrhizal and non-
mycorrhizal plants. I. Responses to superphosphate. N. Z. J. Agr.
Res. 23:225-231.
Powell, C. LI. 1982. Selection of efficient VA mycorrhizal fungi.
Plant Soil 68:3-9.
Powell, C. LI., M. Groters, and D. Metcalfe. 1980. Mycorrhizal
inoculation of a barley crop in the field. N. Z. J. Agr. Res.
23:107-109.
Purcino, A. A. C., C. Lurlarp, and J. Q. Lynd. 1986. Mycorrhiza and
soil fertility effects with growth, nodulation and nitrogen fixation
of Leucaena grown on a Typic Eutrustox. Commun. in Soil Sci. Plant
Anal. 17:473-489.
Purcino, A. A. C., and J. Q. Lynd. 1985. Tripartite symbiosis of
Stylosanthes scabra Vog. influenced by soil fertility treatments of
a Typic Eutrustox. Agron. J. 77:455-458.
Rangeley, A., M.J. Daft, and P. Newbould. 1982. The inoculatation of
white clover with mycorrhizal fungi in unsterile hill soils. New
Phytol. 92: 89-102.

89
Rhue, R.D., and G. Kidder. 1984. Procedures used by the IFAS extension
soil testing laboratory and interpretation of results. Fla. Coop.
Ext. Serv. Circ. 596.
Rotar, P.P. 1983. Legumes in action. P. 21-22. In R.L. Burt et al.
(eds.) The role of Centrosema, Desmodium, and Stylosanthes in
improving tropical pastures. Westview Press, Colorado.
Saif, S. R. 1987. Growth responses of tropical forage plant species to
vesicular-arbuscular mycorrhizae. I. Growth, mineral uptake and .
mycorrhizal dependency. Plant Soil 97:23-35.
Saif, S.R., and A.G. Khan. 1975. The influence of season and stage of
development of plant on Endogone mycorrhiza of field-grown wheat.
Can. J. Microbiology 21: 1021-1024.
Salinas, J.G., J.I. Sanz, and E. Sieverding. 1985. Importance of VA
mycorrhizae for phosphorus supply to pasture plants in tropical
oxisols. Plant Soil 84: 347-360.
Same, B. I., A. D. Robson, and L. K. Abbott. 1983. Phosphorus, soluble
carbohydrates and endomycorrhizal infection. Soil Biol. Biochem.
15:593-597.
Sanchez, P. A., and J. G. Salinas. 1981. Low-input technology for
managing oxisols and ultisols in tropical America. Adv. in Agronomy
34:279-406.
Sanders, F. E., P. B. Tinker, R. L. B. Black, and S. M. Palmerley. 1977.
The development of endomycorrhizal root systems. I. Spread of
infection and growth-promoting effects with four species of
vesicular-arbuscular endophyte. New Phytol. 78:257-268.
Sanni, S. 0. 1976. Vesicular-arbuscular mycorrhiza in some Nigerian
soils and their effect on the growth of cowpea (Vigna unguiculata),
tomato (Lycopersicon esculentum) and maize (Zea mays). New Phvtol.
77:667-671.
SAS Institute Inc. 1982. SAS user 's guide: Statistics. SAS Institute
Inc., Cary, N.C.
Satterlee, L., B. Melton, B. McCaslin, and D. Miller. 1983. Mycorrhizal
effects on plant growth, phosphorus uptake, and N2(C2H4) fixation in
two alfalfa populations. Agron. J. 75: 715-716.
Schenck, N.C., and R.A. Kinloch. 1980. Incidence of mycorrhizal fungi
on six field crops in monoculture on a newly cleared woodland site.
Mycologia 72: 445-456.
Schenck, N.C., and G.S. Smith. 1981. Distribution and occurrence of
vesicular-arbuscular mycorrhizal fungi on Florida agricultural
crops. Proc. Soil and Crop Sci. Soc. of Fla. 40:171-175.

90
Schenck, N.C., and G.S. Smith. 1982. Additional new and unreported
species of mycorrhizal fungi (Endogonaceae) from Florida. Mycologia
74: 77-92.
Schroder, V., D. S. Hayman, and B. Mosse. 1977. Influence of VA
mycorrhiza on plant growth. Rothamsted Report for 1976, part 1, p.
286.
Schubert, A., and D. S. Hayman. 1986. Plant growth responses to
vesicular-arbuscular mycorrhiza. XVI. Effectiveness of different
endophytes at different levels of soil phosphate. New Phytol.
103:79-90.
Snyder, G. H., A. E. Kretschmer, Jr., and J. Alvarez. 1985. Agronomic
and economic response of three tropical legumes to lime and
phosphorus in an acid infertile spodosol. Agron. J. 77:427-432.
Sylvia, D.M. 1986. Spatial and temporal distribution of vesicular-
arbuscular mycorrhizal fungi associated with Unila paniculata in
Florida foredunes. Mycologia 78: 728-734.
Thompson, B. D., A. D. Robson, and L. K. Abbott. 1986. Effects of
phosphorus on the formation of mycorrhizas by Gigaspora calospora
and Glomus fasciculatum in relation to root carbohydrates. New
Phytol. 103:751-765.
Trappe, J. M. 1982. Synoptic keys to the genera and species of
zygomycetous (vesicular-arbuscular) mycorrhizal fungi.
Phytopathology 72: 1102-1108.
Waidyanatha, U. P. De S., N. Yogaratnan, and W. A. Ariyaratne. 1979.
Mycorrhizal infection on growth and nitrogen fixation of Pueraria
and Stylosanthes and uptake of phosphorus from two rock phosphates.
New Phytol. 82:147-152.
Wilson, J. M. 1984. Competition of infection between vesicular-
arbuscular mycorrhizal fungi. New Phytol. 97:427-435.
Yost, R.S., and R.L. Fox. 1979. Contribution of mycorrhizae to P
nutrition of crops growing on an oxisol. Agron. J. 71: 903-908.

BIOGRAPHICAL SKETCH
Onesimo Adonis Medina was born in San Pedro de Macoris, Dominican
Republic, on November 2, 1953. He received his B. Sc. degree in
agronomy in April, 1977, from the Universidad Nacional Pedro Henriquez
Urena, Santo Domingo, Dom. Rep. In June, 1977, he was hired by the
Dominican Agrarian Institute (IAD) as an Assistant Investigator in the
Soil Survey Division.
He began graduate studies in August, 1979, and received a M. Sc.
degree in soil science with a specialization in soil fertility from the
University of Hawaii in 1981. On returning to the Dominican Republic,
he was hired by the National Livestock Research Center (CENIP) as an
Associate Investigator in the Soil Fertility and Plant Nutrition
Division. He also served as an Assistant Professor in the Agronomy and
V
Soil Science Department of the Universidad Nacional Pedro Henriquez
Urena.
In August, 1984, he enrolled to the graduate program of the
University of Florida, Department of Soil Science, and received the
degree of Doctor of Philosophy in December, 1987. He has been married
to Griselda Terrero since 1979 and they have one daughter, Michelle.
91

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.
Dr. D. M. Sylvia, Chairman
Assistant 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.
£
Dr. A. E. Kretschmer, Cochairma:
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.
Dr.G.Kidder
Associate 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.
D^. J. \b!Sartain
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.
Dr^ G. H. Siiyder
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.
A
C'2^S^c/'^CyK~
i, v.
Dr. N. C. Schenck
Professor of Plant Pathology
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December 1987
Dean, Graduate School



ROOT DRY WEIGHT (g) SHOOT DRY WEIGHT (g)
21
2.5
2.0
1 .5
1 .0
0.5
o:o
fi RV RS MR LL SH SG VR
LEGUME SPECIES
Fig. 3-2. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
pasteurized soil in the greenhouse after 45 days. The
legume species were Aeschynomene americana (AA), Arachis
sp. (AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA). Bars
represent the mean of 3 replications. Means with the same
letter within a species are not different (P < 0.05).


84
Bethlenfalvay, G.J., J.M. Ulrich, and M.S. Brown. 1985. Plant response
to mycorrhizal fungi: host, endophyte, and soil effects. Soil Sci.
Soc. Am. J. 49: 1164-1168.
Black, R. L. B., and P. B. Tinker. 1977. Interactions between effects
of vesicular-arbuscular mycorrhiza and fertilizer phosphorus on
yields of potatoes in the field. Nature 267:510-511.
Burt, R. L., and C. P. Miller. 1975. Stylosanthes--A source of pasture
legumes. Trop. Grassl. 9:117-123.
Chulan, A., and P. Ragu. 1986. Growth response of Theobroma cacao L.
seedlings to inoculation with vesicular-arbuscular mycorrhizal
fungi. Plant Soil 96:279-285.
Cooper, K. M. 1978. Adaption of mycorrhizal fungi to phosphate
fertilizers, p. 107. In A. R. Ferguson, R. L. Bieleski, and I. B.
Ferguson (eds.) Plant nutrition 1978. New Zealand DSIR Information
Series No. 134, Government Printer, Wellington.
Crush, J. R. 1974. Plant growth responses to vesicular-arbuscular
mycorrhiza. VII. Growth and nodulation of some herbage legumes.
New Phytol. 73:745-750.
Currah, R.S., and M. Van Dyk. 1986. A survey of some perennial vascular
plant species native to Alberta for occurrence of mycorrhizal fungi.
Canadian Field-Naturalist 100: 330-342.
Daft, M.I., and A.A. El-Giahmi. 1976. Studies on nodulated and
mycorrhizal peanuts. Ann. Appl. Biol. 83:273.
Daniels, B. A., P. M. McCool, and J. A. Menge. 1981. Comparative
inoculum potential of spores of six vesicular-arbuscular mycorrhizal
fungi. New Phytol. 89:385-391.
Daniels, B. A., and H. D. Skipper. 1982. Methods for recovery and
quantitative estimation of propagules from soil. p. 29-35. In N. C.
Schenck (ed.) Methods and principles of mycorrhizal research. Amer.
Phytopath. Soc., St. Paul, MN.
Gallaher, R. N., C. 0. Weldon, and J. G. Futral. 1975. An aluminum
block digester for plant and soil analysis. Soil Sci. Soc. Amer.
Proc. 39:803-806.
Gates, C. T., and J. R. Wilson. 1974. The interaction of nitrogen and
phosphorus on the growth, nutrient status and nodulation of
Stylosanthes humilis H.B.K. (Townsville stylo). Plant Soil 41:325-
333.
Gerdemann, J.W., and J.M. Trappe. 1974. The endogonaceae in the Pacific
Northwest. Mycol. Mem. 5: 1-76.


61
Table 7-2. Analysis of variance for shoot dry weights of Aeschynomene
americana harvested 2 October 1986.
Source of
variation
DF
Mean squares
Block
9
52.57
Fungi (F)
2
1654.72**
Phosphorus (P)
3
6986.26**
Linear (PI)
1
18700.24**
Quadratic (Pq)
1
2250.32**
Cubic (Pc)
1
8.21
F x P
6
91.79
F x PI
2
53.72
F x Pq
2
215.53
F x Pc
2
6.11
Error
99
45.83
Significant at P < 0.01


20
There was no increase in plant growth as a result of inoculation with
G. intraradices in nonpasteurized soil for Arachis sp., leucaena, and
Vigna adenantha. With the exception of leucaena, this may be attributed
to effective colonization by the indigenous mycorrhizal fungi in the
noninoculated soil (Table 3-1).
Only five legume species (Fig. 3-2) were evaluated in pasteurized
soil; three were lost in a greenhouse accident. Siratro, Arachis sp.,
and Vigna adenantha had greater shoot and root dry weights after G.
intraradices inoculation. This increase again was related to effective
root colonization by G. intraradices (Table 3-1). In another study,
Siratro was shown to respond to inoculation with several VAM fungi in
pasteurized soil in the greenhouse (Lopes and De Olivera, 1980).
Inoculation with G. intraradices did not increase either the shoot
or root dry weights of aeschynomene or leucaena in pasteurized soil.
However, mycorrhizal colonization on both legumes was very low (3%).
Leucaena has been reported to be very mycorrrhizal dependent because
it has few root hairs (Huang et al., 1985; Yost and Fox, 1979). The
failure of VAM fungi to colonize it in this study in both pasteurized and
nonpasteurized soil may be due to incompatibility between the plant and
G. intraradices as well as native VAM fungi in the experimental soil,
to inhibitory soil factors on the host-VAM symbiosis, or to the
relatively slow development of the root system. It has been shown that
some mycorrhizal fungi may be less effective on certain plant hosts.
For example, Schroder et al. (1977) reported that Glomus macrocarpum Tul
& Tul increased growth of onions but not of Stylosanthes sp.


62
(Table 7-3). There were no fungi x P interactions for shoot dry weights
of Siratro or aeschynomene.
At all levels of applied P and at all harvests, shoot dry weights
of Siratro were greater for fungal inoculated plants than control plants
(Fig. 7-1). Except for the third harvest, where the effect of fungal
inoculation was less pronounced at all levels of applied P, differences
between fungal inoculated and noninoculated plants were most marked at
intermediate levels of applied P (30-90 kg ha ^) and diminished at the
highest level (120 kg ha ^). The effect of inoculation with VAM fungi
on the shoot dry weights of aeschynomene, at all levels of applied P,
followed the same trend as that of Siratro although the response to
mycorrhizal inoculation was greater (Fig. 7-2).
Inoculated plants of Siratro (Table 7-4 and 5) and aeschynomene
(Table 7-6) had a quadratic relationship between shoot dry weight and P
application. Maximum shoot dry weight of Siratro was achieved between
75-85 kg ha ^ of P and for aeschynomene at 85 kg ha ^ of P, whereas
control plants of both legumes, even with 120 kg ha ^ of P, did not reach
the same shoot dry weight as fungal inoculated plants. Thus inoculation
with VAM fungi resulted in at least a 30% savings (40 kg ha ^) in the
amount of P fertilizer required for maximum yield. In a previous
greenhouse experiment (Chapter VI), I found that inoculation with G.
etunicatum resulted in a 20% decrease in the amount of P required for
maximum yield of Siratro. Some previous reports of VAM field experiments
with legumes in nonpasteurized soil (Black and Tinker, 1977; Khan, 1975)
show responses to VAM inoculation only in the absence of P fertilizer.


53
P APPLIED (mg/kg)
Fig. 6-3. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Aeschynomene
americana.


54
reach the same shoot dry weight as myco-rrhizal plants. Fungal
inoculation with G. etunicatum resulted in a 20% decrease in the amount
of P required for maximum yield. This demonstrates the importance of VAM
fungi in the P nutrition of tropical legumes and would represent an
important savings to the farmer.
Growth responses of Stylo, Siratro, and aeschynomene associated with
inoculation with VAM fungi are closely related to improved P uptake and
^fixation. These results clearly support the findings of earlier
studies that inoculation with VAM fungi not only stimulates plant growth
and P uptake of legumes (Abbott and Robson, 1977b; Habte and Manjunath,
1987; Menge, 1983) but also nodulation and ^-fixation (Asimi et al.,
1980; Lynd et al., 1985; Purcino et al., 1986) which was measured
indirectly in this study by the total plant uptake of N.
Inoculation with effective species of VAM fungi and additions of P
between 25 and 50 mg kg ^, were shown to improve the growth and uptake
of P and N of Stylo, Siratro, and aeschynomene.


root cfiLONiMim m Rnnr fresh veisii (g) snnor dry yeihit (g)
50
(mg/kg)
Fig. 6 1. Effect of fungal inoculation and P applications on the
shoot dry weight, root fresh weight, root colonization, P
concentration, total P, and total N of Stylosanthes
guianensis.


11
between root colonization and spore density, while Hayman and Stovold
(1979) and Giovannetti and Nicolson (1983) found no correlation. This
apparent discrepancy may be due to different sampling methods.
Giovannetti (1985) collected samples within the same plant species and
sites, while the other researchers collected samples from many different
plant species and sites.
Spore production and root colonization are influenced by seasonal
variations (Giovannetti, 1985; Sylvia, 1986), host plant, stage of
development (Saif and Khan, 1975; Schenck and Kinloch, 1980), and soil
type (Lopes et al., 1983). In this survey there was only one sampling,
so it was not possible to separate seasonal or host developmental effects
on root colonization and total spore density.
The 6 species of VAM fungi collected in this survey were: Gigaspora
heterogama (GH) Gerdemann & Trappe, Gigaspora margarita (GM) Becker &
Hall, Glomus etunicatum (ETU) Becker & Gerdemann, Glomus intraradices
(INT) Schenck & Smith, Glomus sp. (GS), and Acaulospora spinosa (AS)
Walker & Trappe. The unidentified Glomus sp. was dark brown to black,
200-250 pm in diam, and had 1 wall of 8-14 pm thickness.
The occurrence of fungal species, as determined by spore numbers,
was affected by the legume host and location (Table 2-4). labal et al.
(1975) and Schenck and Kinloch (1980) also recorded differences in spore
numbers among plant species. The maximum number of spores of G.
margarita occurred at Fort Pierce associated with aeschynomene. Spores
of G. margarita were not found associated with Siratro at any of the four
locations. Spores of G. heterogama, G. etunicatum, and G. intraradices


SHOOT DRY WEIGHT (g)
30
Fig. 4-
ROOT LENGTH COLONIZED (m)
Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments.


Aeschynomene villosa Poir., Stylo (Stylosanthes guianensis SW.), and
Stylosanthes hamata Taub. were increased in pasteurized and
nonpasteurized limed soil in the greenhouse after inoculation with Glomus
intraradices Schenck & Smith.
Inoculation with Glomus etunicatum Becker & Gerdemann and G.
intraradices also increased the growth of Siratro as compared to other
VAM fungi and the noninoculated control in limed, nonpasteurized soil
fertilized with 20 mg kg ^ of P. In other greenhouse experiments, G.
etunicatum and G. intraradices were effective growth enhancers of Siratro
over a practical range of 2.5 to 40 mg kg ^ of applied P in a limed,
nonpasteurized soil. For both fungi, increasing P above 2.5 mg kg ^
increased the percentage and total root length colonized by VAM fungi.
A positive correlation was found between mycorrhizal root colonization
and shoot dry weight. In a limed, pasteurized soil, inoculation with
G. etunicatum increased total P and N of Siratro at 12.5 and 25 mg kg ^
-1
of applied P, but not at 50 mg kg
' .. -s
The effectiveness of G. etunicatum and G. intraradices with Siratro
and aeschynomene was corroborated in a field trial. These fungi
increased the growth and uptake of P and N of both legumes over a range
of applied P from 10 to 80 kg ha
Inoculation of forage legumes with effective VAM fungi enhanced
their growth. Growth enhancement occurred at P and lime levels used in
commercial pasture production and in soils that had a large, but
apparently ineffective indigenous VAM population.
xii


ROOT DRY WEIGHT (g) SHOOT DRY WEIGHT (g)
18
? R
<
' C,* C.
2.0
1 .5
1 .0
0.5
0.0
0.8
0.6
0.4
0.2
0.0
LEGUME SPECIES
Fig. 3-1. Effect of inoculation with Glomus intraradices on the
shoot and root dry weights of tropical forage legumes in
nonpasteurized soil in the greenhouse after 45 days.
Legume species were Aeschynomene americana (AA),
Aeschynomene villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL), Stvlosanthes
hamata (SH), Stylosanthes guianensis (SG), and Vigna
adenantha (VA). Bars represent the mean of 3 replicates.
Means with the same letter within a species are not
different (P < 0.05).


17
The experimental treatments consisted of pasteurized or
nonpasteurized soil, with or without addition of G. intraradices
inoculum. The pots were arranged on greenhouse benches in a completely
randomized design with three replications per treatment. The average
maximum and minimum greenhouse temperatures were 37 and 26C,
respectively. Pots were watered as needed to maintain soil moisture near
field capacity and were re-randomized every two weeks.
Plants were harvested after 45 d. Shoot and roots were dried at 70C
for 48 h and weighed. The percentage of mycorrhizal root colonization
was determined as described in Chapter II.
Significant treatment effects on shoot and root dry weights within
legume species were analyzed by the T TEST procedure of the Statistical
Analysis Systems (SAS Institute, 1982).
Results and Discussion
Inoculation with G. intraradices in nonpasteurized soil resulted in
greater shoot dry weights (P < 0.05) for five of the seven legumes tested
(Fig. 3-1). Greater shoot dry weights of these legumes were positively
related to increased levels of mycorrhizal colonization following
inoculation (Table 3-1).
Root dry weight results were similar to those for shoot dry weight,
except for Stylo, where there was no increase in root dry weight as a
result of mycorrhizal inoculation (Fig. 3-1).


33
Forage legumes are an important component of improved pastures and
must be established rapidly and without excessive cost. When P is a
major factor limiting the productivity of legumes, large applications of
P fertilizer are normally required for legume establishment. However,
with the increasing cost of P fertilizer, alternative strategies for
minimum fertilizer input and efficient use must be adopted. These data
demonstrate that by careful selection of effective VAM fungi, the growth
of Siratro can be enhanced in a P-deficient native soil containing a less
effective native VAM population than the introduced VAM fungi.


8
Table 2-1. Chemical characteristics of the soils sampled for VAM
fungi associated with four tropical forage legumes at four
locations in south Florida.
Location
Legume
species2
O.M.
PH
A1
Ca
Mg
K
P
c > | 1
%
<
-mg kg
soil-
AA
1.4
6.0
23
314
93
8
4
Fort Pierce
DH
1.2
5.3
22
241
15
16
5
VA
1.3
5.5
25
242
21
20
4
MA
1.3
5.2
62
270
25
13
16
AA
3.4
6.1
44
1320
143
64
23
Ona
DH
3.1
5.4
36
920
120
29
8
VA
2.5
4.9
22
480
70
46
8
MA
5.7
4.7
55
' 800
~ 95
43
5
AA
2.5
6.1
66
780
67
40
6
Deseret
DH
2.3
6.0
27
1040
94
27
4
l.n f r {.'1_ 1
VA
2.8
7.2
30
1600
141
55
9
AA
4.7
5.3
28
960
32
28
4
Basinger
DH
3.5
5.2
26
460
100
46
7
VA
3.6
5.1
28
540
111
58
8
MA
4.2
5.2
36
640
129
94
11
ZAA- Aeschynomene americana; DH= Desmodium heterocarpon;
VA= Vigna adenantha; MA= Macroptilium atropurpureum.


32
Schroder et al. (1977) reported that G. macrocarpum increased growth of
onions but decreased growth of Stylosanthes sp.
The most effective fungi in this study were those that colonized the
root most rapidly. Sanders et al. (1977) and Abbott and Robson (1978)
reported that VAM fungi differ in their rates of root colonization.
Abbott and Robson (1981) stated that differences in the effectiveness of
VAM fungi could be due to differences in their ability to (1) colonize
the roots rapidly (infectivity), (2) produce external hyphae, and (3) to
take up and transport P (efficiency). In this study, we measured only
root colonization over time (infectivity) and found, under the conditions
of this experiment, G. etunicatum was the most infective fungus.
Mycorrhizal root colonization, expressed as either percentage or
total root length colonized, was positively correlated with the shoot dry
weight of Siratro. Abbott and Robson (1981) and Plenchette et al. (1982)
also showed positive correlations between the magnitude of mycorrhizal
root colonization and shoot dry weights of plants grown on P-deficient
soils. These results suggest that differences in endophyte effectiveness
may be evaluated on the basis of rates of root colonization. However,
Hayman and Tavares (1985) demonstrated that final root colonization by
VAM fungi may give little indication of the ability of an endophyte to
stimulate plant growth. Abbott and Robson (1978) reported that VAM fungi
which differ in effectiveness and rate of root colonization may have
similar plateau levels of colonization at a late harvest. Therefore it
is not surprising, when colonization is assessed at a relatively advanced
stage of plant growth, that there is often little correlation between
mycorrhizal root colonization and effectiveness.


GROWTH RESPONSE OF TROPICAL FORAGE LEGUMES TO INOCULATION WITH
VA MYCORRHIZAL FUNGI AND PHOSPHORUS APPLICATION
By
ONESIMO A. MEDINA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987


37
The average maximum and minimum greenhouse temperatures during the
experimental period were 34 and 26C, respectively. Maximum
-2 -1
photosynthetic photon flux density was 1800 ju mol m s
Shoot dry and root fresh weights, and percentage and total root
lengths colonized by VAM fungi, were determined after 60 d using
procedures described previously (Chapter IV). Data for all variables
were subjected to ANOVA procedures and regression analysis (SAS Institute
Inc., 1982).
Results and Discussion
Phosphorus applications of 2.5, 10, 20, and 40 mg kg ^ resulted in
Mehlich-I extractable P in the soil of 5.6, 12.8, 21.6, and 38.0
mg kg~l at the beginning of the experiment, respectively. These
differences were still reflected at the end of the experiment when P
concentrations were 4.4, 7.8, 15.2, and 24.6 mg kg *.
Shoot dry and root fresh weights of Siratro were increased by P
fertilization and fungal inoculation (Fig. 5-1). At 2.5 mg P kg--*-, there
was no difference in the shoot dry and root fresh weights among
inoculated and control plants. At all other levels of P, inoculated
plants had higher shoot dry and root fresh weights than control plants.
Shoot dry weights were greatest for plants inoculated with G. etunicatum.
There were no differences in root fresh weights between plants inoculated
with G. etunicatum or G. intraradices. Inoculated plants had a quadratic
relationship for shoot dry and root fresh weights and P application


13
were found associated with all legumes, in at least one of the locations.
Glomus heterogama occurred in greatest numbers at Deseret and Fort Pierce
associated with aeschynomene and carpon desmodium, respectively. The
maximum number of spores of G. etunicatum occurred at Ona and Fort Pierce
associated with carpon desmodium. A high number of spores of G.
etunicatum was also found at Fort Pierce associated with Vigna adenantha.
Glomus intraradices was recovered in greater numbers from carpon
desmodium and Vigna adenantha at Fort Pierce as well as from Siratro at
Ona. The unidentified Glomus sp. occurred in lower numbers than the
other two species of Glomus; it was only found at Fort Pierce, associated
with aeschynomene and Siratro. Acaulospora spinosa was only recovered
from Vigna adenantha at Ona.
Overall root colonization by VAM fungi was low (most values below
20%) which indicates that (1) the native population of VAM fungi is not
very infective and (2) field inoculation may be effective. Attempts to
establish pot cultures of VAM fungi recovered in this survey were only
successful with G. etunicatum and G. intraradices. These two fungi were
shown to be effective in increasing the growth of several forage legumes
and were chosen for further evaluations.


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.
Dr. D. M. Sylvia, Chairman
Assistant 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.
£
Dr. A. E. Kretschmer, Cochairma:
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.
Dr.G.Kidder
Associate 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.
D^. J. \b!Sartain
Professor of Soil Science


CHAPTER VI
EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE GROWTH AND
UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM, STYLOSANTHES GUIANENSIS,
AND AESCHYNOMENE AMERICANA
Introduction
Mycorrhizal colonization is important for legumes because it
increases their P uptake (Abbott and Robson, 1977b; Saif, 1987), and
therefore nodulation and N2_fixation (Asimi et al., 1980; Bergersen,
1971; Gates and Wilson, 1974; Gibson, 1976).
Crush (1974) found that VAM fungi increased the growth and
nodulation of Centrosoma pubescens Benth, Stylo, and Trifolium repens L.
Mosse et al. (1976) showed that effective nodulation of Centrosema,
Stylosanthes, and Trifolium plants in a P-deficient Brazilian Cerrado
soil could be achieved only by introducing both VAM fungi and P.
Mycorrhizal fungi also have been shown to increase nodulation, N2~
fixation, plant growth, plant N and P content in Vigna unguiculata (Islam
et al., 1980; Sanni, 1976), Medicago sativa (Barea et al., 1980),
Pueraria phaseoloides and Stylo (Waidyanatha et al., 1979), Stylosanthes
scabra (Purcino and Lynd, 1985), leucaena (Munns and Mosse, 1980;
Purcino et al., 1986), and Siratro (Lynd et al., 1985). However,
effective tripartite symbiosis (legume-rhizobium-VAM fungus) is
influenced by soil and climatic conditions (Waidyanatha et al., 1979),
45


LITERATURE CITED
Abbott, L.K., and A.D. Robson. 1977a. The distribution and abundance of
vesicular-arbuscular endophytes in some Western Australian soils.
Aust. J. Bot. 25: 515-522.
Abbott, L. K., and A. D. Robson. 1977b. Growth stimulation of
subterranean clover with vesicular-arbuscular mycorrhizas. Aust. J.
Agrie. Res. 28:639-649.
Abbott, L. K., and A. D. Robson. 1978. Growth of subterranean clover
in relation to formation of endomycorrhizas by introduced and
indigenous fungi in a field soil. New Phytol. 81:575-587.
Abbott, L. K., and A. D. Robson. 1981. Infectivity and effectiveness
of vesicular-arbuscular mycorrhizal fungi: Effect of inoculum type.
Aust. J. Agrie. Res. 32:631-639.
Anderson, D. L., and L. J. Henderson. 1986. Sealed chamber digestion
for plant nutrient analyses. Agron. J. 78:937-939.
Asimi, S., V. Gianinazzi-Pearson, and S. Gianinazzi. 1980. Influence of
increasing soil phosphorus levels on interactions between vesicular-
arbuscular mycorrhizae and Rhizobium in soybeans. Can. J. Bot.
58:2200-2206.
Azcon-Aguilar, C., and J. M. Barea. 1981. Field inoculation of Medicago
with VA mycorrhiza and rhizobium in phosphate-fixing agricultural
soil. Soil Biol. Biochem. 13:19-22.
Barber, D. A., and B. C. Lougham. 1967. The effect of microorganisms
in the absorption of inorganic plant nutrients. II. Uptake and
utilization of phosphate by barley plants grown under sterile and
non-sterile conditions. J. Exp. Bot. 18:170-176.
Barea, J. M., J. L. Escudero, and C. Azcon-G. de Aguilar. 1980. Effects
of introduced and indigenous VA mycorrhizal fungi on nodulation and
nutrition of Medicago sativa in phosphate fixing soils as affected
by P fertilizers. Plant Soil 54:283-296.
Bergersen, F. J. 1971. Biochemistry of nitrogen fixation in legumes.
Annu. Rev. Plant Physiol. 22:124-140.
83


47
soil containing bahiagrass. The fungal inoculation technique was
reported in Chapter III.
The legume species used in this experiment were: Siratro,
aeschynomene, and Stylo. Seeds were scarified with sandpaper, wetted,
and sprinkled with type El "cowpea" inoculum (Nitragin Co., Milwaukee,
WI) prior to planting. Three germinated seeds were planted per 620 ml
"Deepots" (J.M. McConkey & Co, Inc, Summer, WA). After emergence,
seedlings were thinned to one per pot.
The experiment was conducted as a 2 x 3 x 3 factorial consisting of
2 inoculation treatments, etunicatum and noninoculated control; three
P levels, 12.5, 25, and 50 mg kg ^; the 3 legume species; and 3
replications. The 18 treatments were arranged on a nonshaded greenhouse
bench in a completely randomized design. The average maximum and minimum
greenhouse temperatures during the experimental period were 28 and 20^,
respectively, and the average maximum photosynthetic photon flux density
was 1793 u mol m-^ s-^.
After 65 d plants were harvested, shoots dried (70C for 24 h),
weighed, and ground in a Wiley mill using a 20-mesh screen. Shoots were
digested by the sealed-chamber procedure of Anderson (1986) and analyzed
for P on a Jarrel-Ash 955 inductively-coupled argon plasma spectrometer
(ICAP). For nitrogen analysis, samples were digested using a
modification of the aluminum block digestion procedure of Gallaher
et al. (1975) and ammonia in the digestate was determined by
semiautomated colorimetry (Hambleton, 1977). Roots were washed from the
soil, air-dried, and weighed. In addition, percentage and total root


56
little soluble phosphate and a small or ineffective native population of
VAM fungi. The experimental site selected for this study contains very
little available P (1 mg kg and a reasonably high indigenous
population of VAM fungi (2 propagules g ^ soil). The indigenous
population, however, was less effective than two of the five introduced
VAM fungi under amended soil conditions (Chapter IV).
Aeschynomene is used extensively in Florida as a forage legume to
supply fixed nitrogen as protein and minerals to grazing animals (Hodges
et al., 1982) Siratro has been used sparingly in Florida (Kretschmer,
1972), but is one of the most widely used legumes throughout tropical
regions of the world (Lynd et al.,1985). However, for the satisfactory
establishment and growth of aeschynomene and Siratro in highly acidic
and P-deficient soils, lime and P fertilizer must be applied (Snyder et
al., 1985). Recent greenhouse experiments have shown that better growth
of forage legumes in these soils can be achieved by inoculation with
selected species of VAM fungi (Chapter IV and V). Glomus etunicatum and
G. intraradices were found to be effective growth enhancers of Siratro
and aeschynomene in a nonpasteurized, limed (3000 kg ha ^) soil similar
to the one used in the present study, over a range of 20-80 kg ha ^ of
applied P. It is therefore of considerable interest to determine whether
inoculation with selected isolates of VAM fungi can improve the
j.1 . r.- .. \ \ 7. '* T
establishment, growth, and nutrient uptake of Siratro and aeschynomene,
under field conditions where soil was limed and fertilized with different
P levels.


7
Results and Discussion j : n
Results of soil pH and chemical analysis of soil samples reflect
the different management regimes (including lime and fertilizer)
(Table 2-1). 'r .'.'.on war
Differences in percentage of mycorrhizal root colonization and total
spore density of air dry soil were significant among locations, legume
species, and location x legume species interactions (Table 2-2). Total
spore density at the four locations ranged from 5 to 679 per 100 g of
air-dried soil, and the percentage of mycorrhizal root colonization
varied from 3 to 41%. Miller et al. (1979) observed variable degree of
mycorrhizal root colonization (4 to 74%) in forage grasses and legumes
in Brazil. Except for carpon desmodium, legume species differed in
percentage root colonization and total spore density among locations
(Table 2-3). Fort Pierce had the highest total spore density for each
legume species sampled except for Siratro.
Attempts were made to relate percentage root colonization and total
spore density to soil P or the other soil chemical characteristics
presented in Table 2-1, but no clear relationships were apparent. Abbott
and Robson (1977a) and Hayman (1978) also reported that spore numbers
were not correlated with soil P or soil pH in cultivated soils.
There was a positive correlation (P < 0.05) between root
colonization and total spore density for all legume species at Basinger
(r = 0;70) and Deseret (r =0.76), but not at Fort Pierce and Ona.
Giovannetti (1985) and Miller et al. (1979) reported a correlation


85
Gibson, A. H. 1976. Limitations to nitrogen fixation in legumes. In
W. E., Newton, and C. J. Nyman (eds.) Proc. 1st Int. Symp. Nitrogen
Fixation, Vol.2. Wash. State Univ. Press, Pullman.
Giovannetti, M. 1985. Seasonal variations of vesicular-arbuscular
mycorrhizas and endogonaceous spores in a maritime sand dune.
Trans. Br. Mycol. Soc. 8A: 679-684.
Giovannetti, M., and B. Mosse. 1980. Evaluation of techniques for
measuring vesicular-arbuscular mycorrhizal infection in roots. New
Phytol. 84:489-500.
Giovannetti, M., and T.H. Nicolson. 1983. Vesicular-arbuscular
mycorrhizas in Italian sand dunes. Trans. Br. Mycol. Soc. 80: 552-
557.
Habte, M., and T. Aziz. 1985. Response of Sesbania grandiflora to
inoculation of soil with vesicular-arbuscular mycorrhizal fungi.
Appl. Environ. Microbiol. 50: 701-703.
Habte, M., and A. Manjunath. 1987. Soil solution phosphorus status and
mycorrhizal dependency in Leucaena leucocephala. Appl. Environ.
Microbiol. 53:797-801.
Hall, I. R. 1978. Effects of endomycorrhizas on the competitive ability
of white clover. N. Z. J. Agr. Res. 21:509-515.
Hall, I. R. 1979. Soil pellets to introduce vesicular-arbuscular
mycorrhizal fungi into soil. Soil Biol. Biochem. 11:85-86.
Hall, I. R. 1984. Field trials assessing the effect of inoculating
agricultural soils with endomycorrhizal fungi. J. Agrie. Sci.
102:725-731.
Hambleton, L. G. 1977. Semiautomated method for simultaneous
determination of phosphorus, calcium and crude protein in animal
feeds. J. Assoc. Offic. Agr. Chem. 60:845-852.
Harley, J.L., and E.L. Harley. 1987. A check-list of mycorrhiza in the
British flora. New Phytol. 105:1-102.
Harley, J. L., and S. E. Smith. 1983. Mycorrhizal symbiosis. Academic
Press, London.
Hass, J. H., and J. Krikum. 1985. Efficacy of endomycorrhizal fungus
isolates and inoculum quantities required for growth response. New
Phytol. 100:613-621.
Hayman, D.S. 1978. Mycorrhizal populations of sown pastures and native
vegetation in Otago, New Zealand. N. Z. J. Agr. Res. 21: 271-275.


82
reported. At all levels of applied P and for all harvests, shoot dry
weights of Siratro were greater for fungal inoculated plants than
noninoculated plants. Differences between fungal inoculated and
noninoculated plants were most marked at 30 to 90 kg ha of applied P
and diminished at 120 kg ha ^. The effect of fungal inoculation on the
shoot dry weights of aeschynomene, at all levels of applied P, was
similar (but more pronounced) as that of Siratro. At the first harvest
of Siratro, plants inoculated with G. etunicatum had higher shoot dry
weights than G. intraradices plants at all levels of applied P. However,
in subsequent harvests of Siratro and for aeschynomene the response of
shoot dry weight to inoculation with the two VAM fungi was similar.
Fungal inoculation resulted in at least a 30% savings (AO kg ha ^) in
the amount of P fertilizer required for maximum yield. Inoculated
treatments had greater percentage of root colonized than noninoculated
treatments at all levels of applied P. Percentage of root colonized by
VAM fungi for the inoculated plants of the two legumes increased linearly
with P additions up to 60 kg ha *. There were no differences in root
colonization between G. etunicatum and G. intraradices at any level of
applied P.
The results of these studies clearly demonstrate that inoculation
with effective VAM fungi can increase the growth of legumes in soils that
may have a high, but largely ineffective native VAM population than
introduced VAM fungi under amended soil conditions. Furthermore, with
highly mycorrhizal dependent crops such as tropical forage legumes, a
mycorrhizal growth response may occur at P levels normally used in
commercial pasture production.


LIST OF FIGURES
Fig. 2-
Fig. 3-
Fig. 3-2
Fig. 4-1
t/.K \ t lac-. i : .3 c ~c"rT -r_;
1. Collection sites for VAM fungi associated with
four tropical forage legumes in south Florida. .
1. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in nonpasteurized soil in the
greenhouse after 45 days. Legume species were
Aeschynomene americana (AA), Aeschynomene
villosa (AV), Arachis sp. (AS), Macroptilium
atropurpureum (MA), Leucaena leucoephala (LL),
Stylosanthes hamata (SH), Stylosanthes guianensis
(SG), and Vigna adenantha (VA). Bars represent
the mean of 3 replicates. Means with the same
letter within a species are not different
(P < 0.05)
. Effect of inoculation with Glomus intraradices
on the shoot and root dry weights of tropical
forage legumes in pasteurized soil in the
greenhouse after 45 days. The legume species
were Aeschynomene americana (AA), Arachis sp.
(AS), Macroptilium atropurpureum (MA), Leucaena
leucocephala (LL), and Vigna adenantha (VA).
Bars represent the mean of 3 replications.
Means with the same letter within a species are
not different (P<0.05)
. Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum
(ETU), or the control (CON) on the shoot dry weight
and root fresh weight of Siratro at two harvests.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05)
Page
5
18
21
. 27
viii


2
effect of improved P nutrition (Daft and ^i-Giahmi, 1976; Habte and Aziz,
1985).
Information concerning the association of VAM fungi with tropical
forage legumes is sparse. Most of the growth response studies reported
were done in either pasteurized soil or in small volumes of
nonpasteurized soil. Except for the work of Saif (1987), little
information is available on growth response of tropical forage legumes
to inoculation with VAM fungi in nonpasteurized soil, especially under
field conditions where introduced species of VAM fungi must compete with
the indigenous VAM population.
Therefore, greenhouse studies were conducted in limed, pasteurized
and nonpasteurized soil to improve the growth of tropical forage legumes
through inoculation with effective VAM fungi and reduced P fertilization.
In addition, the effect of inoculation with selected VAM isolates on
growth and nutrient uptake of two tropical forage legumes under natural
field conditions was investigated at different levels of applied P.
have
lh
L so


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EWSUT5ZQB_TQAXHH INGEST_TIME 2014-12-03T19:30:24Z PACKAGE AA00026465_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


4
associated with four cultivated tropical forage legumes from four
different locations in south Florida.
Materials and Methods
Root and rhizosphere soil samples of four tropical forage legumes
were collected from 11 to 17 October 198A, at four locations in south
Florida: Deseret Ranches, Deer park; Fort Pierce, Agricultural Research
and Education Center (AREC); Ona, AREC; and Basinger Ranch, 109 Ranch
(Fig. 2-1). Most of the soils of the studied area belong to the order
Spodosols. They are dominated by nearly level, somewhat poorly to poorly
drained sandy soils with dark sandy subsoil layers. These soils are used
primarily for pastures, vegetables, flowers, forestry, and citrus.
The forage legumes sampled were: 'Siratro' (Macroptilium
atropurpureum Urb.), (except at Deseret Ranches), aeschynomene
(Aeschynomene americana L.), Vigna adenantha Marechal, Mascherpa and
Stainier, and carpon desmodium (Desmodium heterocarpon DC.). The legumes
were mixed with pasture grasses at the time of sampling. Three
rhizosphere soil samples were collected to a depth of 15 cm for each
legume at each location. Samples, consisting of three subsamples of
approximately 1.5 kg, were placed in plastic bags and transported to the
lu.vt- . i: v*.
laboratory on the same day.
Samples were sieved through a 4-mm screen, and 100 g subsamples were
removed and stored at 5C for spore extraction. Legume roots were
carefully separated manually from grass roots. A portion (0.5 g) of each


CHAPTER V
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH VA MYCORRHIZAL FUNGI IN
NONPASTEURIZED SOIL. II. EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT
DIFFERENT P LEVELS.
Introduction
Forage legumes are an important component of improved grass pastures.
The legumes serve both to increase forage quality and decrease the need
for fertilizer N through N2 fixation. Snyder et al. (1985) studied the
responsiveness of several tropical legumes, including Siratro, to P and
lime in a typical Florida Spodosol. They reported that lime and P rates
of approximately 3000 and 75 kg ha ^ produced maximum yield and maximum
economic return.
In an experiment with red clover in a field containing 10 mg NaHCC^-
soluble P kg ^ soil, plants showed an early response to superphosphate,
but by the end of the second year yields were high in all plots,
equivalent to around 15 t ha ^ dry matter (Hayman et al., 1981). This
result was attributed to one of the introduced endophytes, G. caledonium,
which had spread and sporulated profusely throughout all the plots
(including those inoculated with two other endophytes) and had previously
enhanced growth of lucerne at this site. In upland pastures in Wales,
Hayman and Mosse (1979) found that inoculation of white clover seedlings
with a combination of G. Mosseae and G. fasciculatum "E3" in field plots
given the standard dressing of 90 kg P ha 1 as basic slag doubled plant
34


66
Table 7-4. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P and N
concentrations, and total P and N uptake for the first
harvest of Siratro.
Variable
Regression equations
r2
Glomus etunicatum
Shoot dry wt.
(g)
=
12.36+0.33P-0.0022P2
0.80**
Root coloniz.
(%)
=
5.31+0.77P-0.0043P2
0.91**
Plant P cone.
(%)
=
0.22+0.00072P
0.66**
Total Plant P
(mg)
=
17.73+1.36P-0.0069P2
0.98**
Plant N cone.
(%)
=
2.26+0.035P-0.00025P2
0.52**
Total Plant N
(mg)
=
170.94+20.09P-0.13P2
0.91
Glomus intraradices
Shoot dry wt.
(g)
=
12.03+0.24P-0.0014P2
0.67**
Root coloniz.
(%)
=
10.04+0.59P-0.0034P2
0.86'c*
Plant P cone.
(%)
=
0.16+0.0037P-0.000018P2
0.76"*
Total plant P
(mg)
=
16.08+1.31P-0.0068P2
0.94**
Plant N cone.
(%)
=
2.12+0.041P-0.00026P2
0.56"*
Total plant N
(mg)
=
226.21+15.50P-0.095P2
0.91**
Control
Shoot dry wt.
(g)
=
10.65+0.IIP
0.65**
Plant P cone.
(%)
=
0.18+0.001IP
0.81**
Total plant P
(mg)
=
12.36+0.48P
0.92**
Total plant N
(mg)
=
173.09+3.57P
0.89"'*
P = phosphorus level
Significant at P < 0.01


16
to 3,000 kg ha"l assuming a 15-cm depth of soil ha-^ with a bulk density
of 1.3 g cra'^) and allowed to equilibrate for 30 days before initiating
the study. Solutions of P, K, Mg, Cu, Mn, Zn, B, and Mo also were
thoroughly mixed with the-soil to supply rates of 10, 30, 12, 1.5, 1.0,
1.0, 0.50 and 0.10 mg kg \ respectively. A portion of the soil was
pasteurized at 60C for A h to eliminate the indigenous mycorrhizal
fungi. Then 3 kg of soil was placed into 15-cm-diam plastic pots. The
; T*
pH of the soil was 6.2 at the end of the experiment.
The legumes used in the experiment were: Siratro, aeschynomene,
Aeschynomene villosa Poir., Stylo, leucaena, Stylosanthes hamata Taub.,
cv. 'Verano1, Vigna adenantha, and Arachis sp. Seeds were scarified with
sandpaper, wetted, and sprinkled with type EL "cowpea" inoculum (Nitragin
Co., Milwaukee, Wl) prior to planting. Five seeds of the corresponding
legumes were planted per pot, and plants were thinned to one per pot 10
d after emergence.
Glomus intraradices (isolate S311), used in this study, was isolated
from cultivated Vigna adenantha at the Agricultural Research and
Education Center, Ona, FL. (Chapter II, Table 2-4). Fungal inoculum was
produced in pot culture in pasteurized soil containing carpon desmodium
as the host plant. Pot cultures were 10-weeks old when they were
harvested, mixed and used to inoculate experimental pots. Ten grams per
pot of the soil-root-fungus inoculum containing approximately 200 spores
was spread in a 1-cm-thick layer, at a depth of 3 to 5 cm below the soil
surface. Noninoculated control treatments received 10 g of a soil-root
mixture from noninoculated pot cultures that were free of VAM fungi.


CHAPTER VII
GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND AESCHYNOMENE AMERICANA
TO INOCULATION WITH SELECTED VA MYCORRHIZAL FUNGI IN THE FIELD AT
DIFFERENT P LEVELS
Introduction
The practical goal of studies on plant growth responses to
inoculation with VAM fungi is to obtain increased yield of plants growing
under field conditions. Significant plant growth responses to
inoculation with VAM fungi have been demonstrated in pot experiments
using pasteurized and nonpasteurized soil for several tropical forage
legumes such as: Pueraria phaseoloides (Salinas et al., 1985;
Waidyanatha et al., 1979), leucaena (Habte and Manjunath, 1987; Huang et
al., 1985), Siratro (Lynd et al., 1985), Stylo (Mosse, 1977) and
aeschynomene (Chapter II and VI).
It has been pointed out that inoculation experiments with VAM fungi
should include testing a series of P levels (Abbott and Robson, 1977b;
Hall, 1978; Powell, 1980b) in order to select the optimum P level for a
mycorrhizal response. Except for the work by Saif (1987), little
information is available on plant growth response of tropical forage
legumes to inoculation with VAM fungi in nonpasteurized soil under field
conditions at different P levels. However, there is more data for
temperate legumes. In general, the field sites where plants are most
likely to respond to inoculation with VAM fungi are those containing
55


49
Table 6-1. Mean squares and levels of significance from the analysis
of variance for shoot dry weight, root fresh weight, P
concentration, and total P and N uptake of forage legumes.
Source of
Variation
DF
Shoot
dry wt
Root P
fresh wt cone
Total
P
Total
N
Stylosanthes guianensis
Fungus
1
0.45
1.13
0.011**
6.93^
500.97
P rates
2
0.64
0.21
0.016"
9.13*
790.32
Lineal
1
1.20
0.36
0.032*
17.69**
1511.78
Quadratic
1
0.09
0.05
0.0004
0.57
68.86
Fungus x P
2
0.046**
0.15**
0.00007
0.65
47.14*
Error
12
0.0057
0.026
0.012
0.18
8.48
Macroptilium atropurpureum
Fungus
1
1.40
0.58**
0.0076*
33.78
2360.30
P rates
2
7.28
1.81**
0.028""
163.11
10469.14
Lineal
1
14.54
3.55"
0.056'
325.94
20925.10
Quadratic
1
0.0072
0.05
0.0001
0.28
13.18
Fungus x P
2
0.54
0.10
0.0014
10.37
625.52**
Error
12
0.029
0.027
0.00082
2.47
43.27
Aeschynomene
americana
Fungus
1
0.13
0.042
0.0068**
6.06**
129.34*
P rates
2
0.56
1.07**
0.034""
29.75**
524.16**
Lineal
1
1.09
2.01*
0.064"
58.70*
1035.46**
Quadratic
1
0.027,
0.12
0.0022
0.80
12.85
Fungus x P
2
0.037"
0.030
0.00009
0.10
43.35
Error
12
0.0076
0.015
0.00052
0.25
21.75
Significance at P < 0.05 ''"'Significant at P < 0.01


29
colonization (r^ = 0.83''"). There was a quadratic relationship between
shoot dry weight and length of Siratro root colonized by VAM fungi for
all inoculated treatments (Fig. 4-3).
Plant height and number of leaves per plant were not different among
treatments at any harvest. After 70 d, the mean plant height and number
of leaves for all treatments were 90 and 15 cm, respectively.
There were 2 propagules per gram of soil in the native soil as
determined by the MPN test at the beginning of the experiment.
Inoculum density is known to influence plant growth response to VAM
fungal inoculation (Hass and Krikum, 1985Wilson, 1984). Thus, one of
the problems in comparing the efficacy of VAM fungi is ensuring uniform
inoculum densities (Daniels et al., 1981). In this study, I used the
MPN technique to provide a measure of the inoculum densities of the VAM
fungi, and I adjusted the inoculum densities so that they were uniform
for all inoculated treatments.
There were striking differences in the effectiveness of VAM fungi on
Siratro. The results are consistent with the findings of others (Miller
et al., 1985; Schubert and Hayman, 1986) indicating that different
species and strains of VAM fungi vary considerably in the benefits they
confer to the host plant. This experiment also confirmed previous work
(Chapter III) which demonstrated that the native population of VAM fungi
in this soil was less able to stimulate the growth of Siratro than
effective, introduced species. It is possible that the decreasing soil
acidity obtained by liming changed the native population of VAM fungi
from effective to ineffective as compared to G. etunicatum and G.
intraradices (Hayman and Tavares, 1985). Since it is necessary to lime


59
Soil samples (0-15 cm) were taken at the 4th harvest and analyzed
for pH, P, Ca, Mg, and K using the Mehlich-I extractant method. All
elements were determined by inductively coupled argon plasma (ICAP)
spectrometry. Data for all variables were subjected to ANOVA procedures
and regression analysis (SAS Institute Inc., 1982).
Results and Discussion
Pre-inoculated seedlings were used in this study to ensure the
legume roots were well colonized with the selected VAM fungi, because
this was thought to be the most certain way to ensure establishment of
the inoculum. Once a plant response is ascertained, methods of
inoculation more applicable on a large scale can be tested.
As a result of lime application, soil pH increased from 4.6 to about
6.2, and extractable Ca increased up to about 650 mg kg ^. Phosphorus
applications of 10, 30, 60, and 120 kg ha ^ resulted in extractable P in
the soil of 5.1, 8.6, 17.8, and 34.3 mg kg ^ at the end of the
experiment, respectively. Thus the legume species and VAM fungi in the
present study were exposed to a considerable range of soil P.
Seedlings inoculated with VAM fungi were similar in shoot dry weight
and P concentration to control seedlings at transplanting (Table 7-1).
Seedlings inoculated with VAM fungi were also well colonized, whereas no
VAM colonization was detected on control seedlings (Table 7-1).
Phosphorus amendments and fungal inoculation increased shoot dry
weights of aeschynomene (Table 7-2) and all harvests of Siratro


CHAPTER I
INTRODUCTION
-Forage legumes are an important component of improved grass pastures
and must be established rapidly and without excessive cost. The legumes
serve both to increase forage quality and decrease the need for N
fertilizer through ^fixation.
Newly cleared lands incorporated into pasture production in south
Florida are generally acidic and very low in total and available P
throughout the soil profiles. While improvements to the productivity of
these pastures may be obtained by the introduction of suitable legumes,
effective N2~fixation and establishment:of legumes is frequently limited
by the_low levels of available P in these soils. Snyder et al. (1978)
reported that large applications of P fertilizer are normally required
for legume establishment and optimun growth in these soils. However, .
with the increasing cost of P fertilizer, alternative strategies for
minimum P fertilizer input and efficient use of P must be adopted. One
of these strategies may be via the management of vesicular-arbuscular
mycorrhizal (VAM) symbioses.
r.n -i VeSi ciliar arbuscular mycorthizal fungi can improve the growth of
legumes by increasing P uptake (Bethlenfalvay et al., 1985; Hayman,
1983). Phosphorus is often a growth-limiting factor since many legumes
have P requirements and are poor scavengers of P. The VAM fungi may also
increase nodulation and ^-fixation of legumes, primarily as an indirect
1


ACKNOWLEDGMENTS
It is with deep gratitude that I express thanks to the chairman and
cochairman of my committee, Dr. David M. Sylvia and Dr. Albert E.
Kretschmer, Jr., respectively, for their support, constant encouragement,
guidance, and friendship. I also thank the other members of my
committee, Dr. G. H. Snyder, Dr. G. Kidder, Dr. N. C. Schenck, and Dr.
J. B. Sartain, for their suggestions, support, and editorial comments.
Gratitude is extended to Mr. Tom Wilson for his friendship and
valuable assistance rendered in the field portion of this project.
The moral support, love, and motivation of my brothers, Oquendo,
Miosotis, and Gagarin were essential to the completion of my graduate
program.
This research was funded in part through USDA ARS Tropical
Agricultural Development Grants 83-CRSR-2-2134 and 86-CRSR-2-2846. This
support is greatly appreciated.
Most of all, warmest thanks go to my wife, Griselda, for her
understanding and encouragement during my graduate studies. Her many
hours of assistance in typing the manuscript will always be remembered.
I also thank my daughter Michelle for making her Mommy and Daddy very
happy.
iii


77
C 3
CD
C_J
Q_
I
cr
32
CD!
cd 2
I
CD
CON
ETU
INT
30 60 60 120
P PPLIED (kg/ho)
Fig. 7-5. Effect of P application on P concentration, total P, and
total N of Aeschynomene americana grown in the field
conditions and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). Data
points are means of five replicates.


36
Materials and Methods
The chemical properties of the soil used, liming, and fertilizer
amendments are described previously (Chapter III).
Siratro was used as the host plant in this experiment. The VAM
fungi tested were G. etunicatum (isolate S312) and G. intraradices
(isolate S313). These isolates were maintained in pasteurized soil in
pots with bahiagrass as the host. Soils from 10-week-old pot cultures
were used to inoculate experimental pots. The propagule densities of
the inocula were determined by the MPN technique (Daniels and Skipper,
.1982) and approximately 240 propagules were added to each 15-cm-diam
plastic pot. Details on the origin of the two fungal isolates were
reported previously (Chapter II, Table 2-4). Fungal inoculation
technique, planting, and watering were described in Chapter III.
The experiment was designed as a 3 x 4 factorial consisting of 3
inoculation treatments; G. intraradices, G. etunicatum, and
noninoculated control, and four P treatments; 2.5, 10, 20, and 40 mg P
kg 1 as CaO^PO^^.I^O (equivalent to 5, 20, 40, and 80 kg P ha ^
-1 -3)
assuming a 15-cm depth of soil ha with a bulk density of 1.3 g cm .
Phosphorus was applied in solution one week before planting. Soil
samples from each treatment were analyzed for extractable P at the
beginning and end of the experiment using the Mehlich-I method (0.05 M
HC1 + 0.0125 M H2SO4). The twelve treatments were replicated five times
and arranged on a greenhouse bench in a randomized complete block design


SHOOT DRY WEIGHT (g)
42
ROOT LENGTH COLONIZED Cm)
Relationship between shoot dry weight and length of Siratro
roots colonized by VAM fungi for all inoculated treatments
in nonpasteurized soil.
Fig.
5-2.


68
Table 7-6. Regression equations and coefficients of determination
(r2) showing the relationship of applied P level to shoot
dry weight, percentage of root colonization, P
concentration, and total P and N uptake of Aeschynomene
americana.
Variable
Regression equations
r2
Glomus etunicatum
Shoot dry wt.
(g)
=:
46.59+0.83P-0.0049P2
0.78**
Root coloniz.
(%)
=
7.89+0.88P-0.0047P2
0.89**
Plant P Cone.
(%)
=
0.21+0.0037P-0.000019P2
0.80**
Total plant P
(mg)
=
67.54+4.96P-0.025P2
0.96**
Total plant N
(g)
=
1.35+0.039P-0.00019P2
0.93'*
Glomus intraradices
Shoot dry wt.
(g)
=
40.86+0.97P-0.0056P2
0.86
Root coloniz.
(%)
=
9.26+0.98P-0.0058P2
0.93**
Plant P Cone.
(%)
=
0.16+0.0053P-0.000029P2
0.76
Total plant P
(mg)
=
43.24+5.85P-0.03IP2
0.95"'*
Total plant N
(g)
=
1.18+0.043P-0.00022P2
0.91"x
Control
Shoot dry wt.
(g)
-
43.50+0.31P
0.76**
Plant P Cone.
(%)
=
0.11+0.0045P-0.00002IP2
0.95**
Total plant P
(mg)
=
28.08+3.63P-0.012P2
0.99"*
Total plant N
(g)
=
1.44+0.015P
0.81**
P = Phosphorus level
Significant at P < 0.01


RDOT COLONIZATION CD SHOOT DRY HEIGHT (g)
65
P RPPLIED (kg/ha)
Fig. 7-2. Effect of P application on shoot dry weight and percentage
of root colonized of Aeschynomene americana grown in the
field and inoculated with Glomus etunicatum (ETU), Glomus
intraradices (INT), or not inoculated (CON). Data points
are means of five replicates.


48
length colonized were estimated as described in Chapter IV. A 0-4 scale
was used to estimate the number of nodules per plant, with 1 = 20-50,
2 = 50-100, 3 = 100-150, and 4 = > 150 nodules. Data were subjected to
ANOVA procedures and regression analysis (SAS institute Inc., 1982).
Results and Discussion
At harvest, both fungal inoculation and P applications increased
shoot dry weight, plant P concentration, and total plant P and N of the
three legumes (Table 6-1). Root fresh weight was increased for Stylo
and Siratro but not for aeschynomene. There were fungus x P interactions
for shoot dry and root fresh weights, and total plant N of Stylo.
Siratro had fungus x P interactions for shoot dry weight and total plant
P and N, whereas aeschynomene only had fungus x P interaction for shoot
dry weight.
At all levels of applied P, shoot dry and root fresh weights
and total N of Stylo were greater for mycorrhizal plants than
nonmycorrhizal plants (Fig. 6-1). Differences between mycorrhizal and
nonmycorrhizal plants were most pronounced at intermediate levels of
applied P and diminished with further P addition.
Phosphorus concentration and total P of Stylo were not affected by
fungus x P interactions (Table 6-1), but there was an overall effect of
fungal inoculation (Fig. 6-1). Saif (1987), Mosse et al. (1976), and
Waidyanantha et al. (1979) also reported an increase in plant growth, P
concentration, and total P and N of Stylo in a pasteurized low P soil
following inoculation with VAM fungi and P applications.


44
forage legumes, percentage and total root length colonized by VAM fungi
may be suppressed (Abbott and Robson, 1977b; Schubert and Hayman, 1986;
Thompson et al., 1986).
The result that root colonization by VAM fungi, expressed as either
percentage or total root length colonized, is positively correlated to
shoot dry weight is consistent with the findings of Abbott and Robson
(1981) and a previous study where its implications are discussed
(Chapter IV).
I conclude that, in amended soils where the indigenous population of
VAM fungi is less effective than some of the introduced species of VAM
fungi, inoculation with effective VAM fungi can increase the plant
growth. Furthermore, with highly mycorrhizal dependent crops such as
tropical legumes, growth enhancement may occur at P levels actually used
in commercial pasture production.


22
The failure to obtain good colonization of aeschynomene in
pasteurized soil was unexpected, since this legume was successfully
colonized in nonpasteurized soil. Unfortunately, soil chemical
properties were not determined after pasteurization. Legumes are
sensitive (and hence VAM fungi) to elevated Mn, which is a common
occurrence in heat treated soils.
The results obtained in this study clearly demonstrate that G.
intraradices can successfully compete with some of the indigenous
mycorrhizal fungi present in the experimental soil and promote growth of
several legumes in nonpasteurized soil. This result agrees with earlier
work by Abbott and Robson (1981), Mosse (1977), and Rangeley et al.
(1982), which suggests a potential for successful field-scale inoculation
with effective VAM fungi.


75
Table 7-9. Analysis of variance for P and N concentrations, and total
P and
N uptake of
Siratro for
the first harvest.
Source of
Mean
Squares
variation
DF
P Cone.
total P
N Cone. total N
Block
4
0.00033
26.06
0.42
10491.95**
Fungi (F)
2
0.0059
2556.03
3.03
540732.85
Phosphorus (P)
3
0.039
7248.85
1.47
544117.23
Linear (PI)
1
0.11
19691.07
1.53
1006530.96
Quadratic (Pq)
1
0.0049
2055.48
2.87
625338.95
Cubic (Pc)
1
0.000077
0.010
0.010
487.84
F x P
6
0.00237
175.50*
0.33
33638.14**
F x PI
2
0.0031';
53.34,
0.21,
7291.34
F x Pq
2
0.0039
465.13**
0.75
92589.06x*
F x Pc
2
0.00011
8.02
0.040
1034.01
Error
44
0.00087
26.84
0.14
3606.65
"Significant at
P <
0.01
Significant at P
< 0.05


ROOT FRESH WEIGHT (g) SHOOT DRY WEIGHT (g)
27
Fig. 4-1. Effect of inoculation with Gigaspora margarita (MAR),
Glomus versiforme (VER), Glomus deserticola (DES), Glomus
intraradices (INT), Glomus etunicatum (ETU), or the control
(CON) on the shoot dry weight and root fresh weight of
Siratro at two harvests. Bars represent the means of five
replicates. Means with the same letter within a harvest
are not different (P < 0.05).


89
Rhue, R.D., and G. Kidder. 1984. Procedures used by the IFAS extension
soil testing laboratory and interpretation of results. Fla. Coop.
Ext. Serv. Circ. 596.
Rotar, P.P. 1983. Legumes in action. P. 21-22. In R.L. Burt et al.
(eds.) The role of Centrosema, Desmodium, and Stylosanthes in
improving tropical pastures. Westview Press, Colorado.
Saif, S. R. 1987. Growth responses of tropical forage plant species to
vesicular-arbuscular mycorrhizae. I. Growth, mineral uptake and .
mycorrhizal dependency. Plant Soil 97:23-35.
Saif, S.R., and A.G. Khan. 1975. The influence of season and stage of
development of plant on Endogone mycorrhiza of field-grown wheat.
Can. J. Microbiology 21: 1021-1024.
Salinas, J.G., J.I. Sanz, and E. Sieverding. 1985. Importance of VA
mycorrhizae for phosphorus supply to pasture plants in tropical
oxisols. Plant Soil 84: 347-360.
Same, B. I., A. D. Robson, and L. K. Abbott. 1983. Phosphorus, soluble
carbohydrates and endomycorrhizal infection. Soil Biol. Biochem.
15:593-597.
Sanchez, P. A., and J. G. Salinas. 1981. Low-input technology for
managing oxisols and ultisols in tropical America. Adv. in Agronomy
34:279-406.
Sanders, F. E., P. B. Tinker, R. L. B. Black, and S. M. Palmerley. 1977.
The development of endomycorrhizal root systems. I. Spread of
infection and growth-promoting effects with four species of
vesicular-arbuscular endophyte. New Phytol. 78:257-268.
Sanni, S. 0. 1976. Vesicular-arbuscular mycorrhiza in some Nigerian
soils and their effect on the growth of cowpea (Vigna unguiculata),
tomato (Lycopersicon esculentum) and maize (Zea mays). New Phvtol.
77:667-671.
SAS Institute Inc. 1982. SAS user 's guide: Statistics. SAS Institute
Inc., Cary, N.C.
Satterlee, L., B. Melton, B. McCaslin, and D. Miller. 1983. Mycorrhizal
effects on plant growth, phosphorus uptake, and N2(C2H4) fixation in
two alfalfa populations. Agron. J. 75: 715-716.
Schenck, N.C., and R.A. Kinloch. 1980. Incidence of mycorrhizal fungi
on six field crops in monoculture on a newly cleared woodland site.
Mycologia 72: 445-456.
Schenck, N.C., and G.S. Smith. 1981. Distribution and occurrence of
vesicular-arbuscular mycorrhizal fungi on Florida agricultural
crops. Proc. Soil and Crop Sci. Soc. of Fla. 40:171-175.


Materials and Methods
24
Results and Discussion
26
CHAPTER V
GROWTH RESPONSE OF SIRATRO TO INOCULATION WITH
VA MYCORRHIZAL FUNGI IN NONPASTEURIZED SOIL. II.
EFFICACY OF SELECTED VA MYCORRHIZAL FUNGI AT
DIFFERENT P LEVELS
Introduction
Materials and Methods 36
Results and Discussion 37
34
: i
34
CHAPTER VI
EFFECT OF INOCULATION WITH GLOMUS ETUNICATUM ON THE
GROWTH AND UPTAKE OF P AND N OF MACROPTILIUM ATROPURPUREUM,
STYLOSANTHES GUIANENSIS. AND AESCHYNOMENE AMERICANA 45
Introduction 45
Materials and Methods . 46
Results and Discussion 48
CHAPTER VII
GROWTH RESPONSE OF MACROPTILIUM ATROPURPUREUM AND
AESCHYNOMENE AMERICANA TO INOCULATION WITH SELECTED VA
' MYCORRHIZAL FUNGI IN THE FIELD AT DIFFERENT P LEVELS 55
Introduction 55
Materials and Methods 57
Results and Discussion 59
CHAPTER VIII
Result :.iid D- sou* w
CONCLUSIONS ,
LITERATURE CITED
BIOGRAPHICAL SKETCH -
SELECT!OF OF EFFECTIVE .'A VCO: KHJ..AL M.'rJfF.
79
83
91
v


Fig. 4-2
Fig. 4-3
Fig. 5-1.
Fig. 5-2.
Fig. 6-1.
Fig. 6-2.
Fig. 6-3.
Fig. 7-1.
Effect of inoculation with Gigaspora margarita
(MAR), Glomus versiforme (VER), Glomus deserticola
(DES), Glomus intraradices (INT), Glomus etunicatum
(ETU), or the control (CON) on the percentage
of root colonization and root length colonized
of Siratro at three and two harvests, respectively.
Bars represent the means of five replicates.
Means with the same letter within a harvest are
not different (P < 0.05) 28
Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments 30
Effect of P application on shoot-dry weight,
root fresh weight, percentage of root colonized
by VAM fungi, and total root length colonized
of Siratro grown in limed nonpasteurized soil
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON). 38
Relationship between shoot dry weight and
length of Siratro roots colonized by VAM
fungi for all inoculated treatments in
nonpasteurized soil 42
Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Stylosanthes guianensis 50
Effect of fungal inoculation and P applications on
the shoot dry weight, root-fresh weight, root
colonization, P concentration, total P, and
total N of Macroptilium atropurpureum 52
Effect of fungal inoculation and P applications on
the shoot dry weight, root fresh weight, root
colonization, P concentration, total P, and
total N of Aeschynomene americana. 53
Effect of P application on shoot dry weights
for the first (A), second (B), third (C), and
fourth (D) harvest of Siratro grown in the field
and inoculated with Glomus etunicatum (ETU),
Glomus intraradices (INT), or not inoculated (CON).
Data points are means of five replicates 64
xx


88
Munns, D. N., and B. Mosse. 1980. Mineral nutrition of legume crops.
In R. J. Summerfield, and A. H. Bunting (eds.) Advances in legume
science. Univ. Reading Press, England
Newbould, P., and A. Rangeley. 1984. Effect of lime, phosphorus and
mycorrhizal fungi on growth, nodulation, and nitrogen fixation by
white clover ( Trifolium repens) grown in UK hill soils. Plant Soil
76: 105-114.
Nielsen, J.P., and A. Jensen. 1983. Influence of vesicular-arbuscular
mycorrhiza fungi on growth and uptake of various nutrients as well
as uptake ratio of fertilizer P for lucerne (Medicago sativa).
Plant Soil 70: 165-172.
Pairunan, A. K., A. D. Robson, and L. K. Abbott. 1980. The
effectiveness of vesicular-arbuscular mycorrhizas in increasing
growth and phosphorus uptake of subterranean clover from phosphorus
sources of different solubilities. New Phytol. 84:327-338.
Plenchette, C., V. Furlan, and J. A. Fortin. 1982. Effects of different
endomycorrhizal fungi on five host plants grown on calcined
montmorillonite clay. J. Amer. Soc. Hort. Sci. 107:535-538.
Powell, C.L1. 1979. Inoculation of white clover and ryegrass seed with
mycorrhizal fungi. New Phytol. 83: 81-85.
Powell, C. LI. 1980a. Mycorrhizal infectivity of eroded soils. Soil
Biol. Biochem. 12:247-250.
Powell, C. LI. 1980b. Phosphate response curves of mycorrhizal and non-
mycorrhizal plants. I. Responses to superphosphate. N. Z. J. Agr.
Res. 23:225-231.
Powell, C. LI. 1982. Selection of efficient VA mycorrhizal fungi.
Plant Soil 68:3-9.
Powell, C. LI., M. Groters, and D. Metcalfe. 1980. Mycorrhizal
inoculation of a barley crop in the field. N. Z. J. Agr. Res.
23:107-109.
Purcino, A. A. C., C. Lurlarp, and J. Q. Lynd. 1986. Mycorrhiza and
soil fertility effects with growth, nodulation and nitrogen fixation
of Leucaena grown on a Typic Eutrustox. Commun. in Soil Sci. Plant
Anal. 17:473-489.
Purcino, A. A. C., and J. Q. Lynd. 1985. Tripartite symbiosis of
Stylosanthes scabra Vog. influenced by soil fertility treatments of
a Typic Eutrustox. Agron. J. 77:455-458.
Rangeley, A., M.J. Daft, and P. Newbould. 1982. The inoculatation of
white clover with mycorrhizal fungi in unsterile hill soils. New
Phytol. 92: 89-102.


6
root sample was cleared in 10% KOH and stained with 0.05% trypan blue in
lactophenol (Kormanik and McGraw, 1982). Root colonization by VAM fungi
was estimated by the gridline-intersect method of Giovannetti and Mosse
(1980).
Chemical content of a composite soil sample from each location was
determined by the Soil Testing Laboratory, University of Florida (Rhue
and Kidder, 1984). Mehlich-I solution (0.05 M HC1 + 0.0125 M H2SO4) was
used to extract Al, Ca, K, Mg, and P. All elements were analyzed in the
filtrate by atomic absorption spectrophotometry, except P which was
determined using the ammonium molybdate/ascorbic acid colorimetric
method. Soil pH was determined using a 1:2 (v/v) soilrwater ratio.
Organic matter was estimated by oxidation with 1 N ^0^0-7 in the
presence of H2SO4.
Spores of VAM fungi were removed from soil by the wet sieving method
of Daniels and Skipper (1982) using sieves with 425, 90, and 45 um
openings. Fractions retained on 90 and 45 um sieves were centrifuged
(1000 x g) for 3 min in water. The pellet was resuspended in 40% sucrose
solution and centrifuged for 1.5 min. Spore species were identified
where possible (Schenck and Smith, 1982; Trappe, 1982). In addition,
spores or washed roots were placed in pasteurized Arredondo loamy sand
surface soil (siliceous hyperthermic Grossarenic Paleudult) in 15-cm-diam
plastic pots in the greenhouse and planted with bahiagrass (Paspalum
notatum Flugge), carpon desmodium, or Siratro in an attempt to isolate
VAM fungi in a manner similar to that described by Gerdemann and Trappe
(1974) as the "inoculated pot culture" method.


Table 7-2. Analysis of variance for shoot dry weights of
Aeschynomene americana harvested 2 October 1986. . 61
Table 7-3. Analysis of variance for shoot dry weights
from four Siratro harvests. 63
Table 7-4. Regression equations and coefficients of
determination (r^) showing the relationship
of applied P level to shoot dry weight, percentage
of root colonization, P and N concentrations, and
total P and N uptake for the first harvest
of Siratro 66
Table 7-5. Regression equations and coefficients of
determination (r~) showing the relationship
of applied P level to shoot dry weights for
the first, third, and fourth harvest and
percentage of root colonization for the
fourth harvestof Siratro 67
Table 7-6. Regression equations and coefficients of
determination (r~) showing the relationship
of applied P level to shoot dry weight,
percentage of root colonization, P concentration,
and total P and N uptake of Aeschynomene
americana. . 68
Table 7-7. Analysis of variance table for percentage root
colonized of Siratro by VAM fungi 71
Table 7-8. Analysis of variance table for percentage root
colonized, P concentration, total P, and total
N of Aeschynomene americana. ...... 72
Table 7-9. Analysis of variance table for P concentration,
total P, N concentration, and total N of
Siratro for the first harvest 75
C


86
Hayman, D. S. 1983. The physiology of vesicular-arbuscular
endomycorrhizal symbiosis. Can. J. of Botany 61:944-963.
Hayman, D. S., and K. A. Hampson. 1978. VA mycorrhiza. Field
inoculation trial (white clover in Welsh upland soil). Rothamsted
Rep. Part 1:238-239.
Hayman, D. S., and B. Mosse. 1979. Improved growth of white clover in
hill grasslands by mycorrhizal inoculation. Ann. Appl. Biol.
93:141-148.
Hayman, D. S., R. J. Page, and C. A. Clarke. 1980. Vesicular-
arbuscular mycorrhiza: field inoculation studies with Red clover,
Sawyers I. Rothamsted Rep. Part 1:201-202.
Hayman, D.S., and G.E. Stovold. 1979. Spore populations and infectivity
of vesicular-arbuscular mycorrhizal fungi in New South Wales. Aust.
J. Bot. 27: 227-233.
Hayman, D. S., and M. Tavares. 1985. Plant growth responses to
vesicular-arbuscular mycorrhiza. XV. Influence of soil pH on the
symbiotic efficiency of different endophytes. New Phytol. 100:367-
377.
Hodges, E. M., A. E. Kretschmer, Jr., P. Mislevy, R. D. Roush, 0. C.
Ruelke, and G. H. Snyder. 1982. Production and utilization of the
tropical legume aeschynomene. Fla. Agrie. Exp. Stn. Circ. S-290.
Huang, R., W.K. Smith, and R.S. Yost. 1985. Influence of vesicular-
arbuscular mycorrhiza on growth, water relations, and leaf
orientation in Leucaena leucocephala (Lam.) Dewit. New Phytol. 99
229-243.
Hutton, E. M. 1962. Siratro-a tropical pasture legume bred from
Phaseolus atropurpureum. Aust. J. Agrie. Anim. Hus. 2:117-125.
labal, S. H., K. Sultana, and B. Perveen. 1975. Endogone spore numbers
in the rhizosphere and the occurrence of vesicular-arbuscular
mycorrhizas in plants of economic importance. Biologia 21:227-237.
Islam, R., A. Ayanaba, and F. E. Sanders. 1980. Response of cowpea
(Vigna unguiculata) to inoculation with VA mycorrhizal fungi and
rock phosphate fertilization in some unsterilized Nigerian soils.
Plant Soil 54:107-117.
i i j . ....
Jensen, A. 1984. Responses of barley, pea, and maize to inoculation
with different vesicular-arbuscular mycorrhizal fungi in irradiated
soil. Plant Soil 78: 315-323.
Khan, A. G. 1975. Growth effects of VA mycorrhiza on crops in the
field. P. 149-435. In F. E. Sanders, B. Mosse, and P. B. Tinker
(eds.) Endomycorrhizas. Academic Press, London, New York.


26
Results and Discussion
Plants inoculated with G. etunicatum and G. intraradices had higher
shoot dry and root fresh weights than plants inoculated with the other
VAM fungi or control plants, at 40 and 70 d after planting (Fig. 4-1).
At both harvests, plants inoculated with G. etunicatum had higher shoot
dry weights than plants inoculated with G. intraradices. At the final
harvest, plants inoculated with G. intraradices had higher root fresh
weights than plants inoculated with G. etunicatum.
In contrast, plants inoculated with G. versiforme, G. margarita, and
G. deserticola had shoot dry and root fresh weights that were not
different from the noninoculated plants, except at the final harvest when
plants inoculated with G. deserticola had higher shoot dry and root fresh
weights than the control (Fig. 4-1). At 20 d, shoot dry weights were
not different among treatments (mean = 0.70 g).
rPercentage and total root length colonized by VAM fungi increased
with time (Fig. 4-2). Inoculation with G. etunicatum and G. intraradices
resulted in the highest root colonization at all harvests. At 70 d,
plants inoculated with G. etunicatum had the highest root colonization,
followed by G. intraradices and then G. deserticola. There were no
differences in root colonization among G. versiforme, G. margarita and
control treatments. For the six treatments, total root length colonized
by VAM fungi and percentage of mycorrhizal root colonization followed
the same trend (Fig. 4-2).
Shoot dry weight of Siratro was correlated with total root length
colonized by VAM fungi (r2 = 0.95""") and percentage of mycorrhizal root