Citation
Lime interactions in soils from Costa Rica and the eastern savannas of Colombia

Material Information

Title:
Lime interactions in soils from Costa Rica and the eastern savannas of Colombia
Creator:
Velez Pelaez, Julian, 1944-
Publication Date:
Language:
English
Physical Description:
xx, 199 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Acidity ( jstor )
Adsorption ( jstor )
Ions ( jstor )
Nutrients ( jstor )
pH ( jstor )
Phosphorus ( jstor )
Regression coefficients ( jstor )
Soil science ( jstor )
Soils ( jstor )
Titration ( jstor )
Dissertations, Academic -- Soil Science -- UF
Liming of soils ( lcsh )
Soil Science Ph. D
Soils -- Colombia ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1975.
Bibliography:
Bibliography: leaves 185-197.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Julian Velez Pelaez.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. 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. 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:
028525114 ( ALEPH )
08819315 ( OCLC )

Downloads

This item has the following downloads:


Full Text












CIMT R -CA 1ND THE EASTERN RAVtAS C" COLOE1A By








iJUIAN 3 PEAZ






A O P .. IT'ED TC THE CA COUCIL OF T . ;rE'RSIT OF F....IDA IN PAT IALFULIL-C'NT OF T' UF I E Y OF 'FLO'A






































To my wife,

Helen Margaret Velez
















ACKNOWLLUGNTS


The author wishes to express his deep appreciation to

Dr. W. G. Blue, Chairman of the Supervisary Committee, for his interest, guidance, assistance, patience, and personal friendship throughout the course of the academic and research programs; to Dr. L. W. Zelazny for instruction and valuable suggestions during the overall investigation, and the preparation of this manuscript; to Dr. T. L. Yuan for his advice in preparing laboratory experiments, his clarification of soil acidity and phosphorus adsorption concepts, and preparation of this dissertation; to Dr. G. O. Mott for his relevant comments and advice in the design of greenhouse experiments and interpretation of data, and his

help in reviewing this -work; and to Dr. O. C. Ruelke for his contribution to a greater understanding of crop ecology and for serving on the

Supervisory Committee.

He is indebted to Mr. Edgar Rey and NLr. William Cordero

from the Ministry of Agriculture for their cooperation and aid given during the collection of soil samples in Costa Rica; to Dr. J. 1. Spain from the Center of Tropical Agriculture for his valuable assistance and advice in collecting soil samples in the eastern savannas of Colombia; to Dr'. C. Luna, Ingeniero Agrono:no Victor Vega, the soil classification staff of the Instituto Agustin Codazzi and Dr. E. suarez in Colombia






iii









for their outstanding cooperation in selecting samping sites and storing and shipping soil samples to the United States.

The author is grateful to Dr. C. F. Eno, Chairman of the SoilShience Department, for financial assistance which allowed the coimpletion of this study and provided support for other financial needs; and to Dr. H. L. Popenoe for providing funds used in collection and transportation of soils.

Appreciation is extended to Dr. H. L. Breland, Mrs. Helen

Brasfield and all other personnel of the Analytical Research Laboratory for their assistance in analysis of the large number of samples required for this work; and to Mr. A. Waller and the field personnel for their very important contribution in setting up greenhouse experiments and subsequent collections and preparations of plant material for analysis.

Gratitui is e .rssed to faculty and staff members, especially Mr. J. Gonzales, of the Soil Science Department; to fellow graduate students for their friendship and advice; and to Dr. Marciano Rodriguez for the long hours spent together with the author during the collection and interpretation of data.

Most of all, the author wishes to express his eternal appreciatiori to his wife, Helen Margaet Velez, for all the support, encouragement, understanding, and help provided throughout these long years of hard work; to his parents, Dr. Julio Velez and Mrs. Elvia de Velez, for their immense contribution to the author's education and for their patience in waiting all this time for his return home; to his son, Julian Eduardo Velez, for providing everlasting hours of happiness, especially during periods of stress and uncertainty.



iv
















TAEXl OF CONTENT'
Page

LCKS'NO'V 3 TiE NTz . . . . . . . .. .. . . . . . . . . . .. 11

LISTFTABIES .._....... .. ............. x

S OF ICUR S . . . . . . . . . . . . . . . . . . . . . . . . . xvii

ABhiTRACT. ............. . . . . . . . . . ... xviii

ITlrODCTIO . . .. . . . . . . . . . . . . . . . . . . . . . .

LITERATUiE REViEW . . . . ... . . . . . . . . . . . . . . . . . . 5

otu rct:s of Soil Acidity . . . . .. . . 5

Inorganic Sources . . . . . . . . . . . . . . . . . . . 5

Sourc s . . . . . . . . . . . . . . . . . . . . 6

oluble AcI ds . . . . . . . . . . . . . .. . . . . . . .

Measurement of Soil Acidic Properties. . . . . . . . . . . . 8

Types of Measurable Acidity . . . . . . . . . . . . . .8

Titration Analysis . . . . . . . . . . . . . . . . . . 9

Acid Soils and P ant Growth. . . . . . . . . . . . . . . . . 11

Hydrogen lon Concentration. . . . . . . . . . . . . . . 11

Calcium Deficiecy .......... . . . . . . . 12

Magnesium Deficiency . . . . . . . . . . . . . . . . . 12

Phosphoru Deficiency . . . . . . . . . . . . . . . . 1

Alumium Toxicity . . . . . . . . . . . . . . . . . . . 13

Manganese Toxiitv. . . . . . . . . . . . . . . . . . 1

Molvbd.enum D ciency ... .............. 14

S of oil Ac . . . . . . . . . . . . . . . 15


-










? CNT:TS (Continued.

Page

Soi I Aluminum and Iron ratios . . . . . . . . . ..... 17

Jnn.conill' Oxalate Fraction . . . . .. . . . . . . 17

Cit.rai;e .th. ontn t jicarboiate Fraction . . . . . 17 Sodium Pyrophosphate Fraction . . .. 1 odiam Hydroxide Fraction . ......... . . . 1

Ammonium Acetate Fracticn ...... . . . . 8 Potassium Chloride Fraction . . . . . . . . . . . . . 19

Fractionation Scheme. . . . . . . . . . . . . . . . 20

Effect of Lime cn Soil Alumirum and Iron . . . . . . . . 20 Phosphorus Availability in Acid oils . . . . . . . . . ... 2

Phosphorus Fixation . . . . . . . . . . . . . . . 22

Effect of Aluminum ar.d Lion on Pncsphorus
Availability. . . .................. . . 24

Effect of Lime on Phosphorus Availability . . . . . . . 25

Effect of Calcium Silicate on Phosphorus
Availability. . . . . . . . . .. . ... . . . . . . 27

The Use of the Langmuir Isotherm in Describing
PFosphorus Availablity . . . . . . . . . . . . . . .

Electrochemnical Properties o'" Acid Soils. . . . . . . . . . . O

Tile Isoelectric Point and the Zero Point
of Chargo . . . . . . . . . . . . . . . . . . . . . . 30

Measurement oi' the ZroPir P:int ofC . ae . . . . . . . . 31

Effect of Soil i.nr~lcgy on the Zero Point
of Charge. . . . . . . . . . . . .. . .... ... . 33

Effect of Cation an Antion Adsorption on t:e
Zero Pcint of Cha:. . . . . . . .......... . . .



vi










TAJi3LE OF CONTn.S (Cont.nned)

Page

Practical Implications of the Zero Point
of Charge ..... . . . . . . . . . . * * * * * 35

>Calcium Selectivity in Acid Scils. .-. . ... . . ....... 36

Theoretical ..... . . . . . ...... ......... 38

Effect of Iime on Electrolyte Accunulation
In Acid Soils. 4

Indicator Plant for Greenactse Studies ... . . . . . . . 43 MATR::.AIS AL D "EI THO.. ....... . . .. . . . . .5 .. . . 5

Soil Saples .. ............*.*.. ..5

Laboratory Procedures ..... . . . . . . . . . . . ...147

General Chemical Properties . ...... ...... . 47

Iineralogical Analysis. . . . . . . . . . 4 .

Soil Aluminum and Iron Fractions . .. ......... . . . 9

Lime Requirement . ................ .. .

Sources of Scil Acidity . ...... . ........ . 49

Phospnorus Adsorption . ..... .. .... ...... . 51

Electrochemical Properties. . ............. . 52

Calcium Selectivity . . . . ..... . . . . . 2

Incubation Studies with Lime. . .... .. ....... . 53

Analytical Dte minations . ......... . . . . .

Greenhouse Procedur e. ...... . ............. . . .

General Prleparation ct' Experimcnts .. .......... 54


eri: . '. . .. . ... ........... 55










TABLE OF CONTENT (Continued)

pa-f-Experiment No 3 . . . . . . . . . . 58 REE3ULTS AND DISCJS3ION .. . . . . . . . . . . . . . . . . . . . . 2

Laboratory Experiments. . . .. . . . . . . . . . . . . . 62

General Chemical Properties. . . . . . . . . . . . 62

Extractable Soil Nutrients . . . . . . . . . .. . . . . 4

Total Nutrients. . . . . . . . . . . . . . . . . . . . . 69

Analysis of the Clay Fraction ... . . .. .6... . . 9

Soil Aluminum and Iron Fractions . ........ . .. 7

Characterization of Iron and Aluminum. . . . . . . . . . 75 Lime Requirement . . . . . . . . . . . . . . . . . . . . 7d

Sources of Soil Acidity. . . . . . . . . . . . . . . . . .

Pnosphorus Adsorption . . ..... . .. . . .. 92

Effect of Lime on Phosphorus Adsorption. . . . . . . . . 94 Electrochemical Properties .............. . 98

Calcium Selectivity. . . . . . . . . . . . . . . . . . .

Incubation Experiment5t. . . . . . . . . . . . . . . . .

Greenhouse Experimentz. . . . . . . . . . . . . . . . . . . . 118

Experiment No. 1 . . . . . . . . . . . . . . . . . . . . 118

Experiment No. 2 . . . . . . . . . . . . . . . . . . . . 136

experiment No. 3 . . . . . . . ................... .151

SUIARY Ai CU N .....L IO.... ....... .......... 156

APPEIDCE ....................... .... 161






viAi
.1� - i









TABLE OF C017111 3 Co! .Jinucd



Appendjx B ............ ..... ........ 174

LIT'ERA9i I CI D ................... . ...... 15

BIrORAPH CAL Y''KETCH.. .. ... 198














































ix















I~ST OF TABLES

Table


i Identification number, location, and order
of the soils studied . . . . . . . .................

2 Relation of treatment number to experimental
and coded variables for Experiment No. 1 ... . ... . . 56

Lime and phosphorus treatments for Experiment
N o2 . .................... ...... 59

4 Lime, calcium silicate, and phosphorus. treatments
for Experiment No. 3 .... . . . . . . . . . . . 61

5 Chemical properties of the soils studied . ....... 63 6 Extractable nutrients from the soils sTudied . ..... 65

'7 Chemical properties of the soils studied
related to fertility . . . . . . . . . . . . ...... 68

8 Total elemental analysis of the soils studied. ...... 70

Mineralogy and cation exchange capacity of
the clay fraction from the soils studied . ....... 72 i() Aluminrum fractions of the soils studied. . ..... . . 74 11 Iron fractions of the soils studied. . ......... . o6

12 C0naracterization of iron and aluminum in the soils studied. ... . ............ . . . . . . . . . '77

13 Lime reuiremonts for the soils studied by the exchanleable aluminum, 3MP, and Yuan
tech!ni qu.:cs..... . .... .. ............. 7

l.4 Titratable and ex:tractable acidity in the soils studie . . . . . . . . . . . . . . *.. .......

15 Correlation ac:i extractable acidity, Ii:.. requiren:ent, an,!
c~;:n:icl pro'rtic of i:he1 soils studio~ . ....... . 9


x










IST OF TABILEZS (Continued)

Table Pe

16 Correlation among titratable acidity, cxIractable acidity, l.in requirement,
and alumvniru in tAe coi2s studied . . . . . . . 91

17 Phosphorus adsorption ~ the t soils; Sidicd with an intensity and a quantity itLhod . . . ........ 93

18 Corriation between phosphlorus adsorption ar; ae;ers and selc ed chemical properties
of the soils studied . . . . . . . . . . . . . . 95

19 Regression coefficients and coefficients of determination for the effect of lime (:eql100g)
on the P-sorption capacity (ppm) of incubated
soils measured by the intensity method . . . . . . . . 96

20 Effect of lime on Langmuir t-ype parameters, and coefficients of dletermination for P
sorption measured by the quantity -method
for the soils studied .. ... ........... . . . 97

21 Delta pH-measur ents. or the -oils studied .-..... . 99

22 The negative logarithm of calcium selectivity coefficients for the soils studied in the
presence of potassium, magnesium, and aluminum
in lN solutions . . . . . . . . . . . . . . . . ... .. IC

23 Regression coefficients and coefficients of determination for the effect of lime (meqLi00g)
on the water pH of incubated soils. . ... ..... . 110

24 Regression coefficients and coefficient of determination for the effect of lime (i;eq/100g)
on the IN KC1 puI of incubated soils . ......... 111

25 Regression coefficients and coefficients of determination for the effect of lime (meq100g)
on the exchangeable aluminum (mteq;lOg) of
incubated,(. soils ...... . . .......... . 11.2

2b Regression coeff.cients and coefficients of determination for the effect of lime (meq,1COg)
on the vater extractable potassium (p:n) of
incubated soils . . . . . . . . . . . . . . . . . . . ..113

27 Rtegression coefficients and coet'ffici-,nts of determination for th,:e effect of li:e (mCieqi,!00)
on the water extractable sodium (pmm) of
incubated ;a:u;lls .... . . . . . . . . . .










LIST OF TABLES (Continued)

Table Page

28 Regression coefficients and coefficients of determination for the effect of lime (meq/100g)
on the water extractable calcium (ppm) of
incubated soils . . . . . . . .................... . . 116

29 Regression coefficients and coefficients of.
determination -for the effect of lime (meq/100g)
on the water extractable magnesium (ppm) of
incubated soils. . ..... . . ........... . . . 117

30 .Regression coefficients and coefficients of .
determination for the effect of lime (meqJlOOg) on the water extractable nitrate-nitrogen (ppm)
of incubated soils. . . . . .... . . . . . . . . . 119

31 Regression coefficients and coefficients of
determination for the effect of lime (meq/iC0g)
on the electrical conductivity ()Jmhos/cm) of
incubated soils . . . . . . . . . . . . . . . . . . . . 120

32 Regression coefficients and coefficients-of determination for Pangola digitgrass yield
and nut'ient concentration fro Los Di.manes
surface soil, Experiment No. 1 ........... .. 122

33 Regression coefficients and coefficients of determination for Pangola digitgrass yield
and nutrient concentration from Los Diamantes
subsoil, Experiment No. 1 ............ . . . 123

34 Regression coefficients and coefficients of determination for Pangola digitgrass yield
and nutrient concentration from San Vito soil,
Experiment No. 1. . ..... . ... . . . ......... 124

35 Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration from San Isidro
soil, Experiment No. 1. . .......... . . . . . . 126

36 Regression coefficients and coefficients of determination for Pangola digitgrass yield
and nutrient concentration from Grecia soil,
Experiment No. 1. . ................... 127

37 Regression coefficients and coefficients of determination for Pangola digitgrass yield
and nutrient concentration from Alajuela
soil, Experiment No. 1. . ............. . . 129

xii









110T OF TAiLES (Continued)

Table Page

38 Regression coefficieits and coefficients of determination for Pangola- digitgrass yield
and nutrient concent ration from Agronomy Area
soil, Experiment No. 1 . . . . . . . . .. . . . . 130

39 Regression coefficients and coefficients of determination for Panrgola digitgrass yield
and nutrient concentration from Cararao soil,
Experiment No. 1 . .. .............. . . . 131

40 Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration from Drainageway
soil, Experiment No. 1 . .............. . . 133

41 Regression coefficients and coefficients of determination for Pangola digitgrass yield
and nutrient concentration from East Carinagua
soil, Experiment No. . . . . . . . . . . . . . . . . . 134

42 Lime, magnesium, and phosphorus treatments to the soils studied predicted .to give maximum
Pangola digitgrass yield by the model used in
Experiment No. 1 . . . . . . . . . . . . .. . . 135

43 Effect of lime and phosphorus treatments to Los Diamantes surface soil on Pangola digitgrass
yield and nutrient concentration, Experiment
No. 2. . . . . . . . . . . . . . . . . . . . . . . . . 137

Wh Effect of lime and phophorus treatments to Los Diamantes subsoil on Pangola digitgrass
yield and nutrient concentration, Experiment
No. 2. . . . . . . . . . . . . . . . . .. . . . . . . . 139

45 Effect of lime and phosp oorus treatments to San Vito soil on Pangola digitgrass yield
and nutrient concentration, Experiment No. 2 . . . . . . 140

46 Effect of lime and phosphorus treatments to San Isidro soil on Pan-ola digitgrass yield
and nutrient concentration, Experiment No. 2 . . . . . . 142

47 Effect of lime and phosphorus treatments to
Grecia soil on Pangola digitgrass yield and
nutrient concentration, Experiment No. 2 . . . . . . . 113

I4 Effect of lime and ph ,sphoru; treatments to Alajucla soil on P anola digitciiss yicld and
nutrient concentration, xpK:'jcir7nt No. 2 . . . . . . . .
xiii










LIST OF TABLES (Continued)

Tableage

49 Effect of lime and phosphorus treatments to Agronomy Area soil on Pangola digitgrass yield
and nutrient concentration, Experiment No. 2 . .... . 146

50 Effect of lime and phosphorus treatments to Carararo soil on Pangola digitgrass yield and
nutrient concentration, Experiment No. 2 . . . . . . . . 147

51 Effect of phosphorus treatments to limed Driinageway soil on Fangola digitgrass yield
and nutrient concentration, Experiment No. 2 . . . . . . 149

52 Effect of lime and phosphorus trta;ments to East Carimagua soil on Pangola dig.tgrass vield
and nutrient concentration, Experiment No. 2 ...... 150

53 Effect of lime and calcium silicate treatments to soils from Costa Rica on Pangola digitgrass
yield and nutrient concentration, Experiment
N . 3. . . . . . . . . . . . . . . . . . . . . 153

54 Effect of lime and calcium silicate treatments to soils from Colombia on Pangola digitgrass yield and nutrient concentration, Experiment
No. 3 . . . . . . . . . . . . . . . . * * * . . . . . . 155

55 The effect of lime on the phosphorus sorption capacity of incubated soils measured by the
intensity method . ............... ... 163

56 Effect of lime on the organic matter content of incubated soils. . .................. . 164

57 Effect of lime on the water pH of incubated soils.................. ... . . . . . .. .. . . . . . 165

58 Effect of lime on the 1N KC1 pH of incubated soils ..... .... ....... .. ........ . . 166

59 Effect of lime on the exchangeable aluminum of incubated soils . ... . . . . . . . . . . . . . . 167

60 Effect of lime on the water extractable potassium of incubated soils . . . . . . . . . . . . . . 168

61 Effect of lime on th2 water extractable sodium of incubated soils. ..... . . ... ... 169

xiv










LST O TALE (Continuied)

Table Page

62 Effect of lime on tbh- water extractable calcium of incubated soils .. . .. ......... 170

63 Effect of lime on the water extractable magnesium of incubated soils .............. 171

64 Effect of lime on the water extractaole nitrate-nitrogen of incubated soils. . ..... . . . . . 172

65 Effect of lime on the electrical conduc-tivity of incubated soils . ...... ....... . .. . 173

66 Effect of lime, magnesium, and phosphorus treatments to Los Diarmantes- Surface soil on
Pangola digitgrass yield and nutrient concentration from Experiment No. 1 ............. . 175

67 Effect of lime, magnesium, and phosphor.Is treatments to Los Diamantes Subsoil on
Pangola dlgitgrass yield and nutrient
concentration from Exneriment No. 1 .. .... ... . 1.76

68 Effect of lime, magnesium, and thophorus treatments to San Vito soil on Pangola
digitgrass yield anid nutrient concentration
from Experiment No. 1. . ................. 177

69 Effect of lime, magnesium, and phosphorus treatments to San Isidro soil on Pangola
digitgrass yield and nutrient concentration
from Experiment No. i. . ............ . . . . 178

70 Effect of lime, magnesium, and phosphorus treatments to Grecia soil on Pangola digitgrass yield and nutrient concentration from
Experiment No. 1 . . . . . . . . . . . . . . . . . . . . 179

71 Effect of lime, magnesium, and phosphorus treatments to Alejuela soil on Pangola
diritgrass yield and nutrient concentration
from Experiment No. i. ............ . . . . 180

72 Effect of lime, r!iagncsium, and phosphorus treatments to Aronony Area ;oil on Panu ola dig tgra s yield and nutri ent concentration
fro:i: Experiment No. 1................ . 181



xv










LLIT OF TB3LES (Continued)

Table

73 Effect of lime, magnesium, and phosphIorus treatments to Cararao soil on Pangola
digitgrass yield and nutrient concentration
from Experiment No. 1 . . . . . . . . . . . ... . . . 182

74 Effect of lime, magnesium, and phosphorus
treatments to Drainageway soil on Pingola
digitgrass yield and nutrient concentration
from Experiment No. 1 . .... . . . . . . . . . . 183

75 Effect of lime, magnesium, and phosphorus treatments to East Carimagua soil on Pargola digitgrass yield and nutrient concentration
from Ex eriment No. 1 . ................ 184


















Liming acid soils has been a basic component of soil management systems in temperate regions for many years. This component i. used to maximize growth of temperate legumez that have slightly acid to neutral pi requirements for N fixation, and to keep soil natriencs readily available for plant uptake. In recent years, two schools of thought with respect to limii acia loils hav. developed: McLean (01) at Ohio State University advocates liming soils to p~ 6-5 because of higher nutient availability to plants and higher i. riial activity at .this rH value. These have bee the traditional reason c fo - lmin. Ka nprath (78) at :Torth Carolina State Univ:z ity advocates ucin3 lie quantities that neutralize the exceangeable Al aad& bing Al saturat;ion belo,-w 1.-! in the soil. Reeve nd Sumner (126) i. oath Africa support the exchanjcable-Al school and favor additions of licm suf icient to lower toe uex.cha .ngeable Al1 to 0.2 e eq10lOg of soil. The concept . f l.loed by these two schools raise the question on whether to lime to a favorable pH rarc, ualy between 6.0 and 6.5, or to inactivate toxic substance such as Al and Mc, which takes place around pI 5.5. Evidence in eavor of each theory has been pat forz-ard by the proponents of each line of nout; however, it seems that the idea of limirg to inactivate toxic utance s gaining meomentLtm especially in regions where highly

feathered acid soils containing large amoLunts of toxic substances are cor-::on.


1









When lime was first introduced irto the tropics, the concept of the most favorable pH range was used as the criterion for lime applications. Responses were often erratic and even detrimental to plant growth: Adams and Pearsonu (2) stated that crops respond differently to lime in tropical soils in comparison. t cresponses in temperate soils with maximum yields occurring at pH's from 4.5 to 5.5-. Russell and Richards (129) mentioned that liming in the tropics and subtopics only improves yields on very acid soils and usually reduces yields on moderately acid soils. Numerous experiments in the eastern province of Nigeria (154) indicated good responses to lime applications between 0.5 and 1 ton/acre, particularly in soils with pH below 4.5. In Kenya and southern Rhodesia (154) the application of lime failed to improve the establishment or productivity of grass leys in areas with long dry seasons. Data from Puerto Rico (1), Brazil (110), and Hawaii (157) support lower soil reaction values for optimum plant growth.

All of the work mentioned seems .to substantiate the inactivation of to.cic substances as the best criterion for liming acid tropical soils. However, research by Lucas and Blue (94) with a soil from eastern Costa Rica having low exchangeable Al and Mn showed a depressing effect of lime on plant growth even at rather low lime rates when compared to the high buffering capacity of this soil. Velez, Zantua, and Blue (146) found high electrolyte concentrations as the possible cause for the detrimental effect of lime in the same soil. These authors noted that the negative effect of lime was time dependent and plant yields improved considerably as time increased. This electrolyte effect was first mentioned by Zantua and Blue (164). They observed a dramatic positive growth response of Pangolagrass (Digitaria decumbens stent.) planted to









the limed and unlimed samples of soil used by I;acas and Blue which had been preciously leached with v;ater.

As was mentioned earlier, liming has been used to improve the availability of plant nutrients. Of all the nutrients required for plant growth, P has been singled out as one of the most deficient nutrients in tropical soils (19, 118) and living has been recognized as the best practice to increase P availability in temperate soils. However, work by several researchers (68, 79, 125, 144) has disclosed the possibility that lime, instead of improving P availability, increases the P-retention capacity of highly weathered soils with high Al and Fe content and soils rich in amorphous materials both of which are common in the tropics.

Sound management practices must therefore be developed :reless of the reasons for liming in crder to increase food production to zximum levels in tropical regions where malnutrition and Stavation are comr:mon. Therefore, the following objectives were developed to obtain pertinent answers related to the use of lime in soils from: Costa Rica and the eastern savannas of Colombia:

(a) To determine acidic properties and buffering capacities

of these soils.

(b) To characterize the Al and Fe fractions and their influence

on soil acidity.

(z) To determine the P adsorption capacity, as related to

soil components, and the effect of added lime.

(d) To determine the electrcheiic al properties of the soils,

the pil at which anion or cation adsorption takes place,

and the eff'iect of pH on anion r nd cation adsorption.








(e) To determine th sclectivi~.ty o: th.e soils for Ca,

a component of lime, as i.unu;:ced by the presence of

other cations in soil solution.

(f) To determine the effect of' lime: on the accumulation

of electrolytes in the soils.

(g) To determine the optimum levels of P, Mg, and lime

needed for maximun plant growth.

(h) To determine the effect of CaCO, and CaSiO3 in reducing

the need for fertilizer P.















r'V1EW OF LITERA :r


Sources of Soil Acid ty

Soil acidity was defined by Jackson (72) as the proton yielding capacity of a soil system in going from a given state to

a reference state usually specified as a pH value. The proton sources are, therefore, the sources of soil acidity, and in general they can be divided into inorganic and organic soil constituents (33, 160). Inorganic Sources

The inorganic constituents responsible for acidic properties in soils are layer silicates, oxide minerals, and combinations of layer silicates and oxide minerals and non-crystalline components (33). Yuan (160) proposed a more general grouping of the inorganic components of soil acidity into crystalline and amorphous hydrous oxides and silicates. Protons are generated from these soil constituents by the dissociation

of H ions from structural OH- groups such as those praePt at 1he corners and edges of clay lattices and as basic constituents of hydrous oxide minerals of Fe, Al, and Si (134). Hydrogen ions can also be found satisfying permanent charges on clay minerals (160). These H ions can

be replaced by cations, and thus, they beco"-e sources of acidity. However, H-saturated clays are very unstable and shift to Al-saturated clays spontaneously (33).

The most important source of acidity in mineral soils is the hydrolys�i of Al ions. These ions are displaced from clay minerals by


5









cations and hydrolyze in solution one of zhe products of this hydrolysis reaction is J3' -hich is capable of decomposirn minerals yielding more hydrolyzable Al (33). Jackson (73) described the hydrolysis of Al as

taking place in stages; in the first stage, equal amounts of . ions and mono-hydroxy Al ions are produced below pH 5.0. As the OH/Al ratio increases, polymerization of Al-OIH units take place and large hydroxy Al ions form varying in the OH/Al ratio and in size. The latter increases as the OH content increases. The ultimate product of hydrolysis is Al(OH) which usually precipitates when the OH/Al ratio becomes greater than 2. In general, as pH increases, hydroxy-Al ions tend to polymerize forming a layer lattice composed of stable ring units containing six Al ions with one net positive charge per Al arid releasing H ions to the soil solution (69).

Organic Sources

Functional groups of organic matter such as carboxyls, phenols, enols, amines, and quinones are capable of dissociating hydrogen ions. The type and amount of functional groups vary with soils and organic matter fraction; the nature of the predominant groups present will determine the strength of the acid produced (132). It is recognized that carboxyl groups are acidic enough to dissociate considerably below pH 7.0; nevertheless, phenols and polyphenols are also capable of Hdissociation and, along with the other groups, contribute as well in

yielding protons (33).

It appears that A1+3 and Fe3 in organo-metallic complexes are responsible for the weak acid character of organic matter (96). Interaction of hydrous oxides and organic colloids is regarded as involving the displacement of H ions from carboxylic groups by hydro.y-Al





7


and' - Fe ions (133). Schnitzer r:dc Skinnr (133) showed that the stability of Fe and Al complexes varied with pti. Chelates of Fe with organic compounds were found to be stable in acid media, but decomposed in alkaline media with the production of Fe(OK)3 (44). Complexes of Al and fulvic acid were broken through the replacement of fulvic acid by OH ions at p1 5.0 (44, 85). The hydroxy-Al ions created in this fashion are capable of entering the hydrolysis reaction with the production of H ions. They can also become active in other soil reactions such as P fixation.

Soluble Acids

Coleman and Thomas (33) stated three ways in which soluble acids from biological activity can be of considerable importance in soil acidity. Their considerations can be sumwnarized by mentioning the effect of additions of acid-forming fertilizers to soils, the production of strong acids when soils containing FeS are exposed to oxidizing conditions, and the presence of organic acids from plant residues, root exudates, and litter.

Jackson (73) grouped soil acidity and its neutralization based on the acid strength of the proton retaining site into the

following categorJes:

I. Strong acids, soil pUl h.2 and below:

a. Mineral collcidal electrolytes: Mg, Al-. . .

(HC1 or HI-resin treated clays; unstable, reverting to II).

b. Pree H2SO4 from FeS or S, giving extremely acid soils.

II,. Weak aids, soil 1IH 5 or 5.2 andOC below:









a. -Aluminohydonium cations: Al ( 0 '5 )6 + + 2
H + f.Al(OH)(-2 )] C (KC1 exchangeable; exchangeable protons of very acid soils). b. Possibly some humus carboxyl.

III. Very weak acids, soil pH 5.2 to 6.5 or 7:

a. Humus carboxyl, e.g., in surface soils.

b. Polyaluminchydronium edge Oh5 , e.g., in acid subsoils.

c. H2CO 3; basic aluminum sulfate.

IV. Very very weak-acids, soil pH 6.5 or 7 to 9.5:

a. Humus phenolic.

b. Polyaluminohydronium edge pairs, OH.5+.

OH0 5c. Ca(HCO3)2, ;aHOco3

SV. Extremely weak acids, soil piH above 9.5:

* a. Humus alcoholic hydroxyl.

b. Silicic acid -O1H.

c. Gibbsitic -OH (aluiinate reaction).


Measurement of Soil Acidic Properties Types of Measurable Acidity

Exchangeable acidity is that fraction of'acidity exchanged by neutral, unbuffered salts such as KC1, CaC12, or NaC1. Titratable acidity is the quantity of neutralized acid obtained at a given pH (usually 8.2) (33). Thus, the acidity measured with BaC12_ triethanolamine at pH 8.2 corresponds to titratable acidity, but is sometimes used as a measure of total acidity (106). The latter is also measured by





9


tiUlation with strong bases sch as Ea(CH)2 a-d Na0H (43). Coleman and Thomas (33) stated that the exchangeable acidity, as a proportion of the total acidity, varies with the nature of the soil and the degree of base saturation.

Residual acidity is known as that portion of acidity which is titratable but not exchangeable with neutral, unbuffered salts. It is measured by titrating the soil after extraction with 1N KC1, and it can also be determined by substracting exchange acidity from total acidity. Large amounts of residual acidity are common in soils high in Al and Fe oxides and organic matter (33, 43). Titration Analysis

Potentiometric and ccnductometric titration analysis have

been very useful tools in measuring the components and the quantities of acidity (27, 33, 43, 165). Low (89) clearly showed that potentiometric and conductometric titrations differentiate and measure the influence of H and A]. in acid bentonite. Yuan (159) corroborated Low's findings in true solutions and soil systems. Coleman and Harward (31) showed by potentiometric titration that H saturated montmorillonite behaves like a strong mineral acid. They also attributed the apparent weakacid character of clays to the presence of adsorbed Al. C6ulter (37) using titration techniques found that H-saturated clays became Alsaturated clays with time. Dewan and Rich (43) were among the first to titrate soil systems extensively. They used a(OH)2 and NaOH to measure total acidity, and Na2B407 and NaOAc to titrate A13+ + H+ and Bh, respectively. They could not determine exchangeable H in their samples, and most of the acid character was attributed to hydrolysis of IN KCI extractable Al and non-extractable Al hydrolyzing in place.






10


An initial inflection due to H can be observed in potentiometric titrations followed by a buffer region caused by Al hydrolysis (28, 95, 159). Coleman and Thomas (32) demonstrated that polymeric non-exchangeable Al and Fe hydroxy ions in addition to the exchangeable A13+ ion make large contributions to the buffering capacity of montmerillonite complexed with Al and Fe hydrous oxides in the pHI range 5 to 8. Soils relatively free of organic matter have representative buffer characteristics when dominated by monomeric exchangeable Al

(43). When Al- and I- saturated clays are compared, they also display the same buffer characteristics (89). A second buffer region is found between pH 7 and 8 and is due to dissociation of functional groups of organic matter (165). Soils with dominant fractions of amorphous materials including organic matter do not show sharp buffering regions sir5ilar to crystalline mineral soils or clays, and their buffering capacity to pHI 8 is relatively high compared to that of common mineral soils (142).

Conductometric titrations are of great value in distinguishing between exchangeable H and Al. Marshall (95) showed F. linear decrease in conductivity with increasing amounts of base similar to that of strong acids followed by an increase close to that of a weak acid in titration of mixtures of HC1 and AlC13 ip a pure system. The first inflection point was attributed to exchangeable H and the length of the linear middle section to exchangeable Al. When a H- saturated clay is titrated with a base, Coleman and Thomas (33) pointed out that the conductivity of the system decreases until complete neutralization of the H is obtained. This decrease in conductivity is due to the larger mobility of H ions than of cations of the added base. Conductivity rises rapidly as excess










01.H ions are added because of the high mobility of these ions. Alsaturated clays behave like weak acids; since they have very small amounts of free H ions present, conductivity increases as base is added due to salt formation (33). Dewan and Rich (43) demonstrated that conductometric titration curves for soil systems had the same characteristics as similar curves for pure solutions, and clay systems.


Acid Soils and Plant Growth

The harmful effects of soil acidity have been proven to be caused by secondary effects of the soil reaction, rather than to primary effects such as the activity of the H ion in the soil solution (2, 75, 128). The secondary effects of high acidity are deficiencies of Ca, Mg, P, and Mo, and excess of soluble Al, in, and perhaps other metallic ions. The relative effects of' these factors on plant growth depend on the scil itself as far as the available levels of deficient nutrients and on the susceptibility of the crop to deficiencies of the above mentioned nutrients or excesses of Al and Mn (122). Hydrogen lon Concentration.

Adsorption of many inorganic ions is significantly influenced by variations in H ion concentration; however, plants do grow successfully on acid soils provided the pH does not fall below 4.0-4.5, the nutrient supply is maintained at suitable levels, and the presence of Al and IMn does not reach toxic levels (75). Nevertheless, low pH (5.0) car have detrimental effects on plant growth, and under certain conditions such as lack of toxic levels of Al and Mn, the H ion concentration is responsible for poor plant growth (2, 75). Arnon et al.

(11) showed good growth of bermudagrass, lettuce, and tomato at pH

5.0-7.0 i.r. nutrient solutions. Audus (12) obtained an increase in root








gro ith up to p-. 5.0 for cress, radish, garden res, an. cornL. Hooward and Adams (6') found marked reductions in Snitial Alongation rate of primary roots of cotton below pHI 4.2. Calcium Deficiency

Calcium deficiency is not considered a major factor of acid soil infertility in some areas of the United States except on sandy soils (2, 75); however, acid soils with low CEC values may contain liiting Ca levels for many crops (2). Such is the case in highly leached tropical soils (80). Calciui deficiency in these soils is independent of soil acidity but is influenced by the levels of Ca and other cations; therefore, it is only in the absence of toxic levels of H, Al, or Mn ions that Ca saturation can be considered a good measure of the Ca-availability in the soil (2). However, the Ca requirements of certain crop ; such as peanuts, tom.-atoes, and celery are exceptions because of their inability to taic up Ca from soils that supply adequate Ca for most other crops (2). Magnesium Deficiency

Magnesium deficiency occurs not only on oacid soils subject to leaching, but on calcarcous soils. D-ficient M3 is found in acid soils having low C ,C, high lecchiin, excessive K and Ca, and in soils where crops with high iMg requirements re used (2). Deep ssndy soils are sub4Joct to greater Mg deficiencies than soils high in clay regardless of the p"i (2).

Phosphorus Deiciency

Phosphcras is probably the most deficient nutrient in acid soils (19). Crop establish ent and development in tropical acmd ccils is often inmpossitle without P -ao1iat o n . (16, 2, 118). 1son and





13


Engelstad (118) noted that there a negati.e co.elati n between intensity of weathering and total P. They also exTlained that the rate of P I:ost from tropical sc:il.a is determined by cropping intensity and erosion with slight leaching losses except on very sandy soils. Fertilizer and manure were mentioned as primary Inputs.

The picture of available P in trpMical acid soils becomes

even gloomier due to the high fixation capacity of certain soils such as Andeps. Several tons of P/ha may be needed to satisfy the total fixing capacity (51).

Aluminum Toxicity

Probably the first researcher to suggest the toxic effect of Al on crops was Miyake (111). He demonstrated that 1.2 ppm of Al in solution iwas tczxc to rice and mentioned the possibility of a relatice1lship between Al and the infertility of acid soils. Since then a large number of papers dealing with various aspects of Al toxicity and acid soils nave appeared in the literature. Perhaps one of the most significant papers on the subject was written by Kamprath (77). In this paper he showed the relationship between exchangeable Al and lime. His study led to the recommendation of lime for acid soils based on the exchangeable Al content. He found that when the Al saturation was below 151, the growth and yield of several crops -were maximized. Similar results were obtained by Reeve and Sumner (126) in South Africa. They found that iirne sufficient to reduce exchangeable Al] to 0.2 meq/l00g gave maxi uram growth of Sorghum sudanese Stapf.

The relationship between exchangeale Al and Al in the soll

solution in mineral soils was studied by Evans and Kamprath (47). Tiiey showed that soil solutions contained less than 3 ppm ntil Al saturation








a' ighr than y. . A ~ftcr tis ,utiuratlon 1. as obtianeo, Al in the soil solution ijcrcased sharply to 4.5 ppm at .% saturation. Corn did not respond to lime until Al saturation was 70 or about 3.6 ppm Al in the soil solution; how.evcr, soybeans responded ,ith only 30 saturaton or about 1.3 ppm, Al in the soil ,solution.

Plant species vary widely in their tolerance to Al. Foy

and Brcown (54) found rustard, turnips, barley, and cotton to be sensitive to Al while buckwheat, corn, and soybeans were tolerant.

Adams and Pearson (2) concluded that Al toxicity is largely determJntid by the chemical activity of Al in soil solutions in situ, regard.less of soil type.

Manganze se Toxi city

Adams and. Pearson (2) stated tha, t the amount of easily

reducible Mn. in the soi .ll l emte:mine the potential for tO:xicity to take place and that minimum. levels of 50 to 100 ppii are necessary for !Mn toxricity to occur, however, nutrient solution experiments have shown Mn toxicity symptoms tc occur in tobp:cco at 15 !7mm ':.i (22), in cotton at 10 ppm (3), and in several legu-:.es at less than 1C pm (113).

Plants take up Mn predominantly in the dvaltint state. In

neutral or slightly acid soils only a small fraction is in the divai.,nt state. This fraction becomes lar,er as the3 p of the soil cecrea;es. The maximum soil pI at which Mn can be Ioxic seem: to bc 5.5 (2). Molybdenium Deficiency

A pH of 6.5 is required for maximuiw; availability of soil Mo.

However, at this pH the availa:.bilities of' some of the other micronutrients may limit plant growth more than Mo availability except for 'he growtnr of' leumne. Treating seeds with Mo before plintin in soils with :;w Mo content ;:.y be moe feasible than lImin 4 (iO).








Neutralizatio ;of :Soll ;idersty

The most common method of neutralizig soil acidity is by

the addition of lime either as calcitic limestone (CaCO.3) or dolomitic 3
limestoac (Ca!'.g(CO )2). Calcium s.licate (CW3iO ) is also becoming a very popular living, material.

The reactions ttiking p Lace when lime is added to acid soils can be divided into: (1) dissolution reactions, (2) cation exchange reaction. ad (3) neutralizaticn reaci.Lons.

(1) Dissolution reactions are the hydrolysis of lime in

the soil and according to Seatz and Peterson (134) can be written as:



Ca(HCO3) -- Ca + 2HCOThe solubility of the limestone in the soil is dependent on the partial pressure of C02; the greater the partial pressure of CO2

in the system, the more soluble thie limestone. Dissolution reactions also depend, to a large extent, on the moisture content of the soil and soil temperature. As moisture increases, the air present in the soil is reduced and the concentration of CO2 in the soil air increases resulting in increased limestone dissolution. Water is also a reactant in the first step of the dissolution reaction. Limestone reacts faster at h:igh temperatures; this effect is probably related to diffusion rates of the end products away from the reaction sites (134).

(2) Cation exchange reactions take place when the Ca +ions produced in dissolution reactions replace exchangeable Al3 or H+, if present, held on the exchange complex: KAl + 3/2 Ca + 6.0 i XCa + Al .61 0

2 3




16

3+
(X resents the exchange comply ex in the rejection). The Al .6H20 dis+
placed undergoes hydrolysis with the production of H 30 in solution until it precipitates as Al(0) .3H 0: 2+
[Al.60H2 20 ; [lo).50K2 H3o [Al(OH).50HA + H20 -- [Al(oH) .40I ++ H30+

[Al(OH)2.40H21 + H20 Q� Al(OH)3 .302 + H3+

(3) Neutralization reactions are the last step in dissolution reactions when the end products of the latter react with the H 0+ ions in solution:

H20 + HCO H2CO3 + OH
H3o + + OH" 23 22
The rate of these reactions is directly dependent on the removal of OH- ions from solution. When the H0+ concentration in solution is lowered, the dissolution of lime is reduced and the overall process decreases (33).

Since .the concentration of H O+ ions in solution in acid soils depends on the hydrolysis rate of hydroxy-Al or hydroxy-Fe, lime reactions are also dependent on factors that influence the rate

and extent of hydrolysis reactions (33). Coleman et al. (35) observed an increase in hydrolysis of A13+ and hydroxy-Al with salts and dissolution reactions of lime.

Coleman and Thomas (33) mentioned that a decrease in dissolution reactions of lime occurs when intermediate compounds of Alhydrolysis are strongly held by the exchange complex of the soil reducing the rate of hydrolysis.

Reactions of limestone are also affected by fineness of grinding,

uniformity of mixing with the soil, and the-water content of the soil (33).








Soil Al-minu zno Ircn ra'ti', s

The importance of the diverse for: s of Al and Fe in soil acidity, P sorption, and other so-il chemical properties justifies. their classification and interpretation. Aluminum and Fe become even more important in tropical soils, rich in organic and inorganic forms, since they may be responsible for the majority of the chemical properties of these soils.

The study of the Al and Fe fractions in soils is difficult

in view of the fact that the extractants used lack specificity; however, these extractants give an approximate estimate of the nature and amounts of the predominant Al and Fe fractions. Ammonium Oxalate Fraction

Acid a monium oxalate (oxalate) has been widely used as an extractant for amorphous compounds of Al and Fe. Saunders (131) indicated that oxalate attacks amorphous forms of Al and Fe and very small amounts of crystalline oxides and clay minerals. McKeague (97) extracted Al arnd Fe from some amorphous inorganic substances as well as from soil horizons rich in organic matter complexes of Al and Fe. Hle also mentioned that crystalline Fe oxides were not destroyed. McKeague and Day (98) used oxalate to distinguish spodic horizons from other horizons rich in Fe. Baril and Bilton as quoted by McKeague, Brydon, and Miles (99) found that oxalate extracts considerable amounts of Fe from imagnetite which complicates the fractionation scheme in soils rich in this mineral-.

Citrate Dithioni-te Bicarbonate Fraction

Aguilera and Jackson (4) introduced a citrate-dithionitebicarbonate (CDB) solution as an extractant of free crystalline Al and










Fe oxides for soi's being prepared for mineralogical analysis. They found that CDB solubilizes large amounts of Al from hydroxy Al interlayers of vermiculitic chlorite; however, gibbsite was not attacked. Jackson (72) stated toat CDB extracts Fe oxides without removing iron aluminosilicate minerals, aluminum minerals, gibbsite, amorphous iron-aluminosilicates, :magnetite, and ilmenite. Mehra and Jackson (107) modified the original procedure and their modification is used a great deal in soil analysis today.

Sodium Pyrophosphate Fraction

Aleksandrova (8) extracted humus and its Al and Fe complex salts from soils using 0.112 Na4P207 (Pyrophosphate). Bascomb (15) found that K-pyrophosphate extracts organic Fe and amorphous (gel) hydrous oxides but not amorphous (aged) hydrous oxides. McKeague (97)

evaluated the pyrophosphate procedure. He snowed that some of the Al-, Fe-organic matter products were extracted, and that pyrophosphate dissolves a small proportion of the free Fe, but does not dissolve the inorganic amorphous or crystalline Fe and Al.

Sodium Hydroxide Fraction

Hashimoto and Jackson (60) demonstrated that hot 0.5N NaORf causes dissolution of free amorphous silica, free alumina, and large percentage of amorphous alu:minosilicates provided a high ratio of NaOH volume -o sample weight is used. Jackson (71) indicated that gibbsite is also dissolved by the same treatment. Jackson (74) explained that Al, Si, and amorphous aluminosilicates form soluble Na-silicates and Na-aluminates in NaOH solutions. hAtnonium Acetate Fraction

McLean, Ieddleson, . nd Post (104) found lN INH4OAc at pH 4.8








to be a better Al extractant th .n unbuffere s alt solutions or solutions buffered at pH 7.0 and above. They stated that the Al extracted with acetate (Acet.-Al) was primarily exchange ble. Pratt and Bair (123) working with acid soils extracted less Al from soils of lower pH, arnd more Al from soils of higher pH with acetate as compared to l.l unbuffered solutions of BaC1 2 or KC1. A freshly prepared Al hydroxide solid

was found highly soluble in acetate. McLean (100) explained that acetate removes exchangeable Al plus more soluble or reactive portions of the hydroxy-Al ions and Al polymers. Pionke and Corey (121) showed a high correlation between Acet.-Al and organic matter content in soils, and postulated that acetate is a good extractant for Al associated with organic matter. McLean and Owen (103) also found a close correlation between Acet.-Al and soil organic matter. Igue and Fuentes (70) released considerable amounts of non-exchangeable Al with acetate. Part of this Al was released from organically complexed forms.

Potassium Chloride Fraction

Coleman, Kamprath, and Weed (34) stated that Al extracted

with neutral salt solutions is trivalent, and, therefore, exchangeable. Thomas (140) mentioned that the Al extracted with neutral salt solutions was trivalent from pH 4.4 to 5.2. He found that little Al is extracted with neutral salt above pH 5.3. Lin and Coleman (86) displaced Al3+ ions in anounts close to exchange capacity from Al-saturated soils and clays with IN salt leaching. They concluded that 11' KC1 was the most effective displacing agent for rapid leaching, and that the Al present in the leaciate was primarily exchangeable Al (Ex.-A1). McLean (100) indicated tI t rapid leaching is useful in the procedure so that the




20


acidity formed does not dissolve other forms of Al than the exchar~ieztle. Nye et al. (116) studied the Al-K exchange equilibria and found that K was bound tighter than Al over the entire range of ion saturation in 1N solutions; K ions in IN solutions were able to exchange Alions very readily confirming the work by Lin and Coleman. Fractionation Scheme

Blume and Schwertmann (21) showed that an approximate distinction could be nade between amorphous forms of Fe and crystalline orxides by selective extractions of soils with oxalate and CDB, and that the ratio of oxalate extractable Fe (Ox.-Fe) to CDB extractable Fe (CDB-Fe) could be used as a relative measurement of the degree of aging or crystallinity of free Fe oxides. McKeague et al. (99) recognized the necessity of a more complete separation of the amorphous Al and Fe fractions and proposed the following scheme:

(1) Fe extracted with 0.1 M Na 4P207 (Pyro.-Fe) = Organic Fe.

(2) (Ox.-Fe) - (Pyro.-Fe) = Amorphous inorganic Fe.

(3) (CDB-Fe) - (Ox.-Fe) = Crystalline Fe oxides.

They mentioned that the scheme is less useful in distinguishing forms of Al in soils.

Effect of Lime on Soil Aluminum and Iron

The end products of the lime reaction are exchangeable Ca and Mg, Al(OH)3, and Fe (OH) . The pH increases to approximately 8.3 and complete base saturation is achieved. However, this is an ideal case since most soils are limed to reach pH values between 5.5 and 6.5. Under this situation, large amounts of titratable acidity and unreacted limestone remain. Aluminum and Fe compounds, along with nonionized groups on organic matter and clay, are the sources of this titratable acidity (33).








When lime is added to acid soil, h L:' first ions lost from the soil. are and monomeric exrhangeaole Al. Thus, lime causes a sharp decrease in exchangeable Al, and increaes- hydrolysis of-hydroxyAl and -Fe (33). Aluminum polymerization is also increased as hydroxyAl molecules bridge together to form stable structures (69).

The stability of organic matter complexes of Al and Fe has been shown to decrease as the p1 of the soil increases resulting in Al and Fe ions being released into solution. These ions are capable of entering hydrolysis reactions and polymerization processes (85, 133).

These changes in Al and Fe forms have been detected by a

few workers using chemical reagents. Bhumbla and McLean (17) observed a decrease in pH dependent acidity with lime. More Al was extracted with acetate than with IN KC1 from limed and unlimed soils; the differences between pH dependent acidity and extractable Al were less pronounced when acetate was used as extractant. McLean, Reicosky, and Lakshmanan (105) attributed changes in permanent charge CEC and pH dependent CEC caused by lime to inactivation of Al originally held by organic matter. Their conclusion was based on the high correlation between Acet.-Al. and organic matter. Velez and Blue (144) observed an increase in Ox.-Al with lime in a tropical and a temperate soil. A decrease in CDB-Fe took place in the tropical soil only. No changes were observed in Ox.-Fe and CDB-Al in either soil. They concluded that the amorphous fraction was the most respcrsive to lime additions.


Phosphorus Availability in Acid Soils

In general, available P is less than l of the total soil F at any given tine. It is now generally recognized that P recovery





22


by plants immediately after fertilizer application is usually in the range of 10 to 30p. The reminder which becomes available over long periods of time is precipitated by soluble cations, immobilized by soil micro-organisms, or retained by the soil complex. Phosphate fixation is the conversion of soluble P to a less soluble form, thus reducing its movement in the soil and availability to plants (91). Phosphorus Fixation

The availability of P in soils is controlled by P fixation reactions. -These reactions in acid soils can be placed into three general groups: double decomposition involving solubility-product relations, adsorption, and isomorphous replacement (81). Double decomposition reactions

These reactions are also called precipitation reactions because the end products in acid medium are insoluble P compounds

3 ++
of A13 and Fe3 . In the case of Al) , the aluminosilicates, free sesquioxides, and exchangeable Al may be regarded as the primary sources of Al, and the OH ion concentration dominates the solubility product relation. In all three cases, phosphorus .precipitation can be decreased by increasing the pH of the soil (81).

Russell and Low (130) showed that adsorbed Al or hydrous oxides of Al present on kaolinite exposed surfaces precipitated P. According to Hemwall (62), not much P was fixed as AlPO4 or FePO4. He identified hydroxyphosphates of the form (Al or Fe) (H20) (OH)2i2PO4. However, it has been assumed that P will form variscite-like compounds with Al (AlO4.2H20) and strengite-like compounds with Fe(FePO 4.2H20) with the assumption that Al3 activity was limited by the solubility of gibbite, and the Fe activity by thatof goette (87).
of gibbsite, and the Fe activity by that-of goethite (87).





23


Adsorption reactions

Two types of adsorption reactions are recognized: chemical adsorption and physical adsorption. Both types can be characterized by adsorption isotherms (81). Chemical adsorption can be divided into non-specific and specific adsorption. When P ions are retained as counter ions in the diffuse double layer of positively charged surfaces, the adsorption reaction is called non-specific (63). Specific adsorption takes place when the P enters into the coordination of metal oxides to replace another anion (63). Physical adsorption, ingeneral, is the interaction of forces exerted upon each other by molecules, atoms, or ions; therefore, sometimes it is difficult to make a clear difference between chemical and physical adsorption (138). Physical adsorption has no significance in P fixation according to Hsu (68).

Coleman (36) observed an increase in pH in a clay-phosphate system after P fixation had occurred. He concluded that the exchange of P ions in solution for the OH ions from clay minerals or the hydroxides of Fe and Al increased the pH of the system. Dean and Rubins (40) -illustrated that anion exchange occurs between certain OH ions of the clay minerals or hydrous oxides and P ions in solution. Kinjo, Pratt, and Page (83) showed that NO 3- ions are easily replaced by P ions in soils from Mexico, Brazil, and Colombia. Isomorphous replacement reactions

These reactions are a continuation of specific adsorption

reactions where the latter becomes isomorphous replacement of hydroxyls or silicate ions from the crystal lattice. A new mineral is formed through decomposition of the isomorphously transformed crystal lattice followed by recrystallization (81).








Stout (137) reported the formation of crystalline compounds

in reactions of kaolinite and h'alloysite with phosphates. He mentioned replacement of 01! ions by phosphates as an intermediate step in the reactions. Wada (149) indicated that the reaction of ammonium phosphates

with soil clays produced crystalline armmonium taranakite. Low and Black (90) released considerable Si in the reaction of kaolinite with P. They suggested that the Si released was the result of isomorphous replacement. K1u (68) pointed out that recrystallization of new compounds takes a long tie to go to completion.

In general, P fixation can take place in steps where more

than one type of reaction occurs. 1su (68) explained that P fixation takes place in two steps. The first step was considered as a fast adsorption reaction, and the second step as a slow decomposition-precipitation reaction. When amorphous Al hydroxides of intermediate size are present, precipitation and adsorption become indistinguishable (68). Effect of Aluminu and Iron on Phosphorus Availability

As was pointed out before, Al and Fe compounds are the raw materials for P fixation in acid soils. The importance of such compounds has been discussed in the literature. Saunders (131) showed that P retention was closely associated with Ox.-Al and -Fe, and CDB-Fe in New Zealand soils. Ahenkorah (5) reported significant relationships between P fixation and CD73-Fe in soils from Ghana. He commented that the interaction between pH and CDB-Fe was mainly responsible for the magnitude of fixation. No association with clay or Al was detected. Yuan and Breland (163) observed good correlation between P retention and Al in Florida soils. The best correlations were obtained from oxalate and CDB extractions. Shukla et al. (136) showed a close





25

correlation between P sorption by lake sediments and Ox.-Fe. Velez and blue (144) reduced P sorption by 5o after oxalate treatment of a soil from Costa Rica; but only 15 reduction was observed in an Ultisol from Florida. Phosphorus sorption vas reduced 50% after CDB treatment of both soils. Ballard and Fiskell (14) found acetate to be the most useful extractant of Al to predict P retention in coastal plain soiIs of the southeastern United States. Oxalate was the most valuable of the extractants for Fe.

It appears, from information in the literature, that amorphous forms of Al and Fe are more active in P sorption than crystalline forms, and that there is little distinction between Al and Fe in this regard.

However, the role of pH is very important in differentiating the forms active in P fixation. Hsu (68) noted that at low pHi (4.0) fixation is essentiqaly a precipitation reaction A1ith Al3 and Al hydroxides from crystal lattices; as the pH increases, amorphous Al hydroxide becomes more stable and phosphate is adsorbed at the surface. Fortunately, P adsorbed by amorphous Al and Fe compounds has been shown to be available for plant uptake (76).

Effect of Lime on Phosphorus Availability

Truog (141) emphasized the enhancing effect of lime on P availability in acid soils. HowEver, Velez and Blue (14) observed an increa.: in P retention after lim.ln a soil from Costa Rica. Downer (45) showed that P retention was independent of lime in high P fixing soils from Guyana. fiortenstine (65) found that the relative high P-fixation capacity in a soil from Belize greatly increased as the soil p H increased from limin. Woodruf'f and Kamprath (156) reported a decrease in the P adsorption maxia of' five Ultisols from





26



North Carolina with lime as a result of reductions in exchangeable Al; however, Blue (19) explained that the reduction in P adsorption maxima was not uniforx-i per meq of exchangeable Al neutralized, and that substantial P sorption capacity remained after neutralization of Al. Reeve and Summer (125) failed to obtain a relationship between liming and P desorption isotherms in Oxisols from South Africa. Pratt et al. (124) pointed out that, in general, high Fe oxide and less crystalline soils retained P the most. It seems clear that lime will only decrease P fixation when precipitation is the main reaction taking place and exchangeable Al, crystalline aluminosilicates, and free sesquioxides are the main sources of P fixing

compounds as suggested by Hsu (68).

Studies have shown that increasing pH to 7.0 and above will reduce plant growth and cause P deficiencies. Pierre and Browning (120) found that very high P levels were required at pH 7.0 to increase plant growth. Velez and Blue (145) increased yields of Pangola digitgrass on a heavily limed soil from Costa Rica only after additions of extremely high P rates. Fox, DeDatta, and Wang (51) observed an increase in F uptake by liming an Cxisol from Hawaii, high in Al and Fe, to pH 6.1. Liming to pH 7.0 markedly decreased P uptake by sorghum and desmodium. This decrease was attributed to the increase in available Ca which probably resulted in precipitation of P. Kamprath

(79) suggested that P deficiency is induced in soils with high P fixation capacity that are limed to pH 7.0 and above because of formation of insoluble calcium phosphates. More efficient P uptake has been obtained in Ultisols and Oxisols with lime rates high enough to neutralize









exchange>able Al (51, 125, 156). However, higher lime rates would be required to neutralize active Al in other soils such as Inceptisols even though exchangeable Al may be low in some cases (146, 160). Effect of Calcium Silicate on Phosphorus Availability

Calcium silicate has been shown to be very beneficial to

plants in highly weathered soils (49). Suehisa, Younge, and Sherman (139), and Monteith and Sherman (112) increased sudangrass yield and P uptake on a highly weathered Hawaiian soil rich in Fe with the addition of CaSi03. The beneficial effects of calcium silicate were attributed to improved P nutrition and decreased Al toxicity. Fox et al. (53) in Hawaii, observed an increase in sugar yields from sugar cane using Cal3iO- slag. Increasing rates of CaSiOj on a P treated soil gave increasing sugar yields, but little benefit was obtained by P applications alone. Large amounts of P and lime did not eliminate leaf freckle whereas CaSiO3 did. Lucas and Blue (93) obtained immediate but temporary beneficial effects from CaSiO 3. Calcium silicate, rice hulls, and dry Stylosanthes humilis herbage increased forage yields at the first harvest only. Forage P was increased by all treatments.

Yuan (162) mentioned that even thougn Si is required by

plants, mainly in the grass family, other factors are also involved in the responses to additions of CaSiO, to soils. He pointed out that CaSiO3 reduces soil acidity and Al toxicity, and that P is more available due to saturation of P-fixing sites with SiO=. Laws (84) reported that treating soils with Si decreased their capacity to adsorb P from solution. The quantity of soil P extracted by different solutions increased with the amount of Si applied. Deb and Datta (41) showed that silicate and organic salts such as citrate and tartrate markedly reduced P retention in soils.






28



The Use of the Langmuir Isotherm in Describing Phosphorus Avai lability

Younger and Plucknett (158), 'based on results of a 6 year experiment, indicated that P fixation and low yields from Hawaiian Latosols could be overcome by a heavy initial P treatment to satisfy the fixation capacity. Extra P could be used to maximize yield. They stated that additional P was needed every few years to replace P lost

by crop removal, erosion, and slow residual fixation. Fox, Hasan, and Jones (52) tried to answer the question of how rnch P to add in order to satisfy the fixation capacity of Hawaiian soils based on phosphate sorption isotherms.

Bache and Williams (13) defined a sorption isotherm as "a curve relating the amount of a substance sorbed at an interface to

its concentration at equilibrium in tne medium in contact with the interface. Constant temperature is essential because the relationship is temperature dependent." The Langmuir adsorption isotherm is widely used to describe the relationship between the P sorbed by a soil and the P concentration of the equilibrium solution. The isotherm may be treated according to the following equation called a Langmuir type equation:

C/V = (1/bVm) + (C/Vm),

where V=mg of P adsorbed per 10Og of soil, Vm=adsorption maxim-um (mg/100g of soil), C=equilibrium concentration in moles/i, and b=constant related to the bonding energy of the adsorbent for the adsorbate (156). A plot of C/V against C should give a straight line

of slope 1/Vmn, from which the adsorption maximum can be calculated, and with intercept .,bV:a, from which the constant related to the








bonding energy can be calculat-d once Vm is known (13). In practice, Vm gives the P fixing capacity of the soil and b is used as an index of P availability. The Iangmuir equation is based on the assumption that the energy of adsorption d6.os not vary with surface coverage. This assumption is correct only at P concentrations between 0.5x0 I4M and 5.0xlO- M (13); however Langmuir type equations have been used successflly to characterize linear P sorption over a wide ranLe of P concentrations (52).

Woodruff and Kamprath (156) calculated Vm for a number of soils and then applied P as Ca(H2PO,)2 at rates equivalent to 0, 1/8, 1/4, 1/2, 1, and 3/2 of Vm in a greenhouse experiment. Maximum grcwth of millet was obtained at P rates between 1/4 and 1,2 Vm except for a Norfolk soil were maxi~ yield took place at P rates equal to Vm. Soils with _igh Vm aber able tc supply sufficient P for growth at lower sat rations than soils with low Vm eien when limed. Fox and Kamprath (50) used P sorption curves to adjust the soil solution concentration between 0.01 and 1.A pp:n. in greenhouse studies. Phosphorus was applied as a solution of Ca(I2PO)2. Pearl millet growth approached 95- relative growth when P was 0.2 ppm in the soil solution. This soil solution concentration to obtain 95% relative growth was later called the external P concentration requirement by Fox (49). He stated that P sorption curves can be used to determine fertilizer requirements of highly weathered soils high in P fixing capacity if the external P concentration requirements of crops were known. He pointed out that soil test methods based on extraction of available P may overestimate the P status of soils with high fixing capacity, and, unless calibrated to account for the latter, will underestimate fertilizer requirements by a wide margin.









Electrochemical Prorers'ties of Acid Soils

Soils can be divided into two general groups based on the electrochemical behavior of their colloids: (1) Constant surface charge (CSC), and (2) Constant surface potential (CSP) (143). Constant surface charge soils are dominated by permanent charge colloids such as montmorillonite and vermiculite. Temperate soils belong in this group. Constant surface potential soils are most commonly referred as ph-dependent in terms of charge because of tie variation in CEC with pH (108, 143). Mekaru and Uehara (108) adopted the term CSP since charge variations are not limited to changes in H+ and OHconcentration in the soil, but also may be caused by adsorption of ions such as PO3, Si2-, SO2-, and various organic anions. Highly weathered acid soils of the tropics, rich in oxides and hydroxides of Al and Fe, and soils dominated by amorphous materials belong in this group (i08, 143).

In CSP soils, the potential is determined by potential

detennrmining ions (PDI) (82, 108, .43). Parks (119) defined a potential determining ion as any.ion capable of establishing the surface charge and the potential of a reversible oxide or hydroxide. Van Raij and Peech (143) considered H and OH as the predominant PDI in soils; however, Mekaru and Uehara (108) included P04, Si03, and S304 along with HI and OH as the most common PDI in tropical soils. The Isoelectric Point and the Zero Point of Charge

According to Parks (119), tae zero point of charge (ZPC)

is the pH at which the solid surface charge from all sources equals zero. Isoelectric point (IEP) is a ZPC arising from interactions of H+, Ot-, the solid, and water alone. In other words, the ZPC is a weighted





31


average of the IEF's of its comtionents,. At the IEP the density of positive charges equals the de~:l !.y of negative charges, and the net surface charge equals zero. The IEP equals the ZPC -only in the absence of adsorbed species different from PDI. Van Raij and Peech (143), and Keng and Uehara (82) defined the 2PC of a soil as the pH1 at which the net surface charge is zero. Thus, ZPC is a measurable property of soils having CSP. The soil components will determine the

ZPC of the soil, and in turn, the ZPC of the components will be the weighted average of their IEP's.

Soil components having CSP and, therefore, ZPC include

crystalline and amorphous oxides of Al, Fe, Ti, Mn, and Si, as well as kaolinite, halloysite, allophane, and mnost probably talc and pyrophyllite. Quartz, one of the major soil components, also belongs in this category (82). Van Raij and Peech (143) pointed out that the ZPC of permanent charge components of soils is vrery low or does not exist at all. However, most soils are considered to have a mixture of -CSC and CSP colloids, and the ZPC will reflect the dominant type of colloid present (82).

Measiurement of the Zero Point of Charge

Parks (119) observed that when a solid with finite cation exchange is immersed in a medium containing H+ ions, partial dissociation of the positive counter ion (e.g. Na+) and partial replacement of Na+ by H+ takes place. The extent of dissociation (size of charge) depends upon the Na+ ion concentration [Na+ and 1 ion concentration [I+l. Increasing [H'i at constant [Na+] results in further replacement of Na+ by H+ and fewer dissociations of It. At low [Nat], the

negative charge decreases with decreasing pH!. At hJ-h [Na+]; the





32

exchange of Na4 oy ' is redutc.Cd, and the charge is less pI1 sensitive. This observation confirmed the use of potentic.metric titration curves at different ionic strengths as a valuable tool in determining the ZFC of solids. Van Raij and Peech (143) used this technique for soils and faund that; the ZPC was the pH of the common point of intersection of the titration curves. Parks (119) called these curves H+adsorption isotherms and showed that they followed the Gouy - Chapman model for electrical double l.ayers. Van Raij and Peech (143) indicated that the Stern-model gave a better estimate of the overall charges in soil systems, and that the surface potential due to charges was determined by a Nernst type relation with the pH of the bulk solution and the pH of the ZTC. Keng and Uehara (82) showed an equation relating surface charge (CEC) to pH by combining the Gouy - Chapman equation for surface charge with the Nernst equation for -the surface potential given by Van Raij and Peech. Their equation describes the electrochemistry of many tropical soils when it is not a3ssuned that the surface charge is constant. In this case, salt concentration gives the value of the surface charge at a given pH, and the sign of the charge is obtained by the difference between the pH of the ZPC and the pH of the bulk solution. This difference also affects the amount of surface charge present.

As was mentioned above, the common point of intersection of potentiometric titration curves at different ionic strengths gives the ZPC of a soil. The same principle holds when surface charge is plotted against pH. The slope of the curves is a good measurement of the buffering capacity of the soil (82). The separation of individual curves, the change in pH with salt concentration, indicates the adsorption








capacity of the soil since the intensity -of charge i6 dependent on the ionic strength according to double layer theory (119). Wide separation between curves denotes strong- adsorption for the ions present in the supporting electrolyte solution, while curves that are close together imply low adsorption. The distance between the ZPC and the pH of the soil measured in the supporting electrolyte solution alone is highly correlated with the amount of permanent charge present in the soil (143). In general, these type of curves are very useful in describing some of the most important chemical properties of soils. Effect of Soil Mineralogy on the Zero Point of Charge

Van Raij and Peech (1h3) found that the presence of Al and Fe oxides will increase the ZPC of the soil toward higher pH values. Clay silicate minerals and organic matter will tend to decrease the. ZPC to lower pH values. In general they observed that the electrochemical behavior of the tropical soils studied was similar to that

of netallic oxides in which the surface potential of the reversible double layer is determined by the activity of PDI H+ an. OH- in solution.

Parks (119) indicated that increased crystallinity, dehydration, dehydroxylation, and structural charge of clay minerals decreased 7PC, while surface composition and cleavage habit increased ZPC. He mentioned that permanent charges on clay minerals are independent of pH until the latter exceeds a basic limit, after which the charge increases only if the adsorption capacity of the surface is small to allow for dissociation of cations and exchange reactions which are a measure of increasing charge as in oxides and hydroxides. Effect of Cation and Anion Adsorption on the Zero Point of CharGe

Keng and Uchara (62) observed a decrease in the ZPC of Hawailan








soils measured using CaCl2 'as supporting ~ c LElrolyte in comparison to NaC. The change in pH per 10 fold increase in electrolyte concentration was greater with Cer-l, than with N'aO1. They interpreted 2+
these results as being caused by specific adsorption of Ca2. Calcium ions were strongly preferred to H+ so that a decrease in pH was observed and changes in pH were greater as the CaC12 concentration decreased because of the effect of electrolyte concentration on the double layer thickness. Parks (119) found that cationic species increased ZPC. This is the case when there is no specific adsorption of nations taking place. Hydrogen can replace the cations present cr be adsorbed on the surface; an increase in pH is observed. He also pointed. out that an excess of specifically adsorbed ionic species will remove the pH dependence of the curve or change the ZPC to that of the species. This is common when IN CaC1l is used in measuring ZPC and Ca is speciically adsorbed. The potentiometric titration curve is shifted toward more acid pH's, and additions of acid or base cause small changes in pH.

Specific adsorption of S0e- was suggested by Keng and Uehara

(82) for the increase in ZPC_ of Hawaiian soils with the use of Na2SO as the supporting electrolyte when compared to the ZPC in NaCl solution. A greater change in pH per 10 fold increase in electrolyte concentration also tcok place. In this case, SOL were preferred by the soil over OH- causing an increase in p11 on the basic side of the curve. On the acid side, SO2 replaced coordinated OH ions and pH increased. Parks (119) reported that anionic impurities or pH independent PDI's decrease ZFC. This observation is explained by the fact that nonspecific adsorption of anions occurs on positively charged surfaces leaving an excess of H in solution, and, thus, decreasing pH on the





35


acid side of the titration curve. On the basic side, OH" reacts with dissociated IH and the pH remains less basic than predicted. Practical Implications of the Zero Point of Charge

Keng and Uehara (82) discussed the practical importance of the electrochemistry of tropical soils. They mentioned that the slopes of the curves were steeper with Ca salts than Na salts. This explains why lime is not very efficient in raising pH in some of these soils. The need to use the Stern double layer model suggested that most of the soil charge is found in the Stern layer, and only a small fraction of the total charge .can be used for retention of cations and for soil dispersion. This would explain why cations that are not specifically adsorbed are more likely to leach out, and why some oxidic soils disperse with great difficulty even at high pI's and have very good physical properties in the field. The suggestion was also made that Ca is held with very high affinity in the Stern layer and is not completely removed with a single neutral salt extraction or 1N NH4OAc at pH 7.0.

The small change in pH with lime added in large amounts to a soil low in CEC, and the small amount of Ca in readily exchangeable forms in these soils have sometimes been attributed to leaching. They can also be explained by high charge development and strong Ca retention.

Adsorption of SO4 ions was shown to decrease the ZPC of

Hawaiian soils, and, therefore, increase negative charge (63, 82, 108). This increase in negative charge avoids leaching losses of cations. Phosphate is even more efficient than-SO4 in increasing negative

charge (108). Mekaru and Uehara (108) showed that each mM of PO4/100g





36


sorbed incre sed etion retention by !about C.8 meq/100g. The greater response of plants to CaSiO3, can be explained in part by adsorption
o


Cation exchange capacity measuremenwrts by the usual procedures give unrealistic values for CSP soils due to changes in the net charge caused 6y concentrated salts and subsequent dilution by washings using v~ter and alcohol (82).

By measuring pH in IN KC1 and I2 0, and then taking the

difference, a good description of the net charge is obtained. When the pH in KC1 is greater than the pH in water, the soil lies on the positive side of the ZPC and it has net positive charge. When the pH, in water is greater thaL the p1i in KC1, the soil lies on the negative side of the ZPC and is negatively charged. When the pH in water and KCi are very close or equal, the soil lies very cloce to the 'PC and it has a very small net charge or none at all (143). Mekaru and Uehara (108) called the difference between pH in KCl and water delta pH (6pH1). They found that non-specific anion adsorption is very prominent in soils with a positive p.pH. They pointed out that pH measurements with 1N K2S04 sometimes give positive 8pH values because of increased exchange reactions of SO, with Oil ions. An indifferent electrolyte must be used when the main objective is to determine net surface charge.


Calcium Selectivity in Acid Soils

One of the most important steps in the overall lime reaction

in acid soils is the exchange of Al1+ by Ca' provided by- the liming material. It is through this reaction that the exchangeable acidity

is neutralized. Soils having low affinity for Ca' ions will require





37


larger amountss of lime for acid neutralization than soils with hign affinity. Leaching losses of Ca are much more prcncunced in soils

with poor Ca selectivity, and. most of the exchangeable acidity remains untouched until high lime rates are used. It has also been observed that when Ca2+ ions are present in the soil solution in large amounts due to overliming or low Ca selectivity, electrolytes associated and including Ca will accumulate, depressing plant growth temporarily (146). Therefore, it is of the most importance to study the capacity of soils to adsorb and hold Ca in the presence of other ions in solution and on the exchange complex.

Several models have been proposed to describe ion exchange

processes and predict ion distribution in soils. These models include the Freundlich and Langmuir adsorption isotheirs, the Donnan equilibrium, the diffuse double layer theory, and the law of mass action (42, 155). Deist and Talibudeen (42) felt that the law of mass action described equilibrium reactions in soils better than the other models since it takes into account all ions taking part in the exchange. Wiklander (155) justified the use of the mass action model in view of the knowledge regarding the structure of the diffuse double layer and its relationships to the bulk solution dependent on the activity of the counter ions and its connections with changes in free energy.

The constants obtained using mass action models are considered as selectivity coefficients rather than constants since they vary with the mole fraction of the ions in the equilibrium solution (55, 95, 155), the ratio of the adsorbed ions to the total exchange capacity, the ratio of the ions in the equilibrium solution to the total normality of the solution, and the normal concentration of the equilibrium





3d


solution (38). Selectivity cc;eficients, however, have been used to predict the replacement of a particular ion by another ion on the exchange complex, or to give an indication of the relative strength through which an ion is held on the exchange complex (38, 55, 95).

The model proposed by Gaines and Thomas (55) based on a rigorous thermodynamic treatment of the law of rass action has been used successfully not only to study cation selectivity but also to

investigate the variation of the thermodynamic parameters in soils. Nye et al. (116) found that A.L was preferred over K in dilute solutions (O.05N and below); however, K+ was more strongly bound to the soil in IN solutions. Clark and Turner (30) also showed that in i1I solutions K+ and Na+ were preferred over A13+. Deist and Talibudeen

(42) observed a preference of soils for Ca' over K +n 0.01N solutions. The CEC of the soils did not remain constant and decreased with K saturation. Rhoades (127) reported that soil vermiculites did not show strong preference for Mg+ over Ca waen compared to other vermiculite samples. Coulter and Talibudeen (39) found a strong pre+ +
ference for A13+ over Ca2 in 0.01N solutions and acid soils. Coulter

(38) reported that Y was more strongly adsorbed than Al3+ in 0.01K solutions. Some K was not exchanged by Al3 due to fixation mechanisms. Selectivity coefficients remained constant over the range of cation

saturation.

Theoretical

An exchange reaction can be described as mP + pXKm = mPXp + pM,

and the equation for the selectivity coefficient can be written as





39


(M) [PXp] K
- K' [1]
(p)m [Mmi~ r0ip "

(30), where P and M are cations with valences p and m respectively, X represents the exchange sites, C = mMXm + pPXp, and K' is the' selectivity coefficient. Parentheses and rectangular brackets express molar activities and concentrations relative to the total volume of suspension. Equation 11 can be transformed into the following equation using equivalent fractions instead of concentrations,

(1 - C/Co)P (q/qo)m [2
K' [2]
m ( .- p. m-pm
(C/Co)' (1 - q/jo)p C.o p

(38) where K' is the distribution coefficient; Ym and Yp are the activity coefficients of ions M and F in solution; Co is the normality of the eqaillbrium solution; qo is the total exchange capacity of the soil; (C/Co) is the equivalent fraction of ion M on the exchange complex. This equation is the equivalent fraction equation proposed by Gaines and Thomas (55) where K' is a function of Co and of the relation between (C/Co) and (q/qo). Taking negative logarithms, equation [2] is transformed into the following equation, (C/Co)m (q/qo)m pK' = log - log [3] (1 - C/Co) Cop-m yP (1 - q/qo) or

pK' S - Q [4

(38).

Conventional exchange isotherms give the graphical relationship between (C/Co) and (q/qo). This relationship shows ion preference when (q/qo) is greater than (C/Co). A derived isotherm is the graphical





40


relationship between Q and S. The S intercept gives pK' for the reaction. Derived isotherms are lines of unit slope if K' is constant, and in dilute solutions the intercept does not change with Co. Conventional isotherms have a marked dependence on Co (38).

The equivalent fraction equation has been used in soils

saturated with a given cation to study the replacing "ability of other cations. In natural soil systems, cation pairs can be studied for the same purpose by taking (1 - q/qc) as the equivalent fraction of the companion cation on the exchange complex, qo as the sum of all exchangeable cations present, and (1 - C/Co) as the equivalent fraction of the companion cation in solution. The companion cation is the one that will influence the selectivity of the soil for the cation of interest.

Effect of Lime oh Electrolyte
Accumulation in Acid Soils

The presence of electrolytes after lime applications to

acid soils has been well documented in the literature. Russell and Richards (129) reported leaching of considerable amounts of N0 from virgin soils. Midgley (109) observed a marked increase in water
2+
soluble Ca2 , HCO, and NO- ions by overliming acid soils. Walker and Brown (151) showed accumulation of NO- up to 550 ppm with lime rates
3
as high as 6t/acre. Even lime applied at 3t/acre gave NO- values close
3
to 550 ppm. Ogata and Caldwell (117) also detected high amounts of NO (500 ppm) after soils that had been limed with 16 t/acre were left fallow for 2 years. McLean (102) stated that the concentration of electrolytes in soils increases with dissolution of' lime. He added that in soils high in CEC, electrolytes disappear from solution as





41


CO2 volatilizes, b.ut in soils high in anion exchange capacity (AEC), the increase in OH concentration neutralizes positive charges forcing 2- 2
other anions such as SO into solution. These SO ions may couple with Ca ions to increase the amount of electrolytes in the soil
+
solution. Helyar and Anderson (61) found a large increase in CaL
2
and SO ions in the soil solution of limed soils from New Zealand.
3
However, PO a nd NO ions decreased in concentration with increasing
4 3
amounts of lime. Velez et al. (146) reported substaiitial increases in water soluble Ca 10NO3, and SO2 ions as the lime rates applied to a soil f'omn Costa Rica were increased. In general, electrolyte accumulation in acid soils occurred after high lime rates were applied. In a few cases, small amounts of lime caused a marked increase.

Zantua and Blue (164) mentioned accumulation of electrolytes as the possible cause for the depression by lime of Pangola digitgrass yields in a virgin soil from Costa Rica. Leaching both the unlimed and limed soil increased yields dramatically. The increase was greater in the limed soil. Almost identical results were obtained by Midgley (109) with different plant species when soils from Kansas were overlimed. He pointed out that new seedlings were especially susceptible to injury, and that coarse-textured, acid soils that had not been

cultivated for several years usually produced the greatest injury. He
+
observed that individual electrolytes (Ca2 , HCO3, and NO3) caused no injury, but failed to account for all possible electrolytes. The importance of the combined effect of individual electrolytes was demonstrated by Velez et al. (146). They found marked increases in electrical conductivity (EC) with lime; values harmful to plants were detected in a surface soil from Costa Rica that had been under pasture for a few years, after incubation.





42


Besides acclumlatJon of electrolytes, nutritional disturbances associated with the presence of excessive amounts of ions in the soil solution brought about by large applications of lime are often mentioned as the cause for reductions in yields. Pierre and Browning (120) rzpcrted disturbances in P nutrition of several crops as the major cause of excessive liming. Naftel (115) found ' deficiency in overlimed soils. Yields dccrcsc. sharply r:nen Ca saturations in the soil were higher thai 7 -. McLean (0C2) mentioned that high Ca, adsorbed or in solution, alters tne balance of cations such as micronutrients, K, and Mg. The breadown -of cation balance is reflected in the uptake of these nutrients by plants as was shown by Helyar and Anderson. Naftel (11.4) and Pierre and Browning (10) reported that decreases in available P due to excess Ca in solution causes severe P

deficiencies even in soils with high levels of easily extractable P. High levels of Ca2 associated with Cl" and NO~ in soil solutions were shown to be lethal to orchardgrass (150). According to Eaton (46) 2
Cl- salts are often associated with accumulation of SO , ECO3, and CO2~. Burning and firing of leaf tips and margins, bronzing, premature yellowing, abscission of leaves, and chlorosis are symptoms of excess Cl-. Zantua and Blue (164) observed some of these sympto;.s on Pangola digitgrass grown on a limed virgin soil from Costa Rica.
2
High concentrations of CO and HCO in thie soil may have caused
3 3
phytotoxicity. Iron chlorosis in many plants has been associated with HCO, concentration in soils (122).

The depressing effect of excessive lime on plant growth was

shown to be temporary in all cases (169, 115, 146, 164). Plants usually grew well after the first harvests, without a good explanation as





43


to why. Velez et al. (146) postulated that te temporary harmful effect of lime may be closely related to the persistence of high levels 2
of HCO ions in the soil and the amount of 3044 and NO0 ions released
3 4
by limed soils through organic master decomposition.

As was mentioned above, overliming -is responsible for the temporary harmful effect on plant growth. Acid soils of the tropics tend to be overlimed when recorimmendations are based on raising soil pH to 6.5 or even 7.0 as is the case for most temperate soils. This condition can be avoided if recommendations are made to use lime only to eliminatc- toxic substances or to provide enough Ca for normal plant growth (79, 102).

Indicator Plant for Greenhouse Studies

Pangola digitgrass (Digitaria decumbens Stent) had a consistent negative response to lime in soils from Costa Rica (91, 146, 164). The botanical characteristics of Pangola were described by Hodges et al.

(64). Adequate fertility is required by the plant for sustained growth (64, 148), and its K requirement is high; however, Gammon (56) showed that more than 60i of the K can be substituted by Na without growth reduction. The optimum pH for Pangola on acid flatwood soils of Florida according to.Hodges et al. (64) is 5.5; however, they pointed out, that vigorous growth occurs at low pH's (4.2 to 4.5). This is in line with reports-by Lotero, Monsalve, and Ramirez (88) in Colombia. They found Pangola to be resistant to soil acidity. Hortenstine and Blue (66) reported pH 6.3 as optimal on Puletan loamy sand. Blue et al. (20) obtained appreciable response to N fertilizer on newly cleared land'in Costa Rica after slow establishment of Pangola. Ahmad, Tulloch-Reid, and Davis (3) also obtained significant responses to N on soils from Trinidad.





44


Pangola was shown to be sensitive to Ca and P deficiencies (18, 64, 94, 145). However, lime has depress ~ed yields in tropical soils with soiae evidence that I availability was adversely affected as well. Blue (18), and Hortenstine and Blu: (66) found an indication that the response of Pangola to lime on a soil from British Honduras was dependent on applied P. A larger response to P than lime was documented. Downer (45) obtained yield depression from lime on a brown sand soil from Guyana. Lucas and Blue (94) observed that increasing lime levels decreased Pangola growth on an Entisol from Costa Rica regardless of the rate of P applied from 0 to 450 ppm. Phosphorus concentrations in oven-dry forage were below the critical level of

0.16% proposed by Andrew and Robins (10). Zantua and Blue (164) reduced yields by liming a virgin soil from Costa Rica. Vicente-Chandler (147) obtained positive responses to lime only 5 years after application in Puerto Rico. Figarella et al. (48) reported no yield response or effect on P concentration of Pangola with P rates from 0 to 400 kg/ha. Ahmad et al. (7) did not find a significant response in P content or yield on soils from Trinidad.

















MiTERIALS AND MCTELOD


So~l Samples

Soil samples from Costa Rjca and the Eastern savannas of Colombia were collected during the last week of' August and the first week of September 1972. Five sites were selected in Costa Rica and four in Colombia. Profile samples were taken at each site for soil characterization purposes. Surface samples (0-18 cm) were also collected at each site for laboratory and greenhouse studies. Surface and profile samples from Costa Rica were named after the closest city to the snapling site. Profile and surface samples from Colombia were named to give key locations around the Carimagua experimental station, where sampling was performed. Table 1 gives the soil identification number, its location, and its soil order according to the U.S.D.A. soil classification system.

Descriptions of the sampling areas, sampling sites, and soil profiles except for Los Diamantes were made by Rodriguez (M. Rodriguez-Gomez. 1974. Lime-micronutrient studies wit soils from Costa Rica and the eastern Llanos of Colombia. ih. D. Dissertation. University of Florida, Gainesville). Lucas (91) made the same descriptions for Los Diamantes soils. Hereafter, soils will be referred to by tae location of the sa:.ipling site as shown in Table 1.




45













Table 1. Identification number, iccation, and order of the soils
studied



-Identification Soil
number Location r- order Costa Rica

1 Los Diamantes (Surface) Inceptisol 2 Los Diamantes (Subsoil) Inceptisol

3 San Vito Ultisol 4 San Isidro Oxisol

5 Grecia Ultisol 6 Alajuela Ultisol


Colombia

7 Agronomy Area Oxisol 8 Cararao Ultisol 9 Drainagevay Oxisol 10 East Carimagua Ultisol








Laboratory ProcedureS

G renrl he cal1 b'ooerties

Soil reaction (p1)

Soils were equilibrated in IN KC1 and in water for 1 hour at a 1:1 acil to solution ratio. The pi was measured with a combined calomel-glass electrode attached to a Corning model 12 pH meter. Organic matter (OM)

The vet oxidation method. (Walkley-Black) described by

Allison (9) was used to measure easily oxidized C. A factor of 1.724 was utilized to convert C concentration to OM. Cation exchange capacity (CFC)

Cation exchange caaci~ tv was determined by the NHt saturation method described by Chapman (28). Measurements were made at p3 4.8 and 7.0. Water and 99 ethyl alcohol were used to wash excess sales
+ + +I.
after NI H saturation. Adsorbed NH vas displaced with .a .... an acidified 10 NaCl solution and determined through distillaticn ,w ith a Hoskins steam distillation apparatus (25). For this purpose, 30 mi of acidified solution containing the displaced N~i were neutralized with 20 iml of 50 , Na0IH solution, and N7i, was trapped in 10 ml of i BO3 containir. indicator.

Electrical conductivity (FC)

The saturation extract method for soluble salts explained by Bower and 'ilcc: (23) was used to measure EC. Extractable. soil nutrients

Ixcnanze:iable bases (Ca, 1 ., K, Na) were determined in the pH 4.1 m+ Ni -saturatin solution used for CEC :nasurceents after





48



leaching 250 ml. through soil samples. Aluminum, Fe, and P were also determined in this leachate.

Exchangeable Al was extracted as suggested by McLean (100).

Available P was extracted by the Bray iI method (24).

Water extractable NO -N and S04-S were assayed using a 1:5 soil to water ratio. The suspension was shaken for 30 minutes in a reciprocal sha:ser; a few drops of a saturated KC solution were added to aid in'flocculation. The suspension was then centrifuged (2,000 rpm for 5 minutes) and filtered through No. 42 Whatman filter paper. -Nitrate was measured c olorimetrically as suggested by Bremner

(26), and SO4 was measured gravime rically using the method outlined by Hanna (59).

Total eler.rtal anaiysiL

Total Nutrients (. Na, Ca, ;g, Fe, Al, and P) were extracted following tae HC10L-HF digestion method outlined by Jackson (71). Minersiogical Analysis

Sample preparation for mineralogical analysis was made

according. to Wthittig (153) and Jacsson (72). ualitative analysis of the clay minerals present was made by X-ray diffraction using a General Electric XRD-700 instrument with Ni-filtered CuKa radiation. Quantitative analyses for ,aolinite and gibbsite were made with a Per








The ig saturation method (72) was used to measure the CEC

of the clay fraction. The CEC was taken as the milliequivalents of Mg2 per lOOg of clay displaced with 1N NaNO. Soil Aluminum and Iron Fractions

Ariorphous Al and Fe fractions were determined by the amonium oxalate (pK 3.25) method (131) and the hot 0.5N NaOII method (60). The method proposed by Mehra and jackson (107) was followed to measure the free crystalline oxides of Al and Fe. Organic amorphous fractions of Al and Fe were assayed with the O.1M Na4P207 procedure described by. McKeague et al. (99). The fractionation scheme proposed by McKeague et al. (99) was followed to characterize amorphous inorganic, amorphous organic, and free crystalline oxide fractions of Al and Fe. Lime Requirement

The amount of lime needed to raise soil pH to 6.8 was determined by the SMP buffer (135), and by Yuan's buffer (161). The exchangeable Al method (77) was used to determine the amount of lime required to neutralize the exchangeable Al. Sources of Soil Acidity

Total acidity was measured by the BaC12-TEA method modified by Zelazny and Fiskell (165); however, the suspension was only equilibrated overnight instead of 2 days as suggested.

Potentiometric titrations were carried out following the procedure outlined by Zelazny and Fiskell (165). The procedure was modified by using different soil to lN KC1 ratios and different normalities of titrants (Bap'12 and Na2B407) in order to avoid excessively long titrations. Sample weight varied from 0.2g to 0lg,





50



and the volume of J, KiC1 was -ept constant. Titrant normalities varied from 0.02N to 0.2N. A il i1 soil extract was prepared by incubating 10g of soil in 50 ::l of IN KC1 for 4 days. This suspension was filtered into a 100-mil olumetric flask and the volume of the liquid was made to 100 ml by leaching the samples with 1N KC1 after the first filtrate was collected. This extract was titrated with 0.05N Ba(0H)2. All titrations were made with a Sargent Model D Recording Titrator with a Sargent combination electrode No. 3-30072-15.

Suspensions for conductometric titrations were prepared in the same manner as suspensions for potentiometric titrations. In this case only, 0.271 Ba,(OH),, was used as the titrant, and no 1N KC1 soil extracts were titrated. A Model 751 conductivity meter from Universal laterloc, Inc. was attached to the recorder of the Sargent titrator with an appropriate condue'tivity ce31 to measure conductivity. The cell had a constant equal to one. The conductivity meter was calibrated zo read 5 on the recorder chart at a range of 0.0C1 mho, and set at a range of 10 umho for samples. Tihe total chart expansion was 10 units. The recorder was run at 500 my chart span, and the rate was set to slow (1/3 incAh;minute).

Barium hydroxide titratable acidity was measured poteutiometrically at pTH 5.5, 7.0, and 8.0 wiile Na2B407 titratable acidity was measured only at pH 5.5 and 7.0. Conductometric titrations were allowed to proceed until the slope of the curve was equal to a deionized water blank titrated with 0.2N Ba(OH)2. The end point was taken as 2/3 the amount of titrant in meql00g required to reach the slope of the blank to account for Al(OH)- formation ()13).





51



All acidity measurements were correlated with soil chemical properties and soil Al and Fe fractions to study the degree of association among variables in order to make predictions of where this acidity was originating.

Phosphorus Adsorption

Phosphorus adsorption as determined by the method of hortenstine (65) was called an intensity measurement of adsorption. Phosphorus sorption was taken as the difference between the amount of P added in solution (50ppm) and the amount of P lost from solution (soil basis) after equilibration of a 1:50 soil to solution ratio. Phosphorus sorption was also expressed as the percent of the original amount of P added which was sorbed by the soil.

The quantity measurement of adsorption was the adsorption maximum and the constant related to the bonding energy obtained with a Langmuir type linear equation. The method discussed by Fox and Kamprath (50) was followed to determine data points, and linear regression analysis was used to find values for the intercept and the slope of the regression equation. Added P ranged from 0 to 900 ppm in solution.

The methods were divided into intensity and quantity to differentiate the amounts of P adsorbed at different equilibrium times. Only 16 hours of equilibrium were allowed with the intensity method, while 9 days were allowed with the quantity method.

Correlation analyses were utilized to study the degree of association between P adsorption, soil properties, and Al and Fe fractions in order to estimate the origin of the P sorption capacity.










Amounts of lime needed to raise soil pH to 5.5 were added

to 500g of soil. Limed samples were incubated for 10 weeks in plastic bags at moisture levels close to the inaximum water holding capacity but low enough to avoid s~t araticn. This moisture level is also known as field capacity, although it was not calculated experimentally. After incubation, P adsorption ias determined by the quantity method. Electrochemical Properties

The sign of the net surface charge was obtained as described by Mekaru.and Uehara (108) in 1:1 soil to solution ratios. The ZP' was determined by-. potentiometric titrations following the method outlined by Van Raij and Peech (143). The supporting electrolyte was CaC12 at concentrations of 1.0N, 0.11, 0.01W, and 0.001N. The acid side of the titration -curves was evaluated by adding 10, 5, 2, 0.5, and 0.1 ml of 0.1N HC1. The basic side was assayed by adding 5, 2, 0.5, and 0.1 ml of 0.1N NaOH. Measurements were also taken in the presence of the electrolyte alone. San Vito soil required 15 ml of acid plus the other treatments for a better evaluation of its ZPC.

A combined calomel-glass electrode attached to a Corning model 12 p1H meter was utilized for all pH measurements. Calcium Selectivity

The procedure described by Clark and Turner (30) was used to develop conventional adsorption isotherms for Ca, and the equation reported by Coulter (38) was utilized to calculate the Ca selectivity coefficients. This equation represents a straight line of unit slope when the selectivity coefficient remains constant throughout the range of cation saturation; however, this was not the case for the soils










analyzed, and the Ca selectivity coefficient was calculated forcing the slope of the line to unity. Tis coeff'icint was the mean of all coefficients over the entire ran-ne of Ca saturation for a given system.

Absorption isotherm, and Ca selectivity coefficients were evaluated for equilibrium solution concentrations of 1.0 and 0.001N. Calcium-K, Ca-Mg, and Ca-Al systems were studied. The Ca saturation of the equilibrium solution ranged from 0.1 to 0.9, and the companion cation saturation from 0.9 to 0.1. The sum of Ca equivalents and companioTi cation equivalents was equal to the concentration of the equilibrium solution. The Ca saturation was the equivalent fraction of Ca in solution. The results obtained. with the O.;001N equilibrium solution set were discarded due to failure of the procedure to detect small changes in Ca adsorption since three extractions with IN NH NO were made instead of the five suggested by Clark and Turner (30). Incubation Studies with Lime

Reagent grade, powdered CaCO was applied to 500g of soil

at rates of 0, 1.5, 3.0, 4.5, 6.0, 7.5, 9.0, 10.5, 32.0, and 13.5 meq100.Cg.

Limed samples were incubated for 10 weeks in plastic bas at moisture

levels close to the maximum water holding capacity but low enough to avoid saturation.

After incubation, samples were allowed to air-dry and saturation extracts were obtained as described by Bower and Wilcox (23). The following measurements were made in the saturation extracts: K, Na, Ca, Mg, NO 3-N, and EC.

Exchangeable Al, P sorption (intensity method), pH, and

OM were also evaluated on this set -f samples by the procedures discussed previously.





54



Analytical Determinations

Atomic absorption spectroscopy was uaed to analyze Ca; Mg, Al, and Fe in the different soil extracts described. Potassium and Na were determined by flame emission spectroscopy. Nitrate-N was assayed with a NO3 specific ion electrode attached to an Orion specific ion analyzer unless stated otherwise. Phosphorus was measured by the ascorbic acid method of Watanabe and Olsen (152), and S04-S by Hanna's gravimetric procedure (59).


Greenhouse Procedures

General Preparation of Experiments

Reagent grade, powdered liming materials (either CaCO3 or CaSi03) were mixed with soils in a rotary mixer in sufficient amounts to give desired lime rates. Moisture in the limed soils was adjusted close to the maximum water holding capacity but low enough to avoid saturation. Limed, moist samples were incubated for 6 weeks in the greenhouse. After incubation, samples were air-dried and mixed with other fertilizer materials in a twin-shelled blender to give experimental rates of lime and nutrients required in the experimental designs. One kilogram of soil with the proper treatment was placed in a plastic pot previously filled with coarse gravel to about 2/3 its volume. Pangola digitgrass stolons were collected and placed in distilled water to encourage new root formation. Three grass shoots with recently initiated growth were planted per pot. When new growth was observed, the soils were fertilized with solutions of chosen nutrients other than those involved in the experiment. The amount of water required to





55



keep the pots at a suitable moisture regine for plant growth was calculated by weighing a simple pot from each soil before and after adding water in quantities large enough to avoid saturation and leaching. Plants were cut 3cm above the soil surface at 4-week intervals for a total of three harvests. Nutrient solutions were added to all pots after each harvest.

The fresh plant material for each harvest was placed in a drying room at '70C for 3 days; afterwards, it was weighed and ground in a 20-mesh stainless steel Wiley mill. One gra& of the ground sample from each experimental treatment was ashed at 500C over night. The ash was moistened with a few drops of deionized water and 15ml of 5N HCl were adaed to the samples. This suspension was brought to dryness on a hot plate. The residue received 20ml of deionized. water and

2.25ml of 5N HCl; then it was heated on a hot plate to boiling, and filtered into a 50-ml volumetric flask. The solution was brought to volume with deionized water.

Phosphorus in plant extracts was measured colorimetrically by the ascorbic acid method. Calcium and Mg were determined by atomic absorption spectroscopy, and K and Na by flame emission spectroscopy. Experiment No. 1

A second order, central composite, rotatable response surface design was chosen for this experiment in order to study the effect of independent variables lime, Mg, and P on pangola digitgrass yield and nutrient contents, and to predict the rate of these variables needed for maximum response. The design included 8 factorial points (two levels





56



Table 2. Relation of treatment number to experimental and coded
variables for Experi1ent No. 1



Varables

Experimental Coded
Treatment
number CaC3 iM P CaCO3 P meq/100g------ ppm----1 2.4 48 122 -1 -1 -1 2 9.6 48 122 1 -1 -1 3 2.4 192 122 -1 1 -1 4 9.6 192 122 1 1 -1 5 2.4 48 479 -1 -1 1 6 9.6 48 479 -1 1 7 2.4 192 479 -1 1 1 8 9.6 192 479 1 1 1 9 0.0 120 300 -1.68 0 0 10 12.0 120 300 1.68 0 0 11 6.o 0 300 0 -1.68 0 12 6.0 240 300 0 1.68 0 13 -6.0 120 0 0 0 -1.68 14 6.0 120 600 0 0 1.68 15 6.0 120 300 0 0 0 16 6.0 120 300 0 0 0 17 6.0 120 300 0 0 0 18 6.0 120 300 0 0 -0









and three variables) located at th vertice6s of a cube, 6 points spaced at disttuances t1.682 from the origin aloEn the three axes of the cube in order to make the design rotatable, and a point at the origin replicated 4 times (58). The experimental design was completely randomized.

Lime as CaC03, MgSO , and finely ground ordinary triple

super phosphate were the fertilizer materials. Lime treatments were selected to explore the experimental region between 0 and 12 meq/100g; the level of lime estimated to give maximum response (6 meql0g) was a mean of the lime required to bring p H to 6.4 in all soils calculated with the SMP buffer method. The remaining lime treatments followed design requirements. Magnesium levels were chosen to give a 10:1 Ca to Mg ratio based on the amount of Ca present in the lime treatments selected. A region between 0 and 240 ppm of Mg was studied. Phosphorus levels were adjusted to explore a region between 0 and 600 ppm; the level of P estimated to give maximum response (300 ppm) was based on the

optimun level for pangola yield found by Lucas (91) in soils from Costa Rica. The relationships between experimental and coded variables is shown on Table 2. Multiple regression analysis was used to examine the response, and interpretations were made based on the regression coefficients according to Hader et al. (58).

L.1 pots were fertilized after each harvest with solutions

containing 100 ppm of N and K. Micronutrients were added using 30 ppm of Frit 503 at the beginning of the experiment. Expcri ent Jo. 2

The experimental design used was a split plot factorial

arrangement with 2 lime levels as main plots, 5 F levels as subplots,





58



and 2 reclications. Lime as a800 and finely tg:round triple super phosphate were the fertilizer sources. Lime levels that gave maximum response in Experiment No. 1 were selected. Phosphorus levels were adjusted according to the P adsorption maximum calculated with the quantity method in order to obtain 0, 1/16, 14, 1, and 1 1/4 saturation of this maximum. The purpose of this experiment was to study the potential of the quantity method to predict fertilizer P needs of Pangola in unlimed and limed soils, and to examine the value of Experiment No. 1 to predict lime rates. The amounts of lime and P used appear in Table 3.

All pots were fertilized after each harvest with solutions containing 100 ppm of N, K, and ig.

Experiment No. 3

A completely randomized block design was utilized to compare CaCO3 and CaSiO as lime sources in soils containing adequate P levels
3
for Pangola growth. A level of lime that gave good response in Experiment No. 1 was compared to a level that gave poor response. Phosphorus levels that produced maximum response in Experiment No. 1 were applied as finely ground ordinary triple super phosphate. The primary objective of this experiment was to study the behavior of CaSiO as a lime source
3
and its value in increasing the availability of P applied. The amounts of CaCO 3, CaSi03, and P added in this experiment are shown in Table 4. All pots were fertilized after each harvest with solutions containing 100 ppm of' N, K, and Mg.




59


Table 3. Lime and phosphor-us treatments �for .xperimIent No. 2



Treatments

Soil CaCO3 P


meq/100g ppm

Los Diamantes (Surface) 0 0 10 420 1675 6700
8375

Los Diamantes (Subsoil) 0 0 10 390
156o 6250
7810

San Vito 0 0 7 420 1675 6700
8375

San Isidro 0 0 4 370 1475 5900
7375

Grecia 0 0 4 48o 1925 7700
9625

Alajuela 0 0
18 480 1925 7700
9625

Agronomy Area 0 0 6 137 550 2200 2750





60



Table 3 - Continued



Treatments

Soil CaCO3


meo l 00g ppm

Ce-arao 0 0 4 187 750
3000
3750

Drainageway 7 0 500 2000
800o 10000

East Carimagua 0 0 20 52 207

1037
Sr3





61


Table 4. Lime, calcium silicate, and phosphorus treatments for
Experiment No. 3



Treatments

Soil CaCO3 CaSiO3 P


----------meq'lO1Og------------ ppm

Los Diarmantes (Surface) 2.0 2.0 450 10.0 10.0 Los Diamantes (Subsoil) 1.0 1.0 500
4.0 4.0 San Vito 1.0 1.0 600
5.0 5.0 San Isidro 0.5 0.5 650 10.0 10.0 Grecia 2.0 2.0 500 24.0 24.0 Alajuela 1.0 1.0 650
8.0 8.0 Agronomy Area 1.0 1.0 605
2.5 2.5 Cararao 3.0 3.0 450
16.o 16.0 Drainageway 10.0 10.0 420 24.0 24.0 East Carimagua 1.0 1.0 625
8.0 8.0





62









RESULTS AND DISCUSSION


Laboratory Experiments

General Chemical Properties

Soils from Costa Rica had pH values higher than 5.0 in a 1:1 soil to water ratio (Table 5). Samples from Los Diamantes gave the highest pH values. The measurements ranged from pH 5.1 to 5.9. Lower water pH's were determined in soils from Colombia with a range between

4.2 and 4.9 pH values in 111 KC1 were lower in all cases than those measured in water suggesting the presence of exchangeable sources of acidity such as Al as mentioned by Thomas (1;0). Using the pH in water and according to Jackson (73), the soils from Costa Rica can be classified as very weak acids except for San Vito which fits the category of a weak acid along with all the soils from Colombia with the exception of Cararao that can be classified as a strong acid.

The OM contents of the Costa Rican soils are relatively high, (Table 5); however, Los Diamantes subsoil and San Isidro were not as high as the others. A lower OM content was measured in the soils from Colombia except for Drainageway which contained more than any other soil. Organic matter was probably the major active material in the majority of the soils fro.i Costa Rica and in the Draina,,eway soil from Colombia.

Cation exchange capacities were high for the Costa Rican

soils and the Drainageway soil from Colombia (Table 5). All soils






Table 5- Chemical properties of the soils studied



pH* CEC

Soil H20 KC1 OM pH 4.8 pH 7.0 EC

---------1:1 ---------- ----------meq/100g ---------- mmhos/ cm

Los Diamantes
Surface 5.6 4.8 8.8 10.9 29.0 0.34 Subsoil 5.9 5.2 4.7 10.1 23.1 0.21 San Vito 5.2 4.2 10.0 19.0 37.6 0.32 San Isidro 5.1 4.3 6.3 12.3 23.5 0.14 Grecia 5.2 4.4 11.3 14.6 38.5 0.18 Alajuela 5.5 4.5 10.06.4 38.0 0.20


Agronomy Area 4.6 3.8 4.5 7.8 15.5 O.11 Cararao 4.2 3.6 5.0 9.7 17.3 0.17 Drainageway 4.9 4.3 20.5 21.0 48.4 0.13 East Carirmagua 4.9 4.1 1.7 2.2 4.3 0.17

* n was determined using a 1:1 soil to, solution ratio




64






showed CEC values highly dependent on pd. The pH dependency of CEC

was very marked in the samples from Costa Rica and the Drainageway sample from Colombia. The high amount of amorphous materials found in the clay fraction (Table 9) and the high OM content explained the increase in CEC with pH in the soils mentioned,

Electrical conductivity (EC) was found to be somewhat higher in the soils from Costa Rica (Table 5); although Cararao and East Carimagua from ColombiE showed values similar to those found in the Costa Rican soils. However, all soils were found to be low in electrolyte concentration based on EC measurements.

In general, the soils from Costa Rica had relatively higher pH values, and higher OM content, CEC and EC than the soils from Colombia except for Drainageway which gave values similar to those of the Costa Rican soils.

Extractable Soil Nutrients

Extractable bases (Ca, Mg, K, and Na) were much higher in soils from Costa Rica than in those from Colombia (Table 6). Calcium dominated the distribution of exchangeable cations in the Costa Rican soils. Magnesiumn was substantially higher than K and Na except for Alajuela soil. Exchangeable Al was predominant in the soils from Colombia; however, San Vito and Grecia showed Ex.-Al values considerably higher than those found in the other Costa-Rican soils. Extractable bases and Ex.-A1 values were correlated uith pH and EC values. Soils with high rpi and EC were hiEh in extractable bases and low in Ex.-Al.






Table 6. Extractable nutrients from the 3oils studied



IN NH4OAc(pH 4.8) Bray 1N Water extractable II KC1
oil Ca Mg K aa P P Al S04 -S NOC-N

---------- mehl00g ---------- ------ ppm ------ meq100g ------ ppm------Los Dia:mantes

Surface 3.5 1.11 O.hO 0.19 0.8 8.7 0.17 23.4 37.1 Subsoil 2.9 0.66 0.30 0.16 0.6 9.5 0.06 132.2 8.1 San Vito 5.0 1.70 0.45 0.24 0.4 3.4 1.33 36.6 10.0 an Isidro 2.5 0.79 0.19 0.23 0.4 2.1 0.61 33.3 8,0 Orecia 1.4 0.96 0.57 0.20 0.4 1.6 1.00 41.0 6.0 Alajuela 3.6 1.39 1.50 0.14 0.4 3.2 0.39 64.4 31.6


,gronory Area 0.3 0.12 0.14 0.13 0.2 2.4 3.14 7.8 5.9 Cararao 0.4 0.41 0.21 0.13 0.8 9.6 4.69 60.0 28.9 Drainageway 0.2 0.14 0.30 0.20 0.6 3.1 2.39 37.3 5.6 Eaot CarimajGua 0.1 0.04 0.05 0.12 0.4 2.3 0.89 42.5 5.0









Phosphorus extracted witln 1N , ( -,4.8) was rciatively low for all soils (Table 6); however, F extrc.ater d with th Bray II solution varied considerably and was much hK;: than P extracted with 1N NH 4OAc (pH 4.8). Nevertheless, the extractable P status of all soils was considered extremely low. The r agents present in the Bray II solution are generally recognized to extract more P from Al and Fe compounds, and also from Ca compounds (24), while 1N TI4OAc (pi 4.8) extracts mainly the P found in amorphous Al and Fe fractions (70, 103, 121).

Water extractable SO 4-S was relatively high in all soils except in Agronomy Area (Table 6). Highest amounts were found in Los Diamantes subsoil. A high degree of variation in SO 4-S occurred and a clear grouping of soils by country was not evident. The amount of water extractable NO3-N was relatively fhigh in Les Diamantes surface, Alajuela, and Cararao soils; all other samples were low. Water extractable SO4-S and NO3-N were not individually associated with other soil parameters although their sum related to EC.

The cation and base saturation values shown in Table 7 supported the information given on extractable bases and Ex.-Al. Soils from Costa Rica were higher in base saturation and lower in Al saturation than soils from Colombia; however, base saturation did not reach 50, in the Costa Rican soils and was below 15$ in tae Colombian soils at pHi 4.8. Considerably lower values were obtained for base saturation at paH 7.0. These values were closely associated with those shown in Table 5 for CEC and reflected the p!I dependency of CEC.









The C/N ratio was fcun-d to be adequ,.te for mineralization of N (128) (Table 7). Although ie majority of the Colombian soils and Grecia soil were higher than 15/1, N i:m-aoilization is not likely to be abnormal.

The Ca/Mg ratio for all soils was considered relatively narrow although adequate for plant nutrition (128). The Colombian soils had lower ratios than the Costa Rican soils except for Grecia. The Mg/K ratio was high in the Costa Rican samples except for Grecia and Alajuela soils, and very low in the Colombian samples except for Cararao. A Mg/K ratio of 3.0 is considered adequate for plant uptake of these nutrients; however, Russell (128) explained that in considering exchangeable ca-ion ratios as means of assessing nutrient uptake by plants, one has to take into consideration the crop of interest, the total concentration of cations in the soil, the type of clay, anac the relative contribution of the clay and OM to the base-holding capacity of the soil. He pointed out that, in general, increasing Ca in the soil depresses Mg in the leaves more than K; increasing Mg depresses Ca more than K; and increasing K may depress Ca more tha : Mg, or may decrease them about equally.

The Ca/Al ratio was very low in all soils from Colombia (Table 7). High ratios were calculated for Los Dianmantes soil, and somewhat lower for other Costa Rican soils. Low ratios were used to indicate the dominant effect of Al on the soil exchange complex while high ratios aave a relative index of the dominant effect of Ca. These values were not meant to provide information on plant uptake, although





Table 7. Chemical properties of the soils studied related to fertility



Cation saturation* Base saturation

Soil Ca Mg K Na Al pH 4.8 pH 7.0 C/N Ca/Mg Mg/K Ca/Al



Los Diamantes

Surface 32.0 10.2 3.4 1.7 1.6 47.9 18.2 10.4 3.2 2.8 20.8
Subsoil 28.5 6.5 3.0 1.6 0.6 39.6 17.3 11.3 4.4 2.2 48.0

.an Vito 26.2 8.9 2.4 1.3 7.0 38.8 19.6 11.6 2.9 3.8 3.7

San Isidro 20.6 6.4 1.5 1.9 5.0 30.4 15.9 12.2 3.2 4.2 4.1

Grecia 9.5 6.6 3.9 1.4 6.8 21.3 8.1 18.2 1.4 1.7 1.4

Alajuela 22.1 8.5 9.1 0.8 2.4 40.5 17.5 13.6 2.6 0.9 9.3


Agronomy Area 3.8 1.5 1.8 1.7 41.0 8.8 4.4 16.8 2.5 0.9 0.1

Cararao 4.3 4.2 2.2 1.3 48.3 12.1 6.8 15.8 1.0 1.9 0.1

Drainageway 1.0 0.7 1.4 0.9 11.4 4.1 1.8 12.4 1.6 0.5 0.1

East Carimagua . 5.0 1.8 2.3 5.3 40.4 14.5 7.4 16.2 2.7 0.8 0.1

* Relative to CEC measured at pH 4.8; Al was extracted with i KC1.








liming soil with a Ca/Al ratio greater or si ul to 1 tas been discouraged (19, 49, 77) because the pH is usually higher than 5.4 and the Al Js considered inactive or neutralized at this pH (49). Total Nutrients

Los Diamantes soil was found to have a larger reserve of nutrients than any other soil studied (Table 3). Total K was very

low in San Isidro; total Ca was low in San Tsidro and all the soils from Colombia. Total Fe was considerably lower in the Colombian samples as was total Al; these soils were also very low in P reserves except for Drainageway. In general, soils from Costa Rica, except for San Isidro, had a larger reserve of nutrients than soils from Colombia. According to their total elemental analysis, soils from Costa Rica

appeared to be less weathered than the Colombian soils. Analysis of the Clay Fraction

The predominant crystalline minerals in the clay fraction of the soils from Costa Rica were intergrade vermiculite, kaolinite, and gibbsite (Table 9); however, only trace amounts of kaolinite were detected in Los Diamantes subsoil. The clay fraction of these soils was dominated by amorphous materials. The soils from Colombia showed relatively high amounts of vermiculite along with kaolinite. Relatively small amounts of gibbsite were found in these soils, and the amorphous

components of the clay fraction were much smaller, except for Drainageway, than in the Costa Rican soils. Other minerals present in all soils, except in Los Di tmantes, were quartz and feldspars. Quartz was found in significant amounts. X-ray diffractograms iave indications that metahalloysite was also present in the Costa Rican soils. Intense peaks






Table 8. Total elemental analysis of the soils studied



3cil K Na ca Mg Al Fe P

-------------------------------------------------------------------------Los Diazantes
Surface 0.96 2.7 3.67 1.98 4.8 9.1 0.19 Subsoil 0.98 3 3 3.85 2.23 4.9 9.7 0.17 San Vito 0.35 1.4 0.25 0.45 6.1 8.8 0.08 San Isidro 0.04 1.1 0.04 0.06 9.2 14.4 0.07 Grecia 0.24 1.2 0.20 0.32 9.0 11.7 0.11 Alajuela 0.24 1.4 0.21 0.39 8.6 10.6 0.12


AEronomy Area 0.20 1.3 0.03 0.06 3.0 6.1 0.03 Cararao 0.35 1.3 0.03 0.07 1.0 6.9 0.03 Drainageway 0.33 1.2 0.03 0.06 1.0 7.6 0.08 East Carimragua 0.53 1.3 0.02 0.02 2.1 2.7 0.01





71



were observed at 7.19, 4.4, and 3.59A; the 4.;A peak sloped toward the high angle side of the diffractogram. However, differential thermograms failed to show the associated water expected with halloysite.

Aluminum interlayers were detected in association with vermiculite in the soils from Colombia. The spacing of the mineral was reduced from 14A to 12A by heating a K-saturated sample to 550C. The presence of a 14A peak in the heated sample of Drainageway soil also indicated small amounts of chlorite. A sharp, small peak at 9.6A vas evident in all Colombian clay samples. This mineral was identified as

illite.

The CEC of the clay fraction for the Costa Rican soils except

San Isidro and Alajuela, was higher than that for the Colombian soils (Table 9). Aluminum interlayers probably blocked exchange sites on vermiculite causing lower CEC measurements than expected for the soils from Colombia.

Soil Al.uminum and Iron Fractions

Aluminum fractions

The Ex.-Al fraction was predominantly higher in soils from Colombia, although San Vito and Grecia from Costa Rica were considered high (Table 10). Los Diamantes had the lowest Ex.-A1 content and Agronomy area and Cararao soils the highest. If one takes 0.2 meq

Ex.-Al/100lg as the maximum quantity for optimum plant growth, Reeve and /. Sumner (126), all soils except Los Dia antes would require lime to achieve this value.

The high amount of Ex.-Al found in the soils from Colombia was considered to be closely related to weathering, and high leaching





72





Table 9. Minerology and cation exchange capacity of the clay fraction
from the soils studied



Minerals identified
Amorphous
Soil Vermiculite Illite Kaolinite Gibbsite materials CEC meq/ lOCg
-------------p of the clay fraction--------------Los Diamantes

Surface 10-20 - 6 3 69 35 Cutcil Trace - Trace 3 84 33 San Vito 10-20 - 13 16 26 22 San Isidro 10-20 - 24 4 20 11 Grec a 5-10 - 22 4 19 16 Alajuela 5-10 - 20 18 26 12


Agronomy Area 30-40 5 17 Trace 9 11 Cararao 30-40 5 13 Trace 8 14 Drainageway 10-20 5 3 Trace 38 13 East Carimagua 30-40 5 16 7 11 9





73



of bases. This observation was confirmed by the fact that the more weathered soils from Costa Rica vere also hbigh in Ex.-A1.

In general, the Al extracted with LN NH OAc (pH 4.8) was

markedly higher in Costa Rican soils than in Colombian soils (Table 10); however, the Drainageway soil from Colombia had the highest amount. This fraction of soil Al has been closely related to amorphous soil components -such as OM, Al hydroxides, and Al polymers (70, 100, 103, 13.21, 123); in fict, the amount of Acet..-A1 was closely related to the amount of OM in the soils studied. The distribution of Acet.- Al showed the opposite pattern of that of Ex.-A1.

The fraction of Al extracted with acid ammonium oxalate has been correlated also with the amorphous components of soils (97, 98, 131). The data shown in Table 10 disclosed a good relationship between Ox.-Al and Acet.-Al. The Costa Rican soils were considerably higher in Ox.-Al than the Colombian soils except for Drainageway which had the largest amounts.

The CDB-Al fraction vas considered high in most soils; however, Los Diamantes subsoil from Costa Rica, Cararao and East Carimagua from Colombia had values somewhat lower (Table 10). This fraction gives a good indication of the amount of free Al oxides present in soils (4), and usually highly weathered soils are expected to have a high CDB-Al fraction; however, soils from Costa Rica were higher than soils from Colombia except Drainageway. This discrepancy can be explained by the fact that CDB will also attack amorphous iron-aluminosilicates like the type present in organo-metallic complexes (72).






Table 10. Aluminum fractions of the soils studied



30il Ex. -Al Aet.-Al Ox.*A1 CDB3-A Pyro. -A1 Na0-Al

-------meqjOOg------- ---------------------------------------Los Diemantes

Surface 0.17 18.0 1.2 3.3 0.6 1.3 Subsoil o.o6 18.6 1.6 0.6 0.4 1.5 San Vito 1.33 15.3 0.7 4.0 1.1 0.9 San Isidro 0.61 7.8 0.6 3.9 0.9 1.0 Grecia 1.00 17.7 1.0 6.6 0, 1.3 Alajuela 0.39 11.2 1.1 5.3 0.7 1.4


Agronomy Area 3.15 6.9 o0.3 .6 0.7 0.5 Cararao 4.69 7.9 0.2 1.3 1.4 0.5 Drainageway 2.39 38.4 1.8 5.3 1.4 1.6 East Carimagua 0.89 2. 0. 1 1.3 0.2 0.3


4:-





75



The organic Al fraction (Pyro.-A1) did not separate the soils into well defied groups (Table 10). All soils, except East Carimagua, had large amounts of Pyrc.-Al. In some cases this fraction was le'rger than the Ox.-Al fraction indicating the importance of the organo-metallic complexes. Such was the case for San Vito, San Isidro, Agronoy Area, Cararao, and East Carimagua soils. McKeague (97) found that some of the free oxides are dissolved by pyrophosphate.. The variation observed could have been caused by dissolution of free Al oxides.

A better grouping of soils was obtained using NaOR0 as an Al extractant (Table 10). The soils.from Costa Rica were much higher in NaOQ-Al. than the soils from Colombia except for Drainageway. This was the same pattern observed for soil OM and amorphous materials in the clay fraction, and is explained by the selective dissolution of amorphous aluminosilicates obtained with hot 0.5N NaOlK (60).. Iron fractions

Similar trends as those described for Al were observed for Fe (Table 11). In general, soils from Costa Rica were higher in Fe fractions related to OM and amorphous materials than soils from Colombia. Drainageway, however, was similar to the Costa Rican soils. The CDB-Fe fraction was extremely high in most of the Costa Rican soils, and the NaOH fraction was rather low in all soils. Jackson (72) indicated a preference of CDB for Fe over Al, and Hashimoto and Jackson (60) showed that hot 0.5I NaOH prefers Al more than Fe. Characterization of Iron and Aluminum

Crystalline oxide forms of Al were dominant over amorphous

form in most soils (Table 12). The major exceptions were Los Diamantes





Table 11. Iron fractions of the soils studied



Soil Acet.-Fe Ox,-Fe CDB-Fe Pyro.-Fe NaOK-Fe mM/100g ...........----------------- --. .................

Los Diamantes

Surface 0.12 0.4 2.8 0.3 0.06 Subsoil 0.08 0.5 3.1 0.1 0,04 San Vito 0.14 0.7 10.2 0.2 0.12 San Isldro 0.08 0.4 15.6 0.2 0.08 Grecia 0.09 0.9 14.2 0.1 0.08 Alajuela 0.07 1.1 13.7 0.1 0o09


Agronomy Area 0.08 0.3 6.4 0.1 0o07 Cararao 0.38 0.3 2.0 0.4 0.05 Drain.ageway 0.05 0.1 1.2 0.1 0.08 East Carimagua 0.05 0.1 4.7 0.2 0.04






Table,12. Characterization of iron and aluminum in t~e soils studied



Free
crystalline Amorpious Amorphous Total oxides inorganic* organic amorphous

Soil Al Fe Ai Fe Al Fe Al Fe


-----------------------------------------------------------------------------------Los Diamantes

Surface 2.0 2.4 0.6 0.1 0.6 0.3 1.2 0.4 Subsoil 1.0 2.6 1.2 0.3 0.4 0.1 1,6 0.5 San Vito 3.2 9.5 - 0.5 1.1 0.2 1.1 0.7 San Isidro 3.3 15.1 - 0.2 0.9 0.2 0.9 0.4 Grecla 5.6 13.3 0.3 0.8 0.8 0.1 1.0 0.9 Alajuela 4.2 12.6 0.4 0.9 0.7 0.1 1.1 1.1


Agronomy Area 2.2 6.1 - 0.2 0.7 0.1 0.7 0.2 Cararao 1.1 1.7 - - 1.4 0.4 1.4 0.4 Drainageway 3.5 1.1 0.4 -1.4 0.1 1.8 0.1 East Carimagua 1.2 4.6 - - 0.2 0.2 0.2 0.1

* Values not shown were calculated to be less than zero.





78



subsoil from Costa Rica where the amorphous inorganic forms iominrted, and Cararao from Colonibia which showed a larger proportion of Al in the amorphous organic form; howeve:-, the amount of amorphous Al was significant in all. cases with the organic fraction dominating the inorganic fraction except in Los Diamantes subsoil.

Crystalline oxide forms of Fe were much higher than amorphous forms in all soils (Table 12). Amorphous inorganic forms were larger than amorphous organic in soils from Costa Rica except Los Diamantes surface. Amorphous organic forms of Fe predominated in soils from Colombia except Agronomy Area.

in general, all soils were high in amorphous forms of Al and Fe. .These compounds are considered very active in soil acidity, cation and anion exchange reactions, specific anion adsorption, and several other chemical and physical properties common to tropical soils. Their contribution to some of the processes mentioned will be discussed in some detail later.

Lime Requirement

The amount of lime needed to neutralize Ex.-Al was considerably

lower than that required to raise soil pH to 6.8 (Table 13). There was very little difference in lime requirement by the SMP buffer and Yuan's buffer although the latter gave higher estimates especially for soils high in Or. Both of these methods take into account the acidity and buffering capacity of acid soils. The acidity to be neutralized from

the buffer equilibrium pH to a desired pH is determined by the buffering capacity of the soil; therefore, lime requirement tests using these two





79







Table 13. Lime requirements for the soils studied by the exchangeable
aluminum, SMP, and Yuan techniques



Lime requirement

SM Yuan
Soil Ex.-Al (pH6.8) (pH 6.8)


------------------CaCO3, mt/ha-------------------Los Diamantes

Surface 0.3 15.0 19.0 Subsoil 0.1 10.0 12.3 San Vito 2.2 18.6 22.6 San Isidro 1.0 14.6 19.7 Grecia 1.7 18.6 21.7 Alejuela 0.7 17.2 24.9


Agronomy Area 5.3 14.8 16.8 Cararao 7.9 17.2 20.1 Drainageway 4.0 22.4 39.4 East Carimagua 1.5 3.1 2.7









buffers should give values considerably hi gh.,er than the Ex.-AL method in soils high in buffering capacity. Although the neutralization of Al is the primary function of liming, Yuan (1-0) pointed out that the increase in CEC by liming to a soil pH close to neutrality would greatly improve the soil chemical conditions; nevertheless, success in liming tropical soils has been limited to the neutralization of Ex.-A1 (19, 51, 77, 126). When additional amounts of lime have been used, a temporary condition harmful to several plants has developed (109, i46). Velez and Blue (145) and Zant.a and Blue (164) showed that overliming damage was time dependent in one soil; yields were izparqd subsequently in greenhouse experiments.

Sources of Soil Acidity

Potentiometric titrations

The poteritiometric titration curves for soils from Costa Rica were characterized by lack of inflection points from the initial titration pH to EH 7.0 (Figure 1). A sharp increase in pH occurred, and then, a buffer zone appeared from pH 7.0 to 8.0. Los Diamantes surface and subsoil showed very diffused curves common to amorphous materials

(43). The lack of inflection points, the sharp increase in pH below

7.0, and the small amounts of acidity titrated in the KC1 extracts were attributed to the small amounts of Ex.-Al present and to weak acid character. The slope of this first portion of the curves was caused mainly by dissociation of weakly held H ions such as those present on the exchange complex and on carboxyl and phenolic groups of OM, or H ions created through the decomposition of Al and Fe complexes (96, 133). The buffer zone above pH 7.0 was brought about by dissociation of i ions





81












Lo0s DOnts 9.0 San Isidro

.0 , KC oil EN c---- 6.0 In KCI Sll Ettoct

70 / 7.0

> 0t s oCal sm we E IKCt Sa1ll Sunson


5.0 50

40 I - 40 a I I I * I * .
0 LO 20 30 40 0 30 60 90 120

o90 Los Diomantes 90- Grecio

Sft. E So aue KCI Soil Eaftc"t

7.0- t

1 60 to ac: so 60 agac sa snqewaon


4 0 I , s I 4.0 I I i 1 'a,

0 025 050 075 1O 0 75 150 225 300

9.0 Son ito 9.0 Al-ojuta

No So" Emil=I 8.0 IN KCI Sol Extmal



& s680 5 - sesn, se


7.0,4.0
6. a sc s s s0 1 j ICI $ 1 SuspIns I



0 3 60o o 120 0 50 120 00o 2 Added Base (meq/IOOg)





Figure 1. Potentiometric titrations of soils from Costa Rica.




82



from organic matter f�unctional groaps and -olyluidnohydronium edges

(73). The titration curves obtained with Na 0 were very close to those obtained with Ba(OH)2 except in Los Diamantes soils. This indicated that polymeric and complex Al., not titratable with Na2B'07, were an important source of acidity in these soils. In general, no strong acid character was observed, and the sources of acidity seemed to be associated with Al and Fe amorphous soil components such as OM and organo-metallic complexes.

The titration curves for the Colombian soils were very

different thai -those for the Costa Rican soils except Drainageway (Figure 2). An initial buffer zone was present with an inflection point around pH 5.5. A second inflection point was observed around pH 7.0 after a rather rapid increase in pH from 5.5 to 7.0. An additional buffer zone appeared between pH 7.0 and 8.0, but was not as well delineated as in the Costa Rican soils. Considerable amounts of acidity were found in the 1KC1 extracts. Drainageway had titration curves very similar to the Costa Rican soils; however, the initial slope was not as steep, and the buffer zone was extensive, indicating more buffering and stronger acid character. The initial buffer zone common to most Colombian soils was attributed to hydrolysis of large amounts of Ex.-Al (43). The other portions of the curves were caused by the same sources of acidity discussed for the soils from Costa Rica. Titrations with Na2B40 indicated that polymeric and complexed Al were important sources of acidity in Agronomy Area and Cararao soils. In general, the Colombiaii soils displayed strong acid character, and the sources of acidity were mainly protons yielded by Ex.-Al in tie hydrolysis process. The acidity








90 Agronomy Areao 5D Droinogewoy

80 IN KCI Soil Extroct 8-0 I ~ KCI Soil Extract

7.0- 7.0

60 6.0 a KCt Soil
/ -IN KCI Soil Suspension uSuspsnsion


40 ii I , I 40
T 0 2.5 50 75 100 0 155 31.0 465 620 aC
80 Carorao 9.0 East Carimoguo

IN KCI Soil Extract IN KCI Soil Extract
70- 8.060 7 7.0

5.0 6.0 IN KCI Soil Suspension
IN KCI Soil Suspersion


3.0 1 I I I I 1 I 4.0 - 1
0 375 750 11.25 15.00 0 045 090 L35 180 Added Boase (meq/lOOg)

8o(OH)2
NaoB407
Figure 2. Potentiometric titrations of soils from Colorbia.





8t



of the Drainageway soil was attributed to Ex.-Al as well as amorphous sol components.

Conductometric titrations

The role of Al compounds in soil acidity was emphasized by conductometric titrations (Figs. 3 and 4). In all cases, the titration curves failed to detect the presence of H ions. Acidity from Al compounds was the only type observed. Since the Costa Rican surface soils were low in Ex.-Al, other active forms of Al such as amorphous compounds including complexes with OM were capable. of yielding protons to create acidity. - On the other hand, Ex.-Al was the predominant source of protons in the Colombian soils except Drainageway. In the latter, both exchangeable and amorphous Al contributed to acidity. The term amorphous Al includes organic forms, hydroxy Al formed as an intermediate product of the hydrolysis reaction, and polyaluminohydronium ions created by polymerization of hydroxy Al. Titratable acidity

Small amounts of acidity were measured in the Costa Rican soils taking pH 5.5 as the end point of titration and using Ba(OH)2 as the titrant for potentiometric titrations (Table 14). The Colombian soils except East Carimagua, had considerably higher amounts of acidity than the Costa Rican soils. Drainageway soil yielded much more acidity than any other. A similar pattern occurred taking pH 7.0 as the end point. This distribution broke down at pH 8.0. Los Diamantes (surface and subsoil) and East Carimagua were always lower in acidity than the other soils. Acidity increased markedly from pH 5.5 to 8.0 in most soils froi Costa Rica except Los Diamantes. The largest increase took place




Full Text

PAGE 1

LIMtS iT
PAGE 2

•ro \r:j v/ife, Helen Margaret Ve3.er. . • . i -]

PAGE 3

ACKNO^TLEDGMENTS The author wishes to express his deep appreciation to Dr. W. G. Blue, Chairman of the Supervisary Comrdttee, for his interest, guidance, assisttuice, patience, and personal friendship throuGhout the course of the academic and research programs; to Dr. L. V/. Zelazny for instruction and valuable suggestions during the overall investigation, and the preparation of this manuscript; to Br. T. L. Yuan for his advice in preparing laboratory experiments, his clarification of soil acidity and phosphorus, adsorption concepts, and preparation of this dissertation.; to Dr. G. 0. Mott for his relevant comments and advice in the design of greenhouse experiments and interpretation of data, and his help in revieving this vork; and to Dr. 0. C. Ruelice for his contribution to a greater understanding of crop ecology and for serving on the Supervisory Committee. He is indebted to Mr. Edgar Rey and Mr. Williaiu Cordero from the I^Iinistry of Agriculture for their cooperation and aid given daring the collection of soil samples in Costa Kica: to Dr. J. K. Spain from the Center of Tropical Agriculture for his valuable assistance and advice in collecting soil samples in the eastera savannas of Colorabia to LXi-. C. liina, Ingeniero Agrcnomo Victor Vega, the soil classification staff of the Institute Agustin Codar.zi and Dr. E. S'larez in Colombia iii

PAGE 4

for their outstanding cooperation in selectinK sarnpling -sites and storing and shipping soil samples to the United States. . The author is grateixil to Dr. CF. Eno, Chainfian of the SoilScience Department, for financial assistance which allowed the completion of this study and provided support for other financial needs; and to Dr. K. L. Popenoe for providing funds used in collection and trans' portation of soil'... . Appreciation is extended to Dr. li. L. Breland, Mrs, Helen Brasfield and all other personnel of the Analytical Research Laboratory for their assistance in analysis of the large number of samples required for this vork: and to Mr. A. Waller and the field personnel for their very impoxtant contribution in setting up greenhouse experiments and . subsequent collections and preparations of plant material for analysis. Gratitude is j;x^res3ed to faculty and staff luembers^ especially Mr. J. Gonzales, of the Soil Science Department; to fellow graduate students for their friendship and advice; and to Dr. Marciano Rodriguez for the long hours spent together with the author during the collection and interpretation of data. Most of all, the author wishes to express his eternal appreciation to his wife, Helen I-Iargaret Velez, for all the support, encouragement, understanding, and help provided throughout these long years of hard work; to his parents, Dr. Julie Velez and Mrs. Elvia de Velez, for their immense contribution to the author's education and for their patience in waiting all this time for his return home; to his son, Julian Eduardo Velez, for providing everlasting hours of happiness, especially during periods of stress and uncertainty. iv

PAGE 5

-. . TABli; 0? C0NTEKT3 Page ACKKOV?J.SI>jM£NT? .......... . . . . ... . . ... . . . . iii LIST OF TABLSo. x JJST OF FIGURES . . . ' xvii ASSTRACT ' xvlii irfTHOIlJCTlON 1 LITERATtlTffi REVIEW * 5 Sources of Soil Acidity. < • 5 Inorganic Sources 5 CrgJinic Sources 6 Soluble AcJdt; 7 !!;; a surement of 3o:il Acidic Properties. 8 Typej of Measurable Acidity 8 Titration Analysis 9 Acid Soils and Plant Growth 11 IlydrocRn lou Concentration 11 Calcium Deficiency 12 Magnesiuin iJeficiency, 12 PhOiiphorus Deficiency . , 12 Alujrdnum Toxicity 13 Majit-^ancse Toxicity llf Molytdcnura Dofici;?ncy Ik Kcutralizp.ticn of Soil Acidity I5

PAGE 6

TABIiS 0? CONTENTS (Continued) -. • Page Soil Alii/niaum and Iron Fractions . , 17 Ananoniun Oxalate Fractrlon 17 Citrate Dithionite Bicarbonate Fraction ... . 17 Sod?' urn Pj'rophosphate Fraction . . / . lO Sodium Hydroxide Fraction . , iS Arcmoniuffi Acetate Fraction . . . . . . . '. . . . ' 18 . Potassium Chloride Fraction I9 Fractionation Scheme , 20 Effect of LiwG cn Soil Alumirrini and Iron 20 Phosphorus Availability in Acid Soils pi Phosphorus Fixation 22 Effect of Aluninum and Iron cn Pncspiiorus Availability 2h Effect of Lirne on Phorfphoi-us Availability 2^; Effect of Calcium SiJicate on Phosphorus Availability P7 The Use of the Langrauir Isotherm in Describing Phosphoinis Availability 23 Electroche-nical Properties of Acid Soils 3O The Isoelectric Point and the Zero Point of Charge ^0 Measurement of the Zero Point of Charge 3I Effect of Soil Mineralogy on the Zero Point of Charge , Effect of Cation and /inion Adsorption on tne Zero Point of Ch.nrf^;e " -^-^ vi

PAGE 7

TA13LE 0? C0NTEN73 (Contini.ed) Practical I^nplications of the Zero Point . of Charge , . . . Caj.ciura Selectivity in Acid Scil.s. . . . Theoretical • • . •.. • • • Eff'?ct of Lime on Electrolyte Acc^ir^TaB.ation Acid Soils Itidicator Plant for C-reeniio-ase Stadiee . . . MTFJP.IAI3 AirO METHODS ............... Soil Samples Laboratory Procedures. General Chemical Properties Mineralogical Analysis. . . . ". ...... Soil Alurainum and Iron Fractions. . . . Lime Requirement . Sources of Soil Acidity Phosphorus Adsorption ......... Electrochemical Properties Calcium Selectivity Incubation Studies with Line Analytical Determinations ........ Greenluouse Procedures General Preparation c'* Experiments. . . E>a">erl'uent No. 1, Exi)eri!;icnt No. 2

PAGE 8

TAULE OF COIiTENTo (Continued) . ' ' ' . • Experiment No . 3 ....... INSULTS AND DI3CUS3I0N Laboratory Experiraents . General Chemical Properties. ..... ... Extractable Soil Nutrients Total Nutrients . Analysis of the Clay Fraction. . . .. . ". . . . ... Soil Aluminum and Iron Fractions Characterization of Iron and Aluminuin Linie Requirement Sources of Soil Acidity. Pnosphorus Adsorption Effect of Lime on Phosphorus Adsorption Electrochemical Properties CalcixiTi Selectivity Incubation Experiments . Greenhouse Experiments Experiment No. 1 Experiment No. 2 Experiment No. 3 SUr-ll^lARY Aim CONCLUSIONS , Al'PED DICES Append! .•!; A. , 1 viii

PAGE 9

TABLE OF COIJT^NTS -(Co?vtlnued; . ... .. Fa^e Appendix B . . . 1?^ LITERATuiffi CITED. IS5 BIOGRAPHICAL SKETCH I98 Ix

PAGE 10

IJST OF TABIi:3 , ' " Table .; Ll:&S. 1 Identification numlDer, location, and order _ ^ of the soils studied 2 Relation of treatment number to experimental and coded variables for Experiment Ilo. 1 ? Lime and phosphorus treatments for Experiment, Ka2 . • • • • • k Lime, calcium silicate, and phosphor'as treatments for Experiment No . 3 . . . ol 5 Chemical properties of the soils studied 63 6 Extractable nutrients from the soils studied ...... 65 7 Chemical properties of the soils L^tudied related to fertility , . . . , 08 8 Total elem.ental analysis of the soils studied TO 9 Mineralogy and cation exchange capacity of the clay fraction from the soils studied 12 10 Aluminum fractions of the soils studied 7'+ 11 Iron fractions of the soils studied '(o 12 Cnaracterization of iron and aluminum in the soils studied • 77 13 Lime i-equirements for the soils studied by the exclmnjeable aluminum, oMP, and Yuan tcclmiques ik Titratable and extractable acidity in the soils studied 88 15 Correlation amonc titratable acidity, extractable acidity, li requirement, and chemical properties of the soils studied 9^ X

PAGE 11

LIST OF TABLED (Continued) • Table PP-ge 16 Correlation amcng titratablc ficidlty, extract-able acidity, liir.e requirement, ' . and aluininuni in t'.ie ^;oil3 studied 91 17 Hiosphorus adsorption by the soils oLudied vith an intensity and a uuantity method ........ 93 iB Coi-relation between phosphorus adsorption . parsineters and selected chemical properties . _ of the soils studied. 9^-' 3.9 Eegression coefficients and coefficients of determination for the effect of lime (neci^'-'lOOg) on the P sorption capacity (ppm) of incubated soils measured by the intensity method 9^ 20 Effect of lime on Langmuir t;>pe parameters, and coefficients of determination for P sorption measured by the quantity method ' for the soils studied 97 21 Delta pH -measurements for the soils studied 99 22 The negative logarithm of calcium selectivity coefficients for the soils studied in the presence of potassium., magnesium, and alurrdnum in IK solutions lOS 23 PiCgression coefficients and coefficients of deterinination for the effect of lime (meq/lOOg) on the water pH of incubated soils 110 2h Regression coefi^icients and coefficients of determination for the effect of lime (meq/lOOg) on the IN KCl pii of incubated soils Ill Regression coefficients and coefficients of determination for the effect of lime (meqy'lOOg) on the exchangeable aluminum (raeq^ lOOg) of incubated soils ' 112 26 Regression coefficients and coefficients of determination for the effect of lime (meq/lOOg) on the water extractable potassium (pp:n) of incubated soils 113 27 Regression coefficients and coefficients of determination for the effect of lime (meq/lOOg) on the water extractable sodium (ppm) of incubated samples 11;) xi

PAGE 12

LIST OF TABIoES (Continued) . • -• ' ' ' Table ' Emi 26 Regression coefficients rind coefficients of . determination for the effect of line (meq_/100g) on the water extractable calcium (ppm) of incubated soils . 29 Regression coefficients and coefficients of . ^ .. determination for the effect of li.-ne (meqy'lOOg) \ ' " on the water extractable magnesium (ppm) of incubated soils • • • ^^'^ 3Q . .Regression coefficients «nd coefficients -of • — / ----deterni nation for the effect of lime (meqy^lOOg) on the water extractable nitrate -nitrogen (ppm) of incubated soils • 1^9 31 Regression coefficients and coefficients of ; determination for the effect of lime (i;ieq./100g) . on the electrical conductivity (jimhos/cm) of incubated soils ........... 120 32 • Regression coef ficiehts and coefficients of • determination for Fangola digitgrass yield ,,and nutrieiit concenti-ation froia Los Diaioantts surface soil. Experiment No. 1 122 33 Regression coefficients and coefficients of ' determination for Pangola digitgrass yield and nutrient concentration from Los Diamantes subsoilj Experiment No. 1 1^3 '^k Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration from San Vito soil, Experiment No. 1 ^2.h 35 Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration from San Isidro soil, Experiment No. 1 126 36 Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration from Grecia soil. Experiment No. 1 127 37 Regression coefficients and coefficients of determination for Pan^^ola digitgrass yield and nutrient concentration from Alajuela soil, Exporiment No. 1 129 xii

PAGE 13

rABLES (Continued) -iTable Page 38 Regressioi:i coeff icieiits end coef riclents of determination for Pangoladigitgrass yield and nutrient concentration from i\g"Gnoiny Area soil, PJxperimeDt No. 1 130 39 Regression coefficients and coefficients of determination for PangoJ-a digltgraas yield and nutrient concentration from Cararao soil, Experiment No. 1 13i 40 Regression coefficients and coefficients of determination for Pangola dip^itgrass yield and nutrient concentration from Drainageway soil, Experiment Ko. 1 . . . 133 kl Regression coefficients and coefficients of determination for Pangola digitgrass yield and nutrient concentration frorr. Ea:;t Caririagua soil. Experiment No. 1 13*^hi? Lime, magnesium, and phosphorus treatments to . . . t.b.e ' soils studied predicted .to give riazimum Pangola digitgrass yield by the model used in Experiment No. 1 . . 133 Effect of lime and phosphorus treatments to Los Diatnantes sujrface soil on Pangola digitgrass yield and nutrient concentration, Experiment Ko. 2. . 137 'l-i)-Effect of lime and pbosphci-ias treatments to Los "Diamantes subsoil on Pangola digitgrass yield and nutrient concentration, Experiinent No. 2 ' . ,• . 139 ^5 Effect of lime and phosphorus treatments to Pan Vito soil on Pangola digitgrass yield and nutrient concentration. Experiment No. 2 . . . . . . 2.hO k6 Effect of lime and phosphorus treatments to San Isidro soil cn Pan.":ola digitgrass yield and nuti-ient concentration, Experiment No. 2 142 .Effect of lime and phosphorus treatments to Crecia soil on Pangola digitgrass yield and nutrient concentration. Experiment No. 2 li;3 'l5 Effect of lime and phosphoruf; treatments to Alajuela soil on Pan/'.ola digitgrass yield and nutrient concentration, Kxpcriment No. 2 lh'\ xiii

PAGE 14

LIST OF TABIE3 (Continued) ^ ' Table . Page 1*9 Effect of lime and phoophoms treatments to .. ' Agronomy Ai*ea soil on Pangola digitgrass yield • .. and nutrient concentration, Experiment Ho. 2 • 1^6 50 Effect of lime and phosphorus treatments to Carareo poil on Pangola digitgrass yield and nutrient concentration^ Experiment No. . . 1^-7 51 Effect of phosphorias treatments to limed Di-ainageway soil on Pangola digitgrass yield end nutrient concentration. Experiment No. 2 1^+9 52 Kffect of lime and phospiioras T.rha-.;ments to East Carlmagua soil on Pangola d^-gitHrass yield and nutrient concentration, Ei'Cperiment Ko. 2 . . . ... I50 53 Effect of lime aiid calcium silicate treatments . to soils from Costa Rica on Pangola digitgrass .• • yield and nutrient concentration, Exneriment ' Ho. 3 ' 3^ Effect of lime and calci\m silicate treatments to soils from Colombia on Pangola digitgrass yield and nutrient concentration. Experiment So3 55 'Ris effect of lime on the phosphorus sorption capacity of incubated soils measured by the intensity method I63 56 Effect of lime on the organic matter content of iocubated soils 16^^!57 Effect of lime on the water pH of incubated soils 165 58 Effect of lime on the IN KCl pH of incubated soils 166 59 Effect of lime on the exchangeable aluminum of incubated soils I67 Ei'fect of lime on the water extractable potassium of incubated soils I68 61 Effect of lime on the water extractable sodium of incubated soilo I69 xlv

PAGE 15

LIST OF TAi'I^;3 (Continued) Table . Page 62 Effect cf lime on tbs vater extraclBl'le calcium of incutated soils . , 3-70 63 Effect of lime on the wa,tei-" extractable magnesium of incubated soils . . l?! 6k Effect of lime on the v;ater extractaole nitrate-nitrogen of incubated, soilo 172 65 Effect of lime on the: electrical conductivity of incubabed soils , . . . 173 66 Effect of lirae^ aagnesiurc, and p'nosphorus treatments to Los DiamantesSurface soil on Pangola digitgrass yield and mitrient concentration from Experiment No. 1 175 67 Effect of lime^ raagnesrj:!i, and phosphoms treatments to Los Diamantes Subsoil on Psngo3.a digitgrass yield and nutrient . ' . concentration from Experiment No. 1 .. .". . . . I76 68 Effect of lime, magnesium, and phosphorustreatments to San Vito soil on Pangola digitgrass yield and nutrient concentration from Experiment No. 1. , , I77 6^ Effect of lime, magnesium, and phosphorus treatments to San Isidro soil on Pangola . digitgrass yield and nutrient concentration from Experim.ent No. 1 I78 . . 70 Effect of lime, magnesium, and phosphorus .. treatments to Grecia soil on Pangola digitgrass yield and nutrient concentration from Experiment No. 1 , iVC71 Effect of lime, magnesium, and phosphorus treatments to Alajuela soil on Pangola digitgrass yield and n'.itrient concentration from Experiment No. 1 I80 72 Effect of lime, magnesium, and phosphorus treatments to Agi-onomy Area soil on Pangola digitgra:53 yield and nutrient concentration from Experiment No. 1 I8I XV

PAGE 16

LIST 0? TilBLES (Continued) . ' Table Ji'nge 73 Effect cf lime, magnesimn/ and phosphorus . treatiTients to Cararao soil on Pangola . ' [' digitgrass yield and nutrient concentration from Experiment No. 1 . . l82 7^ Effect cf lime, magnesium, and phosphorus . • • treatments to Drainageway soil on PHngcla . ' ' . . ' ' digitgrass yield and nutrient concentration from Experiment No. 1 I83 75 Effect of lime, magnesium, and phosphoras treatments to East Carimagua soil on Pangola digitgrass yield and nutrient concentration from Experiment Wo. 1 . . , l8^ xvi

PAGE 17

lUTRCnrCTIOK Liming acid soils has been a basic component of soil zcanagement systecis in temperate reeions for many years. This component is used to maximize growth of temperate legumes that have slightly acid to neutral pH requirements for H fixation, and to keep soil na-,;ricnt3 rear3ily eYaij.able for plant uptake. In recent years^ tvo schocls of thought v;ii,h respect to limrig, acj.a i'-oils havi developed: McLean (lOl) at Ohio State Univerrjity advocates liming soils to pll 6.^ because of higher nutrient availability to plants and higliC.i.' microbial activity at. this pli value. Tnese have beer the traditional reasc:?.r. for liming. Kamprath (75) at North Carolina State Univcr.;! by advocates using lime tiuantities that neutralize the exchangeable Al &ad bring I\l jaturation belo',/ j.'-i';' in the soil. Heeve and Sumner (126) in 3outh Africa support the exchangeable -Al school and favor additions of lime sufficient to lover tas exchmigeable Al to 0.?: ;!!eq_/10Cg of soil. The concepts follovod Iriy these tvo schools raise the auestion on vhetiier to lime to a favorable pH rargc^ iicually between 6.0 and 6.5j or to inactivate toxic substances such as A] and Mn, vhich takes place aroand pH 5o« Evidence in favor of each tbeory has been pat forward by the proponents of each line of ' . thought; howeverj it seems that the idea of limirg to inactivate toxic sobstaiices is gaining momeatiLm especially in regions where highly veathered acid soils containing large amounts of toxic substances are common. 1

PAGE 18

V/hen lime vas first introduced into the tropics, the concept of the r-iost favorabie yU. range vas used as the criterion for lime appli' cations. Responses were often ei ratio and even detrimental to plant growth: Adaias and Pearson (2) stated that crops respond differently to line in tropical soils in compari£;on to respoQses in temperate soils vith maxiiaum yields occurring at pH's from k.^ to 5.5-. Russell and Richards (12y) mentioned that liming in the tropics and sut tropics only improves yields on very acid soils and usually reduces yields on n:oderately acid soils. Numerous experiments in the eastern province of Nigeria (15^) indicated good responses to lime applications betveen 0.5 and 1 ton/acre, particularly in soils with pll below k.^. in Kenya and southern • Rhodesia (l5'i-) the application of lime failed to improve the establishment or productivity of grass leys in areas with long dry seasons. Data from Paerto Rico (l), Brazil (lio), and Hawaii (157) support lower soil reaction values for optimum plant growth. All of the work mentioned seems .to substantiate the inactivation of to.:ic substances as the best criterion for liming acid tropical soils. However, research by Lucas and Blue (9^+) with a Goil from eastern Costa Rica having low exchangeable Al and I.ln showed a depressing effect of lime on plant growth even at rather low lime rates when compared to the high buffering capacity of this soil. Veiez, Zantua, and Blue {lk6) found high electrolyte concentrations as the possible cause for the detrimental effect of lime in the same soil. These authors noted that the negative effect of lime was time dependent and plant yields im.proved considerably as time increased. Tliis electrolyte effect was first mentioned by Zantua and Blue {l6k) , They observed a dramatic positive growth response of Pangolagrass ( Pigltarla decumbens stent.) planted to

PAGE 19

the linked and xvnlimed samples of soil u-jed by Lucas and Blue vhicb had "been previously leached vith vater. As vas nientiODfid e£.rliez', lirnins uas been used to improve the availability of plant nutrients. Of ell the nutrients required for plant growth, P h~as been singled out as one of the most deficient nutrients in tropical soils (19, ll8) and liming has been recognized as the best practice to increuise P availability in temperate soils. Eoweyer, work by several researchers (63, 79, 125, 1^^*+) has disclosed the possibility that lime, instead of improving P availability, increases the P-retcntion capacity of highly weathered soils with high Al end. Fe content and soils rich in amorphous materials both of \/hich are cocirnon in the tropics. Sound management practices must therefore be developed rer^rdless of the reasons for liming in order to increase food production to rn.'>>dinum levels in tropical regions i/hcre raaliiutrition and r.tar^^ation are corfiniori. Therefore, the following objectives were developed to ob-:;ain pertinent answers related to the use of lime in soils fron^ Costa Rica and the eastern savannas of Colombia: • : . ' (a) To determine acidic properties and buffering capacities of these soils . ("b) To characti;ri;;e the Al and Fe fractions and their influence on soil acidity. {z) To determine the P adsorption capacity, as related to soil components, and the effect of added lime. ..(d) To determine the clectrcchemicai properties of the soils, the pi{ at which anion or cation i^dsorption takes place, and the effect of pH on anion rnd cation adsorption.

PAGE 20

To determine thr selcctivl.f-.y of these soils for Ca, G component ot lline, as influ;,':ioed by the presence of other cations in soil solution, ^ . To deteiTiine tne ei'feot of lime on the accumulation of electrolytes in the soils. To determine the optiraun levels of P, Mg, and liine needed for maximum plant growth. To determine the effect of CaCO, and CaSiO in reducln.'^ i 3 the need for fertilizer P. ; ' '

PAGE 21

SEVlWvJ OF LITEMTliK2 Sou rces of F>o±l_AcAA-:tY Soil acidity was defined by Jackson (72) as the proton yielding capacity of a soj.l system in going from a given state to a i-eference state usually specified as a pH value. Tlrie proton sources are, therefore, the sources of soil acidity, and in general they can be divided into inorganic and organic soil constituents (33, 160). I norg anic Sourc es . • • ' • The inorganic constituents responsible for acidic properties in soils are layer silicates, oxide minerals, and combinations of layer sixacates ana oxide minerals and non -crystalline components (33). Yuan (160) proposed a more general grouping of the inorganic components of soil acidity into crystalline and amorphous hydrous oxides and silicates. Protons are generated from these soil constituents by tiie dissociation of H ions from structural 0H~ groups such as thooe present at Lhe corners and edges of clay lattices and as basic constituents of hydrous oxide minerals of Fe, Al, and 3i (l3^). Hydrogen ions can also be found satisfying permanent charges on clay minerals (160). These H ions can be replaced by cations, and thus, tliey beco-ie sources of acidity. Hovtver, il-saturated clays are very -unstable and shift to Al-saturated clays spontaneously (33). The most important source of acidity in mineral soils is the hydrolysir; of Al ions. These ions are displaced from clay minerals by 5

PAGE 22

6 cations mid aydi-olyze in so.luti.on, one of rhe products oi" this hydrolysis reaction is T vhich is capatle of decomposing minerals yielding more hydrolyzable Al (33). Jackson (73) described the hydrolysis of Al as taking place in stages; in the first stage, equal amounts of H ions and mono-hydroxy Al ions are produced belov pH 5-0« As the O'd/M ratio increases, polymerization of Al-OII units tal^e place and large hydroxy Al ions foriii varying in the OH/Al ratio and in size. The latter increases as the OK content increases. 'Hie ultimate product of hydrolysis is A1(0K)^ vhich usually precipitates when the OH/Al ratio becomes greater than 2. In general, as pH increases, hydroxy -Al ions tcr:d to polymerize forming a layer lattice composed of stable ring units containing six Al ions with one net positive charge per Al and releasing H ions to the soil solution (69). . . Orga nic Sources Functional groups of organic matter such as carboxyls, phenols, enols, amines, and quinones are capable of dissociating hydrogen ions. The type and amount of functional groups vary with soils and organic matter fraction; the nature of the predominant groups present will determine the strength of the acid produced (132). It is recognized that carboxyl groups are acidic enough to dissociate considerably below pH 7.O; nevertheless, phenols and polyphenols are also capable of H. . dissociation and, along with the other groups, contribute as well in yielding protons (35)« It appears that Al"*"-^ and Fe"*"-^ in organo-metallic complexes are responsible for the weak acid character of organic matter (96). Interaction of hydrous oxides and organic colloids is regarded as involving the displacement of H ions from carbcxylic groups by hydroxy-Al

PAGE 23

and Fc ions {123) • Schnltzer cu;d Skinner ( 133) showed that the stability of ?e and A.l cori!plex--tg varied vlth pii. Chelates of Fe with org^""j-C compounds; were, found to be stable in aciu media, but decomposed in alkaline media with the production of Fe(OE)^ C^'OComplexes of Al and fulvic acid v.'ere broken through the replacement of fulvic acid by OH ions at p:i 5.O {hh, 65). Tne hydroxy -Al ion'j created in this fashion are capable of entering the hydrolysis reaction with the production of H ions. They can also become active in other soil reactions such as P fixation. S olub le Acids _ Coleman and Thomas (33) stated three ways in which soluble acids from biological activity can be of considerable importance in soil acidity. Their considerations can be summarized by mentioning the effect of additions of acid-forrrdng fertilizers to soils, the production of strong acids when soils containing FeS are exposed to oxidizing conditjon^, and the presence of organic acids from plant residues, root exudates, and litter. Jackson (73) grouped soil acidity and its neutralization based on the acid strength of the proton retaining site into the following categories: . I. Strong acids, soil pH ^-,2 and below: a. Mineral colloidal electrolytes; Mg, Al-0. . . "^'k^O (KCI or H-resin treated clays; unstable, reverting to 11). b. I'ree H^SOj^ from FeS or S, giving extremely acid soils. II. Wt'ak acids, soil pII 5 or 5.2 and below:

PAGE 24

a. •Aluminohydronj.um cation?: Al (_0K^''' )6 : H*" + rAl(OH)( -0!]-).-]^ (KCl exchangeable; exchangeabT.e protons of very acid soils). -. b. Possibly some humus carboxj'l. .. • :. • III. Very veak acids^ soil pH 5.2 xo 6.5 or 7: a. Humus carboxylj e.g., in surface soils. 0 -^"^ b. Polyaluminohydroniusi edge OHg'" , e.g., in acid subsoils. c. ^2'^^^3' ''-'^^''^ aluminum sulfate. IV. Vei-y very weai:acids, soil pH 6.5 oi' 7 to 9.5: . . a. Huit^us phenolic . b. Polyaluminohydronium edge pairs, OH . . . . OH^-5-. c. Ca(HCO HalICO . ' ^,'2 3 , V. Extremely veak acids, soil pH above 9.5: • a. Humus alcoholic hy-droxyl. b. Silicic acid -OK. c. Gibbsitic -OK (aLjJiinate reaction). Measurement of Soil Acidic PropcrticG Types of Measurable Acidity . Exchangeable acidity is that fraction of " acidity exchanged by neutral, unbuffered salts such as KCl, CaCl^, or KaCl. Titratable acidity is the quantity of neutralized acid obtained at a given pH (usually 8.2) (33). Thus, the acidity measured \d.th BaCl^triethanolamine at pH 8.2 corresponds to titratable acidity, but is sometimes used as a measure of total acidity (I06). The latter is also measured by

PAGE 25

r 9 titration with strong bases sv,ch as Ba(Cil)Vj a.-.d NaOH (43). Coleman • and Thomas (33) stated that the exchanseable acidity, as a proportion of the total acidity, varies with the nature of the soil and the degree of base saturation. Residual acidity is known as that portion of acidity which is titratable but not exchangeable with neutral, unbuffered salts. It is measured by titrating the soil after extraction with M KCl, and it can also be determined by substracting exchange acidity from total acidity. Large amounts of residual acidity are common in soils high in Al and Fe oxides and organic matter (33j ^3), : Titration Psialysls Potentiometric and ccnauctometric titration ajialysis have been very useful tools in measuring the components and the quantities of acidity (2?, 33, ^3, I65). Low (89) clearly showed that potentiometric and conductoraetric titrations differentiate and measure the influence of H and Al in acid bentonite. Yuan (159) corroborated Low's findings . in true solutions and soil systems. Coleman and Harward (3I) showed by potentiometric titration that H saturated montmorillonite behaves like a strong mineral acid. They also attributed the apparent weakacid character of clays to the presence of adsorbed Al. Coulter (3?) using titration techniques found that H-saturated clays became Al•• saturated clays with time. Dewan and Rich (^3) were among the first to titrate soil systems extensively. They used Ea(OH)^ and NaOH 3+ . to measure total acidity, and Na2B^^0^ and NaOAc to titrate Al + H ai:id H"*", respectively. They could not determine exchangeable H in their samples, and most of the acid character was attributed to hydrolysis of IK KCl extractable Al and non-extractable Al hydrolyzing in place.

PAGE 26

An initial inflection due to K can be observed in potentJo' " metric titrations follov/ed by a naffer region caused ty Al hydrolysis (28^ 95_, 159). Coleman and Tiiomas (32) deTnonr.trsted that polymeric non-e.xchaD.geable Al and Fe hydroxy ions in addition to the exchangeable Al^^ ion make large contributions to the buffering capacity of montmerillonite complexed vitb Al and Fe hydrous oxides in the pH range _ _ 5 to 8. Soils relatively free of organic matter have representative buffer characteristics vhen dominated by inouomeric exchangeable Al ('i-3)« When Aland IIsaturated clays are conpared, they also display the same buffer cnai'acteristics (89). A second buffer region is found ' between pH 7 and 8 and is due -co dissociation of f'anctional groups of organic ciatter (165). Soils with dominant fractions of amorphous materials including organic matter do not show sharp buffering regions similar to crystalline mineral soils or clays, and their buffering capacity to pH 8 is relatively high compared to that cf common mineral soils {lh2). Conductoraetric titrations are of great value in distinguishing between exchangeable H and Al. Marshall (95) showed p. linear decrease in conductivity v;ith increasing amounts of base similar to that of strong acids followed by an increase close to that of a weak acid in titration of mixtures of HGl and AlCl^ in a pure system. The first inflection 3 point was attributed to exchangeable H and the length of the linear middle section to exchangeable Al. When a Hsaturated clay is titrated with a base, Coleman and Thomas (33) pointed out that the conductivity of the system decreases until complete neutralization of the H is obtained. Tl^iis decrease in conductivity is due to the larger mobility of II ions than of cations of the added base. Conductivity rises rapidly as excess

PAGE 27

OH ions; are added because of the nigh mobility of r-hese ions. Alsaturaced clays behave like v.'erUc acids; sixice they have very small amounts of free H ions present, conductivity increases as base is added due to salt foiTnation (33) • Devan and Rich (kj,) demonstrated that conductometric titration cur-zes for soil systems had the same characteristics as similar cur^/es for pure solutions a.nd clay systems. Acid Soils and Pla nt Grcvth The harmful effects of soil acidity ha,ve been proven to be caused by secondary effects of the soil reaction, rather than to primary effects such as the activity of the H ion in the soil solution {2, 75> 128) • The secondary effects of high acidity are deficiencies of Ca^ Ks, and Ho, and excess of soluble Al, v-'in, and perhaps otlier metallic ions. The relative effects of these f3.Gtcr3 on plant growth depend on the soil itself as far as the available levels of deficient nutrients and on the susceptibility of the crop to deficiencies of the above mentioned nutrients or excesses of Al and Mn (128). Hydrogen Ion Conc e ntration • ' ' Adsorption cf many inorganic ions is sirnificantly influenced by variations in H ion concentration; however, plants do grow success fu,lly on acid soils provided the pH does not fall fcelov/ k.O-h.'y, the nutrient supply is maintained at suitable levels, and the presence of Al and Mn does not reach toxic levels (75). Nevertheless, low pli (5.0) can have detrimental effects on plant growth, and under certain conditions such as lack of toxic levels of Al and Mn, the H ion concentration is responsible for poor plant growth (2, 75). Arnon et al. (11) shoved f-ood growth of bormudograss, lettuce, and tomato at pH 5.0-7.0 in. nutrient solutions. Audus (12) obtained an increase In root

PAGE 28

grOT'rth up to pll [3.0 for cress, radish, cnrden vsnz, ariv?. corn. Hovard and Adams (67) found marked reductions In iaitial 33.ongation rate o£ pririiary recti? of cotton telov pH 4,2, . . ' . Cal cium Deficiency Calcium deficiency is not considered a raajor factor of acid soil infertility in some areas of the United States except on sandy soils {2, 75); hovever, acid soils vitb. lov CEC x'alues nay contain lindting Ca levels for jnany crops (2). Sxxch is tUe case in highly leached tropical soils (Oo). Calciiua deficiency in these soils is independent of soil acidity but is influenced by the levels of Ga and other cations; thersforej it is only in the absence of toxic levels of Hj Al^ or Mn ions that Ca saturation can be considered a good riieasui'e of the Ca-availability in the soil (2). Sowever, the Ca requirements of certain crops s-ach as peanuts, tcr.iai;oes, and celery are exceptions because of their inability to take up Ca from soils that supply adequate Ca for most other crops (2). • ' . ' Magnesium Deficiency . . . . • Magnesium deficiency occurs not only cn o,cid soils subject to leaching, but on calcareous soils. Deficient V^, is found in acid soils having lov; ClJC, high leaching, excessive K and Ca, and in soils where crops with high Hg requirements tire used (?). D<;ep sand:/ soils are subject to greater I>!g deficiencies than soils high in clay regardless of the p'l (2). Phosphcrus Deficien cy " ' ' Phosphorus is probably the most deficient nutrient in acid soils (19) • Crop establishment and develoTvr.ent in tropical acid soils is often irapossille vithout P anplj cations {16, 20, il8). Olson and

PAGE 29

13 Engelsted (llB) noted that there i~, a negatire cor.i'elation betwetia intensity of weathering and total P. Tliey also explained that the rate of P lost from tropical soil?, is detenriincd by cropping intensity and erosion, vith slight i.eachiiig losses except on very sandy soils. Fertilizer and manure were mentioned as prlrcarjinputs. The pict^-'-re of available P ixi tropical acid soils becomes even gloomier due to the high fixation capacity of certain soils such as Andepts. Several tons of P/ha may be needed to satisfy the total fixing capacity (5i). AxLUPiK i Ti Toxicity Probably the first researcher to suggest tht toxic effect of Al on crops was Miyake (ill). He demonstrated that 1.2 ppm of Al in solution vas tcxJc to rice and mentioned the possiuiiity of a relaticaship between Al and the infertility of acid soils. Since then a iai'ge number of papers dealing vith various aspects of Al toxicity and acid soils have appeared in the literature. Perhaps one of the most significant papers on the subject was x-/ritten by Kamprath (77). In this paper he shoved the relationship between exchangeable Al and linie. Kis stwxy led to the recommendation of lime for acid soils based on the exchangeable A?i content. He found that when the Al saturation was below 15f;, the growth ixnd yield of several crops were maximized. Similar results were obtained by Reeve and Sumner (126) in South Africa. They found that jime sufficient to reduce exchaj.igeable Al to 0.2 ineq_/100g gave raaximurn growth of SorghTMi Sudanese Stapf. The relationship between exchangeable Al and Al in the -;oil solution in mineral soils was studied by Evans and Kamprath (hi). They showed that soil solutions contained less thjm 3 ppm until Al satui-ation

PAGE 30

4 Ik var. I'-ifhex tlian SO/.. Aj'tcr this ;:,Mluratioii -ias ofctaincdj Al in the noil soiutica iucreased sharply to k.'^ ppm at 80?; zaturation. Corn ' did not respond to lime until Al saturatiori was fOp or about '^.6 ppm. Al in tl'.e soil solution; hovevcr, soybeans responded with only 30^ sakuraLicn or about 1.3 ppra Al in the 3oil solution. Pj.ant species varj' widely in theiitolerance to Al. Foy and Brovrj. {^k) found mustard, turnips, barley,, and cotton to be sensi: tive to Al '-mile buckwheat, corn, and soybeans v.-ere tolerant. Adams and Pearson (2) concluded that Al toxicity is j.argelj,' deteriidned by the chemical activity of Al in soil .-ioluticns in situ , regardlssfi of soil type. Kanga:-.c-s p Toxi city Adams and Pearson (2) stated tht?.t the anioant of easily reducible i4n in the soi3 will oetermne the potential for toxicity to take place and that minin^Jia levels of 50 to 100 ppai are necessary for Mn toxicity to occur. Hovt-ver, nutrient solution e>cperiments have shown Mu toxicity symptoms to occur in tcbecco at 15 ppm Via (22), in ' cotton at 10 ppn\ (3), and in several leguT^es at less than 10 ppm (II3). Plants tsLke up Ito predominantly in the d.lvalent state. In neutral or slightly acid soils only a small fraction is in the divalent state. This fraction becomes larger as the pLi of the soil decreases. • The maximum soil pll at which Mn can be toxic seems oc be "^.^ (2). Molybdenum Deficiency A pH of b.5 is required for maxinrim availability of soil Mo. However, at this pH the availabilities of so'ne of the other micronutrlents may limit plant growth more than Mo availability except for the f^rovtti of Icguincs. Treating seeds with Mo befoi*e plantiu>3 in soils with low Mo content m;.y be more feasible than liming (^0).

PAGE 31

15 ?)eu t rail za tior. Soli l^rtid ity The most commou method of neutralising soli acidity is by the addition of lime either as calcitic Jdmestone (CaCCj) or dolomitic limestone (Cal-ISvCO^)^) • Calciuia. sJlicate (CaSiO.^) is also becoming a very popular Ij.rciag Material. " The reactions fcfiking place vhor lirce is added to acid soils can be divided into: (l) disso.l'n".ior? reactions, (2) cation exchange recctior.:, , ri,vd (3) neutralizatj on reactions. (1) Dissolution reactions are the hydrolysis of lime in the soil and according to Seatz, and Peterson (j.3^) ceui be written as: CaCO^ +li^O + COg ^ Ca(HC0,)2 CaCKCO^)^^ Ca'''^+ aHCO" The solubility of the limestone in the soil is dependent on the partial pressure of CO^; the greater the partial pressure of CO^ in the system, the more soluble the limestone. Dissolution reactions also depend, to a large extent, on the moisture content of the soil and soil temperature. As moisture increases, the air present in the soil is reduced and the concentration cf CO^ in the soil air increases resulting in increased limestone dissolution. Water is also a reactant in the first step of the dissolution reaction. Limestone reacts faster at high temperatures; this effect is probably related to diffusion rates of the end products av-'ay from the reaction sites (13^). (2) Cation exchange reactions take place when the Cations produced in dissolution reactions replace exchangeable Al-' or H+, if present, held on the exchange complex: XAl + 3/2 Ca^+ Gil 0 -z± XCa + A1^.6K 0 XH -I-^.Ca^ -v H 0 rr^ KCa H o"^ 2 3

PAGE 32

. 3+ (X repreocnts t,b.e exchange coniplox in the ^'eaction) . The Al .63.^ displaced undergoes hydrolysis with the production of H^o"* in solution until it precipitates as Al(OK) . 3tl 0: LAI.6OH2J + LAl'OK).50H2j + H^O [Al(0H).50H^r+ H^O [m{OK) ^Jm^ + H3O+ [Al(OH)^.llOH^j''+ H^O — ^ AlCOHj^.SOH^ + H^O"^ (3) Neutralization reactions are the last step in dissolution reactions when the end products of the latter react with the H^C"*" ions! in solution: H^O + HCO; Z=± KoCO^ + OH" H^o"*" + OH" ;z± aigO The rate of these reactions is directly dependent on the removal of OH" ions from solution. When the H 0"*" concentration in 3 solution is lowered, the dissolution of lirae is reduced and the overall process decreases (33)» Since .the concentration of H^O"^ ions in solution in acid soils depends on the hydrolysis rate of hydroxy -Al or hydroxy-Fe, lime reactions are also dependent on factors that influence the rate and extent of hydrolysis reactions (33)' Coleman et al. (35) observed q+ an increase in hydrolysis of Al-^ and hydroxy-Al with salts and dissolution reactions of lime. Coleman and Thomas (33) mentioned that a decrease in dissolution reactions cf lime occurs when intermediate compounds of Alhydrolysis are strongly held by the exchange complex of the soil reducing the rate of hydrolysis. Reactions of limestone are also affected by fineness of grinding, uiiifonin.ty of mixing with the soil, and the. water content of the soil (33).

PAGE 33

Soil AlUiTdimn and Iron Frac tions Tb.e importance of tlie diverse forys of Al and Fe in soil acidity., P sorption, end other soil chemical properties justifies . their classiflcatlcu and interpretution. Aluminum and Fe "become . . even more irriportant in tropical soils, rich in organic and inorganic forms, since they may be responsible for the majority of the chemical properties of these soils. The study of the Al and Fe fractions in soils is difficult in Yiev of the fact that the extractants used lack specificity; however, these extractants give an approximate estimate of ^ the nature axid lanounts of the predominant Al and Fe fractions. AmiA onium Oxalate F raction ... Acid ammonium oxalate (oxalate) has been videly used as ail extractant for amorphous compounds of Al and Fe. Saunders (I3I) indicated that oxalate attacks amorphous forms of Al and Fe and very small amounts of crystalline oxides and clay minerals. McKeague (97) extracted Al and Fe from some amorphous inorganic substances as well as from soil horizons rich in organic matter complexes of Al and Fe. He also mentioned that crystalline Fe oxides were not destroyed. McKeague and Day (9^) used oxalate to distinguish spodic horizons from other horizons rich in Fe. Saril and Bilton as quoted by McKeague, Brjidon, and Miles (99) found that oxalate extracts considerable amounts of Fe from magnetite which complicates the fractionation scheme in • soils rich in this mineral-. Citrate Dithionite Bicarbonate Fraction A^^ilera and Jackson (V) introduced a cltrate-dithionitebicarbonate (CDB) solution as an extractant of free crystalline Al and

PAGE 34

IB Fe oxides for soils being prepared for rainer&logical analysis. They found that CDB solut-ilizes larye amountE; ox' Al from hydroxy Al interlayers of venniculitic chlorite; hovever, giubsite was not attacked. Jackson (72) stated tnat CDB extracts Fe oxides without removing iron aluinj nosiiicate minerals, aluminum mineralSj gibbsite, amorphous iron-al\iniinosilicates, magnetite , and ilmenite. Mchra and Jackson (107) modified the original procedure and their modification is used a great deal in soil analysis tod-'iy. Sodium Pyrophosphate Fraction Aleksandrova (8) extracted h'omus and its Al and Fe complex " salts from soils using O.llA Naj^P^O.^ (Pyrophosphate). Bascomb (15) found that K-pyrophosphate extracts organic Fe and amorphous (gel) hydrous oxides but not amorphous (aged) hydrous oxides. McKeague (97) evaluated the pyrophosphate pi-ocedure. He snowed that some of the A1-, Fe--orgauic natter products were extracted, and that pyrophosphate dissolves a small proportion of the free Fe, but doss not dissolve the inorganic amorphous or crystalline Fe and Al. Sodium Hydroxide Fraction Hashimoto and Jackson (60) demonstrated that hot O.5N NaOH causes dissolution of free amorphous silica, free altunina, and large percentage of amorphous aluminosilicates provided a high ratio of NaOH volume -00 sample weight is used. Jackson (7I) indicated that gibbsite is also dissolved by the same treatment. Jackson (7U) e>rplained that Al, Si, and amorphous aluminosilicates form soluble Na-silicates and Na-aluminates in NaOH solutions. A;n:noniuin Acetate Fracti on AicLean, (leddlcson, find Post (lOij) found IN MLOAc at pH h.Q

PAGE 35

to be a tetter Al extractdat th^.n unbuf fere5 -sslt sclutions or soluivJ.ons buffered at pH 7.O and above. They stated tnnt, the Al extr«.eted with acetate (Acet.-Al) was primarily exchangeable. Pratt aiict Bair (123) working vitii acid soils extracted less Al from soils of lo^.-er pH, ar.>d more Al froji soils of higher pH with acetate as compared to 111 unbuffered solutions of BaCl^ or KCl. A freshly prepared Al hydroxide solid was found highly .'.^oluble in acetate. McLean (lOO) explained that acetate removes exchangeable Al plus more soluble or reactive por-cions of the hydroxy-Al ions and Al polymers. Pionke and Corey (l2l) shoved a high correlation between Acet.-Al and organic matter content in soils, and postulated that acetate is a good extiractant for Al associated with organic matter. McLean and Owen (103) also found a close correlation between Acet.-Al and soil organic matter. Igue and Faentes (70) released considerable amounts of non-exchangetible Al with acetate. P&rt of this Al was released from organically complexed forms. Potassiurg Chloride Fraction Coleman, Kamprath, and U'eed (34) stated that Al extracted with neutral salt solutions is trivalent, and, therefore, exchangeable. Tnoma,s (l40) mentioned that the Al extracted with neutral salt solutions was trivalent from pH 4.4 to 5.2. He found that little Al is extracted 0+ with neutral salt above pH 5.3. Lin and Coleman (86) displaced Al^ ions in amojmts close to exchange capacity from Al-saturated soils and clays with IN salt leaching. They concluded that KCl was the most effective displacing agent for rapid leaching, and that the Al present in the leachate was primarily exchangeable Al (Ex.-Al). McLean (lOO) indicated that rapid leaching is useful in the procedure so that the

PAGE 36

20 acidity formed does not dissolve other forms of Al than the exchangeac3.e. Nye et al. (136) studied the Al-K exchange equilibria and found that K vas bound tighter than Al over the entire range of ion saturation in IK solutions; KT^ ions in IH solutions were able to exchange Al-" . ions very readily confinning the work by Lin and Coleman. . ' Fractionation Scheme . Blunie and Schv/ertmann (21) showed that an approximate distinction could be niade bet-.-een amorphous forms of Fe and crystalline oxides by selective extractions of soils with oxalate and CDB, and that the ratio of oxalate extrac table Fe (Ox.-Fe) to CDB extrac table Fe (CDB-?e) could be used as a relative measurement of the degree of aging or crystallinity of free Fe oxides. McKeague et al. (99) recognized the necessity of a more complete separation of the amorphous Al and Fe fractions and proposed the following scheme;. . (1) Fe extracted with 0.1 M ria^P^O^ (Pyro.-Fe) = Organic Fe. (2) (Ox.-Fe) (pyro.-Fe) Amorphous inorganic Fe. > (3) (CDB-Fe) (Ox.-Fe) = Crystalline Fe oxides. They mentioned that the scheme is less useful in distinguishing forms . of Al in soils. Effect of Lime on Soil Aluminum and Iron The end products of the lime reaction are exchangeable Ca and M^, Al(0H)2, ^^^^3" ^'f^c^^^s^s to approximately 8.3 and complete base saturation is achieved. However, this is an ideal case since most soils are limed to reach pH values between 5.5 and 6.5. Under this situation, large amounts of titratable acidity and unreacted limestone remain. Aluminum and Fe compounds, along with nonionized groups on organicmatter and clay, are the sources of this titratable acidity (33). ..

PAGE 37

. _ • % ' ' 2 When liirte is added to acid soils, Lae f.XTbt ions lost from the soil ejre H and monomeric OKcitangeable Al. Thus, lime causes a sharp decrease in exchangeable Al, and increases hydrolysis of hydroxy Al and -Fe (33) • Aluminun polymerization is also increased as hydroxyAl molecules bridge together to fom stable structures (69) • The stability of organic matter complexes of Al and Fe has been shovn to decrease as the pH of the soil increases resulting in Al and Fe ions being released into solution. These ions are capable of entering hydi'olysis reactions and polymerization processes (85, 133) These cheuages in Al and Fe forms have been detected by a few workej^ using chemical reagents. Bhurnbla and McLean (l?) observed e decrease in pH dependent acidity vith lime. More Al was extracted with acetate than with M KCl from limed and unlimed soils; the differences between pH dependent acidity and extractable Al were less pronounced vhen acetate was used as extractant. McLean, F.eicosky, and J^akshraanan (IO5) attributed changes in peimanent charge CEC and pH dependent CEC caused by lime to inactivation of Al originally held by organic matter. Their conclusion was based on the high correlation between Acet.-Al and organic matter. Velez and Blue (l^vU) observed ' an increase in Cx.-Al with lime in a tropical and a temperate soil. A decrease in CDB-Fe took place in the tropical soil only. Ko changes were observed in Ox.-Fe and CDB-Al in either soil. They concluded that the amorphous fraction was the most responsive to lime additions. P hosphorus Availability in Acid 3oils la general, available P is less than 1'^ of the total soil • F at any given tiije. It is now generally recognized that P recovery

PAGE 38

by plants imraediately after fez'tilizer application is usually in, . . the range of 10 to 3^f^Tae resnainder which bev:orces available over long periods of oime is precipitated by soluble cations, immobilized by soil micro-organisms, or retained by the soil complex. Phosphate fixation is the conversion of soluble P to a less soluble form, thus reducing its movement in the soil and availability to plants (91). Phosphorus Fixation ' • . The availability of ? in soils is controlled by P fixation reactions. These reactions in acid soils can be placed into three -• general groups: double decomposition involving solubility-product • relations, adsorption, and isomorphous replacement (8l). Double decomposition reactions Tiiese reactions are also called precipitation reactions because the end products in acid medium are insoluble P corapoionds of Al-* and Fe-^ . In the case of Al-^ , the alurainosilicates, free sesquioxides, and exchangeable Al may be regarded as the primary sources of Al, and the OH ion concentration dominates the solubility product relation. In all three cases, phosphorus precipitation can be decreased by increasing the pH of the soil (8l). Russell and Low (I30) showed that adsorbed Al or hydrous oxides of Al present on kaolinite exposed surfaces precipitated P. According to Hemwall (62), not much P was fixed as AlPOi^ or FePOj^. He identified hydroxyphosphates of the form (Al or Fe) (HgO) (OH) .Ji,^FOl^ However, it has been assumed that P will form variscite-like compounds with Al (AlP0|j^.2H20) and strengite-like compounds with Fe(FeP0j^.2H20) witn the assumption that Al"* activity was limited by the solubility of gibbsite, and the Fe^ activity by that of goethite (87).

PAGE 39

Ads orption react ions • .... Two types of adsorption reactions are recognized: chemical adsorption and physical adsorption. Both types can be characterized "by adsorption isotherms (81). Chemical adsorption can be divided -i into non-specific and specific adsorption. When P ions are retained as counter ions in the diffuse double layer of positively charged surfaces, the adsorption reaction is called non-.speeific (63). Specific adsorption takes place when the P eaters into the coordination cf metal oxides to replace another anio?.i (63). Physical adsorption, in general, is the interaction of forces exerted upon each other by molecules, atcms, or ions; therefore, sometimes it is difficult to make a clear difference between chemical and physical adsorption (138). Physical adsorption has no significance 5.n P fixation according to Hsu (68) Coleman (36) obser^/ed an increase In pH in a clay -phosphate system after F fixation had occurred. He concluded that the exchaa^xe of P ions in solution for the OH ions from clay minerals or the hydroxides of Fe and Al increased the pH ol" the system. Dean and Rubins (hO) illustrated that anion exchange occurs between certain OH ions of the clay minerals or hydrous oxides and P ions in solution. . Kinjo, Pratt, and Page (83) showed that NO^" ions are easily replaced by P ions in soils from Mexico, Brazil, and Colombia. Isomorphous replacement reactions . • , ." These reactions are a continuation of specific adsorption ' reactions where the latter becomes isomorphous replacement of hydroxyls or silicate ions from the crystal lattice. A new mineral is formed through decomposition of the isomorphously transformed crystal lattice followed by recrystallization (61).

PAGE 40

• 2k Stout (137) reported the formation of ci-ystalline compounds in reactions of kaolinite and halloysite vlth phosphates. He mentioned replacement of Oil ions by phosphates as an intermediate step in the reactions. Wada (l't9) indicated that the reaction of amnonium phosphates with soil clays produced crystalline aromonium tarsmakite. Low and Blade (90) released considerable 3i in the reaction of kaolinite with P. They suggested that the 3i released was the res-vxlt of isomorphous replacement r H3u (68) pointed out that recrystallization of new compounds takes a long tirae to go to completion. In general, P fixation can take place in steps where more than one type of reaction occurs. Hsu (68) explained that P fixation takes place in two steps. The first step was considered as a fast adsorption reaction, and the second step as a slow decomposition-precipitation reaction. When amorphous Al hydroxides of intermediate size are present, precipitation and adsorption become indistinguishable (68). Effect of Aluminura and Iron on Phosphorus Availa bility As was pointed out before, Al and Fe compounds are the raw materials for P fixation in acid soils. The importance of such compounds has been discussed in the literature. Saunders (131) showed that P retention was closelj' associated with Ox.-Al and -Fe, and CDB-Fe in New Zealand soils. Ahenkorah (5) reported significant relationships between P fixation and CD3-Fe in soils from Ghana. He commented that the interaction between pH and CDB-Fe was minly responsible for the magnitude of fixation. No association with clay or Al. was detected. Yuan and Breland (163) observed good correlation between P retention and Al in Florida soils. The best correlations were obtained from oxalate and CDB extractions. Sh'okla et al. (136) shov/ed a close

PAGE 41

correlation betvecn ? sorption by lake sediments and Ox.-Fe. Vele2 and Blue (ihk) reduced P sorption "by 50^ after oxalate treatment of a soil frar.i Costa Bicaj but only 15'^ reduction was obsei'ved in an Ultisol fxxxa Florida. Phosphonis sorption vas reduced 50^ after CDB treatment cf both soils. Ballard and Fiskeil (iJ;) fc\md acetate to be the laost useful extractan-o of Al to predict P retention in coastal plain 3oi3.3 of the southeastern United States. Oxalate -.;as the most valuable of the extractants for Fe. It appears, from information in the literature, that amorphous forms of Al and Fe are more active in P sorption than crystalline forms, and that there is little distinction between Al and Fe in this regard. However, Uie role of pH is very important in differentiating the fcri?.5 active iu F fixation. Hsu (63) noted that at low pli (4.O) fixation is essentia3.3j a precipitation reaction vith Al-^ and Al hydroxides from crystal lattices; as the pH increases, amorphous Al hydroxide becomes more stable and phosphate is adsorbed at the surface. Fortunately, P adsorbed by amorphous Al and Fe compounds has been shown to be available for plant uptake (76). • Effect cf Idme on PhosphorusAvailab ility Trvioz (1^1) emphasized the enhancing effect of lime on P availability in acid soils. How-ver, Velez and Blue {ikk) obser\'ed an increa:3 2 in P retention after liming a soil from Costa Rica. ^ Downer (K-)) shoved that P retention was independent of lime in high P fixing Koils from Guyana. Kortenstine (65) found that the relative high P-fixstion capacity in a soil from Belize greatly increased as the soil riH. increased from lining. Woodruff and FCamprath (I56) reported a decrease in the P adsorption maxi;aa of five Ultisols irom

PAGE 42

26 North Carolina vj.th line as a result of reauctions in exchangeable Al; however. Blue (19) explained that the reduction in P adsoi-ption maxima vas not anj.rcnr; per mea of exchangeable Al neutralized, and that substantial P sorption capacity remained after neutralisation of Al. Reeve and Suinmer (125) failed to obtain a relationship between liming and P desorption isotherms in Oxisols from South ' Africa. Pratt et al. {l2k) pointed out that, in general, high ?e oxide ar-d less crystalline soils retained P the movst. It seems clear that lis-.e vill on.1^' decrease P fixation vhen precipitation is the main reaction taking place and exchanses,ble Al, crystalline aluminosilicates, and free sesquioxides are the main sources of P fixing compoimds as surges ted by Hsu (68). Studies have shovm that increasing pH to 7.0 and above vill reduce plejit growth and cause P deficiencies. Pierre and Browning (l20) found that veiy high P levels were required at pH 7.O to increase plant growth. Velez and Blue (1^15) increased yields of Pangola digitgrass on a heavily limed soil from Costa Rica only after additions of extremely high P rates. Fox, DePatta, and Wang (51) observed an increase in F uptal-.e by liming an Cxisol from Hawaii, high in Al and Fe, to pli 6.1. Liining to pH 7.0 markedly decreased P uptake by sorghum and desniodiura. This decrease vas attributed to the increase ' in available Ca which probably resulted in precipitation of P. Kamprath (79) suggested that P deficiency is Induced in soils with high P fixation capacity that are limed to pH 7-0 and above because of formation of Insoluble calcium phosphates. More efficient P uptake has been obtained in Ultisols and Oxisols with lime rates high enough to neutralize

PAGE 43

Z( excho.ugea'ole Al (51, I56) . However^ higher lime rates vould "be required to neutralisf? active Ai in other soils such as luceptisols even, though exchangeable Al raay be lev in aorne cases (lh6, I60) . . Effect of Calcium Silicate on Fhosyhoras Av ailability . Calcium silicate has been shown to be very beneficial to p3.ant3 in highly veathered soils ('+9). Suehisa, Younge, and Sherman (139). and Monteith and Sherman (112) increased ..sudangrass yield and PuptaJie on a highly veathered Hawaiian soil rich in Fe with tne addition of CaSiO-^. The beneficial effects of calcium silicate were attributed to improved P nutrition and decreased Al toxicity. Fox et al. (53) in Hawaii, obsei-ved an increase in sugar yields I'rom sugar cane using CaSiO slajr. Increasing rates of CaSiO , cn a P treated soil 3 gave increasing sugar yields, but little benefit was obtained by F appli cations alone . Large amounts of P and lime did not eliminate leaf freckle whereas CaSiO^did. Lucas and Blue(93) obtained iinmediate but temporary beneficial effects from CaSiO^. Calcium silicate, rice hulls, and dry Stylosanthes humilis herbage increased forage yields at the first harvest only. Forage P was increased by all treatments. Yuan (162) mentioned that even thougn Si is required by plants, Riainly in the grass family, other factors are also involved in the responses to additions of CaSiG^^ to soils, lie pointed out that CaSiO^ reduces soil acidity and Al toxicity, and that P is more available due to saturation of P-fixing sites with SiO.^~. Laws (8'4) reported that treating soils with Si decreased their capacity to adsorb P from solution. The quantity of soil P extracted by different solutions increased with the amount of Si applied. Deb and Datta {hi) showed . that silicate and organic salts such as citrate and tartrate markedly reduced P retention in soils.

PAGE 44

28 The Us e: of the Tangmuir Isotherm in Describi ag; Phosplioru s Aval lab i li ty Younger and Plucknett (l58)/based on results of a 6 year " e:
PAGE 45

bonding energy can be calculated once Vm io linovn (13)' In practice, Vai gives the P fixing capacity of the soil and b is used as an index of P availability. The I^igniuir equataon is based on the assumption that the cnei'gy of adsorption does not vary vith surface coverage. -1) This assumption is correct only at P concentrations between 0.5x10 M -k . > and 5.0x10 M (13); however Langmair type equations have been used successf-j.'lly to characterize linear ? sorption over a vide range of P concentrations (52). Woodruff and Kamprath (15^) calculated Vm for a number of soils and then applied P as Ca(H2f'0i^)2 at rates equivalent to 0, 1/8, 1/^^ 1/2, 1, and 3/2 of Vm in a greenhouse experiaent. Ifexiimxin growth of millet was obtained at P rates between \jh and 1/2 Vra except for a Norfolk soil vnere maximum yield took place at F rates equal to Vm. Soils vith high Vm vero able to supply sufficienc ? for grov/th at lover sat'jrationy than soi'ls vith lov Vm C'^-en vhen limed. Fox and Kamprath (50) used P sorption curves to adjust the soil solution concentration between 0.01 and l.^^^ ppm in greerJiouse studies. . Phosphoi-n.13 vas applied as a solution of Ca(:l2P0j^)2. Pearl millet growth approached relative growth when P was 0.2 ppm in the soil solution. This soil solution concentration to obtain 95'^ relative growth was later called the external P concentration requirement by Fox (^9). lie stated that P sorption cur\'es can be used to determine fertilizer requirements of highly v;eathered soils high in P fixing capacity if the external P concentration requirements of crops were known. He pointed out that soil test methods based on extraction pf available P may overestimate the P status of soils with high fixing capacity, and, unless calibrated to account for the latter, will underestimate fertilizer requirements by a wide margin.

PAGE 46

Elec trocheraical Frorerties of A.cld Soils _ Soils can be divided into two general groups based on the electrochemical behavior of their colloids: (l) Constant surface charge (CSC), and (2) Constant surface potential (C3P) {ihz) Constant surface charge soils are dominated by permanent charge colloids such as montmorilloni'ce and vermiculite. Temperate soils belong in this group. Constant surface potential soils are most commonly referred as pH-dependent in terms of charge because of the variation in CEC vith pH (108, 143). Mekara and Uehara (IO8) adopted the terra C3P since charge variations are not limited to changes in K"*' and OH" " concentration in the soil, but also may be caused by adsorption of • ' ions such as POj^ , 3i0., 30^ , and various organic anions . Highly weathered acid soils of the tropics,, rich in oxides and hydroxides of Al and FCj and soils dominated by amorphous materials belong in this group (103, 1U3) . , _ /. „ , . In C3P soils, the potential ic determined by potential determining ions (PDI) (82, IO8, 143). Parks (II9) defined a potential determining ion as any. ion capable of establishing the surface charge and the potential of a reversible oxide or hydroxide. Van Raij and Peech (1^4-3) considered H and OH as the predominant PDI in soils; however, Mekaru and Uehara (108) included P0|^, SiO-^, and 30j^ along with H and OH as the most conraion PDI in tropical soils. The Isoelectric Point and the Zero Point of Charge ' • ' According to Parks (II9), tne zero point of charge (ZPC) is the pH at which the solid surface charge from all sources equals zero. Isoelectric point (lEP) is a ZPC arising from Interactions of H*, OH", the solid, and water alone. In other words, the 2PC is a weighted

PAGE 47

avera/je of the TEF's of its components. At the lEP the density of positiva charges eqija.ls the dtnirji^y of negative charges, and the net surface charge equals zero. The IE? cfiuals the ZPC only In the absence of adsorbed species different from PDI. Van Raij and Peech (1^3), and Keng and Uehara (82) defined the ZPC of a soil as the pH at which the net surface charge is zero. Thus, ZPC is a measurable property of soils having C3P. The £;oil components will determine the ZPC of the soil, and in turn, the ZPC of the components vill be the weighted average of their lEP's. Soil comijcnents having CSP and, therefore, ZPC include ciystallins and amorphous oxides of Al, Fe, Ti, I4n, and Si, as well as kaolinite, halloysite, allophane, and most probably talc and pyrophyllite . Quarts, one of the major soil components, also belongs in this category (82). Vaii Raij and Peech (I'V?) pointed out that the ZPC of peraanent charge components of soils is -"-ery low or does noc . exist at all. However, most soils are considered to have a mixt-ure of CSC ar.d CSP colloids, and the ZPC will reflect the dominant type of oolloid present (82). . Meas^rre^ent of the Zero Point of Ch arge -J Parks (119) observed that when a solid with finite cation exchange is immersed in a medium containing H"*" ions, partial dissociation of the positive counter ion (e.g. Na"^) and partial replacement of Na" by H takes place. The extent of dissociation (size of charge) depends upon the Ka"*" ion concentration [Wa"*"] and h"^ ion concentration [ll"*"J. Increasing [h"*"] at constant [Na"*"! results in further replacement of Na"^ by h"'' and fewer dissociations of if*" , At low [Na"*"], the negative charge decreases with decreasing pil. At hJ "rh [Na"^]," the

PAGE 48

-1• exchangeof Ila" cy K' is reduced, and the chai-ge is less pll sensitive. This observation confiirced the use of potenticiaetric titration curves &t different ionic strengtiis as a valuable tool in determining the ZFC of soT-ids, Van Raij and Peech (1^3) used this technique for soils and found thai the ZFC was the pH of the conirnon poi.nt of intersection of the titration cur\'-es. Parks (119) called these curves H^adsorption isotherms and shoved that they followed the Gouy Chap.-nan model for electrical double layers. Van Raij and Fetch (li;-3) indicated that the Stem model gave a better estiraate of the overall charges in soil systems^ and that the surface potential due to charges vas determined by a Nemst type relation vith the pri of the bulk solution and the pH of the Z?C. Keng and Uehara (82) showed an equation relating surface charge (CSC) to pli by combining the Gouy Chapman eqiaation for surface charge vlt?i the Nemst equation for the 3ur"^"ace potential given by Van Raij and Peech. Their equation describes the electrochemistry of many tropical soils when it is net as^uned that the surface charge is constant. In this case, salt concentration gives the value of the surface charge at a given pH, czid the sign of the charge is obtained by the difference between the pH of the ZPC and the pH of the bulk solution. This difference also affects the amount of surface charge present. . As was mentioned above, the common point of intersection of potentiometric titration curves at different ionic strengths gives the ZPC of a soil. The same principle holds when surface charge is plotted against pll. The slope of the cui-ves is a good measurement of the buffering capacity of the soil (82). The separation of individual curves, the change in pH' with salt concentration, indicates the adsorption

PAGE 49

c-apacibjr of the soil since the intensity oi' charr.c is d^^pendent on the ionic strength accoif-diiig to double layer theoi-y (II9). Wide Sv'j paration between curves denotes strons adsoi'ption for the ions present in the supporting electrolyte soJ.uticn, while curves that are close together iraply low adsorption. The distance between the ZPC and the pH of the soil measured in the supporting electrolj'te solution alone is highly correlated with the amount of perinanent charge present in the soil (1^3) • In general, these type of curves are very useful in describing some of the most important chemical properties of soils. Effect of wSoil Mineralogy on th e ZeroJ.^joint_ of Charge Van RaiJ and Peech (l-'i3) found thct the presence of Al and Fe oxides will increase the ZPC of the soil toward higher pH values. Clay silicate minerals and organic natter will tend to decrease the• ZPC to lower pH values. In general they obser^'-ed that the electrochemical behavior of the tropical soils studied was similar to that of netailic o::ides in which the surface potential of the reversible double .layer is determined by the activity of PDI H*" and OH' in solutic Parks (119) indicated that increased crystallinity, dehydration, dehydroxylation, and structural charge of clay minerals decreased ZPC, while surface composition and cleavage habic increased 2iPC. He mentioned that permanent chai'ges on clay minerals are independent of pH until the latter exceeds a basic limit, after which the charge increases only if the adsorption capacity of the surface is small to allow for dissociation of cations and exchange reactions : . which are a measure of increasing charge as in oxides and hydroxides. Effect of Cation and Anion Adsorption on th e Zero i^ojnt of C hange Keng and Uehara (82) observed a decrease in the ZPC of Hawaii

PAGE 50

soils meatsured using CaCl,^ as ^\\^ aupportiij,^ electrolyte in coiupai-ioOD to NaCl. Ills change in pH per 10 fold increaoe in electrolyte concentration vas greater with CcCl,^ than with KaCl. They interpreted these resu].ts as being caused by specific adsorption of Ca^"*". Calcium ions were strongly preferred to H"^' so that a decrease in pH vas observed and changes in pH were greater as the CaCl2 concentration decreased because of the effect of electrolyte concentration on the double layer thickness. Parks (II9) found that cationic species increased ZPC. This is the case when there is no specific adsorption of cations taking place. Hydrogen can replace ths cations present or be adsorbed on the surface; an increase in pH is observed. He also pointed out that an excess of specif ica3.1y adsorbed ionic species will remove the pH dependence of the curve or change the ZPC to that of the species. This is common when IN C&Ol^ is used in measuring ZPC and Ca"^ is specifically adsorbed. The potentiometric titration curve is shifted toward more acid pH's, and additions of acid or base cause small changes in pH. Specific adsorption of 30^" was suggested by Keng and Uehara (82) for the increase in ZPC of Hawaiian soils with the use of NagSOi^ as the supporting electrolyte when compared to the ZPC in NaCl solution. A greater change in pH per 10 fold increase in electrolyte concentra2 tion also took place. In thir. case, SOj," were preferred by the soil ovsr OH" causing an increase in pH on the basic side of the curve. On the acid side, 30^" replaced coordinated OH ions and pH increased. Parks (119) reported that anionic impurities or pH Independent PDI's decrease ZPC. This observation is explained by the fact that nonspecific adsorption or anions occurs on positively charged surfaces leaving an excess of HT*" in solution, and, thus, decreasing pH on the

PAGE 51

acid side of the titration cui-ve. On the basic side, OH" reacts with dissociated H~ and the pH remains lass basic than predicted. Practical Im p lications of the Z er o Point of Charge Keng and Uehara (82) discussed the practical importance of the electrochem.Lstry of tropical soils. They mentioned that the slopes of the curves were steeper with Ca salts than Na salts. This explains why lime is not very efficient in raising p^^ in some of these soils. The need to use the Stem double layer model suggested that most of the soij. charge is found in the Stern layer, and only a small fraction of the total charge can be used for retention of cations and for soil dispersion. This would explain why cations that are not specifically adsorbed are more likely to leach out, and why some oxidic soils disperse with great difficulty even at high pH's and have very good physical properties in the field. The suggestion was also made that Ca~ is held with very high affinity in the Stern layer and is not completely removed with a single neutral salt extraction or IN NHj^OAc at pH 7.0. The small change in pH with lime added in large amounts to a soil low in CEC, and the small amount of Ca in readily exchangeable forms in these soils have sometimes been attributed to leaching. They can also be explained by high charge development and strong C a retention. Adsorption of SOj^ ions was shown to decrease the ZPC of Hawaiian soils, and, therefore, increase negative charge (63, 82, 108). This increase in negative charge avoids leaching losses of cations. Phosphate is even more efficient thaB-S0|^ in increasing negative charge (108). Mekaru and Uehara (108) snowed that each miM of PO, /lOOc,

PAGE 52

tjoroed incrciS^A 'ration retention by about 0.8 rDGo/iaOg. The greater response of plants to CaSiO,^, can be expie.ined in part by adsorption Of SiO^ on the soil surface arid subseauer.t increase in viatica retention Cation exchange capacity measurements by the usual procedures give unrealistic values for C3P soils due to changes in the net charge caused uy concentrated salts and suboequent dilution by washings usir^ vater and alcohol (82). By measiiring pH in m KCl and E^O, and then taking the difference, a good description of the net charge is obtained. When the pH in KCl is greater than the pH ir vater, the soil lies on the positive side of the ZPC and it has net positive charge. IVhen the pH in vater i3 gieat^r thai, the pH in KCl, the soil lies oj. the negative side of tile ZPC and is negatively charged. When the pH in vater and KCl ai-e vei-y close or equal, tne soil lies very close to the ITC and it has a very s:tic:11 net charge or none at all (l^j). Mekaru and Uehara (108) called the difference between pU in KCl and water delta pH (Capd) . They found that non-specific anion adsorption is very prominent in soils with a positive ApH. Tliey pointed out that pH iceasure.-nents with IN K„3Ci^ soraetines give positive values because of increased exchange reactions of SO^^ with OH ions. An indifferent electrolyte must be used when the main objective is to determine net surface charge. Calcium Selectivity In Acid Soils / : One of the most important steps in the overall line reaction in acid soils is the exchange of Al-^ by Ca'provided bythe limir^ material. It is through this reaction that the exchangeable acidity is neutralized. Soils having low affinity for Ca^"^ ions will require

PAGE 53

37 ?,.3rger cuaounts of lime for Gcid neutralization than soils vith higa afx'jnity. Leaching losses of Co. are much iriore pronounced in soils vith poor Ca selectivity, and most of the exchan^^eable acidity remains untouched until high lirae ratos are used. It has also been obser/ed 2+ that vhen Ca ions are present m the soil solution in large amounts due T.O overliniing or low Ca selectivity, electrolytes associated and includirig Ca will accumulate, depressing pl".nt growth temporarily {lh6). Therefore, it is of the most importance to study the capacity of soils to adsorb and ho3.d Ca in the presence of other ions in solution and on the exchange complex. .... Several models have been proposed to describe ion exchange processes and predict iou distribution in soils. These models include the Freundlich and Langmuir adsorption isothei-ms, the Donnan equilibriiira, the diffuse double layer theory, and the lav of mass action (42, 155). Deist and Talibudeen (42) felt that the lav of mass aciion described equilibrium reactions in soils better than the other models since it takes into account all ions taking part in the exchange. Wiklander (155) justified the use of the mass action model in view of the knowledge regarding the structure of the diffuse double layer and its relationships to the bulk solution dependent on the activity of the counter ions and its connections with changes in free energy. Tiie constants obtained usii^g mass action models are considered as selectivity coefficients rather than constants since they vary with the mole fraction of the ions in the equilibrium solution (55, 95, 155), the ratio of the adsorbed ions to the total exchange capacity, the ratio of the ions in the equilibrium solution to t!ie total normality of the solution, and the normal concentration of the equilibrium

PAGE 54

solution (38). Selectivity coefficients, however, have been used to predict the replacement of a particular icn by another ion on the exchange complex, or to give an indication of the relative strength through vhich an ion is held on the exchange complex (38, 53, 95). . • The model proposed by Gaines and Thomas (.55) based. on a rigorous thermodynamic treatment of the law of nass action has beer, used successfully not only to study cation selectivity but also to investigate the variation of the therraodj-namic parameters in soils. Nye et al. (lib) found that Ai-' was preferred over k"^ in dilute solutions (0.051^ and below); however, was nore strongly bound to the soil in IIT solutions. Clark and Turner (30) also showed that in U solutions K*" and Na"*" were preferred over Al^ . Deist and Talibudeen (k2) observed a preference of soils for Ca^ over k"^ in O.Om solutions. The CibC of the soils did not remain constant and decreased with K saturation. Rhoades (12?) reported that soil venniculites did not show strong preference for ^^6^'^ over Ca^"*" vnen compared to other vermiculite samples. Coulter and Talibudeen (39) found a strong preference for Alover Ca^ in O.om solu::ions and acid soils. Coulter (38) reported that k"*" was more strongly adsorbed than Al^^ in O.OU^ soiucxons. Some K vas not exchanged by Al-^ due to fixation mechanisr Selectivity coefficients remained constant over the range of cation saturation. The oretica l ' " " " An exchange reaction can be described as raP + pNtvm = mPXp + pM, and the equation for the selectivity coefficient can be written

PAGE 55

39 s (30), vhere ? and M are cations with valences p and m respectively, X represents the exchange sites, C r inMXm + pPXp, and K' is the selectivity coefficient. Parentheses and rectangular brackets express liiolar activities and concentrations re'lative to the total voJ.ume of suspensionEquation (jLj can be transformed inlo the following equation using equivalent fractions instead of concentrations. (1 C/Co)P (q/qo)"-' m / , vHi / / \P m-p 111 (C/Co) (1 q/qor Co K' [2] (38) where K' is the distribution coefficient; and Yp are the activity coefficients of ions M and F in solution; Co is the normality of the equilibriiim solution; qo is the total exchange capacity of the soil; (C/'Go) is the equivalent fractiorj of ion M on the exchange complex. This equation is the equivalent fraction equation proposed by Gaines and Thomas (55) where K' is a function of Co and of the relation between (C/Co) and (q/qo). Talcing negative logarithms, equation [2I is transformed into the follov;ing equation, * (C/Cof -l (q/qof pK' = log log [3] (1 C/Co)P Co?-™ (1 qyqo)P m or pK' = S Q M (38). Conventional exchange isotherms give the graphical relationship between (C/Co) and (q/qo). This relationship shows ion preference when (qyqo) is greater than (C/Co). A derived isotherm is the graphical

PAGE 56

relatioash-ip between Q and S. The S intercept gives pK' for the reaction. Derived isotherms are lines of unit slope if K' is constant, and in dilute solutions the intercept does not change vith Co. Conventional isotherms have a marked dependence on Co (38), The equivalent fraction equation has been used in soils saturated vith a given cation to study the replacing 'ability of other cations. In natural soil systems, cation pairs can be studied for the same purpose by leaking (l q/qc) as che equivalent fraction of the companion cation on the exchange complex, qo as the sum of all exchange able cations present, and (l C/Co) as the equivalent fraction of the companion cation in solution. The companion cation is the one that will influence the selectivity of the soil for the cation of interest. Effect of Li r.e oh Electrolyte . " Accumulation in Acid Soils "Hie presence of e3.ectrolytes after lims applications to acid soils has been v/ell docximented in the literature. Russell and Richards (129) reported leaching of considerable anounts of KO^ from virgin soils. Midgley (109) observed a marked increase in water 2+ soluble Ca , HCO^, and NO^ ions by overliming acid soils. Walker and Brown (I51) showed accumulation of NO^ up to 55O ppm with lime rates as high as 6t/acre. Even lime applied at 3t/acre gave NO^ values close to 550 ppm. Ogata and Caldwell (II7) also detected high amounts of NO" (500 ppm) after soils that had been limed with I6 t/acre were left fallow for 2 years. McLean (102) stated that the concentration of electrolytes in soils increases with dissolution of lime. He added that in soils high in CEC, electrolytes disappear from solution as

PAGE 57

COg volatilizes, but in soils high in anion exchange capacity (AEC), the increase in OH concentration neutralizes positive charges forcing other anions such as 30j^ into solution. These SO^ ions may couple 2with Ca ions to increase the amount of electrolytes in the soil sclution. Helyar and /uaderson (61) found a large increase in Ca'^ and SO" ions in the soil solution of limed soils frora Nev Zealand. However, and ions decreased in concentration with increasing amounts li(he. Velez et al. {2.k6) reported substaiitial increases in 2'* _ 2vater soluble Ca , IIO^, and 30^ ions as the lime rates applied to a soil fi'on Costa Rica vere increased. In general., e.i.ectrclyt.e accumulation in acid soils occurred after high lime rates vere applied. In a few cases, small amounts of lime caused a marked increase. Zantua and Blue {l6k) mentioned accumulation of electrolytes as the possible cause for the depression by Ij.me of Pangola digitgrass yields in a virgin soil from Costa Rica. Leaching both the unlimed and limed soil increased yields dramatically. The increase was greater in the limed soil. Almost identical results were obtained by Midgley (109) with different plant species when soils from Kansas were overlimed. Ke pointed out that new seedlings were especial3.y susceptible to injury, and that coarse-textured, acid soils that had not been cultivated for several years usually produced the greatest injury. He observed that individual electrolytes (Ca^^, HCO", and NO") caused no injuiy, but failed to account for all possible electrolytes. The importance of the combined effect of individual electrolytes was demonstrated by Velez et al. (ikb) . Ttiey found marked increases in electrical conductivity (EC) with lime; values harmful to plants were detected in a surface soil from Costa Rica tiiat had been under pasture for a few years, after incubation.

PAGE 58

Besides acconwlation of electrolyten, autritioaal disturbances associa'csd with the presence of excessive araounts of ions in the soil solution brought about by large applications of lime are often mentioned as the cause for reductions in yields. Pierre and Browning (l20) rspcrted disturbances in P m'.trition of several crops as the major cause of excessive liming. JIaftel (II5) 'found 3 deficiency in cverliiaed soils. Yields dccrcciscd sharply vnen Ca saturations in the soil were higher than 'J'^-fi. McLean (lC2) mentioned that high Ca, adsorbed or in solution, alters tne balance of cations such as micronutrients, K, and Mg. The breaKdo-wn of cation balance is reflected in the uptalie of these nutrients by plants as vas shown by Helyar and Andei-son. Naftel (ll4) and Pierre and Brcvning (120) reported that decreases in available P due to excess Ca in solution causes severe P deficiencies even in soils with high levels of -.j&sily extractable P. 2 High levels of Ca associated with CI" and KO^ in soil solutions were shown to be lethal to orchardgrass (150). According to Eaton (kS) 2 CI salts are often associated with accuraulation of SOi^", dCO^, and 2CO^ . Burning and firing of leaf tips and margins, bronzing, pre-. mature yellowing, abscission of leaves, and chlorosis are sj-mptoms of excess CI". Zantua and Blue (16U) observed some of these sympto.r.o on Pangola digitgrass grown on a limed virgin soil from Costa Rica. 2High concentrations of CO^ and HCO" in the soil may have caused •, 3 J) :,.>:: phytotoxicity . Iron chlorosis in many plants has been associated with 'dCO'^ concentration in soils (122). The depressing effect of excessive lime on plant growth was shown to be temporary in all cases (109, 115. 1^6, l6k) . Plants usually grew well after the first harvests, without a good explanation as

PAGE 59

. k3 to v/hy. Velez et ai. {iko) postulated that the, temporary .harnifal effect of lime nay be closely related to ttia persistence of high levels of FCO~ ''ons in the soil and the amount of 307. and K0~ ions released 3 i by limed soils through organic matter decomposition.' ' ^ As vas mentioned above; overliming 'is responsible for the temporary harmful effect on plant growth. Acid soils' of the tropics tend to be overlirr-ed vhea recomiendat j ons are based on raising soil pK to 6,'; or even 7.0 as is the case for most temperate soils. This condition can be avoided if recoinmendations are made to use lime only to climinat-e toxic substancetj or to provide enough Ca for normal plant growtn (79, 102). Iiidicator Plant for Greenhouse Studies Pangola digitgrass ( Pigit aria decumbens Stent) had a consistent negative response to line in soils from Costa Pvica (91, 1^6, 164). Tiie botanical characteristics of Parigola vere described by Hodges et al. (6^'l-)» Adequate fertility is required by the plant for sustained grovth {'Sk, l'43)_, and its K requirement is high; however. Gammon (5^) shoved that more than 60^ of the X can be substituted by Ila without growth "reduction. The optimum pH for Pangola on acid flatwood soils of Florida according to .riodges et al. {ok) is 5.5; however, they pointed out, that vigorous growth occurs at low pH's {k.2 to h.^). This is in line with reports by Lotero, Monsalve, and Piamirez (SS) in Colombia. They found Paiogola to be resistant to soil acidity, ilortenstine and Blue (66) reported pll 6.3 as optimal on Pule tan loamy sand. Blue et al. (20) obtained appreciable response to N fertilizer on newly cleared land 'in Costa nica after slow establishment of Pangola. Ahmad, Tulloch-Reid, and Davis (i) also obtained significant responses to K on soils from Trinidad.

PAGE 60

• . • ' hk Pangola vus Gbown to be sensitive to Ca aiid P deficiencies (18, ohf Sh, Ik'}). However, line has depressed yields in tropical soils with oo.ae evidence that r availability was adversely affected as veil. Blue (18), and Hortenr.tlne and Bhxs (66) foiuid an indication that the response of Pangola to lime on a soil from British Honduras vas dependent on applied P. A larger responise to P thr3in lime was documented. Downer {k'p) obtained yield depression from lime on a brown sand soil from Guyana. D-icas and Blue (^k) observed that increasing lime levels decreased Pangola growth on an Entiscl from Costa Pica regardless of the rate of P applj.ed from 0 to ^1-50 ppm. Phosphoras concentrations in oven-dry forage were below the critical level of 0.16=*; proposed oy Andrew and Robins (lO). Zantua and Blue (l6i;) reduced yields by liming a virgin soil from Costa Rica. Vicente-Chandler (ikj) obtained positive responses to lime only 5 years after application in Puerto Rico. Figarella et al. {hS) reported no yield response or effect on P concentration of Pangola with P rates from 0 to 14-00 kg/ua. Ahmad et al. (7) did not find a significant response in P content or yield on soils from Trinidad. .

PAGE 61

• . MATERIALS /IKD METEODS ' "-' ' ~ ' Soil San^ples " ' . v Soil samples from Costa Rj ca and tiie e-istern savannas of Colombia vera collected during the last veek of August and the first week of September 1972. Five sites were selected in Costa Rica and four in Colombia. Profile samples vere taken at each site for soil characterization purpose-. Surface samples (0-l3 cm) were also collected at each site for laboratory and greenhouse studies. Surface and profile samples from Costa Rica were named after the closest city to the sa:npling site. Profile and surface samples from Colombia vere named to give key locations around the Carimagua experimental station^ where sampling was performed. Table 1 gives the soil identification number, its location, and Its soil order according to the U.S.D.A. soil classification system. Descriptions of the sampling areas, sampling sites, and soil profiles except for Los Diaraantes were made by Rodriguez (M. Rodriguez -Gomez. 197^. Lime-micronutrient studies vita soils from Costa Rica and the eastern Llanos of Colombia, rh. D. Dissertation. University of Florida, Gainesville). Lucas (91) made the same descriptions for Los Diamantes soils. Hereafter, soils will be referred to by tae location of the sampling site as shown in Table 1. '45

PAGE 62

Table 1. Identification number, Iccation, and order of the soils ctudied Identification n^.imber C osta Rica X ;.Los Diamantes (Suri'ace) Inceptisol 2 Los Diamantes (Subsoil) Inceptisol -3 San Vito Ultisol k San isidro Oxisol 5 Grecia Ultisol 6 Alajuela Ultisol Colombia 7 Agronomy Area Oxisol 8 Cararao Ultisol 9 Drainagevay Oxisol 10 East Carimaeua Ultisol _ . ... .. ..... .^^"ii location -order

PAGE 63

L£"boratory Procedure s Gen errtl Cheiiiical Prope rties ' ^ " Soil reaction (pil ) Soils were "equilibrated in IN KCl and in water for 1 hour at a 1:1 3cil to solution ratio. The pli vas measured with a combined calomel-gla'ss electrode attached to a Corning model 12 pH meter. Organic ma tt er (6m) . . The wet oxidation method (Walkley-B7Lack) described by ' Allison (9) vas used to measure easily oxidized C. A factor of l.'JZh v.'as utilized to convert C concentration to CM. • Cation exchange capacity (CF C ) Cation exchange capacity was determined by the saturation method described by Chapman (23). Measurements were T.ade at pH k.8 and 7.0. V/ater and 99^ ethyl alcchcl were used to wash ezicezs salts after WJ^ saturation. Adsorbed KHjJ" was displaced with Ka*^ frc") an acidified 1C> Y.s.01 solution and determined through distillation with a Iloskins steam distillation apparatus (25). For this purpose, 30 ml of acidified solution containing the displaced were neutralized •' with 20 ml of 50;i KaOH solution, and m~. v/as tracced in ""O ml of H^BO^ containir.f~ indicator. • • Klec trlcal condactivlty (FC) The saturation extract method for soluble salts explained by Bower end Wilcox (23) vas used to measure EC. Plxtractable soil nutrient s Lxcnangeable bases (Ca, I-Is, X, Wa) were detennined in the pH k.Q Wd^ -saturating solution used for GEO measurements after

PAGE 64

leaching 2^0 la]. through soil saraples. Aluminum^ Fe^ and P verc also deterrtilned in this leachate. Exchangeable Al was extracted as suggested by McLean (lOO). Available P vas extracted by the Bray II method ( 2h) . Water extractable NO^-N and SO]^-S were assayed using a 1:5 soil to water ratio. The suspension was shaken for 30 minutes in 3 reciprocal sha;^er; a few drops of a saturated KCl solution were added to aid in 'fiocculatioh. The suspension was then centrifuged (2,000 rpm for 5 minutes) and filtered through No. h2 vrnatTian filter paper. Nitrate was measured coloriraetrically as suggested by Brenner (26) J and 50^ was measured grarimetrically usi:7g the metuod outlined by Hanna {59) • . • Total el er.i rintr.l analysis • • Total Nutrients (K. Jlaj Ca, ^g, Fe. Al, and p) were extracted following tne HClOii-HF digestion method outlined by Jackson (TI) . Min era lor i c al Analy s i s ' ^ Sample preparation for mineralcgical analysis was made according to Vhittig (153) sxid Jac cson (72)Qualitative analysis of the clay minerals present was made by X-ray diffraction using a General Electric XRD-7CO instrument witn Ni-filtered CuKoc radiation. Quantitative analyses for caolinite and gibbsite were made vdth a Fer<;inEljner D3C~2 differential scanning calorimeter. The amounts of iiaolinite and gibbsite in the clay fraction were calculated from their respective differential thermogram peak areas (72). The amount of vermiculite was estimated based on the peak area of X-ray dif fractogrsims (l53), and amorphous components of the clay fraction were measured oy loss of weignt after treatraeno with hot 0.5N NaOtI (60).

PAGE 65

The Mg saturation method (72) was used to measure the CEC of the c.lB.y fraction. Tne CEC was taken as the milliequivalents of Mg per lOOg of clay displaced with IN NaNO^. Soil Alurainum and Iron Fraction s Amorphous Al and Fe fractions were determined by the aiiaioniuin oxalate (pil 3.25) method (131) and the hot O.5N NaOII method (60). The method proposed by Mehra and Jackson (IO7) was followed to measure the free crystalJ.ine oxides of Al and Fe. Organic amorphous fractions of Al and Fe v;ere assayed with the O.IM Nai^P^O,^ procedure described by. McKeague et al. {99). The fractionation scheme proposed by McKeague et al. (99) followed to characterize amorphous inorganic ^ amorphous organic, and free crystalline oxide fractions of Al and Fe. Lime Requirement ... , . The amount of lime needed to raise soil pH to 6.8 v;as determined by the 3MP buffer (135), f^nd by Yuan's buffer (161). The exI changeable Al method (TT) was used to determine the amount of lime required to neutralize the exchangeable Al. • Sources of Soil Acidity ' Total acidity was measured by the BaCl2-TEA method modified by Zelazny and Fiskell (165); however, the suspension was only equili. brated overnight instead of 2 days as suggested. Fotentiometric titrations were carried out following the procedure outlined by Zelazny and Fiskell (165). The procedure was modified by using different soil to IN KCl ratios and difi-erent normalities of titrants (B^lt)^ and Na^Bi^O,^) in order to avoid excessively long titrations. Sample weight varied from 0.2g to lOg,

PAGE 66

End the volume of IJ^ KCl was kept constant. Titrant normalities varied, frora O.ca^ to 0.2H. A U] KCl soil extract was prepared by incutatins lOg of soil in 50 rtil of IJ^ KCl for h days. This suspension was filtered into £i 100-rnl voluraetric flask and the volume of the liquid was made to 100 nl by leaching the samples with IN KCl after the first filtrate was collected.This extract was titrated with 0.05N Baf 011)2. titrations were made with a Sargent Model D Recording Titrator with a Sargent combination electrode No. 3-30072-15 Suspensions for conductometric titrations were prepared in the same manner as suspensions for potentiometric titrations . In this case only, 0.2TJ Ba(0H)2 was used as the titrant, and no IN KCl soil extracts were titrated. A Model 75I conductivity meter frora Universal luterloc. Inc. was attached to the recorder of the Sargent titracor with an appropriate conductivity ce]l to measure conductivity The cell had a constant equal to one. The conductivity meter was calibrated zo read 5 on the recorder chart at a range of O.Cl^niho, and set at a range of 10 _^mho for samples. T:ie total chart expansion was 10 units. The recorder was run at 5OO mv chart span, and the rate was set to slow (I/3 inca,minute) . ' .. Barium hydroxide titratable acidity was measured poteutiometrically at pH 5-5, 7-0, and 8.0 wnile NagB^O.^ titratable acidity was measured only at pH 5.5 and 7.0. Conduc tome trie titrations were allowed to proceed until the slope of the curve was equal to a deionized water blank titrated with 0.2n Ba(OH)^. The end point was taken as 2/3 the amount of titrant in meqy lOOg required to reach the slope of the blank to account for Al(OH)r formation (^i3).

PAGE 67

All acidity measurements vere correlated v;ith soil chemical properties and soil Al and Fe fractions to study the degree of associa tion among variables in order to make predictions of vhere this acidit vas origiiiating . Phosphorus Adsox-ption Phosphorus adsorption as determined by the method of Hortenstine (65) vas called an intensity measurement of adsorption. Phosphorus sorption was taken as the difference betveen the amount of P added in solution (50ppm) and the amount of P lest from solution (soil basis) after eq^uilibration of a 1:50 soil to solution ratio. Phosphorus sorption was also expressed as the percent of the original amount of P added which vas sorbed by the soil. The quantity measurement of adsorption was the adsorption ; maximiXLr and the constant related to the bonding energy obtained with a Langmuir type linear equation. The method discussed by Fox and Karaprath (50) was followed to determine data points, and linear regression analysis was used to find values for the intercept and the slope of the regression equation. Added P ranged from 0 to yCO ppm in solution. The methods were divided into intensity and quantity to differentiate the amounts of P adsorbed at different equilibrium times. Only I6 hours of equilibrium were allowed with the intensity method, while 9 days were allowed with the quantity method. Correlation analyses were utilized to study the degree of association between P adsorption, soil properties, and Al and Fe fractions in order to estimate the origin of the P sorption capacity.

PAGE 68

Amounts of lime needed r,o raise soi.L pK to 5.5 were added to 500g of 30il. Liciea samples vere iucuta-:ed for 10 weeks in plastic bags at moisture levels close to the raaxinam water holding capacity but low enough to avoid sataraticn. Tliis Moisture level is also kno^'ra as field capacity, although it was not calculated experimentally After incubation, P adsorption was determined by the quantity method. Electrochemical Properties " . ; rThe sign of the net surface charge was obtained as deccribed by Mekaru.and Uehara (103) in 1:1 soil to solution riitios. The ZFC was determined bypotent iometric titr-stions following the method outlined by Van Raij and Peech (1^3) • The supporting electrolyte . was CaClp at concentrations of l.OH, 0.3^, O.OM, and O.OOM. The acid side of the titration -curves was evaluated by adding 10, 5j 2, 0.5, and 0.1 ml of ClIJ HCl. The basic side was assayed by addi^ig 5, 2, 0.5, and 0.1 ml of O.IH NaOH. Measurements were also taken in the presence of the electrolyte alone. San Vito soil required 15 ml of acid plus the other treatments for a better evaluation of its ZPC. A combined calomel-glass electrode attached to a Corning model 12 pH meter was utilized for all pH measurements. Calcium Selectiv ity , • ' The procedure described by Clark and l^'j.rner (30) was used to develop conventional adsorption isotherms for Ca, and the equation reported by Coulter (38) was utilized to calculate the Ca selectivity coefficients. This equation represents a straifrht line of unit slope when the selectivity coefficient remains constant throughout the range of cation saturation; however, this was not the case for the soils

PAGE 69

analyzed, aad the Ca selectivity aoefficieut vas calculated forcing the slope af the line to unity. Lhis coefficient vas the mean of all coefficients over the entire ran/je of Ca saturaticn for a given syctem. Ad.sorption isotheraif. and Ca selectivity coefficients were evaluated for equilibrium solution concentrations of 1.0 and O.OOIK. Calcium-Kj Ca-Mg, and Ca-Al systems were studied. The Ca saturation of the . ecLuillbrium solution ranged from 0.1 to 0.9, and the com' panion cation saturation from O.Q to 0.1. The sum of Ca equivalents and coKipanicHi cation equivalents vas equal to the concentration of tne equilibrium solution. Tlie Ca saturation vas the equivalent fraction of Ca in solution. The results obtained with the 0.00 IK equilibrium •. solution set vere discarded due to failure of the procedure to detect sraall changes in Ca adsorption since three extractions with ITi NIL, NO 4 3 vere made instead of the five suggested by Clark and Turner (30). Incubation Studies with Lime Reagent grade, powdered CaCO vas applied to 500g of soil 3 at rates of 0^ I.5, 3.0, 4.5, 6.0, 7.5, 9.0, 10. 5, 12.0, and 13.>raeq./3.00g Limed sairiples were incubated for 10 veeks in plastic bags at moisture levels close to the maximum water holding capacity but low enough to avoid saturation. After incubation, samples were allowed to air-dry and saturation extracts vere obtained as described by Bower and Wilcox (23). The following measuirements were made in the saturation extracts: K, Na, Ca, Mg, EO^-N, and EC. Exchangeable Al, P sorption (intensity method), pH, and CM were also evaluated on this set ->r samples by tiie procedures discussed previously.

PAGE 70

Analy tical D et errainations . . Atomic absorption spectroscopy vas used to analyze Ca,Mg, Alj and Fe in the different soil extracts described. Potassium and Na vere determined by flame emj.ssion spectroscopy. Nitrate-N vas assayed vith a NO^ specific ion electrode attached to an Orion specific ion analyzer unless stated otherwise. Phosphorus was measured by the ascorbic acid method of Vatanabe and Olsen (I52), and S0;^-3 by Henna's gravimetric procedure (59)* ' Greenhouse Procedure 3 v. . General Preparation of Experiments . Reagent grade, powdered liming materials (either CaCO^ or CaSiO^) were mixed vith soils in a ro-oary mixer in sufficient amounts to give desired lime rates. Moisture in the limed soils vas adjusted close to the maximum water holding capacity but low enough to avoid . saturation. Limed, moist samples were incubated i'or 6 weeks in the greenhouse. After incubation, samples were air-dried and mixed with other fertilizer materials in a twinshelled blender to give experimental rates of lime and nutrients required in the experimental designs. One kilogrsun of soil with the proper treatment was placed in a plastic pot previously filled with coarse gravel to about 2/3 its volume. . . Pangola digitgrass stolons were collected and placed in distilled water to encourage new root formation. Three grass shoots with recently initiated growth were planted per pot. When new growth was observed, the soils were fertilized vith solutions of chosen nutrients other than those involved in the experiment. The amount of water required to

PAGE 71

keep the pots at a surLtable raoioturc reginc for plaut ^rovth van calculatsfl by veighirig a sample pot from each soil before and after adding vater in quantities large enough to avoid saturation and leaching. Plants j-rers cut 3cra above the soil surface at Ji-week intervals for a total of three han/ests. Nutrient solutions were added to all pots after each harvest. ' . . The fresh plant material for each harvest was placed in a drying room at 'TOG for 3 days; eftei-wards^ it vas veighed and ground in a 20-me3h stainless steel wiley mill. One gram of the ground sample from each experimental treatment was ashed at ^OOC over night. The ash was moistened with a fev drops of deionized water and 15ml of 5N HCl were adaed to the samples. This suspension vas brought to dryness on a hot plate. Tne residue received 20ral of deioni zed. water and 2.25ml of 5N HCl; then it was heated' on a hot plate to boiling, and filtered into a 50-nil volumetric flask. The solution was brought to volume with deionized water. Phosphoi-as in plant extracts was measured colorimetrically by. the ascorbic acid method. Calcium and Mg were determined by atc.iic absorption spectroscopy, and K and Wa by flame emission spectroscopy. Experiment Ko. 1 A second order, central composite, rotatable response surface design was chosen for this experiment in order to study the effect of independent variables lime, I^lg, and P on pangola digitgrass yield and natrient contents, and to predict the rate of these variables needed for maximum response. The design included 8 factorial points (two Icve

PAGE 72

56 Table 2. Relation of treatment, nnmber to experimental and coded variables for Experin-.-.it No. 1 Variables Experimental Coded Treatment number GaCO-, Mg P CaCOMg P meq/lOOg ppm^ 1 2A k8 122 -1 -1 -1 2 9.6 k8 122 1 -1 -i 3 2.k 192 122 -1 1 -1 k 9.6 192 122 1 1 -1 5 2.h 48 ii.79 -1 -1 1 6 9.6 U8 479 1 -1 1 7 2.k 192 U79 -1 1 . 1 8 9.6 192 479 1 1 19 0.0 120 300 -1.68 0 10 12.0 120 300 1.68 0 0 11 6.0 0 300 c -1.68 0 12 6.0 2H0 300 0 1.68 0' 13 6.0 120 0 . 0 0 . -1.68 6.0 120 600 0 1.68 15 6.0 120 300 0 0 0 16 6.0 120 300 0 0 0 17 6.0 120 3000 0 0 18 6.0 120 300 0 0 0

PAGE 73

and three variables) located at. the vertices of a cubr, 6 poixitf; spaced at distances 1'1.682 from the origin along the three axes of the cube .. .. in order to make the design rctatabie, and a point at the origin replicated i|tines (58)' 'i'i^e experimental design vas completely randomized. , " LiEie as CaCO^, Mg30^, and finely ground ordinary triple . super phospiiate were the fertilizer materials. Lime treatments were selected to explore the experimental region between 0 and 12 raeqylOOg; the level or lime estimated to give maxjiiTom response (6 raeqylOOg) was a mean of the lime required to bring pH to 6.k in all soils calculated with the 3MP buffer method. The remaining lime treatments followed design requirements. Magnesixim levels were chosen to give a 10:1 Ca . to % ratio based on the amount of Ca present in the lime treatments selected. A region between 0 and 2k0 ppm of lAg was studied. Phosphorus levels were adjusted to explore a region between 0 and 600 ppm; the level of P estimated to give maximum response (300 ppm) was based on the optimu;ii level for pangola yield found by Lucas (91) in soils from Costa Rica. The relationships between experimental and coded variables is shown on Tai)le 2. Multiple regression analysis was used to examine the response, and interpretations were made based on the regression coefficients accordir.g to Hader et al. (58). Axl pets were fertilized after each harvest with solutions containing 100 pp.-n of N and K. Micronutrients were added using 30 ppm of 5'rit 503 at the beginning of the experiment. . . Ex periment Ho. 2 . . Tee exx>eriraental design used was a split plot factorial arrangement with 2 lime levels as main plots, 5 P levels as subplots.

PAGE 74

mv\ 2. replications. Lime e,s CaCO.^ and finely ground triple super phosphate vere the fertilizer sourue.'i. Lime levels that gave maximum response in Experiment No. 1 vere selected. PJiosjihorus levels were adjusted according to the P adsorption maximum calculated with the quantity method in order to obtain 0, l/l6, ih, 1, sxxd. 1 1/4 saturation of this maximum. The purpose of this experiment was to study the po. tential of the quantity method to predict fertilizer P needs of Pangola in unlimed and limed soils, and to examine the value of Experiment No. 1 to predict lime rates. The amounts of line and P used appear in Table 3' All pots were fertilized after each harvest with solutions containing 100 ppm of N, K, and I4g. . . ' Experiment No. 3 " " ' _ A completely randomized block design was utilized to compare CaCO_ and CaSiO as lime sources in soils containing adequate P levels -> 3 for Pangola growth. A level of lime that gave good response in Experiment No. 1 was compared to a level that gave poor response. Phosphorus levels that produced maximum response in Experiment No. 1 were applied as finely ground ordinary triple super phosphate. The primary objective of this experiment was to study the behavior of CaSiO^ as a lime source and its value in increasing the availability of P applied. The amounts of CaCO^, CaSiO^, and P added in this experiment are shown in Table k. All pots were fertilized after each harvest with solutions containing 100 ppm of K, and Mg. ' ^' ' . .

PAGE 75

Table 3« I.tae and phosphoiia;? trsetnents yor ixperinent IIo. 2 59 Treatraeut.3cil CaCOIjOs Diaczantes (Surface) Los Diaasiites (Subsoil) meqy lOOg 0 10 0 10 0 k20 1675 6700 8375 0 390 1560 6250 7810 San Vito O 7 0 h20 1675 6700 8375 San Isidro 0 k 0 370 5900 7375 Grecia Alajuela 0 k 0 18 0 1925 7700 9625 0 kSo 1925 7700 9625 Agronomy Area 0 6 0 137 550 2200 2750

PAGE 76

60 Table 3 Continued Treatments Soil CaCO^ P !neq,/l00g ppm Cai*arao 0 h 0 187 750 3000 37^0 Drainageway 7 n \J 500 2000 8000 10000 East Carimagua 0 20 0 52 207 83c 1037 i i J

PAGE 77

Tii'ole k. Lirae, calcium silicate, and phospMorus treatments for Experiment No. 3 Treetments Soil CaCO^ CaSiO^ P meqy'lOOg ppm -"1 r\ 450 10.0 10. .0 1 n J500 4.0 4.0 OcUi VlOO 1.0 1.0 600 5.0 5.0 cin ± s J. ur o 0.5 0.5 650 10.0 10.0 Grecia 2.0 2.0 500 24.0 £4-0 Alajuela 1.0 1.0 650 8.0 8.0 Agronomy Area 1.0 1.0 . 605 2.5 2.5 \ .-. Cararao 3.G 3-0 • 450 16.0 l6.o • DrainEigeway 10.0 420 24.0 24.0 ' V East Carimagua •1.0 '.• 1.0 625 8.0 8.0 -

PAGE 78

RESULTS AKD DISCUSSION ' " Laboratory Experiraents General Chemical P roperties • Soils from Costa Rica had pH values higher than ^.0 in a 1:1 soil to vater ratio (Tatle 5). Samples frora Los Diaraantes gave the highest pH values. The measurements ranged from pH 5'1 to 5'9« Lower water pH's vere determined in soils from Colombia with a range between k.2 and i)-.9 pH values in 3^ KCl were lower in all cases than those measured in water suggesting the presence of exchangeable sources of 3+ acidity such as Al as mentioned by Thomas (lUo). Using the pH in vater and according to Jackson (73)j the soils frora Costa Rica can be classified as very weak acids except for San Vito which fits the category of a weak acid along with all the soils from Colombia with the exception of Cararao that can be classified as a strong acid. The OM contents of the Costa Rican soils are relatively high (TabT.e 5); however, Los Diamantes subsoil and San Isidro were not as high as the otners. A lower OM content was measured in the soils fi-om Colombia except for Drainat^evay which contajned more than any other soil. Organic; matter was probably the major activs material in the majority of the soils frora Costa Rica and in the Drainageway soil frora Colombia. ' ' • ^ Cation exchange capacities were high for the Costa Rican soils and the Drainageway soil from Colombia (Table 5). All soils

PAGE 79

=1o 0 o o .-j 0.1 00 o H CI CJ rH r-i • • • • » d d O o d o o O O o CO t!0 O O H S o fH ir\ ir\ O 8^ CO CVJ m r>o CM CO CO CO CO H CO ON H O CO -i. CO c Cvl « • • • • o O ON a* VO OJ H H H H H H CM o CO t-CO o CO VO CO o H o O OJ rH o 1^ CO CM 0! CO -4^ CO oo VO CO CO -rio OJ VO ON CJ OJ UN VO Cvj cr\ o CO n o 0) o MS CO o M X» n to o p > 05 m o H to n CO CtJ H O 05 H Q) OJ o n o u O CO u oJ ^* (0 o •H 0) (•5 03

PAGE 80

shoved CEO values highly dependent on pH. The pH dependency of CEC was very raarked in the samples from Costa Kica and the Drainagevay sample from Colombia. The high amount of amorphous materials found in the clay fraction (Table 9) and the high CM content explained the increase in CEC with pH in the soils mentioned. Electrical conductivity (EC) was found to be somewhat higher in the soils from Costa Rica (Table 5): althougn Cararao and East Carimagua from Colombic showed values similar to those found in the Costa Rican soils. However, all soils were found tc be low in electrolyte concentration based on EC measurements. In general, the soils from Costa Rica had relatively higher pH values, and higher OM content, CEC ana EC than the soils from . Colombia except for Drainageway which gave values similar to those of the Costa Rican soils. . . _ . Extractable Soil Nutrients Extractable bases (Ca, lAg, K, and Na) were much higher in soils from Costa Rica than in those from Colombia (Table 6). Calcium dominated the distribution of exchangeable cations in the Costa Rican soils, hlagnesium was substantially higher than K and Na except for Alajuela soil. Exchangeable Al was predominant jn the soils from Colombia; however, San Vito and Grecia shoved Ex.-Al values considerably higher than those fo-ond in the other Costa' Rican soils. Extractabl bases and Ex.-Al values were correlated with pH and EC values. Soils with high pil and EG were high in extractable bases and low in Ex.-Al. ••

PAGE 81

-P o a M X Q) 0) 03 I ca M CO o -4s o cn I o r-l 03 6 ft Pi o O o vo OS VO O • • • • « • t-CO o CO vo H CO ro H CM m OJ CVJ « ro H vo VD CO m • oo CO c • H —J • vo CO • 6o.o m • CVl 6 ? CO VD -4 JOJ CO VD o o O O o o o O o On H MJ H CVl OJ o Cvl -4 H H PO H o C-VJ OJ H O O o o O O C o O O o O CI J. — I Lr\ o -Jr-l H OJ O 0-1 LTN O o o O O o o O iH o o O o OJ s VO VO o c> [~.. CN CN PO OJ i-H r-l rH o H o o O rH o . O o o ON -rt cn -:t OJ r-l OJ CO o O o o O 0) i) +> •H « 5 o 03 cn. o to CO o O ^ •P -H •H W > M CO CO OJ 03 r-l CD rH & o o o (0 u a u as CJ a •H U 03 I B T^ ct5 o p d

PAGE 82

Phosphorus extracted Kith IN Nilj^OAc (pH k.8) vas relatively lov for fA,ll soils (Table 6); hov/ever, F exti-Ji-ted with the Bray IT. solution varied considerably and vas much hl.(;ht'r tnan P extracted vith IN NH|^OAc (pH k,8). Nevertheless;, the extractable P status of all soils was considered extremely lew. The rc-ugents present in the Bray II solution are generally recognised to extract more P from Al and Fe compounds^ and also from Ca compounds {2k), vhile UJ NHj^OAc (pii U.6) extracts mainly the P found in amox-phous Al and Fe fractions 103, 121). " _ • Water extractable SO^-S vas relatively high in all soils except in Agronomy Area (Table 6). Highest amounts were found in Los Diamantes subsoil. A high degree of variation in SOj^-S occurred and a clear grouping of soils by country was not evident. The araount of water extractable NO._^-N was relatively high in Lcs Diamantes surface, Alajuela. and Cararao soils; all other samples were low. V/ater extractable SO^^-S and NO^-N were not individually associated vith other soil parameters although their sum related to EC. The cation and base saturation values shown in Table 7 supported the information given on extractable bases and Ex.-Al. Soils from Costa Rica were higher in base saturation and lower in Al saturation than soils from Colombia; however, base saturation did not reach in the Costa Pican soils ami was below l^'ji in tae Colombian soils at pK 1(.3. Considerably lower values were obtained for base saturation at pli 7.0. These values weixclosely associated with thoae shown in Table 5 for CEC and reflectrd the p!I dependency of CEC.

PAGE 83

TSie C/K ratio was fcund to be adetx-iate for minerali'iation of K (120) (Table 7). Alt.hougl?. uxe majority of the Colombian soils and Greeia soil were higher than 15/lj K i'rmcoilization is not likely to be abnorrnal. . , -v-. The Ca/Mg ratio for all soils was coasidered relatively. narrow although adequate for plant nutrition (128). The Colombian soils had lower ratios than the Costa Rican soils except for Grecia. The Mg/K ratio was high in the Costa Rican samples except for Grecia and Alajuela. soils, and very Iom in the Colombian samples except for Cararao. A Rg/K ratio of 3.0 is considered adequate for plant uptake of these nutrients; however, Russell (l?8) explained that in considering exchan^^eable ca^oion ratios as means of assessing nutrient uptake by plants, one has to take into consideration the crop of interest, the total concentration of cations in the soil, the tj'pe of clay, and the relative contribution of the clay and OK to the base-holding capacity of the soil. Re pointed out that, in genei-al, increasing Ca in the soil depresses Mg in the leaves more than K; increasing Mg depresses Ca more than K; and increasing K may depress Ca more than ' Mg, or may decrease them about equally. • :• ''r.':'";'' Ttie Ca/Al ratio was very low in all soils from Colombia (Table 7), High ratios were calculated for Los Dia-Tiantes soil, and somewhat lower for other Costa Rican soils. Low ratios were used to indicate the dominant effect of Al on the soil exchange complex while high ratios save a relative index of the dnminant effect of Ca. These values were not meant to provide information on plant; uptake, although

PAGE 84

(d o u o o •H t+J c3 U -P 03 CO 0) w to p< 03 o a> Q CO o CCi r-i CO H rH H H • • • • • « i J0\ o O S O CO CM CO CvJ On CO • • « • « • CM CM _-t H o o rH O CO CVJ OJ ^jO ITN O vo f• « • • * • CM OO H CM OJ H rH OJ CO CVJ CM CO CO JOJ • « • • o H CM CO fO ^o OJ VD r-l rJ rH r-i rH r-i (—1 rH OJ OO CT\ H LfN CO CO -d• • • « • 00 ON ccj VD H , — f r-i On CO LTN CO H H 1 — C7\ CO (J) Q CO CM JdO OJ H H o O CO * O OO • • « • • rH c vo OJ H CO o -f ON OO t — OO CN 0^ • • • • H H H H c H rH 6 U-N _::} _:}• CJN CO CM OO « • • 0-) OO CM H c^ CM CM CM CTN vo LTN LTN OJ CO O VD OD VO vo CO -t o H H O Lr\ CVJ VD LTN CO fO o O t • • • CO CO O C7N Cvj cr, H LTN CM CM CM CVJ OJo a tin u o w . o rJ +' CO tH > to O •H W M o 0) o 05 r-) 0) (0 H o c o o 03 (< 05 U o •H 0) •H M 03 O d

PAGE 85

liming soils vith a Ca/Al ratio greater or eauul to 1 tias been discouraged (19, ^9, 77) because the pH is usually higher tharx 5.^1and the Al is considered inactive or neutz'alized at this pH ('f9). Total Nutrients ^ ' •' Los Diamantes soil was found to have a larger reserve of nutrients than any other soil studied (Table 3). Total K was very low in 33X1 Toidro; total Ca was low in 3an Inidro and all the soils from Colombia. Total Fe was considerably lower in the Colombian samples as v/as total Al; these soils were also very low in P reserves except for Drainagevay. In general, soils from Costa Rica^ except for San Isidro, had a larger reserve of nutrients than soils from Colombia. According to their total elemental analysis, soils from Costa Rica appeared to be less weathered than the Colombian soils. Analysis ot' t he_Cla y Fraction The predominant crystalline minerals in the clay fraction of the soils from Costa Rica were intergrade vermiculite, kaolinite, and gibbsite (Table 9); however, only trace amounts of kaolinite were detected in Los Diamantes subsoil. The clay fraction of these soils was dominated by amorphous materials. The soils from Colombia showed relatively high amounts of vermiculite alon^: with kaolinite. Relatively small amounts of gibbsite were found in these soils, and the amorphous components nf the clay fraction were much smaller, except for Drainageway than in the Costa Rican soils. Other minerals present in all soils, except in Los Diamantes, were quartz and feldspars. Quarts war, found in sif^nificant amounts. X-ray dif fractograms ijave indications ttiat raeteihulloyGite was also present in the Costa Rican soils. Intense peaks

PAGE 86

70 t — 'JO Cvj CO cn CO H H 'r-i o o H o . o o O • • « • * V * • • • O o o o O o O o o d H CO iH ON VO • « • • • ON ON 00 H O H VD CJ H < 00 a. H o VO • VO cr> ON CO o CO o I-l CJ CO VO OJ ON VD VD CM ON CVJ o CO O o O O • • * • • • • H cj O o O o o o o o 05 O LTN Jo H on ro no CM VD CO o CV) CM o O O O • • m • • • CO d o d O d o o o oj CO CO OJ CO CO OJ CO • • • H H VO CO l/N -=} o LTN CO CO ON ON o OJ OJ OJ CO b-N « • • • • d d O o d d o o o d 0) rH m Eh o 10 0) I •H Q w 3 o o O tn (0 o +> •.H > CJ to O CO o OJ u o 05 H 0) 05 I O a o H o 05 05 U 05 O 05 a OS 65 •H 05 CJ P to 05 W

PAGE 87

j' • vere observed at h.k, and 3o9A; the ^.'.A peak sloped tovard the high aojle side of the dif fractogram. Hovever, differential ther;r.ograms failed to show the arsociated water e;
PAGE 88

72 Table 9« Minerology and cation exchange capacity of the clay fraction froiK the soils studied M inerals identified Amorphous Soil , . Vermiculite Illite Kaolinite Gibbsite materials CEC ineqy ICC] ^ of zhe clay fraction Los Diamante s Surface 10-20 6 3 69 35 Gut ceil Trace Trace 3 81; San Vito 10-20 13 16 a6 22 San Isidro 10-20 2U \ 20 11 Grecia 5-10 22 % 19 16 Alajuela 5-10 20 18 26 12 Agrcnoniy Area 30-1+0 5 17 Trace 9 XX. Cararao 30-1+0 5 13 Trace 8 lif Drainagevay 10-20 i; y 3 Trace 38 13 East Carimagua 30-^0 5 16 7 11 9

PAGE 89

of bases, 'i'his obser
PAGE 90

1 I o u >» o X -.-4 o to o o CO o O A LTV H d to 0) (1) -p O •H o O i tr) J-1 •H ro CO d o -A r-i « o a) d H o ON H o VD o no • > d o H O H d o 4-> o > M CO CO o u o cd r-l d d o VO oo 00 ON • • CO l-Pv H VD H H H H I o a o u to o u Ej o vo ro <-H d IJ• • • O H H CD o a •H OJ o oo v£) o ON VO oo vo CO OO 0'-) • • • • • fl « • oo O (O VO i/N vo o H • • * • d d A C c rH o OJ IfN C ON ON H VO OO 00 • oo cj O P. H n] O -P V) 05

PAGE 91

75 The orgrjdc Al rraction (Pyre. -All did not separ-ite the soils into vtll dsfined [-roups (Table 10). All soils, except East Carina^Taa, had large aii.ounts of Pyrc.-Al. In sone cases this fraction vas larger tiiajn the Ox.-Al fraction indicating the inportance of the organo-rietailic complexes. Such was the case for San Vito, San Isi-drOj A^^ionoxcc/ xirea, Cararao, vna. East Carimagua soils. McKeagus (9?) found that scae of the free oxides are dissolved by pyrophosphate. The variation obser'/ed could have been caused by dissolution of free Al oxides. A better grouping of soils was obtained using KaOH as an Al extractant (Table 10). The soils . from Costa Rica were much higher in I\IaOH-A3. than the soils from Colombia except for CraJ-nageway . This vas the same pattern observed for soil OM and amorphous materials in the clay fraction, and is explained by the selective dissolution of aiDorphoas aliirainosiiicates obtained vitb. hot 0,5N NaOK (6o).. Ir on fractions , . . / ; Similar trends as those described for Al vsre observed for Fe (Table 11). In general, soils from Costa Rica were higher in Fe fractions related to OM and amorphous materials than soils from Colombia. I>rainageway, however, was similar to the Costa Rican soils. The CD3-?e fraction was extremely high in most of the Costa Rican soils, and the KaOH fraction was rather low in all soils. Jackson (72) indicated a preference of CDE for Fe over Al, and Hashimoto and Jackson (6o) showed that hot O.pN IlaOfl prefers Al more than Fe. Characterization of Iron and Aluminum Crystalline oxide forms of Al were dominant over amoi^phous forms in most soils (Table 12). The major exceptions were Los Diamantes

PAGE 92

76 I O a? o u I CO o 1' ! o o o to o O tf) (U -f-> 3 o G O CO CO t-CO o r-K o o o o o o o • « • o C C o o o 6 d o O Jo d o O o d o o o 1-4 o •H eS O o O :< -i -P :.o Vi tr. CO oj to •H O C3 H <-3 < r-4 « O d CO o O 55 < I" o o o CIS u •a ea o H o CVJ * o . v"' CO H cu VD ^O cu • « • OJ o IP* H -if H H H d cu CO CO ON o CO CO H o H o o o CO o o • • • • • * d o o d o o o o o d aJ 3 f^ o c cd a

PAGE 93

77 •p o in o & o 0 h 1 ° * •0 o S -H o a o 0) •H (U H fi, -P r-l < 0) &4 o CO o p § •H Q (0 o CJ * • • * t • • > o o d o o rH O c o d CJ On o r-l CO Cv) If • • • • r-( 1-1 r-l O r-! O H H O i-i d d o SI n o Cd I-i o o H f o JON CO • • • • o d H o o o d o o w CO o CO 6 CO o 0) O -dr-: • « o o d rH H CO in CM CO CVI • • • • • • o o o o o o o a 0) o a o u o CJ O a •ri a CVi CJ o J0-) H H • • • • • • oj ITS H H CVl H O c CO CM OJ r-l • • • • cj .Joj H u a +> to

PAGE 94

78 subsoil frcra Costa Rica where t.-ie amorphous inorsanic forms dorr.irtated, and Cararao from Coiomcia which showed a larger proportion of Ai la the amorphous organic form; however, the ainount of amorphous Al was significant in all cases with the organic fraction dominating the inorganic fraction except in Los Diaaantes subsoil. Crystalline oxide forms of Fe were much higher than amorphoun fonas in all soils (Table 12) . Amorphous inorganic forms were larger than amorphous orgarJlc in soils from Costa Kica except Los Diasantes surface. Anorphous organic forms of-Fe predominated in soils from CoD.onbia a:-ccept Agronomy Area. ' * In general, all soils were high in amorphous forms of Al and Fe. . These compounds are considered very active in soil acidity, cation and anion exchange reactions, specific anion adsorption, and several other drienical and physical properties common to tropical soils. Their contribution to some of the processes mentioned will be discussed in some detail later. ; Lime Requirement The amount of lime needed to neutralize Ex.-Al was considerably lower than that required to raise soil pH to 6.8 (Table 13). There was very little difference in lime requirement by the 3MP buffer and Yuan's buffer although the latter gave higher estimates especially for soils high in OM. Both of these methods take into account the acidity and buffering capacity of acid soils. The acidity to be neutralized from the buffer equilibrium pH to a desired pH is determined by the buffering capacity of the soil; therefore, lime requirement tests using these two

PAGE 95

79 Table 13 . Lime requirements far the soils studied by the exchangeable alunilnun, 3MP, and Yuan techniques Lime requirement :x>;-;^. J;---1^ 3M? • Yuan Soil j; Ex.-Al W (pH6.8) (pH 6.8) CaCOj, mt./ha Los Diamaates Surface 0.3 15.0 19,0 Subsoil 0.1 10.0 ^ 12.3 San Vito 2.2 ^ l8.6 22.6 San Isidro 1.0 ik.o 19. 7 Grecia I.7 ' ' ; ' ; ' 18. 6 21.7 Mejuela 0.7 I7.2 2h.9 Agronomy Area 5.3 14.8 I6.8 Cararao 7.9 / 17-2 20.1 Drainageway k.O , . 22. U 39. l^ East Cariraaiiua I.5 3.I 2.7

PAGE 96

80 bafi^srs shciild give values considerably higher thaa the Ex.-Al method i-i soils liigh in buffering capacity. Altho'j^h the neutralization of Al is the priaiary function of lii!iin(j^ Yvian (l60) pointed out that the increase in CEC by liming to a soil pH close to neutrality would greatly improve the soil chemical conditions; nevertheless, success in limj.ng tropical soils has been limited to the neutralization of Ex.-Al (19, 51, TTj 126)\Ihen additional arounts of lime have been used, a temporary condition harmful to several plants has developed (109;, 1^6). Velez and Blue (1^1-5) and Zantu.a and Blue (l64) showed that overliming damage was time dependent in one soil; yields were iraproved subsequently in greenhouse experiments. Sources of Soil Acidity Potenti ome trl c t i trat ions . ' • The potentiometric titration cur\'-es for soils from Costa Kica were characterized by lack of inflection points from the initial titration pH to cH 7.0 (Figure l) . A charp increase in pH occurred, and then, a buffer zone appeared from pH 7.0 to 3.0. Los Diamantes Eurfece and subsoil shewed very diffused curves common to amorphous materials (^3). The lack of inflection points, the sharp increase in pH below 7.0, and the small amounts of acidity titrated in the KCl extracts were attributed to the small amounts of Ex.-Al present and to weak acid • character. The slope of this first portion of the curves was caused mainly by dissociation of wea}:ly held H ions such as those present on the exchange complex and on carboxyl and phenolic groups of OM, or H ions ci-eated through the decomposition of Al and Fe complexes (96, 13j). rae ouffer zone above pll 7.O was brought about by dissociation of il ions

PAGE 97

81 . ~ Added Base (meq/IOOg) Figure 1. Potentiometric titrations of soils froia Costa Rica.

PAGE 98

82 froKi organic matter j;\mctioaai grocips £:nd pciy?,luirJ.noiiydroniuru ed^es (73). Tlie titration curves obtained with were very close to those obtained with Ba(OH)p except in Los DianiEintes soils. This indicated that polymeric and corauleK Al, not titratable with Na^Bi «i if 7 were an important source of acidity in these soils. la general, no strong acid character was observed, and the sources of acidity seemed to be associated with Al and Fe amorphous soil ccnponents such as OM and organo-metallic complexes. The titration cur\-es for the Colombian soils were very different than tiiose for the Costa Rican soils except Drainageway (Figure 2) . An initial buffer zone was present with an inflection point around Si" . pH 5«5« A second inflection point was observed around pH 7.0 after a rather rapid increase in pa from 5.5 to 7.0. An additional buffer zone appeared between pH 7.O and 8.0, but was not as well delineated as in the Costa Rican soils. Considerable amounts of acidity were found in the KCl extracts. Drainageway had titration cur-zes very similar to the Costa Rican soils; however, the initial slope was not as steep, and the buffer zone was extensive, indicating more buffering and stronger acid character. The initial buffer zone common to most Colombian soils was attributed to hydrolysis of large amounts of Ex.-Al (43). The other portions of the curves were caused by the same sources of acidity discussed for the soils from Costa Rica. Titrations with Na2B|.0 indicated that polymeric and complexed Al were important sources of acidity in Agronomy Area and Cararao soils. In general, the Colombian soils displayed stronacid character, and the sources of acidity were mainly protons yielded by Lx.-Al in tie hydrolysis process. The acidity

PAGE 99

83.

PAGE 100

8!* of the I'i'o.iriagew&y soil vas attributed, to .E;c,-Al as veil as anorplious soil components. Conductoraetrie titrations The roie of Ai compounds in soil ecidity vas emphasized by ccnductoT-etric titrations (Figs. 3 and U). In all cases, the titration curves failed to detect the presence of H ions. Acidity from Al coriipounds was the only type observed. Since the Costa Hi can surface soils were low in Ex.-Al, other active foms of Al such as a;tiorphous compounds including complexes vith OM vere capable, of yielding protons to create acidity. On the other hand, Ex. -/J. vas the predominant source of protons in the Colombian soils except Drainagevay, In the latter, both, exchangeable and amorphous Al contributed to acidity. The tern anjorphcus Al inclxides organic forms, hydroxy Al formed as an intermediate product of the hydrolysis reaction, and poiyaiuminohydronium ions created by polymerization of hydrorvy Al. Titratable acidity Small amounts of acidity vere measured in the Costa P.ican soils takir^g pH 5*5 as the end point of titration and using Ba(0H)2 as the titraat for pctentiometric titrations (Table lU) . Tne Colombian soils except East Carimagua, had considerably higher amounts of acidity than the Costa Rlcan soils. Drainagevay soil yielded much more acidity than any otaar. A similar pattern occurred taking pH 7.0 as the end point. This distribution broke dovn at pH 8.0. los Diaraantes (surface and subsoil) and East Carimagua vere alvays lover in acidity than the other soils. Acidity increased markedly from pH 5.5 to S.O in most soils froiu Cojta Rica except Los Diomantes. The largest increase tooic place

PAGE 102

36 (ujD/soLjiiJLU) A|!Aipnpuo3

PAGE 103

«7 lYom pH 7,0 to 8.0. These increases were caxised by the buffer zone shovn on the titration curves between pH 7,0 and 3,0, The ColorAbiau soils, except Drainageway, db'-d not suffer r;uch a big increase. A lai'ge proportion of the titratable acidity vas measured at pH 5.5; however. Ife-ainagevay had the largest increase fr03 p3 7.0 to 8.0. Patterns close to those observed vith Ba(OH),. develoiied using Na^Bj^O^ as the titrant (Table 1^0 . In general, soils from Costa Rica yielded less acidity than soils from Colombia, and an increase occurred between pH 5.5 and 7.0. The two titrant s neutralized about the same amounts of acid from pH 5.5 to 7.0. Titrations with Ka^Bi^O^ were not carried out beyond pH 7-0 because higher pH's are not attained with this base (^3). The results obtained vith conductometric titrations (Table j-k) were also very similar to those nientioned above. The aacunts of titratable acidity were verv large due to the high pH required for the formation of alurainate and insoluble silicate precipitates (U3). The use of Ba(CH)2 titrant for conductometric titrations was dictated by the extreinelj' high pH values needed for completion of the reactions involved. Ext rac table acidity ' ' . •*> ' . Soil extracts obtained vith IH KCl were used to measure the quantity of exchangeabls acidity. The results were very close to those already discussed where soil suspensions were titrated with Ba(0H)2 and ^^2^U°7 5.5 (Table l^i) . Extractions of acidity with BaGlg-TEA produced somewhat erratic values. This reagent is used to estimate total acidity and should be comparable to potent! c~e trie titrations

PAGE 104

O 0} U X (A H -P 4-1 v4 Eh i CM H •H CO CO •H Q > CO o •H M CO OS •H o OJ C5 4) UN H JCO CO CM • CO VO ON O o a o O f4 Id u a > a •H CO u OJ a a •H o W cd

PAGE 105

09 wltia Ea(OH)^ to pll 8.0: hovevor, values obteined with EaCl^-TEA for Los Dlamantas ar:d San Vitc; were rrrach higher than those obtained with Sa(OIl)^. Grecia^ Cararao, and Drainegeway gave lower estimates. San Isidro, Alajuela, Agronomy Area, and East Carimagua yielded comparable amounts of acidity. The lower values were explained by the fact that samples for BaCl^-TEA raeasuremen-cs were incubated overnight, and samples for potentiometric titrations were incubated for h days. Higher values ve?^e the result of chelating ligands in TEA (I65). The relative acid yielding capacity of the soils seemed to be an important factor for these soils that gave higher and comparable values; in this context, soils from Costa Rica were, in general, higher than soils from Colombia, except Drainageway. Correlation of acidity with soil properties and soil aliurdnum Organic matter provided a high degree of association with all measurements of acidity except those used to determine acidity due to Ex.-Al (Table I5). Cation exchange capacity correlated well with estimates of the total amo^ont of acidity. Water pH w<»3 related to general measurements of acidity caused by Ex.-Al; however, IN KCl pH correlated only with those measurements specific for Ex.-Al acidity. Exchangeable Al was highly associated with exchangeable acidity. Amorphous fractions of Al such as Acet.-Al and Pj'ro.-Al gave the best correlations with measurable acidity (Table I6) . Aluminum extracted with oxalate was related only to conduc tome trie and BaClg-TEA measurements of acidity where the total amount of Al plays a role. The same was true for the Al extracted with CDB and NaOH.

PAGE 106

90 I ON • try • On » CO • o • o 0\ ON o o o -p a u •H -< & © s p •H O 0} 0) •§ 4J O TJ OQ 0) u X V -P 0) •d o •H to o ca o xs H -S ca -H p Oj « S-t i) 4-> P fH 0) o u i H d ca o o Tit a o H .C 0) O ^^ -d o d H EH O o OJ o ft CO ft o n to OS * (0 4-> a o •H Vi <« O o c O •H +> ca H (U u o o OJ • o • o • O vD CM O ' CO o\ • • o o •d "H o 0] V H -5 P ca m ^o I • ro CO o I CO • • IfN • • CM • • • o ITN • • ON • • • • • H • • NO • • CO • ON • • • • * * • • • • • o 1 • • o 1 • • o 1 • • • o 1 CO H ' CO • cn O • CV-) LTN • CO ON • • • • o o V u> O O CvJ . • . , — ir, P --i U 4^ l/N O OJ f• • C o ij> ^o -H n ^ +} a ft ft ft CM ft ft 'J V a ca Ti +f w a o o o • • On • O •00 . <-l On O ' o CO ' ON •H t> c3 ^ Fh P O CM t) \-; .H ca +> H pq W CO CO • • o o VO 1^ •* VO ON • CO CO • • « • o o • CO VD CNOJ t~On • • o o +3 d e •H B rn •H ca I o X 0) > +> a) o f-i ft l/N o p (CI s o 1-i Vi H c •H to

PAGE 107

I x o a 15 • • t\y • •CO • • I* • • o o r-l << I O u * m -p C ir\^ O t-rH ITN • • • o o o VO ON VD f• O O ^ H C CJ • \ o o vovr> ^CO t-VD • * » o o o < Q> O o c o •H +J OS H OJ U u o o ' o 1 o o • VO • H • CO • • • o o o r-i < I X o > idit p d TV o 0) ej B o bl ^ !5 in «] o 03 iH 1 1 P O CM O SI •H c d ci P H « a CO X w w 1^

PAGE 108

92 The role cf amorphous Al in soil acidity vas highlighted by the close association betveen variables describing thei^e Al forms and soil acidity. The high correlation betveen acidity and OM was probably caused by the presence of Al in organo-metallic complexes. Phosph o rus Adsorption ^ The intensity factor of P adsorption vas described by using the quantity of P scrbed and the percentage sorption of that added.' The P adsorption maximujn and the constant related to the energy of adsorption calculated vith a Langmuir type linear equation described the quantity factor. Intensity was used as a measure of "active adsorption" where short term precipitation and anion exchange reactions seemed to be the adsorption mechanisms involved (76); quantity vas used as a measure of "potential adsorption" where long term specific adsorption and isomorphous replacement reactions seemed to be the adsorption mechanisms involved (76). " Soils from Costa Rica and the Drainageway soil from Colombia were very active in P adsorption and also showed a high potential for it (Table I7). The constant related to the energy of adsorption was also high for these soils. In the case of Drainagevay, all aasorption indexes were extremely high. Correlation analysis showed a high aegree of association between P adsorption parameters, OM, CSC, and amorphous forms of Al (Table I8). Correlation analysis explained the high affinity of the Costa Rican ooilj and Drainageway soil for P since these soils had the highest amounts of amorphous components. A point of interest was that the soil amorphous components, especially amorphous Al, were also fo'ond to bo very active in soil acidity. Apparently, thcje types

PAGE 109

93 6 > 1 -P •H P 3 * p •H tr> C -P M O O to o e Pi t3 a; •H ft ft o s ft ft -p (0 o VD VO O tCVl VO CM QD c.-. CO CJ 3 -t O CO l/N OJ VO -? C\J on ir\ (O sr-i ITN ^.l VO OJ H O cu CM VO CM CO VO H 00 H H H r-J 0) o •H o rn o p o •H W M s to O o r-i 05 O CvJ OJ •vO CJ CO VO ON I & o a o O d a u 03 O on CO H ON c5 0) f-l VO OJ CJ CJ CO cry cn ON OJ VO CO o o • CO CO ro cn VO VO LTN CJ cn CJ c o •H P & • H fl o O :o •H 4' Oj eS S ''m D< O V e M ft u >> 0) +' a •H d) E +J C O 0} +5 OS C) ^ d +> H >3 -H 3; o XJ 4J ft
PAGE 110

of acid soils have an inlierent potential for F adsorption, and their liiauasement beccrries even more ccnipiex than soils where Ex.-Al is tae dorainarit factor of acidity. An encouraging feature is the fact that the P found associated with non-crystalline fractions of Al has been shown to be available for plant uptake (76). Effect of Lime on P hosphorus Adsorption Intensity factor Lime reduced P retention in the majority of the Costa Rican soils; hovrever, retention was linearly increased in Los Diamantes subsoil (Table 19). In most cases the regression coefficients indicated a tendency to increase retention after a critical lime application was exceeded. This application for each soil can be approximated from the data in Table 55 j Appendix A", or can be calculated from the minimum P retention value using the regression equation. Lime increased the retention capacity of the Colombian soils except Cararao. In the latter, retention decreased linearly. The coefficients of determination showed that the lime effect on the intensity factor of P adsorption was not significant for raost of the soils. Ksu (68) suggested that lime will only decrease P retention when precipitation reactions with Ex.-Al and crystalline Al compounds are the main mechanism involved; in fact, P retention decreased linearly and significantly only in San Vito and Cararao soils which have large amounts of Ex.-Al. The significant linear increase in East Carimagua was probably caused by saturation of the exchange complex with Ca and subsequent precipitation of P by Ca and adsorption by free CaCO^ particles (51). The extremely low CEC of this soil (Table 5) could be saturated with a relatively small CaCO.

PAGE 111

95 V V B 53 U rt ft ft s > G O •H -P & o m ft ft 03 5-1 o (M \D OJ CVl £*— Lr\ >n ir\ rr, CO OD CO • • • * • o O o d o O * tn -P a 9) 0) o V o •H P (U rH CD u U O U H OJ H CO CO CO ITS • • • • • • o O d O O O O e ft ft >> •p •H o cn ft ft o H CVI CO CO ON vo JON CO Jo c, CO 00 On 00 CO CO • • • • o o o o d d On CO H ON H H ON vo CO ON CO CO cc • « • d o O o d o o CO ft o f1 • -p 1 1 t) • OQ o Q 6 O I a] H > a 4-: •H cd rO O U ft :A O X! P P OS o •H *

PAGE 112

96 4CM * * * * * f * ir\ -t H Jvo CO H Cj ro -d 00 o
PAGE 113

97 > o CO to to 9> VD cf; O m CO CO 00 CO r~ no ON a\ o\ o\ ON CA ON co On o o O o O O o o o O CO H on \o vn o OJ t~ir\ o r-i ON ro "cvj rVD ^H CO > H ON ro ITN ^ \0 CO CvJ U-\ CN CO On ON • tOn On • CN m CO ON • NO CN ON CO On O O d o c> d o cH O H LTN CO H ON OJ CO ON On M 0) U rH -P O •H o: o o CO ^ u o o zs r5 p •H V) CO •H CO Q M CO CO eS O a> ca 0) .-H ON CVl rH NO irN I? O ES O O CO ^< d (0 o LfN H ON 3LfN OJ c '3 u LTN H ON m o\ ON ON 0\ CN ON CN d d d o IPv JCO o CO (7s NO ON IPV H ITN rH NO o CVJ NO CVl H CVl OO •vO NO NO LTN L — CVJ CVj fiO (0 6 o p fO ea

PAGE 114

• . application and the remainin,:; li-.e and Ca could become active for P retention. Amorphous Al activated by lime was responsible for the significant increase in retention shown by Los Diamantes subsoil (68j $6, 135)As mentioned previously, this soil was doniinated by amorphous Al compounds. Quantity factor Lirae increased the P adsorption maximum in the majority of the soils from Costa Rica (Table 20). San Vito and Alajuela suffered very little change. A decrease took, place in the soils from Colombia except Drainageway. The increase in P adsorption of the latter was the highest among all soils. The constant related to the adsorption energy decreased sharply with lime in all cases. These decreases meant that although P adsorption became higher with lime in soils having large amounts of amorphous Al v;hich was active in adsorption,, plants could absorb P from these sources (76). It was also mentioned by Hsu (69) that lime decreases adsorption in soils high in crystalline forms of Al. This was the case in the Colombian soils. Electrochemical Properties Delta pH . ' • Negative ApH values were obtained for all soils by subtracting the pK in HgO from the pH in IE KCl (Table 21). The values were larger than 0.5 pH units, and in some cases very close or equal to 1.0. Negative values were also calculated by difference between pH in iJJ KgSO and H^O; however, the ApH was smaller than or equal to O.5 pH units. Mekaru and Uehara (108) explained that the ^pK measured with KgSO^^ is in some cases smaller than the one measured with KCl and someti es positive because of exchange reactions of 30., with OH ions.

PAGE 115

99 Table 21. Delta pH measurements for the soil s studied Soil pH* HO IN KCl IIJ KCl-HgO IN K^SOi^-HpO Los Diamante s .• 'Surface 5.6 h.& 5.3 -0.8 -0.3 h./ Li IwT o ^ X 5.7 -0.2 Ran '/"Stn -1.0 -0.5 Sari y f^r*'^ ? • -L -0.8 -O.Jf Greela 5.2 k.k -0.8 -0.3 Alajuela 5.5 5.0 / • -1,0 -0.5 Agronomy Area U.6 3.8 -0.8 -0,k Cararao k.2 3.6 i^.l -0.6-0.1 Drainagevay h.9 -0.6 -0.2 East CarimEigua h.9 k.l 4.5 -0.8 -O.U * pH was determined using a 1:1 soil to solution ratio.

PAGE 116

100 Based on *ip}[ vork by Van Raij and Peech (1^3) and Mekaru and Uehara, all soils studied l:.ad net negative charges; SOj^ adsorption occurred apparently through exchange vith OH ions. T^ie zer o point of charge Potentiometric titration curves in the presence of CaClg at different concentrations for the soils from Costa Rica were characterized by steep slopes and a good separation between curves on the basic side of titration (Fig. 5)Los Diamantes subsoil, however, shoved st&ep slopes only in the presence of high CaClp concentration.?. ^ and the separation of the cur>/ej. vas more pronounced although the curves with the two lowest concentrations overlapped. The point where all cur\'cs crossed (ZPC) was not well defined, and there v/as a tendency for the IK CaCl^ curve to shift toward more acid pli's. The ZPC for the Costa Rican soils was estimated between pH k.Q and U.5 except San Vito where the ZPC was approximately at pH 3-5It took larger amounts of acid to reach the ZPC in this soil. The potentiometric titration curves for the soils from Colombia compared to those for the soils from Costa Rica were characterized by flatter slopes; however, Drainageway showed steeper slopes similar to those of the Costa Ricau soils. A good separation on the basic side of titration was observed between curves at different CaCl2 concentrations except for East Carimagua where the curves were very close together. Separation was less marked at the lowest CaCl^ concentrations. The ZPC was poorly defined, and the curve for IN CaCl _ 2 was shifted toward more acid pH values. The ZPC of these soils was estimated to be between pll 3.^3 and 4.5; however, the ZPC of East Carimagua showed a very low buffering? capacity at ail concentrations.

PAGE 117

101 Figure 5« zero point of charje determination for scilr. from Costa Rica.

PAGE 118

102 (6ooi/bauj) (6oOi/b9Lij) 9soap9ppv povpappv asog p9ppv pov psppv

PAGE 119

The rather steep slopes obseri'^ed for the soils from Costa Rica and Drainageway vere an indication of high buffering capacity. The latter warj especially high for the more concentrated CaClg solutions. The fact that the IN CaCl^ curve was shifted toward mors acid pH's reinforced the validity of the above statement. Specific adsorption of Ca ions at high Ca concentrations caused the high buffer character. Calcium was preferred to H on the acid side of the curves developing low pH, and was also preferred to Na on the basic side of the curves; on this side, Ca exchanged acidity keeping pH lower than expected. The high Ca adsorption of the soils at high Ca concentrations was supported by the degree of separation of the curves on the basic side of titration. The location of the ZPC showed that the soils will become positively charged only at pH values lower thaii 3-5? and tliat most of the net charge is negative as was disclosed by ApH measurements. Based on this fact, anion adsorption due to positive charges was considered minimal; precipitation and specific adsorption mechanisns were found more likely to occur. . . The results suggested that additions of high amounts of lime will not be very efficient in raising soil pH. Lime in small quantities will be more advisable and will also neutralize toxic substances. The fact that Ca was selectively adsorbed at high concentrations seemed to substantiate the point made by Keng and Uehara (82) on tae importance of the Stern layer. They stated that most of the negative charge in tropical soils was found in the Stern layer. This property means less charge for cation retention and soil dispersion, but good soil aggregation and good physical properties. -

PAGE 120

10k According to the presence and iocuticn of the ZPC, all soils were foimd to have constemt surface potential fC3P) as well as constant surface charge (C3C) colloids. Trie low pH values at vhich the ZPC vas found reflected the mineralogy as well as the CM content of the soils. Parks (119) observed a ZPC between pH 3.5 and h.^ for clay minerals, and Van Raij and Peech (1U3) pointed out that OM depressed the pH of the 7Jr'C in several, soils. The location of the ZPC in the Costa Rican soils was probably influenced more by the high OM content them by the clay mineralogy. The latter seemed to be the dominant factor in the Colombian soils except Drainsgeway. This soil behaved more like the Costa Rican soils. . Calcium Selectivity At low Ca equivalent ratios in solution, Al was preferred to Ca in Los Dianantes surface soil (I'ig. 7). The soil became more selective for Ca as the equivalent ratio in solu-^ion increased to 0.3. Beyond this point, Ca was adsorbed preferentially over Al. Los Diamantes subsoil shoved the ssune dependence on Ca equivalent ratios in solution; however, Ca did not become dominant oh the exchange complex vintil the ratio was higher than. O.U. San Vito and San Isidro displayed a strong preference for Al over Ca. The Ca equivalent ratio on the excheJige complex was never greater than 0.2. Grecia and Alajuela did not select Ca or Al, although ail .data points were below the equal selectivity line indicating a slight preference for Al. All soils from Colombia showed no special preference for Ca or Al (Fig. 8). Los Dianantes surface soil adsorbed more I4g than Ca at low equivalent ratios in solution (Fig.' 7). Calcium was preferred at ratios

PAGE 121

105 1.0 r OS 06 0.4 02 0 Los DiOiTiantes (surface) . A * * o • o / • o / • o 3 c J. 0 02 0.4 06 08 lO o cr o O cr 1.0 08 0.6 0.4 02 oL Los Diomantes .(subsoil) * AAA/ Va / / A • /A 4 5 ° 5 I JJ 0.2 0.4 06 08 1.0 1.0 OS 06 04 Q2 San Vito OL_A_±_£ ^ X J 02 04 06 08 10 «.0f08 0.6 0 .4 02 Son isidrc A A 0 L a...^-* t ^ t ^ J 0 1.0 08 0.6 0.4 0.2 0.4 OS 0.8 1.0 Grecia / A' 0.2X 4 X X 0 lOr 0.8 0.6 h 0.4 02 0.2 0.4 06 08 10 Ok Alajuela X. 9^6 /I /t 0 02 0.4 06 08 10 Cco/Co Co -K • Co Mg o Co-AI A Ficure 7. Calcium adsorption isotherms for soila from Costa Rica.

PAGE 122

106 1.0 r 0.8 0.6 0.4 02 0 Agronomy Area / 9 / /ft / /a \^ L. P" 0 02 04 0.6 06 10 " Corarao y'^ a Stor 00 06104 0.2 h 0 L 4 4 /O -L J 02 0.4 06 08 10 1.0 r 00 0.6104 02|Droinageway i / OtZ L JJL 1.0 0.8 06 0.4 02 0 0 02 04 06 03 1.0 East Corimaguo / / c ^ / o / J02 0.4 06 08 1.0 CCo/CO Co K • . . ; Co Mg o Co-AI * ?i,-;ure 8. Calcium adsorption iirotherr.s for soilc fron Colombia.

PAGE 123

greater than 0.7« Los Diaraantes subsoil shoved similar behavior; hcvever Ca did not become dominant on the exchange corap3.ex. The other soils from Costa Rica and the soils from Colombia (Fig. 8) had no selectivity for either Ca and ^!g. Potassiu.m vas dominant over Ca on the exchange complex of Los Diaraantes surface and subsoil at low Ca equivalent ratios in solution only (Fig. 7). The re^aaining soils, except Agronomy Area, did not select for either Ca or K. Calcium vas strongly preferred to K in the Agronomy Area soil. — ' The Ca selectivity coefficients for the Colombian soils in the Ca-K system were higher than the coefficients for the Costa Rican soils, although Cararao had similar lover values (Table 22). Selectivity coefficients for Ca were high for the ;aa.jority of the soils in the CaMg system except for Los Diajnantes surface and subsoil. San Vito and San Isidro had very low coefficients in the Ca-Al systems. AH other soils shov/ed rather large coefficients. These results suggested that Ca vas adsorbed the most in tae presence of Al, and the least in the presence of K. This effect was attributed to valence and concentration; in fact, low valence cations are preferred over high valence cations in -concentrated solutions (38^ II6) . However, Ca was not discriminated against, and was always present in significant amounts on the exchange complex. Tlie fact that Ca was preferred over Al in some soils and vas always present on the exchajige complex in significemt amounts in all soils substantiated the results of the potentiometric titration curves

PAGE 124

Table 22. The negative logarithm of calciurri selectivity coefficients for the soils studied in the prer.ence of potassium, mag. ^ nesiura, and aluminum in IN solutions . ., Soil' -Ca-Mg Ca-Al p!v' Los Diamantes 1^. .. • "^.v. ' * ^ Surface -0.53 -O.67 -3.22 Subsoil -: -C.6h -0.99 . -3.20 San Vito ; . -O.83 -l.hl > . 2.15 San Isidro . . -0.88 .. -I.37 . 2.kl Grecia '/ -O.77 -' -1.30-' -l.kk AlajuelG -0.70 -1.25 -l.kl Agronomy Area -2.10 -I.3I . • -1.68 Cararao . -0.75 -1.29 . -I.63 Drainagevay " -1.00 -I.50 -1.73 East Cariraagua -O.9I -l.kk -1.86

PAGE 125

109 ) using GaCl2 at different concentrations. It was evident that Ca vas specifically adsorbed, and in this process created acidity through exchange reactions . ' . : , ' Incubation Experiments Water pH increased linearly -^^ith lime in all soils except East Carimagua where the increase was logarithmic (Table 23). The pH in IN KCl increased with lime exactly the same way as the pH in water (Table 24). Exchangeable Al decreased logarithmically as lime applications increased except in Los Diamante s subsoil where the decrease was linear (Table 25). Adams and Pearson (2) also observed a logarithmic decrease in Ex.-Al with increasing lime rates. They pointed out that this behavior is common to acid soils high in Ex.-Al that are limed. Lime had no significant effect on the water extractable K found in saturated extracts in Los Diamantes subsoil, San Vito, and San Isidro (Table 26). Water extractable K decreased as lime applications increased in all other soils except East Carimagua. However, a tendency for water extractable K to increase after an initial decrease was observed in Alajuela soil, and was attributed to replacement of K by Ca from the exchange complex as the Ca saturation became increasingly higher. The same explanation was given for the increase observed in East Carimagua. In this case, the low CEG was rapidly saturated with Ca which in turn displaced K even when small amounts of lime were used. The general decrease in the K concentration of saturated extracts caused by lime has been usually attributed to the increase in pli dependent CEG ( 103 ) .

PAGE 126

\4 110 cvj p o CJ o o (0 o o •H p o H c p •H ^^ tC tc o •H p ft (U O u Q) •P rH •rl o C'5 (0 P C 0) o o •H to m 0) to o VD H o OJ 0\ 00 ON 0\ CO 00 vo a\ CN On CO • • • • • d d d d o o o d O o rH CO 9 o •H W M CO o C3 OJ H IS •rj fli I o c o o CO u
PAGE 127

Ill CVi H as CO CO c^ (M CO as • OJ as as as CO as • ON • ON x> • CO 00 • CO • « o * o d o d o o o o o p o V (VI o o o •rf P 05 p to Cj (U in -P ft a o o H 10 to 0) ^ 0) O OJ CO rH O H ON o on o CO H H r-l H o H OJ OJ o o O O O o O o o o VO oo oo JOJ oo irs oo oo -It OJ iSS o ca CO •H o CO o o o CLT ca •H ^^ CO •H W «j 0) > M H () a 0) ctf (h H CO to O -a: at (1) o o o u (J cC a; c CO u a & 0.1 H •H ca o p to 03 W O d p p . CO P s o •H CiH •H G to •H to o « 1) P 0) o in p C c •H y •H a.' CJ o H fj >^ a 4-> O -H •H rH 0) -H to p 0) CO tu:. o

PAGE 128

112 •H H o p o * iO to 3> o cS s o H p o H (!) o to 0 i •H •H O •H tH tH c3 O H •§ a.> to X! a 0) •H t> •H tiO 0) ,Si CO o •H -p •H f-l c3 bD O J-1 (D •H 4-> o (U a M o CO « p a (U •H O •H tH o M CO CO •H o o o a OJ «8 0) e o d o o U cS u CO o cS O •H T-i CO 00 03 t! CJ u H +J 1) 33 ?< 03 O W *

PAGE 129

113 CJ * * * * * * * CO H CO o JCO H H CO H CO ON • « • • • • d O o o o d o d O o 0! CV| O O a o o •H -P ca M o 0) P a 4) •H O •H 5 p p p< a; o ^1 O! P c M ft CO o CO o CVJ vo ON r-l Cvj VO VD H •H O ft rH O 9) P o CO w P c cS § P (I) t) nJ u rn •H O to in o p •H > CO o •H w M C! a) to CJ 05 O O o ca u
PAGE 130

uk Water extrac table decreased as lime applications increaced in the majority of the soils (Table. 27). Ir.e lirae eff&ct v&s not significant in Los Diamantss surface and San Vitc soils. An increase followed by a decrease in water extractable Ma took place inCararao soil as lime increased. This behavior was e:Kplained by exchangereactions occurring between Ca and Na as the Ca saturation reached a critical point. ^ ' ' ; '• • ^yIncreasing applications of lirae caused a linear increase in water extractable Ca in most soils (Table 23). Calcium increased iogarithniically in East CarirrBgua because of the rapid saturation of the low CEC with Ca. The slopes of the regression equations for Ca in Los Diaraantes, Grecia, and Alajuela were considerably higher than the slopes for the other soils indicating a fast accumulation of water extractable Ca brought about by lime. The Mg concentration of saturated extracts as influenced by lime did not follow a definite pattern (Table 29). In general, water extractable Mg is decreased as line applications increase because of the increase in pH dependent CEC (103); however, a decrease was observed for a few of the soils studied. Water extractable NO,-K generally increased as lime increased (Table 30). A decrease was observed in San Vito, Drainageway, and East Cariraagua. San Isidro showed an initial increase, but the concentration eventually fell. The increase in HO^-N was very marked in the soils from Costa Rica where it took place. These soils yielded high amounts even when no lirae was added. Drainageway, surprisingly, did

PAGE 131

115 en * * * * * VO H O ITN o:) CO vo • • • • o O o o d o o * * CO o CM * rH ON O •H -r— ' c3 Pi 5 * CO o • o I 6 (U 0) O o C3 u •H -p •H Vt Cfl bO Q •H O •rH l+H tH d) o u o H CO (0 0) f^ bO 0) * CM -4I U S3 •H 1-q I * * * 1 CO * m * vo LTN o H CM H CM • d d o d d d d * * cr\ o d I p Pi i p w 3 o CO o CO o •r4 CO o •H w M C! CD •rS o u .-I (U "-3 (L) O a o o a 03 O 0) •H

PAGE 132

116 s •H H O -P O o rH •H (U O to p •cJ u 4) o P ss -£ o c •l-t s:: p •H •H o lU a -p p) 4.1 p — 'd tH o •w w o iJ >H CJ a} o o rH 5 C3 a> H-> o O o 0$ u -d X (0 10 u -p 0) -p (U g3 •H ? o •H ej tM Cm -p (U O d O o C o "fcO -H o (0 o rH «• CM P o 0) o O a u •P ca -d 03 •tH B -p •H c6 o 5h ca a; d •H p Pi i) o (U 4J C M O 7) C 0) «H 01 O o c o •H to i) h « P. * * CVJ rH r»-> ON 0^ • • o c CO o ON CO t-ON rH ON H CJN CM ON * o c:n C7N * * LTN CO * * vo rH * c o CVJ CVJ LfN H LA CM ^o Ov > 0) rH H >5 >, P •H •r-l H rH •H •H XI X3 ca ca x> o o !-i p< LfN rH O o C o OJ 0) xi -p -p -p ca ca p -p c: a ca ca o o •H •H
PAGE 133

117 CO a o o o a u 4^ u d -rH •.-< c3 to ft cS (Li a o t-i -p o m P C
PAGE 134

not show an increase even thcyij^h the OM content vas the highest of any soil. Apparently this OM was resistant to biological decomposition clue to the large amounts of Al and Fe associuted with it. An absence of nitrifying organisms was also considered since the Costa Rican soils with high amounts of organic Al and Fe did show an increase. The electrical conductivity of saturated extracts increased markedly with lime in all soils (Table 31); however, the lime effect was not significant in San Vito perhaps due to the decrease in water extractable NO -N, Trends for EC were very sir.:ilar to those observed 3 for Ca and NO^-N. Los Diaraantes surface, C-recia, and Alajuela showed the largest accumulation of Ca and NO^-N as well as the highest EC values. Data in Tables 62, ok, and 65 in Appendix A shew the accumulation of water extractable Ca and NO^-N and the increase in EG in these soils as lime additions increased. Values for Los liiamantes surface soil were not as high as those reported by Velez et al. (iko) ; however, they used much higher lime rates and some of their samples remained in storage for several years before incubation. Nevertheless, accumulation of electrolytes, especially with additions of high lime rates, was considered a major negative short-term aspect in liming those soils from Costa Rica mentioned above. This problem was not considered of e^i'eat importance in liming the soils from Colombia. Greenhouse Experiments Experiment No. 1 The experimental variables in this trial were lime, Mg, and P. Nitrogen, K, and micronutrients were kept constant at values considered adequate for Pangola digitgrass growth based on previous experience (1U5, Iho, 16k).

PAGE 135

119 % H H CO H •H O O m P o u •p ca a> ^ «; V a o y — -V. E o Qi 4^ 65 j3 Ti CJ PS hr\ tjj Q C* 'H CJ 1 3 0} m 5 •P u •H (U £3 •H CJ H H tH «iH a; o b o -P § X a u •P a •H > o •H C) X! P o c o o a ^ o to •H o to o H U (1) O • o (U iH n O CJ -p a a; Cm a O o c3 U o •H e •H a O a c •H 1-1 ft o P c M O CO to ci i> •H O •H tH t!H (U o V O •H to to 0.1 u E * 1 t % 1 * CO CO VD C\J H CO H rON o 00 CO CO G • • • • • • • o o o o d O O d o no CO * H CO VD CO -4m O 0-) H CO H r-i CJ -aCVJ 0\ c O ON 00 o • • • • d o vo VD LTN VD H CVJ H CO H H o 0} cn H •H O w to o p o •H to HI r! (I) to 03 V 0) u o (0 H 0) rH o c o u '
PAGE 136

120 Cvl t * 1 m H c -=)• CO m CVj CO CO ON CO • • • • • o o o O d o * LTN OO * CO OO 4J O O o O o •H P •H fH CS O I I I I 1 I I I I I I rft .ij Q (l> •H y •H til CJ o o c •H CO w OJ U bO 0) * d rH * C no (D * CD OJ ON Q") CO * ITN * CO CO 00 JH +5 Pi OJ o •p d H S O — . o (M OS o VD OJ OJ ON VO lf\ JCO no vo rH UN OJ H OJ H H H o CO 0) o O o p a to o •H n M 05 H o o 05 rH (D 0) o ci o u o 03 U oJ 05 05 > •H 05 U O u u +> to 05 W

PAGE 137

121 t Phosphoras vas the raoct i:nportant variable affecting; yield and P concentration of Parigola dljritgrass in Los Diamantes surface soil (Table 32). The coefficients of determination showed that the quaiiratic model vas a poor predictor of yield and nutrient concentration, and suggested that a linear model vould have been better for these purposes. Based on the regression equation, maximum yields would occur using 10 meqyiOOg of lime, 87 ppra of Mg, and 423 ppra of P (Table k2); hovever, these estimates were biased because of the poor fit of the model. Phosphorus effects (linear and quadratic) were the most prcluinent in determining yields, and P, Ca and 1% concentrations of Pangola digitgrass in Los Diamantes subsoil (Table 33). Tbe quadratic effects of lime, Mg, and P plus the linear effect of P contributed significantly to the Mg concentration in the plant. A negative linear effect of lime and Mg on yield vas suggested. The quadratic model described very veil the variation in yield and nutrient content of Pangola from this soil; therefore, lime, Mg, and P must be ta;cen into consideration for improved growth and adequate nutrition of this plant in this soil. Maximum yields were predicted to occur with applications of 8 meq/lOOg of lime, 130 ppm of I-fe, and hhO ppm of P (Table h2) . ; . In Sen Vito soil, P had highest effects on yield of Pangola (Table 3-'+). Lime and Mg showed a negative non-aignificant effect. The quadratic model proved to be a good predictor of yield. Both the linear and the quadratic effects of P were important in determining the Ca concentration of Pangola, and the quadratic model in general gave a good description of the variation in Ca concentration; however, the model was only adequate for predictions of the lJ\g concentration in the plant

PAGE 138

122 0) i ! 1 1 -3' t rVO ei ' Ui ! ro i CM 1 iA O <-i Or") 1 1 -V i-i 1 ..it' ^] r-I CM cj rH ! CJ 1 d i 0^1 rn m VO vn VD 1 1 ix; ~t 1 1 W o 1 [a, 1 '-^ Oj 1 W 1 it t i > — { H r-i 1 o r-l O a) p< •H ~ H fcO CJ O to to >-i to ^ ^ -H fj v. « > •J5 •P c •rH •H tfH VI aj o o o •r-I tJ) V3 u a) i ( -dt (X) vO t — "l 1 W .'J 1 S W ON 1 r-i v6 OJ C3^ tCVJ I LJ'^ p d) u u dj p a ! r-i I i e <^j I I W I * I VO I VO 1 ^ I cc. I o ON I S (X. CVl OJ I 1 VO W i w ^ CO H oj * * m 1 t -:}VO w ^ 1 W A 1 w 0\ OS oj H H OJ tu E VD I oj I 9 I J1 CM Oh X 0) o I w r-j VO CO I W o I I OJ I .^0 § •H p C aj +> O •H O •rl •m 0) o o VD O ON -a1-1 > CJ H vo d P •H rH •H A3 Cj .O O u ft VO o r-i O 0) +» +> OS +> o H •H •H CO OJ *

PAGE 139

123 o rH 'H O O CO 'A p ri c5 tjy CO S 0) a 0) X! •r-l o CO o +' 1-^ a a > • CO H O •fH \ X to r! 0 w vH 0) Ul ,--1 01 ,Q d) a U -fH I I I I I r I I t/3 P !-1 0! •r-! O •H "Vh O o -) q O •iH to ) I -r f.] w 1 . w ! 1 W CA CO VD O" 0.1 C) r'i -i CV H 1 ro ITS ) I I 1 ri vb /~\ OJ <\! 1 CM ( * * * * H 1 ! .::( k 1 w ro » ^' t 1 H — OJ ; * 1 1 W [x3 1 1 r-q CO OA CO tvo VD cO rH O.! I CO P Pi t) o fH a> p d M CO •H -5 e C\J I •vD I I CO I w o or) ! H 1 m 1 w 1 o OJ oj CM CV •H 1-1 * cI PH CO oj I * CvJ t d~\ CO W 1 ("> LTN oj oo OJ ) i w vo vo I t I r W) CM :^ (X, o •H vo t I I \0 OJ -Js a) ..H H ! 1 vo CO w CO • r-j r-H VO I ^ ^ a C7\ I Oh o 1-1 CO • * • — r 1 • i 1 a o •H P * c VO •H » • o 1 p O fH P. o 0) P P 03 P tJ •H V< •H to

PAGE 140

121^ ! r-l O C ft 'ri "^-^ O VI CT t -i CI Ul a D :^ 1 t 1 1 1 I < I i 1 It: W L-'l S3. W t~ ir\ < — 1 on o .c o vo CO 1 1 1 1 1 j Q * j 1 1 1 LTV vo 1 1 1 1 1 t 1 f 1 1 * w 1 CO ir\ a;' O o <^ iH j t— r-l ^5 r-l 1 1 1 H H 1 1 d o •H * P r-H rn vO o rc5 1 1 1 1 1 w 'A !^ •H CvJ o CO vo OJ a3 H trH a; c iH -4VO 1 1 il 1 1 M 1 s. t CO t — VO V3 CM H r-H CO VO VO 1 CJ 1 1 1 1 1 1 oa 1 -d--tI' — K! w 1 VO i 1 OJ CVJ H OJ i 1 OJ 1 1 no a' 1 * * OJ OJ •-0 LA 1 L(^, 1 • t m rH 1 1 'A rH 1 OJ • 1 w 1 1 w A w CO 1 1 ^' rH i -P P< 01 CI u 0) a -1 s (1^ £1 CVJ •rH (-5 OJ Oh Pi X X 0) (U X B E3 •H Li -p 0) •d O CO p C •H O •H Vl D O o OJ CO VO o * CO, o OJ

PAGE 141

125 al.though the quadratic effect of was significant. I'he Lfe x P iuteraction and the linear effect of Mg had the highest effect on the K concentration of Pangola. A good estimate of thib nutrient was given by the model. I»3agnesium and the lima x Mg interaintion had the largest effects on the Na concentration of Pangola dry tissue. The model was a good predictor of this nutrient. Lime at 8 raeq^/lOOg, Mg at l66 ppm, and P at 851 ppm were the rates estimated to give maximum yields (Table m2) . The effect of P was most important in the yield, and P and Iv5g concentrations in Pangola from the San Isidro soil (Table 35). Lime h&,d a linear non-significant negative effect on yield. The quadratic effect of and the lime x Kg interactions also influenced P concentration in forage to a large extent. The lime x P interaction played a do:ninan-c role in K concentration in Pangola. Tbe quadratic model, in general, was accurate in predicting yield and nutrient concentrations in the grass except Ca and Na. Maximum yields were predicted from applications of 10 raeq/lOOg of lime, 2k2 ppm of Mg, and ppm of 'P (Table h2) . The P concentration of Pangola planted in Grecia soil was affected significantly by Mg and all of the quadratic variables (Table 36) . The quadratic effect of lime was important in determining the K and Na concentrations of the plant, and the quadratic effect of P influenced the Ca concentration the most. The quadratic model was suitable for predicting only the concentrations of P and Mg. Pangola yieJ.ds were not affected significantly by the variables, although lime aivi Mg showed negative linear effects. Maximum yields were predicted '.vith additions of 8 meq/'lOOg of lime, 120 ppm of Iv\g, and 585 ppm of P (Table k2) .

PAGE 142

126 "J +3 rH O •H H to o o to a M o oj CO CO +> 0 o O a Q CO to ti3 OJ I 00 Pi a) o (U p CJ I I H I CO \o 1 1 .OE1 w 1 w 1 vo 1 rH r-l I OJ 1 1 o ir\ >D irN vo t 1 w 00 1 1 M 1— s 1 W 1 w CO 1 * ' ^? 1 OJ 1-1 1 — i 1 J1 CVl 1 d 1 vo VO w 1 £3 vo CO JH t vo I I 0^ 1 r-l I w OJ H OJ I w ON oj * VD OJ I * vo I CO I H OJ CVJ 1 W 1 w 1 w ITS C\J 1 cu I •vO J1 1 W 1 w 1 OJ o CO -3ON \o CVl Oi fiO OJ 0) 1-3 ^o I >3 &3 a CVl I w CO 9 0< >^ s •rt I ON VO •D ^O .IE" ,1E1 H I vo rH ! VO I ON I I o •H +> Ctf C •H n OJ o 43 a) •H o o o * CO d o d * CO c 00 d CVl

PAGE 143

127 o Li p O d u 0) S § ^< •ri 0) P '0 a w •ri cJ ^ C H 0) •H jj O 0/ Q '^:! r! > crt o is} p a 0) V d o p a ^ f-i 1 -T M 1 W 1 fx.) CO r-! H CI OJ H I W ON cj CO CVJ P o I I I rvo s > H I CO I * ro I I ! I I < 1 VD 1 CO 1 w 1 W VO 1 m 1 • cO 1 d i 1 1 ^.4• w On 1 r-l l r-i .-1 Oj H 1 m t CO t VO vo w 1 1 oj I W 1 Ed 1 LTv oj oj H ^ I I w I JI 3 * * VD I w * VO r-5 * I w VO I H 1 on OJ 1 CVI -^1J.3 • • 1 W 1 w o 1 rH t VO H • r j • rH OJ X 0) u CJ ^ M CM ST a, •H I w o ft4 a o VI p (0 c a o p a) «H o (0 -p c (U •rl o •H > * 0) CD * rH H o >J >J P P d •rt -H H rH •H P< ft O Lr\ r-i O d • • o o x: Si p p p p 05 cU P P c c cJ Id o o H •H a a •H -H to fO * * *

PAGE 144

128 Yield of Peirigola was determined largely by P applications to Alajuela soil (Table 37). Lime har] an apparent linear positive effect on yield although it vas not significant. The P concentration in the plant was dependent on ohe quadratic effect of P. The ?ig effects on the concentrations of Ca and Mg in forage vere the nost important. The latter was also influenced by the quadratic effect of P. The lime :: P interaction along with the Mg effects controlled to a large extent the K content of Pangola, and applied P in association with the quadratic effect of lime determined the levels of Na found in the grass. The quadratic model described yield and nutrient concentration with a high degree of accuracy except in the case of P. Applications of 13 mcq/ lOOg of lime, ?11 ppm of Mg, arid 353 ppni of P were predicted to produce maxiimira yield (Table h2) . Phosphorus vas found to affect yields the most in A,5rcnomy Area soil (Tj:ble 36). Lime had an apparent positive linear though nonsignificant e^'fect on yield. Lime and P influenced the variability in P concentrations in Pangola forage. Lime affected Ca concentrations the most^ and Mg vas the most iraportaiit variable influencing Mg levels in the plant. Yield and concentrations of P^ Ca, and Mg vere estimated closely with the quadratic model. xMaximura yields were predicted with applications of h meqy lOOg of lime, 36 ppm of N^g, and 56O ppm of P • (Table k2) . Phosphorus vas the only variable shown to affect yield, and Ca and concentrations in Pangola forage from Cararao soil (Table 39). Lime showed a non-significant linear positive effect on yield. The model vas a good predictor of yield, and P, Ca and K concentrations. Maximum

PAGE 145

129 to CO c3 U HO •P •H aJ H r-l O /•> \J ^ t T i-4 Gi o 'rH ? , J-t n io & W r-T g •H O GJ t/1 +> a; W r-~I O ^» O it5 ty CO . 1 -P t! aKJ o 5h •H 6h Vi M rj O C o 4J ai H 4'> C! (D CO o p C! O O 0) •H O P •H tH 0) tH •H GJ u o C> -< a o v^ to CD n ^^ tvD O •H PC 0) rH ,Q OS CO a o •H P ^ -P aa o o -p a; •H 4' cS 0-, rH O a) P< •H £3 O « •H O n rH c cs w u P a ! JO OJ ci B tH u CO I w o. f — t CO 1 * "l "r 1 1 i 1 * Cl4 W w 1 H on CO o 1 CO 1 • a) V -J 1 O 1 * 1 * 1 VO * O. w ir\ n 1 M 1 W CO 1 O C\i en OJ ! r-i t t— 1 CO H 1 — 4 1 1. 1 $ 1 'i * Cij w IV 1 6.2E• l M CA tCO oj Cm Lr\ 1 ! CVl 1 w rH I I rH I I CO 1 1 1 H w ca. 1 W CC) ON CO o a\ 1 — i rH rH * I I t no I W CO CO on I w ON CO I OJ I I I I VO ! CVl J5? E OJ M OJ Si (X, 0) I On I I ><; 0) H •r< CO I C£| o OJ H I Ok X d o •H 4J (S G •H a 0) +i (U i 1 VD 1 ) nt: w OJ o o CI rH H rH rH •tH <^H 8 * OS !V * * CO CO d * OO CO d • • H H CJ 0) > > OJ OJ ON r-H H d p 4-) •H •H rH H •H •rl ^ X) c3 XI O u ^4 * ft * rH o LTN CO o o d d a 0) _r| 4-> p -P ca c r* CO CJ o Ti •H <»H vH •H a a •H CO ro Ol 1

PAGE 146

130 m to P •H ( 'd c Ci O « GO O K -i-i O £1 C r-J •n Tl 05 O -rl 0 +1 o o o u 03 fi) 0) K o •H 0 o O -H 4-1 C.) 4 ' •H Ci '(4 -rl a; o +> o :3 a R _^ o -d t'j erf a.' "c! U H to ill Q) -H 00 H EH O 4^ c; S-) d o ci o o 4' CI tH "D" 4-'> H O 0) ft •H o to t4 0.' W ^ 0) (0 K > P a •H y H !t4 o c o •H U) W 4-1 1 1 LTv 1 yo VO <>. w 4 M I w 1 1 1 \.r> 1^ o H oj H H f — H 1 no Jvo o-> .lE1 1 W a\ t H H ...H I w CO I 4J P< C.I p a * I (a I CO I ^ H VD 1 t w w 1 W ;rN LTN oj H -f1 1 1 VD vo §' W ON w ON W JW On -—I CJ H vo H cvj % -=t.:t * VD 1 VO VO 00 , IE00 CM CM VO no I 00 * CJ J1 vo ro VO 1 W 1 no ^ « • • • H ITS 0-N vo CJ 0) E (1< CM CJ R 1^ X R Oh g •H +> eC a> <« o w +> a 0) •H o •H 64 > ai a; iH r4 4:" +> •H -H iH rH •H aJ 03 o o Sh U p. p< O irN H O • • o o 4-'' 05 03 § § C> CI iH "H '+4
PAGE 147

'v 131 I si i o •H P eC 0) o ri o o Ti -P H O
PAGE 148

* 132 > . ' \ ' yields vere predicted with additions of h mcqy'lOOg of lime, 67 ppm of 1/^5 J and koh pp'ii of P (Table k2) . Phosphorus effects vere the most inportant to determine yields and P concentrations of PaTigola fro:n Drainagevay soil (Table kO) . Lime shoved a non-sicnificant positive linear effect on yield. Tlie Ca concentration of the grass was largely influenced by lime. Only P and Ca concentrations were best described by the quadratic model. Axjplications of 11 meqy'lOOg of lime, 121 ppm of Mg, and 38I ppm of P were predicted for raaxiimim yield (Table h2) . The concentration of P jn Pangola grown on East Carimagiaa soil was affected largely by applied P (Table 4l) . Lime and P effects and the Mg x P interaction vere important in determining Ca concentration. I»Iagnesium controlled Mg concentrations, and added F was influential in the variation of t^e K concentratj ons . Lime had an apparent negative nonsignificant effect on yield. The quadratic model was a good predictor of the concentrations of P, Ca, Mg, and K found in the grass. Maximum yields were predicted with 2^1:r.eqy'100g of li:ae, 620 ppn of Kg, and I70 ppm of P "(Table k2) . However, these esti;nates were biased due to the failure of the model to account for yield variations, and were extremely high for lime. In general, lime had a negative effect on Pangola yields in most of the Costa Rican soils, except Los Disunantes surface and " Alajuela, and East Carimagua from Colombia. The remainder of the Colombian soils shoved positive responses to lime; however, the lime effect was not significant although it contributed to the quadratic models for yield. Magnesium never had a significant effect.

PAGE 149

13'i a o Q • 3 (-» , 1 e. lr1 5 (If ft, r..-! •n W Tl ri >-f r\ L> *<—( OJ E M Ss. rO Al t > +-' •> n> UJ M u "H «rl CQ If.* c , i; +5 w C> o o f-, •n fM o; o o •H o P Qj 'd I. +J a 0) is: o -|J c! £3 O (U V •H O -p •H a '+H (U <»H •H O J-l o o :s o H CO tn o !h U5 u O « >> a o •H -P to u -p o 0 o o +3 c (U •H a. d +J H O o to •H O in H w XJ a) ^1 -H D 03 K > 1 H 1 1 1 ' W W 1 w ! U-\ o • ; cO 1 !.0 4> O o o •H W W > OJ i) H ,-l >> >s. P -P •H -H H •<-l «] X) O rH •H rC 03 o ft ft C ITN H O d d O (U ^ J3 p -p -p p § § u o •H -H d (3 •H Ti ro CO * *

PAGE 150

13^ (A H O •H o to t-l U (1) cs5 K > + o H o •H vH «H 0) o n to 01 U a) 1 1 1 t — L I'-l ^ r-'i I ( > o H H CO oJ ri ir\ 1-1 O 1 -4H t n.! 1 '.r\ 00 ! O * * W VO 1 X I ! 1 1 1 1 ! 1 * Ci -7* w ) r^l I ^ • • • • • o H I CC) ft o (U p i •*: I tn 1 t>-) rn • OJ • 0> I CvJ 1 ti "l ^W {i} !! 1 W 1 ) r\ fj a) ^ VD OJ CJ r-l I I VO vo 1 1 1 'X) CO OJ -Jvo * vo I o--. JI H 1 1 OJ vo W 1 W t w CO \S\ • H CO I vo 1 -4 1 K-l w 1 W LfN o ON « -4OJ vr OJ PL. OJ X H P-. X S3 •H !> 0) (1) >>>-. +> -P •H -H H r-i •H -H rO ^ tS Crt rO .O o o u u p( p. O LTx r-l O d c-j •P 4i -P 03 aj V o Vi v-1 vH -i-t •H -h" CO to * *

PAGE 151

Table h2. Ldree, niagnosjuni, and phosphoras t.r5>c L-iients tc ths soil; studied predicted tc give maximum Pan^^ola digitgrass • • yield by the model used in Experiment. JJo. 1 Tr;eat.rneno5 Soil CaCC, Mg meqy'lOOg ppm Los Diamantes Surface 10 -87 . h23 Subsoil ' 8 . 130 kkO an Vito . 8 I66 . . 85I San Isidro 10 2k2 Jjh Grecia . . . ' 8 120 5^5 /aajuela 13 211 353 /sgrononiy Area k 3^ . 5^0 Cararao h 67 k6k Drainagev/ay 11 121 381 East Carimagua 2k 620 I70

PAGE 152

136 but the responseo of the Cocta Rican soils to Mg were negative except In a few cases. However, a resjionse to Mg was evident in the Colombian soil:-:. Phosphoi-us dotainated all of the responses, and was considered essential for iraproved yields and adequate nutrition of Pangola. • The positive responses to lime in the Colombian soils probablyresulted from neutralization of Ex.-Al and additions of Ca in adequate amounts. The soils from Costa Rica had low amounts of Ex.-Al and high amounts of extractable Ca and the response to lime in these soils was, in general, negative. Experiment No. 2 The effect of lime on yield in Los Diamantes surface soil was nonsignificant (Table ^3); however, a negative trend occurred. Phosphoz*us had a marked positive effect. An increase in yield over the no P treatment was observed using P rates equivalent to l/l6 and 1/h of the P adsorption maximum. The P concentration in the plant was not influenced by lime, but increased considerably with increasing P rates. The second P increment gave values in the forage higher than O.l&p considered by Andrew and Robins (lO) as the critical P level for Pangola. Forage Ca was not affected by lime; maximum Ca concentrations were obtained with P at 1/h the P adsorption maximum and were within the range of 0.50^ given by Gomide et al. (57) as an average concentration in k-vee\i old Pangola. The first two increments of P increased Hg concentration in the plant, but these were always lower thaji the average of O.kO'f} reported by Gomide et al. (57) as the IAq concentration for 4-veek old Pangola. Phosphorus rates of h20 and 1675 ppm gave lowest K concentration of Pangola when no lime was applied. Forage K decreased

PAGE 153

137 Table k}. Effect of lirce and isurface soil on Pangols dig.ltgrar,s concentration, Experineut No. 2* rhoz-^'noTMSi treatiixe.Dts to Los Diamantes leld and nutrient Apixlied Nutrient concentration CaCOo — — • p Yield P Ca K Na ir.eq/lOGg g/pot 0 0 3.9 0.07 0.27 0.23 2.C2b 0.27b Ji20 Ik.k 0.13 0.36 0.32 1.12a 0.15a 17.4 0.24 0 . 50 C.30 1-39? 0.21a 6700 k.6 0.35 0.33 0.15 2.38 b 0.16a 11 . 0 0.37 0.2J 0.13 2.26b 0.18a 10 0 3.2 0.06 0.53 0.19 1.76a 0.21a ii20 9.0 0.12 O.il'; 0.27 1.61ia 0.26a 1675 l4.2 0.l4 0.67 0.27 1.07b 0.25a 6700 6.6 0.36 0.)f9 0.13 2.28c 0.23a •J . 3& U . 00 0 . i.y 1 . 50a 0.17 a 0 10.3m 0.23;n 0.34ni 0.22in 1.83 0.19 10 6.9m 0.21m 0.51ra 0.20m 1.65 0.23 0 3.5w 0.06v O.3OW 0.22x 1.89 0.2illt20 11. 7x 0.13VX 0 . Ulwx 0.29Y 1.38 0.20 1675 15. 8x 0.19X o.59y 0.28y 1.2k 0.23 6700 5.6w 0.35Y O.itlw 0.i3w 2.33 0.20 6375 6.2w 0.38y o.ii3^-. O.lkv 1.90 0.18 * Values vithin columns within subtables followed by the same letters within letter groups a to d, m to n, and w to z are not significant at the 0.05 probability level.

PAGE 154

slightly vit'i 10 pieq/lOOg of liue "udedj and tiie second P increment yielded tlie lowt^st concentration. Values i'ov K in the pj.ant vere higher than values given by Gomi.de et al. (57). Phosphoi-us decreased forjige Ka concentrations with no i.lme added., and had no effect at the high liipe rate. Overall, lime increased Na coucentrstions i:\ the pltint. In Lg3 Dis-jnantes subsoil, lime had a significant negative effect on.Pangola ,/leld (Table kl) . The latter increased considerably vith applicatj.ons of P equivalent to 1/k of the P adsorption maximum; higher rates depressed yield dramatically. Wien no lime war. applied, P concentrations in forage increased consistently; however, when 10 meq/lOOg were applied, the effect of P fertilizer cn P concentrations in forage was quadratic. The maxiraura P concentrations occurred at P application rates of 390 and I56G ppra. The second increment of P brought forage F to values higher than O.liS^.. IJo effect of linve was obser-/ed on the Ca concentration in Pangola. Phosphonas at the highest rates increased forage Ca to values greater than 0.50^;. A significant increase of Mg in the plant occurred with lime. Phosphorus had a quadratic effect on this nutrient with maxiraura values obtained with the first two increments of added P. The concentration of K and Tla in the plant were not affected by either lime or applied P. Lime had a nonsignificant negative effect on yield of Pan^^ola in San Vito soil (Table i+5). A very marked increase in yield tooli place with P rates equivalent to 1/ 16 and 1/k of the P adsorption maxiraura. The P concentration in the plant was not affected by lime, and increased substantially with applied P. Values higher than 0.18-;^ were obtained with the second P increment. Forage Ca did not show any response to lime, and increased only with the highest P rates to values

PAGE 155

Effect of l.lme and pnoipJiorus trfcsv'.ents to Los Diamante subr.oil on Pangcla di(-;i tyrass yield and nutrient concentration. Experiment Uo. 2* 139 Applied ' ; Nutrient concentration CaCO., P ^' "'Yield P Ca Mg K Na meqyiOOg pprn ti/pot 0 0 2.1 0.06a 0.31 0.25 1.33 0.l3 390 9.0 0.11a 0.27 0.22 1.74 0.23 1560 15.6 0.24b 0,46 0.23 i.25 0.20 6250 1.1 0.32c 0.49 0.16 1.63 0.20 iQio 1.7 0,34c 0.51 0.17 -1.93 0,20 10 0 1.9 0.G9a 0.)i0 0.20 I.95 0.I8 390 4.3 0.10a 0.51 0.25 1.98 0.18 1560 11.2 0.23b 0.55 0.28 1.96 0.27 6250 1.4 0.42c: 1.02 0.20 1.94 0,25 7810 1,6 0.30c 0.76 0.14 1.75 0.23 0 5.9ra 0.21 0.4lm 0.19ni 1.68ia 0.20ra 1'^ ^-In 0.23 0.6>. 0.2.1-n 1.9l!a 0.22ra 0 2.0w 0.07 0.35V 0.l8vx 1.891-/ 0,l8w 390 6,7x 0,10 0,39^' 0.23VX l,86w 0.21w 1560 13 Ax 0.24 0.50^'x 0.25x I.6OW 0.24w 6250 1.3w 0.37 0.75y O.lSwx I.78W 0.22W 7810 1.7v 0.32 0.64xy O.lDw 1.84w 0.22TV * Values within coluraas within subtables followed by the same letters within letter groups a to d, ni to n, and v to z are not significant at the 0.05 probability level.

PAGE 156

Tatle; • Effect of lirr^e and phosphoras treat.Tients to San Vito soil on Pangola dii'-itgraso yield and nutrient concentration, . Experiment iio. 2* ,.• • ihO Applied Nutrient concentration CaCO, P Yield P IS^ ' tfe K IJa meq/lOCg ppra S/pot v 0 0 3.9 0.09 0.2h 0.1^ 2.13 0.20 '+20 16.6 0.15 0.36 0.22 1.20 0.15 1675 14.8 0.37 0.33 0.20 1.56 0.26 67CO 3.2 0.)i7 0.67 0.18 2.32 C.25 8375 2.0 0.36 0.'i7 O.li^ 1.99 C.I9 7 0 " 2.8 0.09 O.kh 0.16 1.72 0.16 • i|20 11,9 O.lh 0.39 0.19 1.77 0.17 1675 15-7 0.33 0.58 0.19 1.52 0.22 6700 2.7 O.kh O.U9 0,1k 2.1|-5 0.22 8375 k.k 0.55 0.53 0.17 2.66 0.22 0 5.1m 0.29.m O.klm O.lBra 1.8km 0.21ra 7 7.->Ti C.31n 0.1|i|rii 0.17;n 2.02ra 0.20m 0 3.ifw O.O9W O.3W O.15W 1.93xy O.lSwx k20 Ik.^x 0.15W 0.37W 0.21X l.k8v 0.l6w 1675 15. 2x 0.35X 0.35W 0.20x I.5WX 0,.2\v 6700 3.OW 0.U5y O.^Sx 0.l6w 2.382 0.2i;y 8375 3.2w O.koy 0.50v;x 0.15w 2.33yz 0.20x " Values within colarans within subtables followed by the same letters within letter groups a to d, m to n, and w to z are not significant at the 0.05 probability level.

PAGE 157

close to 0.505^. T\\e first two incz-ements of P increased Mg In the plant; other treatments had no c-ffect. Fnosphoras applications substantially increasea forage K; lirr.e snoved no effect on either K or Na concentration in fae plant. A slight increase in Na concentration took place v;ith added P. Li!.ie had & small non-significant negative effect on Pangola yield in San Isidro soil (Table kS) . Phospborua rates equivalent to 1/16 and 1/k of the P adsorption maximum had the iai-gest effect on yield. Adeouate P concentrations in the plant were obtained with P rates equal to or higher than 1^75 ppm. There was no effect of either lime or P applied on the Ca concentration in the plant. Tlie higliest values for forage Mg and Na occurred with the first tvo increments of P. Lime and P had no effect on forage K. Lime had no effect on the yield of Pangola. in Grecia soil (Table hj) . Best yields were obtained with P rates equivalent to I/I6 and 1/k of the P adsorption maxiiaiim. The P concentrations in Pangola were^ adequate after applications of I925 ppm of P. Forage Ca increased with added ?, altaough values were lower than O.^OfJi, Phosphorus had a quadratic effect on forage Mg with the maxinram level occurring with the first increment of P. No effect of applied lime and P was observed for forage K. Phosphorus also had a quadratic effect on the Na concentration in tae plant; the maximum Na concentration was found at P rates of 1925 ppm. Lime decreased the yield of Pangola significantly in Alajuela soil (Table kQ) . Phosphorus rates equivalent to I/I6 and 1/1+ of the P adsorption maximum gave the best yields. Forage P increased with

PAGE 158

Table ii-6. Effect cf lime and. phc-phonis treatr-.^nt^ to San Isidro soil on Pan^ola digitgrass yield and riutrient concentration, Experiraent No. 2* Applied Nutrient concentratjon CaCO.-, P Yield P Ca K Ka Dieq/' lOOg ppn. 6/?ot .0 0 1.7 0.11 0.2<'+ 0.16 1,9J 0.18 370 6.6 O.lb 0.25 0.20 1.71 0.26 1-475 8.0 0.36 0.28 0.16 1-93 0.25 5900 3.2 o.i^5 0.29 0.12 2.00 0,16 7375 2.0 0.38 0 .-26 1.79 0 . 11 k 0 2.6 0.09 0.31 0.13 1.73 0.17 370 5.0 0.17 0.37 0.18 1.73 0.21 1*^75 6.k 0.36 o.ho 0.16" 1.98 0.25 5900 2.7 0.56 0.38 0.14. 2.54 0.21 7375 1.8 O.kk 0.39 0.11 1.99 0.15 0 0.29 0 . 26m 0.15ra i .0 1 in O.lQra 4 3.7ni 0.32 0,37pi O.lW 2.00ra 0.20m 0 2.1w O.lOw 0.28v O.lW 1.82w 0.l8wx 370 5.8y O.lbw O.3IW o.l9y 1.7i^v 0.23x li^75 7.2z 0.l6\ O.3UV 0 . i6x I.96W 0.25X 5900 3.OX o.5iy 0.33W 0.13W 2.27V 0.19WX 7375 2.3w 0.33V 0.12>v I.89W 0.13W * Valuer, within colurans within subtables followed by the same letters within letter groups a to d, n to n, and w to z are not significant at the 0.05 probability level.

PAGE 159

Teblc; '^7. Effect of J.iine and paofjphoriis treF.tments to Grecia soil on Panj ?ole dig! t^s.z:i y ield and nut rient concentration. Experimenu. No. Applied concentration P Y < #^ 1 '1 p Pa Mg K Na 0 0 _) • 0.07 0.18 0,13 2.0U 0 kSQ 9.5 o.i4 0.21 0.23 2.08 0 28 C.19 2.09 0.26 7700 3.3 0.52 0.30 0.15 2.60 0.41+ 0. ^8 0.13 2.29 0.15 h H Q <^ • J. 0.11 c.29 0.11 1.83 it8o 0.15 0. 36 0.27 2.17 0.23 1925 10.6 0.33 0.35 0 . 30 7700 3.3 0.43 0 . 3-'4 0.13 2.10 0.17 9625 2.8 0.50 O.Ul 0.13 2.kl 0 1 7 0 6.2m 0.29ra 0.28ra 0 . 17ra 2. 22m 0.20m 5.7in 0.30m 0.35ni 0.17in 2.19a 0.22ra 0 2.5w o.09w 0.23w 0.12W I.9W 0.21VX kSo 9.6x 0.15W 0.29w 0.25y 2.12W 0 . 25vx 1925 11. 5x 0.31X O.3IX 0.20X 2.26v 0.28x 7700 3.3v O.iwy 0.36xy 0.1'4V.' 2.35W O.lSw 9625 2.7w 0.47y 0.39y 0.13w 2.35W O.low * Values vithin columns within subtables followed by the same letters within letter groups a to d, m to n, and w to z are not significant at the 0.05 probability level.

PAGE 160

V, Ihk Increasirjt?; rates of and P iricrexeni-s eqjxe.l r,o or higher than 1925 ppra produced adequate values. Galciux in the plsnt increased significantly v,'ith lime ar.d P rates higher than i-SO ppm. The lime effect on concentration was ncnsignificant; only the first two increments of P . . increased forage Mg. Potassium in "che plant was not affected by lime and P add<--d. Applied P had a quadratic effect on forage Na; maximum levels occurred at 770G ppm. • . Ho effect of lime on yield of Pangola was observed in Agronomy Area soil (Table k^) . Phosphorus applications increased yields significantly in comparison with the no P treatment, but increments of P higher than I37 pprn had no further efftct. The forage F concentration increased with increasing applications of P. Adequate P levels were found with P rates higher than I37 ppm. Calcium in the plant was increased with P in the unlimed soil; there was a quadratic effect of P in the Ijnied soil, and Cs concentrations were higher with the first two Prates. Phosphorus had a quadratic effect on the % concentration in the plant. The first two P increments gave the highest values. Forage K and Na increased significantly at the highest P rates. Lime showed a significant negative effect on Pangola yield in Cararao soil (Table 50). Phosphorus applications equivalent to I/16 and 1/k of the P adsorption maximiim gave the highest yields. Forage P increased as P rates increased. Additions of P equal to or higher than 187 ppm proved to be sufficient to increase P concentrations to adequate levels. A quadratic effect of P on the Ca concentration in Pangola was observed; inaximuTi Ca levels were found with P applied at 3OOO ppm. The first two increments of P affected forage Mg tae most in a positive way.

PAGE 161

Effect, of lime and phcsphonis treatments to Alajuela soil on Fangola di>~itgras3 yield and nutrient concentration. Experiment JJo. 2* Applied ... . , . Nutrient concentratio n CaCO^ P Yield P Ca Ife K Na meq^'-'lOOg ppm g/pot —-^ 0 0 6.1 0.11 0.20 0.l6 2.9& O.ll^ Uao il+.2 0.13 0.21 0.17 2.71 O.OS 1925 18.9 0.26 0.2'7 0.20 2.32 0.16 7700 4.2 0.52 C.36 0.15 2.18 0.20 962^ h.3 o.h3 0.28 0.12 2.28 0.12 18 0 7.5 0.12 0.35 C.21 2,83 0.07 ^\&C 9.3 0.16 0.31 0.21 3.00 0.10 1925 12.1 0.27 O.3B 0.22 2.88 0.15 7700 6.k 0.i|7 O.UO Q.lh 3.03 0.18 9625 k.9 0.U2 0.1+2 0.13 2.I12 O.lll 0 9.5ni 0.29m 0.27m 0.16m 2.1i-9m O.lkrc. 18 8.2n 0.29ra 0.37n O.lSm 2.82m 0.13m 0 6.8wx 0.12\; 0.27W O.lSx 2.91v O.lOw k80 12.0xy O.I5W 0.26w 0.19x 2.85w O.C9w 1923 15. 0.27X 0.32WX 0.21X 2.59W O.lo.xy 7700 5.3W 0.it9y O.38X G.lkv 2.59W 0.1?y 9625 k.'Sv 0.1+ 3y 0.3:ix O.I3W 2.35W C.I3WX Values v?itiiin columns vithin subtables folloved by the same letters within letter groups a to d, in to n, and w to z are not significant at the 0.05 probability level.

PAGE 162

Tatle ii9. Effect of li;ne and phosphorus treatxents to Agronoini' Area soil on Pangola digitgrass yield and nutrient concentration, Experiment No. 2* CaCO,, P Yield P Ca Mg K Na . me(3_/''lC02 ppm g/pot ^0 0 1-9 0.06 0.23a 0.1k 1,77 0.l8 137 6.7 0.13 0.21a 0.19 1-^ 0.l8 550 7-5 0.31 0.28b 0.21 1.59 0.25 2200 9.3 O.5O 0.36d 0.17 2.13 0.33 2750 5.7 0.55 0.32c 0.15 2.61 0.27 6.0 U.5 0.09 0.k2a 0.13 1.32 0.20 137 6.5 0.13 0.50b 0.17 1.42 0.17 550 7.6 0.32 0.55c 0.20 1.39 0.23 2200 5.8 0.53 0.1^7b 0.12 1.91 0.26 2750 6.k 0.53 0.i^3a 0.12 2.18 0.26 0 6.2m 0.32ra 0.28 0.17ra 1.91ra 0.2W 6 . 6.2ra 0.32ra 0.^7 0.15m 1.6i+m 0.22m 0 3.2w 0.08w 0.33 O.I3W l.^kv 0.19V 137 6.6x 0.13V 0.35 0.l8xy I.45W O.lSw 550 7.6x O.3IX O.hl 0.20y l.ij-9w 0.2Wx 2200 7.6x 0.55y O.hl 0.15wy 2.02x O.3OX 2750 6. Ox 0.5% 0.37 0.13W 2.39X 0.26x * Valuco within columns vithin subtables followed by the same letters within letter groups a to d, ra to n, and w to z are not significant at the 0.05 probability level.

PAGE 163

1^7 Table 5'^). Effect: of lime and phosphorus treatments to Cararac soil on Pangola digitgrass yield and nutrient concentration. Experiment Uo. 2* ^ipp.l.ied Nutrient concentration CaCO, ,J p Yield P Ca Mg K Ka rneay lOOg ppm g/pot 4... C 0 3.1 0.10 0.22 0.15 1.88 0.18 187 7.1 0.18 . 0.18 0.19 1-79 0.19750 9.0 o.ii5 0.2h 0.20 1.90 0.25 3000 5.8 0.58 0.3k 0.16 2.57 0.29 3750 H.3 0.50 O.3U 0.13 2M0 0.23 h 0 187 750 3000 37SO 3-3 6.1 8.1 h.3 k.Q 0.11 o.ig 0.k2 0.58 0.62 0.29 0.3k o.ko 0.50 0.32 0.16 0.18 0.18 C.I6 0 . Ik 1.-85 1.71 1.63 2.51 2.21 0.18 0.19 0.22 . .. 0.26 .. 0.21 : 0 5-9m 0.36m 0.26m 0.17rn 2.11m 0.23m ^ ' 5.^n 0.39m 0.37m 0.16m 1.98m 0.21m 0 3.2w 6.6x O.llw 0.26w 0.l6vx l.86v 0.l8w 187 0.19W 0.26w 0.l8x 1.75W 0.19V . 750 8.6y 0.J+3X 0.32W 0.19x 1.77V 0.24w 3000 5.IX 0.58y 0.U2X 0 . l6wx 2.5i+x 0.27W 3750 U.6vx 0.56y 0.33WX 0.13W 2.3OX 0.22W Values within columns within subtables followed by the same letters within letter groups a to d^ m to n, and v to z are not significant at the 0.05 probability level.

PAGE 164

Only the highest rates of P increased K in the clar.t, and no effect of either liine or P applied vas deterniined for forage Na. Phoophorias had no effc,ct ;:n Pangola yield in limed Drainagfc-wa\/ soil (Table pi). Forage P increased as P ratee increased. Adequate P concentrations in forage wei'e obtained with applications of 5OO ppm of P. Very little effect of P vas otasi'ved on the concentration of Ca, K, and Na in the plant.Forage I«5g increased vith the first P increment, and no further effect occurred. Yields of Pangola inci^eased considerably vith added F in unjlimed East Carimag.ua soil (Table 52); however, no increase was present vith P applications higher than 207 ppm. Yields were generally lov^er when lime vas added, but an increase occurred vith added P. Phosphorus applicatioas equivalent to 1/k of t?ie P adsorption maximum gave the beat yields in the unlirned soil; however, higher P rates were required in the limed soil. The same trends were observed' vhen forage P vas the dependent variable. Both lime and ? increased the concentration of Ca in the rjlant. Forage Mg was not altered with P in the unlimed soil, but increased with increasing P rates in the limed scdl. Potassium in the plant was not affected by lime or P. Foi'age Ka j.ncreased with the highest P rates. In general, lime had a negative effect on the yield of Pangola, This indicated the failure of the response surface design used in Experiment uo. 1 to predict the best lime rates needed. The small response due to lime was probably the cause of this failure; however, the large response to P api-lications was confinaed. The P adsorption maximum vas a very good index for P fertilization, and best results were obtained

PAGE 165

Table 51. Effecrof phosphorus troatKn-'nts to lireed Drainageway boil on Pangola digitgrass yield anrl nutrient concentration, Experiment No . 2^:; Applif Yield Nu.trient concentration P P Ca MB K meq_/lOOg ppm g/pot 7 0 2.3a 0.078 0.37s .0.10a 1.45a 0.30a ^00 G.ka 0.18b 0,ii0a 0.17b 1.85a 0.3l^a 2000 S8a 0 . 36c 0 . 34a 0.16b 2.59a O.i^Ta 8000 2.9a 0 . ^^a 0 . i5b 2.39a 0,17a 10000 1.6a 0.55d 0 . kha 0 . 16b 2.40a 0.21a Values within columns followed by the same letters from a to d er° not significant at the 0.05 probability level.

PAGE 166

. r . . . , * >. Table ^j2. Effect of llwe and phosphorus treatinents to East Carimagaa soil on Pan^ola d.lgit»3rass yield and nutrient concentration. Experiment No. 2*150 App lied Nutrient concentration CaCO^ P Yield P Ca Mg K Na .J meq/lOOg ppm g/pot ffo 0 0 3.2a 0.08a 0.22 0.20a l.?8 O.I6 52 5.9b C.lla 0.19 0.20a l.lv8 0.14 207 8.2c 0.20b 0.2^ 0.21a I.I6 O.IJ+ 830 7.9c C.oUc 0.36 0.20a l.k?. 0.19 1037 8.5c 0.75d O.hk 0.21a 1.64 0.22 20 0 1.4a 0.08a 0.55 0.09a 1.30 0.16 52 3-6b 0.13ab 0.?5 0.12ab 1.42 0.l4 207 3.2b 0.19b 0.54 0.15b 1.57 C.I5 330 5.0c 0.32c 0.79 0.28c 1.72 0.17 1037 5.8c 0.43d 0.84 0.29c 1.52 O.IQ 0 6.8 0.35 0.29m 0.21 1.46c 0.17m 20 3.3 0.23 0.65n 0.19 1.51ra O.lbm 0 2.3 0.08 O.38W 0.15 1.44w 0.l6wx 52 4.8 0.12 0.37V 0.16 1.45W 0.l4w 207 5.7 0.19 o.40v 0.18 1.37V 0.l4w 830 6.5 0.43 0.57X 0.24 1.57V O.lSxy 1030 7.2 0.59 0.64x 0.25 1.58V 0.20y * Values vithin columns vithin subtables folloved by the same letters vithin letter groups a to d, m to n, and v to z are not significant at the 0.05 probability level.

PAGE 167

151 vith P rates equivalent to l/lo and 1/ h of the P adsorp-ion na;cirnu:n. The latter vas superior in providing adequate nutrients for good growth. Forage P values were equal or higher than 0.18'^ only vhen P rates larger than lOCO ppra were used in the soils from Costa Rica. The P source was also a good Ca source, and could be used instead of line to provide needed amounts of Ca. This obser'/ation suggested that the effect of lime was not due to pH, but rather to nutritional disorders; the lime levels chosen were probably in excess of what is required for Psmgola. The r^g in the plant was always lower than O.ifO^o reported by Gomide et al. (57). Magnesium fertilization using rates higher than those applied should be considered although the I-lg effect in Experiment No. 1 was not significant. The P effect probably masked the lAg effect. The levels of K and Ila in the plant were always adequate for growth (56, 57). Experiment No. 3 A clear superiority of CaoiO^ over lime in providing better yields of Pangola was not observed in the soils from Coota Rica except in Los Diamajites surface soil where the low atj-olication of CaSiO save 3 the best result (Table 53). Only in Grecia soil did GaSiO^ give a higher forage P concentration than lime. Lime was a better source of Ca. Forage Mg was not affected appreciably by either one of the materials. The K concentration in the plant was significantly lower where CaSiO^ was used at the highest level in Grecia and Alajuela soils. No treatment effect was observed in the other soils. Much higher Ka concentra. . tions in the plant were found with Ca3i0although no Ila was present in • the fertilizer material.

PAGE 168

Calcium silicate gave better Pangoia yields in the Colombian sciis excep't Agronomy i\rea. This -.'as especi;iliy trut: v/hen high rates of lime vere added (Table • Phor-phorus concentrations in the plant vere hi£;her with CaSiO^, except from the Agrononiy Area soil. This effect was nonsignificant in some cases. Higher concentrations of Ca in forage were obtained with lirae. The treatment effect on Mg was very small. Foraf^e K did not vary appreciably due to treatments, and the concentration of Ka in the plant was higher V7ith CaSiO^. In general, the effect of CaSiO on yield and forage P was 3 not as proaounced as suggested by other studies; howeve?.', application rates were based on the response to lime observed in Experiment No. 1. The fact that yields using CaolO, vere in most cases superior to yiej.ds using the highest rate of lime, and Ca levels in the plant were much higher vn".th lime, suggested that electrolytes do not accumulate as readily with CaSiO,. It is possible that. Ca, NO-,, and KCO, ions generated to a large extent by CaCO^, are the main contributors to the short-term negative lime effect observed before in Los Diamantes soils in greenhouse trials (1^5^ l6'+).

PAGE 169

153 c r-l to m ^• Q .--i -D fO C\j ro LfN d o d d a? aJ c(5 ^ H '.r ^vo CO OJ OJ _:} » • • • o o o o 03 tC Ci 03 CJ OJ «^ <^ o o d o as a; S3 ^ C^ O VC CJ r-« rVi OJ 'J~\ d c o d Si c! c3 13 ctf j~ ir\ ir. t> OJ f-CO • • • * t~! O; r-i eS c6 r! cfl VO O Cn 01 -JCO t-a) * * • r~i r i iH r~l Oj 05 03 0? CO CO 0 -. • 4 • f-i H f"i r-; 03 e9 c3 «5 O O .H 0\ O fco OJ H H o a o o -1-5 c OJ O 0} d 0} oo CM OJ Cvi d o C) o a; o o d o d d /5 s\ ro f) r-l H O w ,Q cJ b3 p3 ^ l>-_4 J c ", CM r-l cu ': j OJ O C3 O O rH •H •p o p. (Jo * ,a cfl CO ^ to CO O rH C7N • * • * t-CM CO Cfl CO Cfl o3 JH -di-H cr\co cTv o cS 03 cfl cfl H CO O O l-H r-J t-1 flj ^ O ^ CM O CJ OJ -aCO CO CO 'C3 rH tPv O d d d d rH

PAGE 170

15U a o •H +•> tr! (-1 -P C! 0) o fl o o -p c a; f^ •P a; 2 •H a o o 00 'd P. o •H o 00 o c > 03 o a! •H O 0) TV b to to o o a) a I ,0 aJ X) t> t~o ra\ CM u-\ • * • « o o o o ^> ^ A3 ca {t-) r>-u r-; OJ oa CM oj ,Q P P tfl LTv CM VO W CM CM iH C3 o d d <0 p C3 cC H n-) unco 00 lTn CM t\i • « • • 0000 CO 03 (S p t— O CO rH H CM CM • • « • 0000 1 rC eJ CvJ 05 o ft-VsO CO vo o o Cvj ^ CM o o • • CM ^ CM 05 CJ cS vD 0.1 rH r-i H I 'l « • • t 0000 P P C5 O O O Lr\ CO (^-^ CM H CA VC; CM CM r-i rH 0000 0) p cd c3 \£) vo ro UA CM 00 01 CJ d d d d Rj Cj tfl (3 O U N rH CM CM CVJ .^J d d o d p c8 b5 c3 VO 00 O CO • • • « H ON O t-H .-H O O rHCC o o tH 00 •P o ca o o P •P o c; -d § • rH H • O H H-( a; > 'J) 0) P >> OJ -P P -H -9 W ,Q 05 g O U ft LA o o u c •H r-« P iH 0 rJ H C3 > 0 -p p cd -p a o •H •iH rJ bO fH w

PAGE 171

15'; CO — i o !S o 03 Tl B O H o o * G O rH o p r~i El •H (D O E •H !•< o O p R< X CO ;^ J" s o +•> •H P 0) CJ M U 4-3 P a v y >-4 o C) •H o H •H p ;n f1 (LI B :i •H •H c, 03 o 'rf «} § X) a' •H H w CO o r3 Ih •p bD O +> o •H O OJ VO OJ OJ no • • • o o o o • • • • CO CO O O f)vr) o o l-l ctf C) 03 S u ft cS 05 -O ^ VC L'^CO ON 0-) t»-i >x; o o o d q3 cd 03 05 COJ rH Ll> ON CT\CO vo • • • • 1 — I r-i rH r— I oJ 0? 0} 05 rO rH OJ Cvl rH H rH H o d d d -O O o 03 03 LTx OJ LA CVJ o o o o cC o3 o3 C!5 ONCD CNO rH H rH CVl o d d d •so 03 s 05 o +> £1 ^ 03 05 01 ;3 rH CO On oo • • • « lA LTx tQD 03 s! o3 ^ On'vO rn rH iH CV rn • • • • o o c o CJ ^ CJ oJ V) O H O CU 00 oo r-l • • • « .-I rH H H CJ 05 03 f"'! C7N 'H « OJ O.I OJ rH • • k • o o o o o O rH rr, CO -r\ o o d d ftj OS 03 ^ d d d d 05 05 ^ ^ O LA ON o^ O CD d H OJ 00 C CO CC' o o r-f CO o o rH OJ o o H CO O 03 o cn -p 0) r-l 0) § H H • OH > in (U OJ rH •H ^ >J +> P -r-t ,a H m ,Q
H 4-> A •H ;» LA o rH^ o p C 0] P a 0! rH rH '.-D 05 % t > 01 *

PAGE 172

.. SUMMARY AND C0KCLaSI0N3 . ' ' ' \ Soil properties ai? fee ting p?^ant respoixses to lime were examined in soils from Costa Rica and the eastern .oavannar. of Colombia. The soils froi'i Costa R.ica showed higher pH^ CM, CEG., EC, and exchangeable base values than the soils froni Colombia.. Exchangeable Al vas considerably higher in the latter. Drainagevay 3oil from Colombia had chemical • '• properties similar to the Costa Ric£i.n soilr.. Total C&... Mg^ Al, -Fe, and P were lower in the soils from Colombia, , ,. . • • • Amorphous materials tind kaolinite were the predominant clay minerals irr the Costa Rican soils^vhile ver;n:" ciilite and kaolinite were dominant in the Colombian soils. . • • Amorphous forms of Al and Fe were fcxuid to a larger extent in the Costa Rican soils; hovever, the crystalline oxides of these elements were the largest fraction present in all soils. . Lime requirement, based on Ex.-Al, vas higher in the soils from Colombia, but, in general, much lower than the lime requirement calculated with the Sf-IP and Yuan buffer methods. The buffering capacity of the soils from Costa Rica vas attributed to OM and amorphous Al compounds including organo-metallic complexes and polymetric Ai. Exchangeable Al was the main source of acidity in the Colombian soils. Tltratable and extractable acidity correlated best vith OM and amorphous forms of Al. 156

PAGE 173

157 PhoKpiioriis adfiorption v/a? hij-;!! in the so.llr. from Co3ta Rica and Da-airiagewaj* froT, Colombia. Cox-? •elation Analysis showe
PAGE 174

158 The ? Gdsoirptlon iraxinura vas a very valuable tool in determining P fertilizer needs of Pangola in all soils. Verw large ainounts of added P vcre necessary to increase forage P to valvies adequate for grovtii in the Costa Rican soils. A response surface experiment failed to predict adequate linie and Mg needs, and Ca:3iO_, did not )iave a definJ) ite advautage over GaCO, in, improving yields and forage P concentrations. The negative effect of lime was associated with the accun-ulatioa of electrolytes in Los Dianantes surface, G-recia, and Alajuela soils fro:n Costa Rica. The lime effect vas very small, and, in general, soils high in exti"actab].e Ca exid lov in Ex.-Al had the largest negative response to lime. ' " . Basid on the experimental results obtained in the latorato;ry and' the gree.uhcuse, the folloving conclusions can be made to satisfy the objectives sez forxvard at the beginning of this vork: (1) ./\i;iorphou3 forms of Al vere very important in determining the buffering capacity of soils high in organic matter and non-crystalline components. Exchangeable Al was more important otherwise. (2) Adsorption of P v;as very large and v;as closely related to OM anu amorphous forms of Al. Lime increased P adsorption, but also increased P availability through . , its decreasing effect on the constant related to the P adsorption energy. ' . . (3) Neb negative charges were present in all soils. Extensive positive charges were found to develop only at pH values lower than 3 •13.

PAGE 175

159 [h) High concentraticno of Ca in CcClg solutions vere found to be impractical to raise soil pH, but were very efficient in increasing adsorption of Ca. (5) Specific adsorption of Ca vas found to take place at high Ca concentrations in solution-. (6) Calcium. NO^-N, and EC of saturated extracts increased markedly with increasing applications of li:ae in Loa DiaiTiantes, GreciS;, and Alajuela soils from Costa Rica. (7) Amounts of lime required to raise pi to values suggested for temperate soils (6.8) would likely be impractical. These quantities are probably not required to supply adequate Ca and/or,i-!{5 as nutrients. (8) Phosphorus was the principal limiting factor for Pangola digitgrass growth. Nitrogen effects were not ueasured. (9) The effect of lime on plant growth was very small and was related to the amount of extractable Ca and Al in the soil. The primary beneficial effect resulted from improved Ca nutrition and detoxification of Al. (10) The negative response of Pangola digitgrass to lime nay have been caused by nutritional disorders associated ; with the accumulation of electro3.ytes such as Ca, r;C-,-N; and HCO^'s which are by-products of the lime reaction. (11) Magnesium was needed to maintain plant grov;th and development, but its effects were masked by more important factors such as P.

PAGE 176

l6o (12) CalciuiR silicate va--? not sho-'.-n to be superior to CaCO in improving yields and P coucenfcrations oi Pangola digitgrass. • ' . -• — • •

PAGE 177

APPENDICES

PAGE 178

APPENDIX A TABLES 55 TrIROUGH 65

PAGE 179

il l.r^ CO H H O a ate M o >> +i o 0} o CI o •H o m » p( O to X! o Xi CJ p< Q) >i 43 -P •H to d a o o +J e •H •H rH O j +> ^3 O 0) •d Vi ^ o o o 1 o o o o O o O 1 o o p o o OJ C CM CVJ 1 ITv ^7 rH OJ CO to o to H CO eS •H o ID u o 0." •o 01 o C O LPv O •M tr\ LPi OJ ir\ Ol O tir\ H H r-l H LPv O . O O O . ir\ OJ f — Xh O cv; CM G CO ir\ CO rH rH H H H rH o a o u o cd a u a o ol cti a OJ CM o o Cvi o 1 IJPV o O O in, O O 1 ' — 1 fn CM O O CJ CVI ir\ CM C cA VD OJ O -4vo OJ 1 rH H H H H r-l H o O O O C Lr\ O O o Lf\ O OJ '_r\ cn r-i H CO vO VD l/N CO rH rH H rH H rH H rH o O O LTN co OJ rH o o CM o o o O O VO ITN CvJ rH o .r\ o l/N o CM ir\ OJ vo vo VD r-l H ir\ o CM OJ r — VD hH H rH CO O O —J o VO ^.FN ON CO OO LTN CO O -4VD CO CO rH rH H r-) H H H rH 05 s •H vy a)

PAGE 180

o r-t 0.) V •H r-H o Qi w 1 6 T CO rH rH t0.! ro vo ro CO O L ir\ ^oo oo ON H H or> to ro c Lr^ CVJ • • m vo vo oo O rH cu CO w (O o OO m N U"\ 00 a"? ro o H OJ CM o on a) vo r-l vo Jir\ ^CO -4 ON rH H X) o rH OJ VO ITN VO CO CO 00 rH H r*Tv t — \S\ On ro • vo Jvo t— oj OO ON. rH • ^00 ON IPv vo CO vo ^CO O rH Ol V • r — o OO O o LfN t• • (O d VC A o rH H H CVl U cS to o o u o d p •H H in -H t', cS > M Ti O ••-3 n 0) 03 OJ Cj u rH CO to o o o o 0 U a o c •H 03 ^< u3 O in

PAGE 181

165 c v.") CO IfN CO On O o rH (S o o OJ rH o r-l O ON o ri O •H C M CO .~r VO VO iTn iH OO H C3N VO VO vc VO ITN ir\ 1*" VO VO to o LTN ITv tVO Of) o cc t~ VO irN VO VO VO ON UN CVJ m VO LTN VO Lr\ VO VO r>, LPv ON ir\ CO ^VO ITN o VO OJ VO Oi IP. Jir\ CVJ VO CO LPv CO LTN tJOn O H OJ CO" fi/N VO VO LTN VO iTN PO OJ CO CO ON o OO if\ VO ir\ LTN L'-N LTN ITN t~H o VO ON VO H ITN O LTN VO t/N L.'^ LTN IT-, t/N ON ON irN co OJ CN J\f\ l/N ITN -ciITN VO ro OO no ITv ON OO ^' ITN * ITN .» l/N LTN OJ H OJ LTN VO OJ ON ON ITN LTN LTN LTN ITN H T^ O CO to a4-> o •H CO 6-1 o CO o i u O d -i-> •H cn •H M Q >• M to O 03 •H O 0) o 05 H 0) •J OJ r-) < I I a o o 0} 0) cd o >1 03 U cC to 03 R •H ^^ 03 O +-> o 03 W

PAGE 182

3.66 1.0 o o H 7?* CD •M ri Oi Pi «5 c O o C3 O O Cv'l O O « O v6 O or, fJ4 on [ ir\ VO UN VC Lr\ ..-\ vo ' ^H VO UN VO • o VD lA -:} l.'N (A. VO CO lA OJ • UN Cn ir\ CO H OO o VD <-% ITv VO ir\ ITN ITN lA vD lA lA rVD • -:t 'sO VD ITN O ft LA H irN CC « l/N lA irx o • • C" — H <\1 VO CO O CO H ON OO VD ur\ L<-\ lA L'A lA r — • OS • On OO iTN CO r'i • lA • .4" CO a-) ^O ir\ CO ir\ On (M LA • .4VO • H ON VD VD O -iVO CO H LA -•t -4VD ITN CO H ON ir\ -4OO LN VO • rn LA rH J• • on VD » --IO a •H CO fNJ LA C\J -J00 VO rH O to p 5 OJ o u O m :i to o to o •H t'l n o O 0$ H a; •o 0} rH < 0 0) 6 a o o (-1 c3 O •H a) ft •H ^^ tj OS

PAGE 183

l67 q H cr* E5 •-I '^ P, O s o .-1 CM H d O ON O O no r-i 03 •H P •H a M o I X r-i o to to +> Q 5 o o o o 6 o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o c o o o o o o o o o o o c o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o r-i o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o H o o o o o o o o o o o o o o o o o o o a CM o o o o o o no o o o o o o o o o o o o o o o o CO o o o o o o oo o o o o o o o o o o o CVJ c ON CO ON H CO o o o o CO OJ CO ^1 o o o o o o o r-l o o ir\ o 00 ON o H o o ro r4 crj ITN o o o o o H o H CM r-i o o U \ CO OJ o i-i H LPv o fo o rH o CO H OJ o r-H o C\ On 0^> ON H o fo o r<-i CO • d o H d H d o-) CVJ d IS (0 0) H o •H •s a; o o i CO > •H ,o o 7S p rH o CO to CO •H w (S OJ o 0} o M •H a M a o o cd •H p 0 a' 00 'fl

PAGE 184

168 o u •H J1 C5 f O o o ! ON fO >.o -TN UN ON • • 1 • H J1 oo .-H •r-l O 01 o 1 fYN •'"'N C\J C\! J VD CM CM LfN H CM >— i Li A CN ON O ITN hU> On UN on • • Cj\ I — J" -4. r"( Cvi H { Cvi O j 1 G\ VD fv-) o. Li% ON o. • « • 0\ CO CM CM j m H J\ J { o o 00 UN VO t— H CD VD CM no 1 — ! k Pi o •s -=^ O CV CM ON CM t\o cO LTN on [ CJ 1-1 L/N 1 1 OJ I OJ ^\ \ V _ Li \ ON ^N -4. 1 CO VO o o r-{ CM m ^ OJ OJ irv o l/N VO ITN CO ON LPV ro H ITN H VD OJ Ol LPv CM OJ UA O O VD ON ^i/N VD ON LTN « r-l t~iH LTN d H ITN oo nn a! C5N ON
PAGE 185

169 bC O .0 H (? 0) Oi) O O O OJ O r-i o ^o O H O i. ft 0 0 UN 0 ir\ 0 4 1 o\ H 0^ rH CO ro ir\ 0 0 0 oUA LTN • • • Cxi Lr> m 'TO 00 -:t 0 0 0 U\ ur, 0 VO 1 — 1 Lr\ 0-) 0^ o-> LTN JJ0 0 0 0 L/^ 0 ITV 0 ...1 1 — 1 Lr\ 0\ 0 0 0 0 0 0 r-l '1 •0 d ( — 1 vl:? VO 0 ir\ 0 0 LTN 0 0 ir\ • CO lA H rH 'd VO ir. 0 0 0 UN 0 0 LfN OJ 1 — t VO oj rH d , 0 0 0 0 CO r-i a^ H CO ir\ 00 LPi 0 0 0 0 Ui'N ir\ C\J oj rH CO co' OJ 0 0 m d H ON H CO ir\ VO 00 o -p § •H Q m O a u H •H o to cn 0 u 0 c« +3 •H r-i •rl OJ > n n u 0) 0} CD Ctl CO ro 0 o a o u o U ca crt o 0} a •ri 03 O 3, «] O P (0 ca

PAGE 186

170 cvj LTV IfN O r-i o vo 0-) OJ C\i ON Cvl CO OJ o o CO d ON OJ Li \ OJ OJ 0\ OJ <\J CO O O H 0) 0) O o o o o OJ OJ cr. cn OJ CM H OJ H H LT. CO ON -rN H CM CM H 0^ H CJ H OJ O OJ orCV! 0\ o CM CO ON 0\ CO OJ CM r-l C3N i-f r-i 0-) CM CvJ H O H ^rH LTN r-l o VD O CM LTN LfN OJ rH CO CVJ LfN. OJ OJ O rH ON o C7N r-i O rH m CT\ o -3VO o CM CO CM OJ o rn w •p a H o o in o CO O -p > en o OJ •H O V u o r< 03 o Ih ;>i R o a c o o CO n 03 03 O w

PAGE 187

171 to o o r-l "'^ V 11) CD •H rH P, a ni O O 05 U o o O o O ft L I -*-> L l/N t CM OJ H CT> VO rH 6 d CM r^CO C — (' # CM H OJ i7n r-i ON .H O H O o o ir\ CO CM CM VO • « ON CM CO H rH (Si CM CM Ol d r-l d OJ rr) vo i>t-^t• « • 6 CO H H CM CM ON rH O H O d t-CO LTN O O LTN rH JCO OJ OJ CV) CM H OJ O 0.1 o o O VC _-t b-\ CM LfN 0') LTN LfN rH CM oo CO O CM O o O 1-1 VO ir\ (TN rH VD ITN CO no CM on VD CM OJ O CM o O O O ITN OJ LTN >X/ Lr\ ON cn CO H CM OO OJ OJ O OJ O O o VO o O ITN VO CM LA VO LTN \o OJ CI 1^ CM o CM O o OJ OJ JON OJ CO ON u\ VD CO H H CM JCO CM o H o o rH •H O CO 01 a i Q 0) O o C/3 o n to O (-1 O CS 4J •H H •H tn cS QJ > M •H o ^ d OJ (A OS .— 1 CO o 10 I i? o c o o 0} u a o (0 CS H

PAGE 188

1 o o ro H j 1 rH o f [ O o'l O <-i [ 1 1 1 1 o [ /~\ y — ^ 1 1 1 H o 1 * O • On 1 Q o 1 H o 1 (—1 1 t 1 o 1 O CJ ir\ O , j -H i-i P< C o O a\ 00 1 o o Ki ! j t ON o t ; o I co 1 i o o H o H O O CO -p •H a H o o CO CO O c o c ^o o CN o CVJ o o H m O CM r-l CO •r) O .Q CO CO CO vo o rH O tO rH o o o o CJ o UN CO o CO o 0\ c:i o o H o o vo O o o o o o O cvj o o rH o o vo rH CO o -{ > R CO CO o x) •H to M ON o vr> rH rH t l-l CM rH l/N C ro vO 1/N ON VO CM OJ ^or; OJ VO C7N LTN (7N CO o o O CO r-H VO ON LTN rH ir\ 00 vo O vo r^ CM H -d^o CO O O O VO o VO on vo CM ON VT) H LTN l/N LTN ON -4vr.> o LTN CO CO CO ? CO 0! rH c o hi (0 CO
PAGE 189

173 w o o .-I El U) & O o O ITS O CV) r-f O O o 0-1 LPv S o -P OJ o o o oa J! CM * • « o '/\ o rn CO o r-; CvJ m CVl rH CVJ r-l r-i o O l-l o o o o OJ as OJ OJ OJ c H CO CVJ • H d o d H r-l d o d d CN a
PAGE 190

APPENDIA B TABLES 66 TimOUGlI 75

PAGE 191

Table 66. Effect of lime, rr.apcne-ii-.iin, and phosphorur, treatments to Los Biamante,o surface so:!l on Pangola digitcrass yield a.ad nutrient concentrat.ion from Experiment Mo. 1 Ilut rlent concentration-x-"Treatment Total number yield* P Ca K Na 1 P.i^ 0.10 0.68 0.37 1.54 0.18 2 5.6 0.11 0.53 0,30 l,5^i0.17 3 6.3 0.10 0.k3 0.3^+ 1.27 0.14 k 6.6 0.13 0.59 0.44 1.84 0.20 5 6.1 0.11 0.''i8 0.27 1.33 0.15 6 6.4 0.10 0.53 0.31 1.53 0.21 7 5.1 0.10 0.55 0.32 1.38 0.16 8 ' 6.5 0.13 o.6i 0.38 1.69 C.22 9 5-7 0.11 0.45 0.39 • 1.47 0.19 10 5.9 0.13 0.57 0.36 1.50 0.15 11 5-9 0.13 0.52 0.30 1.62 0.15 12 6.7 0.12 0.U6 0.35 1.66 0.15 -:> 3.2 0.08 0.48 0.31 2.15 0.17 Ih 6.li0.12 0.55 0.35 1.29 0.16 15 5.2 0.12 0.50 0.35 2.00 0.19 16 6.7 0.10 0.48 0.36 1.33 o.i4 17 7.1 0.11 0.64 0.41 1.55 0.l4 10 6.3 0.12 0.59 0.38 1.62 0.19 * Total of three har'v'csts. ** Mean of three harvests.

PAGE 192

176 Table 67. Effect of limej map;ni.;ulujrij and phosphorus treatments to Los D.iarnantes sufcsoi.': ot; Pancola dic; tgrass yield and nutrient concentratiori frci Experiffcnit No. 1 Nutrie nt c or-ce nti' ation** Treatment Total number yield-*-' P Ca Mg K g/pot 1 3.9 0.11 0.53 C.34 2.07 0.17 2 3.9 0.10 0.60 0.30 1,69 0.14 3 k.2 0.11 0.50 0 . 45 2.10 0.16 k 3.9 0.10 0.54 0.39 2.01 0.17 3 5.5 0.10 0.4? 0.3c lJi8 0.11 6 4.7 0.13 0.60 0.29 1.77 o,i4 7 5-3 0.11 0.50 0.37 1.51 0.12 8 1^.1 0.13 0 . 60 0-35 1.60 0.l4 9 5.6 0.10 0.45 0.35 1.59 0.10 10 4.8 0.13 0.57 0.30 i.63 O.li^ J.X 4.2 0.12 0.64 0.28 2.09 0.17 12 5.1 0.12 0.42 0.40 1.53 0.12 13 0.4 0.08 0.72 0.37 2.52 0.19 lit 5.3 0.12 0.60 0.33 1.52 0.09 1? 3.9 0.12 0.51 0.28 1.56 0.11 16 4.3 0.13 0.49 0.29 2.03 0.16 17 4.7 0.13 C.44 0.31 2.15 0.17 18 4.4 0.11 0.44 0.26 1.53 0.l4 * Total of three harvests. Moan of three harvests.

PAGE 193

Table 68. Effect of lime, rnagnealurrL, and X'^iosphorus treatments to San Vito Soil on Pan'-^ola digitgrnss yield and nutrient concentretion from Fxperiraent Ko. 1 Nu t r .i e n t c cn -:• e nt rat i on ** Treatment Total number yield* P Ca Mg K Na g/pot 1 0.10 0.'42 0.28 l.DO 0.12 2 k.6 0.12 0.56 0.31 1.67 0.14 3 5-5 0.10 0.36 0.26 1.95 0.15 k 5.1 0.12 0.52 0.30 1.84 0.1? 5 7.0 0.11 O.ho 0.30 1.3i^ 0.10 6 0.11 0.50 0.26 1.87 O.lU 7 6.6 0.12 0.39 0.27 1.27 0.10 8 6.6 o.lh 0.56 0.31 1.29 0.09 9 6.2 0.12 o.Ui 0.30 1.55 0.11 iO h.l 0.12 0.60 0.28 1.77 0.13 11 5.7" 0.11 0.50 0.26 1.70 0.13 12 5.3 0.13 0.59 0.38 1.53 0.10 13 l.h 0.03 o.kk 0.21 2.ii9 0.17 lU 6.7 0.13 0.U2 0 . 23 l.kk 0.12 15 5.0 0.10 o.kk 0.28 1.53 0.12 16 it. 8 O.lU O.kk 0.29 2.00 0.16 17 h.9 O.lU 0,38 0.22 1.86 0.13 18 5.1 0.12 O.kQ 0.29 I.U9 0.11 * Total of three riarvests. Mean of three harvests.

PAGE 194

178 Table 69. EE'fect of lime, iriacnt-^r.liLn;, and phosphorus treatments to San Isidro coil on Paui-,ola digitf^rar,r. yield and nutrient concentration from Ezcpv^rinent Ho. 1 Nutrient concentration** Treatment Total number yield* P Ca Mg K Na g/pot c". 1 2.4 0.11 0 . 32 0.23 1.9*+ 0.13 2 1.8 O.lU O.U9 0.30 2.10 O.lif 3 3.2 0.11 0.29 0.22 2.1J. 0.12 k 3.6 0.J2 0.''+7 0.36 2.12 5 0.13 O.iH 0.26 1.50 0.12 6 2.5 0.16 0.59 0,29 2.70 0.25 7 5.0 0.13 o.Uo 0.28 1.66 0.13 3 3.8 0.14 0.35 • 2.07 0.15 Q U.9 0.13 0.27 C.23 1.79 O.IJ+ 10 2.0 0.13 0.59 0.38 2.30 0.16 11 2.5 0.1^1 C.4l 0.23 2.08 0.15 12 3.7 O.li^ o.ui 0.32 2.19 0.15 13 0.5 0.06 0.50 0.19 2.01 0.13 1^4 5.0 0.36 0.25 1.73 0.15 15 2.5 0.13 0.U8 0.33 2.36 0.16 16 4.2 0.12 0,60 0.37 1.91 0.13 17 h.2 0.13 0.U3 0.31 1.85 0.12 18 0.11 O.Ul 0.26 1.77 0.12 * Total of three harvests. ** Mean of three harvests.

PAGE 195

'.'ble 70. Ei'fect of ?.inf', magnitdu/':, and phus;..-ionis treatnients to Grecia soli on Pari,3cl.5i digltgruss yi^ild ar-d nutrient coiicentration from Expei-iment No. .1 179 tr n t cone entrati ori''--* Treatment Total nuniber yield* P Ca J^g K Na g/pot 1 0.09 0.38 0.30 1.82 0.12 .2 .4.2 G.iO 0.46 . . 0.37 ^ 1.93 0.11 3 5.1 0.08 0.31 0.33 1.75 0.11 h 4.8 0.10 0.i;8 0.38 1.94 0.13 5 6.0 0.10 0,42 0.29 1.77 0.12 6 5.7 0.22 0.50 0,31 1.73 0.11 T 5.6 0.11 0,40 0-35 1.69 0.10 8 6.2 0.14 0 , 45 0.33 1.89 0.12 9 0.10 0.26 0.32 2.04 0.14 10 k.o C.I3 0.36 0.28 2.12 o.i4 T 1 5.1 0.10 0.42 0.29 1.83 0.12 "t ^ 5.0 0.12 0.39 0.38 1.91 0.13 13 0.2 0.05 0.56 0.34 2.07 0.12 ll! 6,5 0.12 0.35 0.30 1.82 0.13 IP i).7 0.11 0,35 0.32 1.84 0.12 16 5.3 0.11 0.31 0 . 30 1.64 0,10 17 if. 8 0.03 o.4o 0.33 1.58 0,09 18 6.0 0.10 0.38 0.32 1.63 0.09 * Total of three harvests. * Meein of three harvests.

PAGE 196

Table '{1. Effect of" lime, magni^fiiur:, . and phosphoras treatments to _ Alajaola soil on Par.y-ola digitgrasc yield '
PAGE 197

181 Table 72. Effect 01* lime, roagnosiiim, and phOoP -lOrue treatments to' /igrono^Tiy Area so;"l on Fangola dig;i.tf7;ras3 yield and nutritnt concentration iroii Experinicnt No. 1 l!u t.ricnt c cu centryttion*^" Treatment Total number yield* P ' Ca hig K Na 1 2.9 0.09 0.50 0.20 1.53 0.09 2 3.2 0.11 0.73 0.20 1.47 0.09 3 3.8 0.13 0.37 0.27 1.97 0.11 k 2.8 0.12 0.52 0.28 1.89 0.10 5 4-. 2 0.19 0.51 0.17 1.56 0.09 6 4.2 0.16 0.77 0.22 1.48 0.10 7 h.Q 0.18 0 . 47 0.27 1.48 0.10 8 30 0.l4 0 . 69 0.31 . 1.70 0.10 9 . 3.9 0.15 0.29 0.27 1.60 0.12 10 3.2 0.20 0.79 0.31 1.52 0.10 11 3.5 0.15 C.65 0.12 1.84 0.10 12 3.9 0.13 0.54 ' 0.33 " 1.52 0.09 13 0.9 0.06 C.49 0.18 2.42 0.16 14 0.19 0.71 0,26 1.29 0.09 15 3.7 0,12 O.Si 0.22 1.31 0.13 16 3.5 0.14 0.59 0.25 1.84 0.12 17 h.k 0.l4 0.54 0.26 1.57 0.10 13 3.5 0.16 0.60 0.31 1.89 0.13 * Total of three harvests. Mean of three harvests.

PAGE 198

1 182 Table 73. Effect of lime, n.a;.;:;:.-3iur'', and phoaphoras. treatments to Cararao soil on Pa-nf'Ola cigltgra.is ^ lelsi and nutrient coiaceiitration from Experiment No. 1 jV'utr ieat con centration-^'* Treatment Total nu!.ib.;r yield* P Ca Mg K Ha g/pot .1 3A O.lh 0.3'9.21 1.95 0.09 2 U.2 0,12 0.59 0.32 . 1.63 0.07 3 3A 0.12 0.29 0.23 1.83 0.10 k 2.5 0.12 0.65 0.21 2.29 0.15 5 k.b 0 . 19 0.-V9 0.19 I.5U 0.06 6 h.6 0.21 0.56 . 0.22 1.62 0.08 7 0.21 0.36 0.27 1.51 0.06 8 3-5 0.20 0.69 0 . 26 . 1.61 0.09 9 o.i6 0.23 G.2h l.i+1 0.06 10 2.8 0.l6 0.96 0.28 2.20 0.11+ 11 . h.O . O.lU . 0.65 0.20 1.45 .0.07 12 l+.O 0,15 O.I18 0.28 1.70 0.12 13 1.5 0.07 0.38 0.18 2.i|-5 0.11 Ik k.k 0.21 0.58 0.25 1.61+ 0.10 15 3.8 0.1'^ 0.5U 0.24 1^99 0.09 l6 3.6 0.15 0.53 0.25 1.63 0.09 1? 0.15 0.56 0.28 1.64 0.09 10 ii.8 0.17 0.70 0.33 1.83 O.lU * Total of three harvests. ** Mean of three harve;jts.

PAGE 199

183 Tabic7'''-. Effect of lime, main-Lfu-liin, and phosphorus treatmeints to Dratinaf^.eway soil on ir'ansola digitri;ra3s yield and nutrient concentration, froia ?,>.periment No. 1 _ N utrient eoncenrrati on*-'^ Treatment Total liomber yield* : ' P?.; ' • Cai Mg K Na g/pot 1 4.0 O.OQ r 0.33 0.17 1.96 0.20 2 . k.i^ C.iO 0.49 0,17 1.-82 0.17 3 5o 0.08 0.36 0.25 1.89 0.21 k 3-9 0.09 0 . 42 0.18 1.52 0.16 5 4.1 0.l4 0.34 0.17 1.92 0.16 6 4.5 0.15 0.45 0.16 2.C4 0.22 T 4.6 0.11 0.28 0.18 1.64 0.13 8 4.6 0.13 0.4o 0.18 1.68 0.16 9 3-0 0.12 0. 22 0 . 22 1.91 0.20 10 4.0 0.13 0.48 0.22 2.01 0.19 11 4.0 0.11 0.52 0.09 1.92 0.19 12 4.2 0.13 0.43 0.26 2.00 0.20 15 0.4 0.03 0.50 0.10 2.00 0.33 14 4.1 0.15 0.4i 0.21 1.94 0.17 15 5-1 0.10 0.42 0.21 1.66 0.16 16 4.6 0.13 0.35 0.16 2.03 0.21 17 4.1 0.13 o.4o 0.18 1.99 0.19 18 4.2 O.IC 0.33 0.15 2.14 0.20 >''• Total of three harvests. Mean of three harvests.

PAGE 200

184 T'voJ.c 75* Effect of liiiie, ntagnosiuni, and phos.i/noi-^j"treafcnien+j3 to -— East Carimagua soil on P6.rigola digit^jrasr:. yield anr' nu . _ ' trient concentration fro-ii Experiment lie. 1 Nutrient c on e ent rat i on** Treatment Total number yield* p Ca Mg K Ka g/pot 1 3.3 0.10 0.56 0.18 1.76 0.08 2 3.0 0.16 0.57 o.i4 2.05 0.07 3 3.7 0.10 0.46 0.38 1.70 0.06 k 2.5 0.15 0.64 0.36 2.32 0.07 5 h.k 0.29 0.B5 0.20 1.57 0.08 6 3-0 0.31 0.95 0.17 1.85 0.07 7 4.1 0.28 0.62 0.39 1.47 0.08 8 2.6 0.24 0.64 0.4o 2.23 0.09 9 3-9 0.20 0.32 0.25 1.32 0.06 10 2.3 0.19 0.57 0.27 2.56 0.11 11 1.6 0.37 0.82 0.09 2.66 0.12 12 3.5 0.27 0.66 0.45 2.19 0.10 13 O.k 0.07 0.47 0.22 1.34 0.17 1J> 3.9 0.35 0.73 0.32 1.86 0.11 15 3.2 0.25 0.67 0.32 2.12 0.09 16 3.4 0.20 0.81 0.31 2.l4 O.IC 17 2.5 0.14 0.55 0.26 2.30 0.09 18 3.0 0.24 0.62 0.28 2.33 0.11 * Total of three harvests, Mean of three haivests.

PAGE 201

LITER^TJIS ClTiX 1. Abiv-ua, F., and J. Yiccinte-Chaadler. I96'/'. Siigar cane yields as related to acidity of a huinid tropic Ultisol. A^ron. J. 59:330-332. 2. Adams, F. , and R. Fearsor. I967. Crop response to liae in the southern United States and Puerto Rico. In R. V . Pearson (ed.) Soil acidity and liming. Agroijortii' 12:T6l-2C6, Aruer. 3oc. of I'^TCn.y Madlocn,, Wis, 3. Ao^a^/s, F,, and J. I. Wear. 1957. lYanganese voxicity and soil acidity in relation to crirzlile leaf of cotton. Soil Sci. Soc. Ar.sr. ProG. 21:305-508. . ; ' h. Aguilera;, N. H., and M. L. Jackson. 1955. Iron oxide removal from soiir. and clays. Soil Sci. Soc. An^.er., Proc. 1": 359-36'4-. 5. Al-ienkorah, Y. 196c3. Fnosphoruii rete.ntion capacities of some cocoa-£rov?ing soils of Ghana ajid their relationship with soil prou-rties. Soil Sci. 105:2^^-30, 6. Ahmad, K.. L. I. Tullcch-Reid, and C. E. Davis. 1969a. Stadies on pai-igola grass ( Pigitari a decun bens Stent) in Trinidad. I. Description of the expcrinients and' effects of nitrogen. Trop. .Igric. (Trinidad) lf6: 173-178. 7. Ahmad, K., L. I. TullochrReid, and C. E. Davis. 1969b. Fertiliser studies on pangola grass (Pir-itaria de ciLmtiens Stent) in Trinidad. II. Piffects of phosphoras, potassium, aiid magnesivjn. Trop. .\^7ric. (Trinidad) it6: 379-I86. ' 8. Aleksandrova, L. N. I96O. The use of sodium pyrophosphate for isola'oir^s free huinic substances and their organic -mineral compounds frcn soils. Soviet Soil Sci. 2:190-197. 9. Ajlison, L. E. I965. Organic carbon. In 0. A. Black (ed.) Methodr. of soil analysis. Agronomy 9:13^7-1378. An;er. Soc. of Ar;ron., Madison, V/is. 10. Andrew, C. 3., and K. F. Robins. 1971. The effect of phosphorus on the growth, cht.-:iical composition^ and ciitical phosphorus porcc:rit£v-:c3 of some tropical pasture grasses. Aust. J. Agr. Res. 22; 693-706. 11. Arnon, 1). I., V. E. Fratzke, and C. M. Ooiinson. 19'+?. Hydro^;cn ion eoncenl. ration in relation to absorptiou o" inorganic nutrients by hii:,rier plants. Plant Physiol. 11 :'jV;~'j2k. , , , 185

PAGE 202

156 .12. Aiidu3, L. J. 19'+9' Studies oji the pH.' relationship of root growth and its iahibition by 2: ij-dichlorophenoxyac-etic acid and coumarin. Kev PhytoloG. ii5: 97-114. 13. Bache, J3. V/., suid E. 0. Williams. 1971' A phosphate sorption index for soils. J. Soil Sci. 22:289-301. l^t. Ballard, R., tind J. G. A. Fibkcll. 197^' • Fhosplioi'us retention in coastal plain forest soil;-,: I. Relationship to soil properties. Soi3 Sci. 3oc . Aner. ?roc. 'i8:2\^0-2')^. ].5. BascouiD, C. L. I96S. Distribution of pyrophosphate -extractable irOi.) -iod organic carbon in soils of various groups. J. Soil Sci. . 19:25.1-263. . . • ' ; : • — 16. Bazan^ R. I969. The coastal pine ridge soils of British Honduras and their fertility status. Ph.D. Dissertation. Univ. of Florida, Gainesville, Fla. (Diss. Abstr. 70:12226). 17. Bhumbla, D. R., and E. 0. McLean. I965. Aluminum in soils. VI. C'aanges in pH-dependent acidity, cation exchange capacity, and extractable aluminum with addition of lime to acid surface soils. Soil Sci. Soc. Amer. Proc. 29:370-374' 18. Blue, W. C. 1969« Fertilizer response v/ith pangolagrass on Fuletan loamy sand, British Honduras, as indicated by pot experiments. Trcp. Agric. (Trinidad) 46:25-29. 19. Blue, W. G. 197'^-« Management of Ultisols and Oxisols. Soil and Crop Sci. ooc. Fla. Proc. 33:126-132. • 20. Blue, W. G., L. Andrade, E. Rey, M. Ramirez, L. L. Larson, and W. E. Schaefer. 1963* Investigations of the potential for pasture developLient in the Atlantic zone of Costa Rica. Soil and Crop Sci. Soc. Fla. Proc. 23:208-221. 21. Blume, H. P., and V. Schwertmann. 1969* Genesis evaluation of profile distribution of alujninum, iron, and manganese oxides. Soil Sci. Soc. Amer. Proc. 33:436-44i4-. 22. Bortner, C. E. 1935' Toxicity of manganese to tur^ish tobacco in acid Kentucky soils. Soil Sci. 39:15-33' 23. Bower, G. A., and L. V. Wilcox. I965. Soluble salts. In C. A. Black (ed.) Methods of soil analysis, /vgronoinj' 9:933-951. Amer. Soc. of Agron., Madison, Wis. 2h. Bray, R. u,, and L. T. Kurtz.. 1945Deter:nination of total, organic, and available forms of phosphorus in soils. Soil Sci. 59:39-45. . 25. Bremner, J. M. 1965a. Inorganic forms of nitrogen. In C. A. Elack (ed.) Methods of soil ai\alysis. Agronomy 9:1179-1237. Amer. Soc. of Agron., Madison, Wis.

PAGE 203

26. Eremner, J< M. 1965b. Total nitrogen. In C. A. Black, (ed.) . Methods or soil analysis. rVjroaoiiiy 9;ll't9-117S. Aroer. See. of Agron.,, Madison, Viis, Z; . Ch'cio, T. T., and M. E. Hari/ard. I962. Mature of acid clays and rrlationsaipo to ion activltieand ion ratios in equilibrium solutions . Soil 3ci. 93:2^16-253. 28. Chapraan, H. D. 19t>5. Cation-exchani-^e capacity. In C. A. Black (ed.) Iletiaods of soil ai;alysis. Agronomy 9:S9i-9"l. Amer. Soc. of A;^ron., Madison, Wis. 29. -oheiiuey^ 11. A. D. 1972. Response of Pi git -aria se tj vslva to nitrogen, phosphorj-S, ECgnefiuni, and calciura on E'oini sandy loaiii, Guyana. I. Effect on yield; tissue co.nposition and nutrient upt3>.e. 'rrop. Agric. (Trinidad) ii-9: 115-124. . .. 30. Clark, J. B., end R. C. Txirner. 196?. Extraction of exciiangeable cations and distribution constants for ion exchange. Soil ' Sci. Soc. .Amer. Proc. 2S:Z(l-2(h. • 31. Coleman, N. T., and M. E. Hardvard. 1953. Tbe heats of neutral' ization of acid clays and cation exchange resins. J. Amer. Chera. Soc, 75:6oi45-6o-^6. 32. Co.ler.ai.\, IJ. T., and G. w. Thomas. 1954. Buffer curves of acid clays a~> affected by the presence of ferric iron and aluininom. 3ci.l Sci. 80C. Aiaer. Proc. 28:167-190. 33. Coleman, N, T., and G. V.'. Tho^aas. I967. The basic chemistry of soil acidity. In R. V7. Pearson (ed.) Soil acidity and liming, .rigrono'-.iy 12; 1-^+1. A-ner. Lloc. of Agron., Madison, l-^is. 3^1. Coleman, H, T., E. J. Kamprath, and 3. E. V/eed. I958. Liming. Advan. /vgron. 10:^75-522. 35' Coleraan, N. T., G. W. Thomas, F. H. leRoux, and G. Bredell. 196''^ Salt -exchangeable and titratable acidity in bentonitesosquioxide raixtures. Soil Sci. Soc. Ainer. Proc. 28: 35-37. 36. Golo:>an, R. P. 19^*5 • The mechanism of phosphate fixation by Konti'iorilionite and kaolinite clays. Soil Sci. Soc. Araer. Proc. 9: 72-70. 37. Coulter, B. S. 1969a. 'i'he chemistry of hydrogen and alumimom iouf: in soils, clay minerals, and resins. Soil Pert. 32:215-223. 38. Coulter, B. S. 1969b. The -quililria of K:A1 exchange in clay minerals and acid soils. J. Soil Sci. 20:72-83. 39. Coulter, 3. S,. and 0. Talibudeen. I968. Calcium: a.luminura exchange equilibria in clay minerals and acid soils. J. Soil Sci. 19:237-250. ;

PAGE 204

188 ^0. Dean, L. A., and S. J. Rub.' no. 197^ • /jrion exchange in soils. I. Exchiingeabie phosphor.;;; TiUd the anion -ez'-^harige capacity. Soli Sci. 63; 377-387. In. Deb, D. L., and N. P. Datta. I967. Effecv. of associating anions on phoophoras retention in soils. 1. Under various phosphorus, concentrations. Plant ana ocii £6:503-516. '+2. Deist, J., and 0. Talibudean. I967. Ion exchan,t5e in soils from the ion pairs K-Ca, K~Kb, and K-Iia. .3, Soil Sci. 18:125-137. h3. Devan, H. C, and C. I.Kich. 1970, Titration of acid soils. Soil i'.ci. 3oc /cner. Proc. 34; jS-U't-. kh. iXovner, A. V., and W. G. Blue. 19 ('1. Lindng in relation to crgeriOmetallic complexes in soilo. Soil and Crop Sci. Soc. Fla. Proc. 31:204-207. V^. Dovrner, A. V. 1972. Factors affecting fertility of selected bro'.vn oand sci-ls of Guyoaa. Ph.D. Dissertation. Univ. of Floridii, Gainesville, Fla, (Diss. Abstr. 73:552). ho, Eaton;, F. M. I966. Chlorine, p. S1Q--I35. In II. D. Chapman (ed.) Diagnostic criteria for plants and soils. Div. /igric. Soi., Univ. California, Bez'keley. h'J . Evans, C. E., and E. J. Kamprath. 1970. Lime response as related to present Al saturation, solution Ai, and organic .matter content. Soil Sci. Soc. /i.T:er. Proc. 3i4-: 893-896. he. Figarella, J,, J. Vicente-Chandler, S. Silva, and R. Caro-Costas. 19bk, Effects of phosphorus fertilization on productivity of intensively manaf^ed grasses under numid tropical conditions in Paerto Rico. J. Agr. Univ. of Puerto Rico 40;23o-2U2. it9^ Fox, R. I,. 197^Chemistry and management of soils dominated by amorphous colloids. Soil and Crov. 3ci. Soc. Fla. Proc. 33:112-119. 50. Fox, R. L., and E. J. Kamprath. 1970. Phosphate sorption isothenas for evaluating bhe phosphate requirement of soils. Soil Sex. Soc. A:aer. Proc. 34:902-907. 51. Fox, R. L., S. K. DeDjitta, and J. M. Vang. I96U. Phosphorus and alu-'.-dnum uptake by plants from latosols in relation to liming. Int. Congr. Soil Sci., Trans. 8th (Bucharest) IV: 595-tx)3. Fox, R. L., 3. M. Hasan, and R. C. Jones. 1971. Phosphate and sulphate sorption by Latosols. Proc. Int. Siiap. Soil Pert. Evalu. (New Delhi) l:857-o6i4-.

PAGE 205

189 53" J'ox, K. L., J. A. Silva, 0. R. yciirt<';ej r>. L. Flucknett, and G. D. Sherman. I967. Soil end plant silicon and silicate responses by sugar cane. Soil Sci.. Sec. /jaer. ?roc. Vx: 775-779. 5^. Fey, C. D., and J. C. Brcwii. 196^-. Toxic factors in soils: II -Differential alumlniim tolerance of plaxit species. Soil 3ci. Soc. Arier. Proc. 26:27-32. 55Gaines, G. L., and H. C. Tlionias. 1953Adsorption studies on clay minerals: II. A fonwalation of the thermodynamics of exchacige-adscrption. J. Chem. Phys. 21;71^-'i'i3. 56. Car,L"on. N., Jr. 1953' Sodium and potassium i-equirements of panjrola and other pasture grasses. Soil Sci. 76:81-90. 57. Gomide, J., C. H. Noller, G. 0. Mott, J. H. Conrad^, and D. L. Hill. 19^9' Mineral composition of six tropical grasses as influenced by plant age and nitrogen fertilization. Agron. J. 61:120-123. 58. Eader, R. J., K. E. Harvrard, D. D. Mason, and D. P. Moore, 1957' An investigation of some of the relationships between copper, iron, and molybdenum in the growth and nutrition of lettuce: I. Experim'irntal design ejid statistical methods for characterizing the response surface. Soil Sci. Soc. Amer. Proc. 21:59-6i|-. 59' HaTiiia, W. J. I96U. Methods for chemical analysis of soils, p. k'{h-'}01. In F. E. Sear (ed.) Chcnistry of the Soil. Reinhold Pub." Corp., Ilev York. 60. Kashitnoto, I., and M. L. ^Jackson. I96O. Rapid dissolution of aliophane and kaolinite-halloysite after dehydration. Clays and Clay.Min. 7:102-113. -^1. Helyar, K. R., and A. J. Anderson. I97U. Effects of Calciu-'ti carbonate on the availability of nutrients in an acid soil. Soil Sci. Soc. Amer. Proc. 33: 34I-5U6. 62. Herawall, J. B. 1957. The fixation of phosphorus by soils. Advance. Agron. 9:95-112. 63. Kingston, F. J., R. J. Atkinson, A. :;. Posner, and J. P. Quirk. 1967. Specific adsorption of anions. Nature 215: 1^59l46l. 6k. Kodi^es, E. M., G. B. Killlnger, J. E. McCaleb, 0. C. Ruelke, R.^J. Allen, Jr., S. C. Schan};, and A. E. Kretschner, Jr. 1967. Pan.30lafr:rass. Agr. Expt. Sta. I.E. A. 3., Univ. of Florida, Gainesville, Fla. Bull. 718. « 65. Hortenstine, C. C. 1956. Phosphorus fixation and phosphorus fractions in sandy soils. Soil and Crop Sci. Soc. Fla. Proc. 26:136-1^+2.

PAGE 206

19c 56. Hortenstiiie , C. C, and W . G. Blue. jO^'B. Growth responses, in. three plant species to liine and phosptioius applied to Puletan loamy fine saiid. ooil nnd Crop 3ci. 3oc. Fla. Proc. 28:23-2-3. 67. Howard, D. D., and Adanis. I965. Ctlciuir. requirement for penetration of subsoils by pri:nary cotton roots. Soil Sci. Soc. Amer. Proc. 29:558-561. 66. Hsu J P. H. 1965. Fixation of phosphate by aluraiaun and iron in acidic soils. Soil Sci. 99:398-^1-02. 69. Hsu, P. H., and C. I. Pvich. I96O. Alumi-auin fixation in a Sj'nthetic cation exchanger, ooil Sci. Soc. Mier^ Proc. 2ii-: 21-25. 70. Igue, K., and R. Fuentes. 1S)72. Characterization of Aluminum in volcanic ash soils. Soil Sci. Soc. Aicer. Proc. 3^:292-296. 71. Jackson, M. L. 1956. Soil chemical auialysis •• Advanced course. 2d Edition. Published by the author, Eepartj.isnL of Soils, Jniv. of Wis., Madison, V/is. 72. Jackson, M. L. 195Q' Soil chemical analysis. Prentice -Hall, Inc., Englevooci Cliffs, S. J. 73» Jackson, K. L. 1963Aluminun bonding in soils: A unifying principle in soil science. Soil Sci. Soc. Amer. Proc. 27:1-10, 7^. Jackson, M. L. 1965. Free oxides, hyvlroxides, and araorphous aluL-inosilicates . in C. A. Black (ed.) Methods of soil analysis. A^rronoray 9:578-603. Aiaer. Soc. of Agron., Madison, V/is. 75* Jackson, W. A. I967'. Physiological effects of soil acidity. In S. VJ. Pearson and F. Adams (ed.) Soil acidity and liming, /agronomy 12:^3-12^;. /oner. Soc. of Agron., Madison, v;is. 76. Juo, A. S. R., and B. G. Ellis. 3.968. Chemical and physical properties of iron and aluminum phosphates and their relation to phosphoras availability. Soil Sci. Soc. Amer. Proc. 32: 216-221. 77. Kamrirath, E. J. 1970a. Exchangeable aluminum as a ci'iterion for aindng leached minoi-al soils. Soil Sci. Soc. Amer. Proc. 3^^:252-25^. 7w. Kariprath, E. J. 1970b. Lime requirement of soils-Inactivate toxic substances or favorable pH range. Soil Sci. Soc. Amer. Proc. 3^:363-i6'f. .79. Ka-nprath, E. J. 1972a. Potential detrimental effects from liming highly vcath;'n-ed soils to neutrality. Soil and Crop Gci, ;:oc, Fla. Proc. 31:200-203.

PAGE 207

191 60. ?:a:nprath. Y-. J. 1972b. 3oi± acidity c.nd .lindr^T;. p. 136-l'l9. In Coiasittee on Tropical :iollh (ed.. ) Soilr; of the h-ojaid tropics. Kat. Ac&d. Sci. 81. ?:araos, L. T. 196'». Soil fixation of plaat nutrients, p. 36939^1. In F.. E, Bear (ed.) Chemistiy cf L/ie soil. 2d. Edition. Reinhold Pal:. Corp., Kev York. 62. Ken~, J. C. W., and G. Ueliara. 197^. Chemistrj'., mineralogy, and taKonomy of Oxiscls and Ultisols. Soil ai:d Crop Sci. 3oc. Fla. Proc. 33:119-12^0. " / . . 83. Kin.jo, T., P. ?. Prattj and A. L. Page. 197I. ^iitrate adsorption: Jj . In conip.etition with chloride, sulfate., and phosphate. Soil Sci. Soc. ,%ier. Proc. 35:725-728. 8^i. J^z'-fs, W. D. 1950. Water-soluble silicate application to a calcareouc clay soil and effect on noil properties and nutrient uptajic b> plants. Soil Sci. Soc:. ;\!ner. Proc. 15:09-92. 35. Levssque., M. , and M. Schnitzer. I967. Organo-metallic interactrlor.s in soils: 6. Preparation and properties o:? fulvic acidinetal piiosphates. Soil Sci. 103:163-190. 86. Lin,, E.^, and T. Colerjim. I96O. The measurement of exchangeable alui!u.nura in soils and clays. Soil Sci. Soc. Atner. Proc.^ 87. Lindsay, L., and E. C. Moreno. I960. Phosphate phase equilibria in soils. Soil Sci. Soc. Aner. Proc. 24:177-182. bS. Lotero, J., S. A. Monsalve, and A. Ramirez. I97I. Respuesta ds gramineas y legurcinosas forrajeras al encalaiaiento. In Sociedad Colorabiana de la Clencia del Suelo (ed.) Acide~y encalaTu.enT.o en el tropico. Priaer coloquio de suelos. Suelos Ecuatoriales. p. 210-239. 89. Lov, P. E. 1955. The role of aluminum in the titration of b?ntonite. Soil Sci. Soc. Aaer. Proc. 19:135-139. 90. Lew, P. t:._, and C. A. Black. I950. Reactions of phosphate vitii kaolinite. Soil Sci. 70:273-290. 9.1. Lucas, L. N. I969. Phosphorus availability in alln.vial rainforest fioiis from Costa Rica as affected by phosphorus fertilization arid soil aaendinents. Ph. P. Dissertation. Univ. of Florida, Gainesville, Fla. (Diss. Abstr. 70:122^7). 92. Lucas, L. n., and VJ. G. Blue. 1972a. Effects of llrne and phosphorus on selected alluvial Entisols from eastern Costa Rica. I -Phosphorus retention and soil phosphorus fractions. Trop. A,'jric. (Trinidad) '49:287-295.

PAGE 208

192 93> JuQ&s, L. N., and W. G. Blue. 1972b. Paru^olacrays (nif^Utuvla decu mbens Stent) growth as affectea by or;2;ariic raau-riais "and " calcium silicate applied to a aoil from Los Dia:nantes, Costa Rica. Trop. Agric. (Trinidad) U9:26l-2.?6. 9k. Lucas, L. II. J and W. G. Blue. 1973. Effects of lime and phosphorus on selected alj/avial Entisols from eastern Costa Hica. II-Forage plant responses. Trox). A^rlc. (Trinidad) 50:63-71+. 95. i'arshal, C. E. IQSk. The physical chemisti-y and mineralogy of soils. Vol. I: Soil materials. John Wiley and Sons, Inc., New York. 96. Martia, A. E. I96O. Chenical studies of podzolic illuvial horizoas: V. Flocculation of hixnius by ferrous iron and nickel. J. Soil 3ci. 11:382-393. 97. McKeague, J. A. I967. An evaluation of G.m pyrophosphate and pyrophosphate-dithionite in comparison v;ith oxalate as extractants of the accuiTiulation products in Podsols and some other soils. Can. J. Soil Sci. ^+7:95-99. 98. McKeaguo, J, A., and J. H. Day. I969. Oxalate -extrac table Al as^ a criterion for identif>'ing Podzol B hori",ons. Can. J. Soil Sci, 49:161-163. 99. McSee<:5ue, J. A., J. E. Brydon, and N. M. Miles. I971. Differentiation of forms of extractable iron and aluminum in soils. Soil Sci. Soc. Amer. Proc. 35:53-38. 100. McLean, E. 0. I965. Alujninum. In 0. A. BlacK (ed.) Methods of soil analysis. Agronomj' 9:978-998. Araer. Soc. of Agron., Madison, Wis. 101. Mclean, 0. 1970. Liine requirement of soils-Inactivation of toxic substances or favorable t)H range. Soil Sci. Soc. Amer. Proc. 3h:^63-y3k. 102. McLean, .E. 0. I97I. Potentially beneficial effects from liming: Chemical and physical. Soil and Crop Sci. Soc. Fla. Pxxic, 31:189-196. 103. McLean, E. C, and E. J. Oven. I969. Effects of pll on the ccntributions of organic matter and clay to soil cation exchange capacities. Soil Sci. Soc. Amer. Proc. 33:855-858. lOh. >.;cU;an, 2. 0., M. R. IleddlcGon, and G. J. Pojt. 1959. Aluminum in soils. III. A comparison of extraction rcethcdo in soils and clays. Soil Sci. Soc. ;\!;ier. Proc. 23:289-293. 105. McLean, L. 0., D. C. Kcicosky, and C. I-akshmanan. I965. Aluminum in soils. VII. Intei'relataonships of organic matter, limin^^^, and oxtractablo Al vith ponncuient charge (KCi) and pll u. pendc-nt^CEC of surface soils. Soil Sci. Sec. .nmer. Proc. 2:9:374-378.

PAGE 209

193 106. Mehlich, A. 19'+3, I)C'terr.-.iration-of cjiticn and anion exchanfje pi'operties of soils. Soil 3cic 6'^:h7i9-kk'} . 107. Kehra, 0. P., and M. L. Jackscn. i960. Iron oxide removal from soils and clays by a dithlcrJ-te-cltrato system with sodium bi-i^. carhonate b-iffer. Clays and Clay Minerals 7:317-327. lOO. Kokaru, T., and G. Uehara. 1972. Anion adsorption in ferruginous tropical soils. Soil 3ci. 3oc. /irner. Proc. 36:296-300. — ICQ. Kidgley, A. R. 1932. Overliming acid soils. J. Amer. Soc. Agron. 2^^:822-836. 110. Mikkelsen, D. 3., L. M. M. deFreitas, and A. CMcCl'ong. I963. : Effects of liming ond. fertilizing cotton, corn, soybeans on campo cerrado soil;.-Statc of 3ao Paulo, Brazil. IRI Research. • . Institute, Inc. Ball. 29, p. Uo. 111. Itlyaice, K. I916. The toxic action of soil aluminum salts upon the growth of the rice plant. J. Biol. Chen. 25:23-28. 112. Monteith, N. H., and G. D. Sherman. I963. TYie. comparative effects of calciiuu and of calcium silicate on the yield of sudan grass grown in a ferruginous latosol and a hydrol humic latosol. Hawaii Agr. Exp. Stat. Tech. Bull. 53. p, ilO. 113. Morris, H. i)., andW. H. Pierre. 19-^9Minimum concentrations of manganese necessaiy for injury to various legumes in culture sol^-iticns. Agron. J. 41:107-112. llh. Kaftel, J. A. I936. Soil liming investigations: II. The influence of lime on the sorption and distribution of phosphorus in aqueous ai:;d soil colloidal systems. J. Amer. Soc. Agron. 28:74.0-7:32. 11.5. Naftel, J. A. 1937 • Soil liming investigations: V. The relation of boron deficiency to over-liming injury. J. Ainer. Soc. Agron. 29:761-771. 116. Kye, P., D. Craig, N. T. Coleman, and J. L. Ragland. I961. Ion exchan;3e equilibria involving aluminum. Soil Sci. Soc. Amer. Proc. 25:l4-17. 117. Ogata, G., and A. C. Caldwell. 196O. Nitrate content of soils and nitrogen content of oat plants as affected by rates of liming, /igron. J. 52:65-68. 118. Olson, R. A., and 0. P= Engelstad. 1972. Soil phosphorus and suirar. p. 82-101, In Coimnittee on Tropical Soils (ed.) Soils of the humid tropics. Nat. Acad. Sci. " , " ~ ^ . ... * ; '* ' ' . 119. Paries, G. A. I967. Aqueous surface chemistry of oxides and complex oxide minerals. Isoelectric point and zero point of charge. In R. F. Gould (ed.) Advances iu chemistry series 67:121-100. . . , ,(-. V' .

PAGE 210

' \ r 'r . 19^^ 120. Pierre, V. 'A., and G. M. Brovning. The temporary injurious effect of excv^iinive lining of eclcl i^oil:.; and its rclat.iou to the phosphate nutriiioa of plants. J. ;v:aer. Soc. iXgron. 27: 7 '12-759. 121. Piohke, H. 3., and R. 3. Corey. I967. Relations between acidic alumxnuir. and soil pll, clay, and organic inatter. Soil Sci. Soc. /iiuer. Proc. 31:7i+9-752. 122. Pratt, P. F. 1966. Carbonate and bicarbonate, o. 93-97. In H. D. Chap.Tian (ed. ) Diagnostic criteria :Cor plants and soils. Div. /igric. Sci., Univ. California, Berkeley. 123. Pratt, ?. F., and F. L. }3air. I961. A cc.niparison of tne rea/'ento for extraction of Al froia soils. Soil Sci. 91:357-359. 12'+. Pratt, PP., F. F. Peterson, and C. S. Holzhey. I969. Qualitative .'iiineralogy and chemical properties of a few soils from Sao Pavilo, Brazil. Turrialba 19:^+91-1496. 125. Re.-ive, N. C-1, andM. E. Sunmer. 1970a. Effect of aluminum toxicity and phosphorus fixation on crop growth on Oxisols in Ilatal. Soil Sci. Soc. Araer. Proc. 3l+:263-267. 126. Reeve, K. G., and M. E. S'oriner. 1970b. Lime requirements of • Katal Oxisols based on exchahfieable aluminum. Soil Sci. Soc. lMx.e.r. Proc. 3l+:595-59S. 127. Rhoades, J. D. IS^l • Cation exchange reactions of soil and specimen vermiculite. Soil Sci. Soc. Amer. Proc. 31:361-365. 126. Russell, E. W. 1961. Soil conditions and plant growth. 9th Ed. John Wiley and Sons LTD, Kew York. 129. Russell, E. J., and E. H. Richards. 1920. The washing out of nitrates by drainage water from uncropped and unmanured land. J. Agr. Sci. 10:22-43. 13c. PJussell, G. C, and P. F. Low. 195^. Reactions of phosphate v.itn kaolinite in dilute solution. Soil Sci. Soc. Aaier. Proc. 131. Saunders, W. H. I965. Phosphate retention by Sew Zealand soils and its relationship to free s'^squioxides, organic matter, and other soil properties. N.Z. J. /vgric. Res. 8:30-57. 132. Schnntzer, M., and 3. I. M. Skinner. I963. Organo-raetallic interactionc in coils: 1. Reactions between a number of metal ions and the organic matter of a Podzol Bh horizon. Soil Sci. 96:86-93. 133. Schnitser, M., mid S. I. M. Skinner. 1964. Orr.anc-metal.lic Interactions in soils: 3. Properties of iron find alu-ninum org-anic matter complexes, prepared in the laboratory and extracted f.roiii a soil. Soil Sc.l. 9y:.197-203.

PAGE 211

134. Seatz, L. F., K. B. Peter3on. 196i<-. Aoid^ a.licaline, saline, and sodic soils, p. 292-519* In F. E. Bear (ecL.) Chemistry of the soil. RfirJiold Puo. Corp., New York. 135. Shce^naker, H. E., E. 0. Mclean, and P. 7. Pratt. I96I. Buffer methods for determining lime requirement of soils with appreciable amounts of extractable aluminiairi. Soil 3cl. Soc. .^er. Proc. 25:27^^-277. 136. Shukla., S. S., J. K. Syers, J. D. H. Williams, D. E. Armstrong, and R. F. Harris. 1971^ Sorption of inorganic phosphate by lake sediments. Soil Sci. Soc. Amer. Proc. 35:2^4^1-2^+9. *. 157' Stout, P. R. 1939' Alterations in the crystal structure as a result of phosphate fixation. Soil Sci. Soc. Amer. Proc. h: 177-182. 138. St'jimra, W., and J. J. Morgan. 1970. Aquatic chemistry. John VJiley and Sons, Inc., Nev York. 139. Suehisa, 0. R. YCTonge, and G. D. Sherman. I963. Effects of silicates on phosphorus, availability to Sudan grass grown on Hawaiian soils. Hawaii Agr. E:
PAGE 212

196 14?, ' Vicc-nte -Chandler, J. 196?. TliC role of fvertilir4ci-s in hot. huiiiid tropical paGtures. rio;Il and Crop ^ci, 3oc. Fla. Proa. 26:328-360. 148, Vicente-Chandler, J., J. F.igarella, and 3. 3ilva. I961. Effects of nitrogen fertiii?.:!tion and frequency of cutting on the yield ajid composition of pangol5J(gras3 in Puerto Rico. " J. Agr. Univ. Paertc Rl-co" ^^:37«' " ' IJ+9. Wada, K. 1959 . Reaction of phosphate with allopheine and halloysite.Soil Sci. 87:325-330. 150. V/adleigh, C. H., H. G. Gau-jh, ajid H. Kolisch, I95I. Mineral composition of orchard grass grovn on Pachappa loan salinized with various sialts. Soil Sci. 72:275-282. 151. Wallr.er, R. H., and P. E. Brown. 1935. Nitrification in the grundy silt loam as influenced by liming. J. Anier. Soc. Agron. 27:357-363. 152. Vfatanabe, P. S., and S. R. Olsen. I965. Test of an ascorbic acid method for determining phosphorjs in water and NaHCOo extracts from soils. Soil Sci. Soc. Amer. Proc. 29:677--675. 153' VThittig;, L. D. 1965. X-ray diffraction techniques for mineral identification and mineralogical composition. In C. A. Black (ed.) Methods of soil analysis. Agronorn,y 9:^?71-698. Amer. Soc. of /igron., Madison, V.'is. l^k. v/hyte^ R. 0., T. G. Moir, and J. P. Cooper. I968. Grasses in agriculture. P'ood Agr. Orgn. UK, Rome. p. 66-67. 155. V/iklander, M. 1964. Cation and anion exchange phenomena, p. 3.63-205. In F. E. Bear (ed.) Chemistry of the soil. Reinhold Pub. Corp., 'New York 156. Woodi'uff, J. R., and E. J. Kainprath. 1965. Phosphorus adsorption maxinnim as measured by the langmuir isotherm and its relationship to phosphorus availability. Soil Sci. Soc. /uner. Proc. 29:1^1-8-150. 157. Yoxxnser, 0. R., and D. L. Plucknett. 196^+. Liming materials. Hawaii Farm Sci. 13: i' -8. 158. Youn^jer, 0. R., and D. L. Plucknett. I966. Quenching the high phoijphoras fixation of Hawaiian latosols. Soil Sci. Soc. Ajuer. Proc. 30:653-655. 159. Yuan, T. L. 1963Some relationships among hydrogen, aluminum, and pH in solutions and soil systems. Soil Sci. 95:115-163. 160. Yuan, T. L. 1970. Interpretation of soil pH in liming practice. Soil and Crop Sci . Soc. Fla. Proc. 30:200-210.

PAGE 213

.197 161. Y^xnn, T. L. 19lka. A double buffer mc.z'noa for Vac determination of lime requirenitnt of acid soll.v. 3oll Sol. Soo. Mev, Pros. 38: il37-i<-40. 16c. Jaan, T. L. I'^'jh'h. Chemistry and minerllcgy of Andepts . Soil and Crop Sci . 3oc. Fla. Prcc . 53:101-108, 163. Yuan, T. L. , and H. L. Breland. 1969' Correlation of Al and Fe ad extracted by different reagents with phosphate retention in several soil groups. Soil and Crop Sci. Soc. Fla. Proc. 29:73-86. . . 16k. Zantua, M. I.and G. VJ. Blue. 3-971 • Plant response to iron, " leaching, and time after liming a virgin alluvial Entisol from eastern Costa Rica. Soil and Crop Sci. Soc. Fla. Proc. 31: 165-169. 165. Zelazny, L. W., and J. C-. A. Fiskell. 1971. Acidic properties of some Florida soils." II. Exchangeable and titratable acidity. Soil and Crop Sci. Soc. Fla. Proc. 31:1^^9-154.

PAGE 214

1 BIOaRWnCj\L SKETCH Julian Velez, sea of Dr. Julio Velez Marquee and Mrs. Elvia Pelae?. de Velezj vas bom October 30, l^kh, in Libano (Tol.), Colombia, South Ani.erica. He attended primary school in Ibague (Tel.) at Liceo Val f.i-on 1951 to 1953, Colegio Cooperative in and Colegio ds San Situcn in 19^)'^. H:i 3 hi.^h school education vas completed at Co.legio San Siraon from 1956 to I962. In I965 he moved to the United States vhere he entered iJcrthv-'estern State University of Louisiana in June J.965, as a freshiTian after copiple^ing a course In English and orientation at Louis'Jnna State University frotr. J".riuary to June of the same year. In September 19'-'''', he transferred to Louisiana State Uiiiversity where he secured the Bachelor of Science degree, niajoring in Agronomy, in January I969. In FebrJiar;.' of the same year, he entered the Graduate School of Louisiana State University, and completed the Master of Scienc program as a Soil Science major in January 1971. ' . . . He becaTie a graduate student in the Soil Science Department at the University of Florida in March I97I. Presently he is a candidate for the degree of Doctor of Philosophy under the supervision of .Dr. W. G. Blue. ' . He married Helen M. Kotard on March 2, I969. A son, Julian Eduardo Veles, came to his home on February 23, 1973 . He is a member of the following honorary fraternities: Phi Kappa Phi, Alpha Zeta, 198

PAGE 215

199 Gararaa Sigraa Delta, aiid Sif,raa Xi. Ke is alco an active meniber of tiie American Society of i^^ronomy, the Soil Science Society of America, and the International Soil Science Society. Along with Dr. W. G. Blue, he has published several scientific papers in recognized asricultural journals. . He served as President of the Soil Science Department Graduate Students Association and va& a member of the University of Florida Soccer Club. ; .}:...;

PAGE 216

I certify that I havo read this ctudy and that in Kiy opinion it confonns to acceptable staiiJardo of scholarly presentation and is fully adequate, in scope and quality, as a di'isertation for the degree of D-rjctor of Philosophy. William G. Blue, Chair.Tian Frofesf.or of ooil Science I certlf;/ thst I have i^ead this ato.dy and that in nij' oxjinion it confoims to acceptable standards of scholarly presentation and is fully adequate, in scope and quaiity. as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that in .t^^' opinion it conforin.3 to acceptable standard*, of scholarly presentation axid is fully adequate, in scope and quality, as a d:; ssertaticn for the de^-iree of Doctor of Philosophy. Lucian W. Zelazny"^ Associate Professor of Soil Science I certify that I Iiave read this study and th.at in my opinion it confoxTns to acceptable standards of scholarly preoenuation and ic fully adequate, in scope and quality, as a dissertation for the decree of Doctor of Philosophy. _ X? /Gerald 0. lott ^ Profesotir of Ar.-^or.'oniy

PAGE 217

I certify that I have read this study and that in my opinion it conforms to acceptable stanciarciri of schclariy presentation and is fully adcnuate, in scope and naiviio/^ as a diswcrtatioa for the degree Of Doctor of Pnilosophy. otto C. RueDice Professor of Agronomy This dissertation was su'br.ix.ted to the Dean of the Ccilef.e of Agriculture and to the Graduate Council;, and was accepted as i-artial fulfillment of trie requirements for the degree of Doctor of Philosophy. Dean, Graduate School


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 ET3TXV168_SWBC05 INGEST_TIME 2015-03-09T20:09:32Z PACKAGE AA00028895_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES