Micromorphology and genetic interpretations of selected Colombian andosols

Material Information

Micromorphology and genetic interpretations of selected Colombian andosols
Calhoun, Frank Gilbert, 1939- ( Dissertant )
Carlisle, Victor W. ( Thesis advisor )
Caldwell, Robert E. ( Reviewer )
Popenoe, Hugh L. ( Reviewer )
Blanchard, Frank N.
Lugo, Ariel E. ( Reviewer )
Reynolds, John E. ( Degree grantor )
Place of Publication:
Gainesville, Fla.
University of Florida
Publication Date:
Copyright Date:
Physical Description:
xv, 198 leaves : ill. (some col.) ; 28 cm.


Subjects / Keywords:
Adhesives ( jstor )
Allophane ( jstor )
Feldspars ( jstor )
Halloysite ( jstor )
Ions ( jstor )
Minerals ( jstor )
Nodules ( jstor )
Silts ( jstor )
Soil horizons ( jstor )
Soils ( jstor )
Dissertations, Academic -- Soil Science -- UF
Soil Science thesis Ph. D
Soil physics ( lcsh )
Soils -- Colombia ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )
Spatial Coverage:
South America -- Colombia -- Andean Highlands


Ten soil profiles from the Andean highlands of Departamento Marino, Colombia, were investigated by use of chemical, physical, mineraological, and micromorphological techniques in order to elucidate pedogenetic processes in soils derived from andesitic volcanic ash and similar parent materials. Six Andosol profiles were selected in an altitudinal sequence from 2,000 m to 3,510 m as representative of maximum ranges of weathering severity experienced by aeolian pyroclastic materials in the altiplano of Marino. four additional soil profiles, developing in parent materials with varying degrees of ash influence, provided supplementary pedogenetic information. all of the soils were influences by annual rainfall range of 500-2000 mm and an annual isothermal temperature rang of 6C to 18C which was closely related to altitude. The following parameters were determined on selected horizons of each soil: (a) particle size distribution, (b) bulk density, (c) fifteen atmosphere water content, (d) pH in water and N KCl, (e) exchangeable cations, (f) cation exchange capacity, (g) organic carbon, (h) free iron oxides, (i) X-ray diffraction of fine clay (<0.2 u) and coarse clay (0.2-2u), (j) infrared analysis of fine clay and coarse clay, (k) NoF test for allophane, and (l) light and heavy mineral contents of the sand (50-500 u) fraction. Thin sections were prepared from selected horizons of Andosol profiles 1-6 (2,000-3,510 m) and described in detail. The above mentioned analyses of andosol profiles 1-6 revealed the following trends with increasing elevation: (1) increases in exchangeable K and Al; organic carbon; volcanic glass, pyroxenes, and andesite lithorelects in the 50-500 u sand fraction; allophane in <2 u clay fraction; and isotropism of plasmic fabrics in subsurface horizons; (2) decreases in pH values; exchangeable Ca and Ne; free iron oxides; quartz, feldsapr, and hornblende in 50-500 u sand fraction; hydrated Halloysite in the <2 u clay fraction; sesquioxidic nodules; and bire-fringence of plasmic fabrics in subsurface horizons. Varying amounts of hydrated halloysite, metahalloysite, vermiculite, cristobalite, gibbsite, feldspar, and quartz were identified by infrared and X-ray analyses of the coarse clay fraction of Andosol profiles 1-6. Hydrated halloysite, metahalloysite, and possible allphane A were identified in the fine clay fraction. Andosol surface horizons generally had isotic plasmic fabrics while subsurface horizons exhibited weakly undulic and silasepic plasmic fabrics which were replaced by isotic plasmic fabrics at higher elevations. The term "Andic" was proposed as a modifier for this sequence of fabric types. A clay mineral weathering sequence of allphane--->metahalloysite--->hydrated halloysite was postulated. the maturation of this sequence increased with profile depth and with increasing altitude. Coarse clay vermiculite was hypothesized as being transient or insufficient time has elapsed for it's transformation to other 2:1 phyllosilicates. It was further postulated that vermiculite pedogenesis occured rather rapidly with some modification from a primary mineral source.
Thesis (Ph. D.)--University of Florida, 1971.
Bibliography: leaves 190-196.
General Note:
General Note:
Statement of Responsibility:
by Frank Gilbert Calhoun, Jr.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000310796 ( AlephBibNum )
07561591 ( OCLC )
ABT7491 ( NOTIS )


This item has the following downloads:

Full Text

i'i crc. orphol ogy' a, ] GC~n C ic Inter protal lous
or' Selcc.ced Colunbian Andosols




1') 7


t .au or wi ,:s to exprc-s his sinar p: ,prieci ion to 1r.

Victor W. Carlisle for his unid!.nce, crncouragci-ent, and a:sistancP.

,'urin] th' progress of this i vesti ati on. Appreciations are extended

to Drs. R Caldwill . L, Popenoe, F. l. Blanchard, and A. F. Lugo

for their interest and partici pation on the SnLuper,'isoy/ Cor,.li ttec and

r-view of this manuscript.

Supreme grati ude is expressed to Dr. Carlos Luna For hi hospi-

Ality and guidance during the Coloimian portion of this research.

i'ho author is grateful to !n nioo Agronoini Vic or. _Vega and Dr.

r niicisco Si lva for labor Lory fci;i tie-s t.hey ,,:ost kinil' y miade Freely

avail blie. Appreciation is expresoJ to Insi ti to Georjrofi co ":'Lgustin

Ccoazzi" for travel arr, ngn'nts, !Lot'i lotis tic and financial within

'olorbia Gratitude is exinded to Sr. L.,-o ;nid s taj a foir 1 ils assis-

IIance in mi cromorpho o ical a nai /es d ur i td'e el hr'1onC Ic Y tage of

this inves tigation. Sincere tanlk na nre ex.pre,'d the laboratory

tcchnicians in the "Laboroorio Je Sucos" n fo hil irig LoIdI!Ct misny

of thS routine chemical and physical analyses. Th'e au ,or i so cannot

fully express his dcup appreciation to th"e gi'o-iomi.'ts ao oth r

Colombian citizens for the many kinmrnesses i-,' e" -'.rd-' .o hiim.

The author is indeb -eJ o Dr. C. r. '.no for ajrcring the i'OEA

Ti i!e 1V [Fail .snlip which tacili at, this stuiy. Gr 'ci .u e is -l xten

to the. Ce.itr for Ti'o iOcai /,ric':i itui for pro i:;i, g :d. di ioi'i finan-

rcal support for th;s work both ;n Colo,,bia ard at the University of

i1o c- a.

Thanks are c 'd to Dr. 0, H. iubbell for access to p.oto-

graphic equipij' a i !1: oratory space during micron orlphol :icl i;n

-,etigations. Appreciation is expressed to Dr. L. I. rZel ny fcr

operational assistance in X-ray diffraction techniques :rd Icli~pl

consul tatli. pertaining to this study.

A special thanks to Sr. 'Juliali Velez for occasional Spanish trans-

lation assistance.

The author wishes to express special commiendar ion to Mrs. Lillian

S. Ingenlath for excellent typing and proof'ng of this dissertation.

Above all, the author takes this opportunity to express deepest

gratitude to his wife, Nancy, whose unyielding encourajeisent, labcr,

and understanding contributed beyond measure to the successful coaple-

tion of this dissertation.




ABSTPAC1 . . . .

INTRODUCTION . . . . . . .


Andosols . . .

Definition . . . . .
Occurrence . . . . .
Morphology . . . . .
Physical . . . . .
Chemical . . .
Mineralogy . . . . .
Microfabric . .

Al lophane . . . . . . .

Introduction . . . . .
Structure and Compesi Lion
X-Ray Diffracltio Data ..
Electron Microscopy
Differential Thermal Analysis
Infrared Spectrosco .
Dehydration and .':ehydialionr
Surface Area . . . . .
Optical Prope-. ties ......
Ion Exchange and Sorption
Allophana-Orrqganc re.cLior.
Fedocrnetic a iherng .
t i,>ti-,. ..... ..........

MATERi .i.3 AND IHETHMIs ....... .

Si .c P3 roce: n . . . .
fielc Proce,-' r.s . . . .. .


Physical I -thods .... .. .. . . . 26

Fa S-i;2 Size Dist ibut n . . . . . 26
Bulk Cc i t . . . . . . . . . 27
F; fteen-Atmosphare Water Cciat-.t . . . . .. 27

Chem; ca Metods . . . . . . . . . 27

pH Determin-tion . . . . . . . . . 27
Cation Exchange Capacity . . . . . . . 27
Exchangeable Calciurn and !agnesium . . 27
Exchanjgeable Potassium . . . ... . 27
Exch r.,geable Sodium . . . . . . . . 28
Organic Carbon . . . . . . . . 23
Exchan enable Alumiu . . . . . . 28
Free iron Oxides . . . . . . . . 28

Mineralogical Methods . . . . . . . . 28

Dispersion an" Fri-ciotionon of Clay . . . . 28
X-Ray Diffractiojn . . . . . . . .. . 30
DSfFerential Thermal Analysis . . . .... .. 32
Infrared Spectro-,etry Analysis . . . . . . 33
Sodium Fluoride Test for Allophane . . ... 34
Sand Miieralogy . . . . . . .. . . 34

Mi cron;iorphclogi Thin Seccicn Techniques . . . . 35

in regnat l or ........ . . . . . . 35
Preparation oF Thin Sections . . . . .. .. 35
Apaiysis i . . . . . . . . . . 36

DFS.RIPTIOi; COF HASRli STUCY AREA . . . . . .. 37

Location . . . . . . . 37

Physiography . . . . . 37

Bej,-otk and Surfi;ia1 Geology . . . . . . . 39

Cli ;H ate. . . . . . . . . . . . . .

Plant Eco!ogy . .. . . . . . . . . 45

Dry Sub-raopical Forest . . . ... ... 45
Dry Lowa-er Mlontae crest . . ......... 4
mi Lw r ontane ores . . . . .
:umlid icpta;a Forest . . . . . . . . 17
Sub-Alpin' Grassl a ds . . . . . . . 49


uLI A D CIS i ON . . . 51

V-th :y . 'r ; . . ............. 51

O lnic M li ter Uxi4.tion . . .. . . . 51
Oi'o;rsion of Ando Clays . . . . . . 51

Morphclogy . .. ............ . . . 56

Chemical Properties . . . . . 60

Soil Reaction . . . . . . . 60
Organic rbon .. ... . . . ... . 65
Exchlaiige-JJ cases ................. 65
Exchani., aJb e Aluminum . . . .. .. .. 66
Cation cxchSnge Capacity . . . . 67
Free lion Oxides . . . . . . . . . 67
Sodium Fi oiride Test or Al lophane . . ... 67
Genetic Ilterr-'latiorships of Chemical
Properties .. .. .. .. .. ...... 67

Physical Proper'ies ...... ......... .. . 69

Sand Hilin ralogy . . . . . .. . . . . 72

Light Minerals . . . . . . . . . . 72
iHea'v7i !- nerals .. . . . . . . . . . 777
Genetic Relations of Sand liinera!
Distrihut ons ...... . . . . . 78

Clay Mlineralogy . ... ... .............. 79

X-Ray D fraction . . . . . . . .. 79
S rrar i P'nalysis ... . . . ... . . 102
Differ -en ial Therma Analysis. . . . .. 117

Mi' pholo .y . . . . . . . . . . 119

Pescri ti vc Features . . . .... .. . . 124
Cen-Lic lntarpreta tions . . . . . .. i 28

SUMMARY .. . . ... .. . . . . ..... . 33

CONCLUS ICGrS . . .......... . . . ... 139

APFENDI X ................... . ....... 143


Lli 'l ATURE CITED .. . . . . . 190

i OGRAPH CA. SKETCH . . . . . . 197


.hbl e Page

0 Cli .I: rc ationshi p in
De a .,. ., Colo.A i . . . . . I

2. Fe, Sij, AI a 1ol.nis in "xLlact:, after
treal.r,;it of certain Andosol horizons by
iN'OCI o, H . . . . . . 52

2. EFF .ct of pr tiadtii:nt aIid dispersing
ac nt on particic size distr ib tion . . 55

i. Chti-nical prop ,r ti- s of selec.eJ Coloi bian
soils . . . . . 6

5. Ph',i;c l r -'pe '1 ies of selected Colonbian
oil s . . . . .. . . 70

G. Sad (50-500 ,; f' ,.'tion) miniralogy oF
seei ctcd Colo:rbian s.i ls . . . . . . 773

7. Cry. all ine *,ii;Cer: l recognized froi m
X-ray d;ffrct on itIcerns of the f 'n
cliy (<0.2 ,) a .oirse clay (0,2 2 .)
fractions oi sel',:itd Colombian soils . . . 92

8, 1ecta i;d Field description of s-eleced
Colom.ibian soils . . . . . . . . .. 143

9. The d/n spacing of crystal ne minerals
re olii zed firoi X-ray dii fraction pfatL rns
of 'th fine clay (<0,2 p) fraction of
s,.ccted Colombian soils . .......... 15/

!0. Tl'h d/ri s:pacing of rristalline minerals rec-
oni.--ed fron X-ray i fraction patterns oi
ith; oarce clay (0.2-2 p) fraction of selected
Colcrbian ooils . . . . . .. ... i0i

I1. Kicr c /ph-ology of selected Colombian
Ando ol s . . . . . . . . . . 166


1. Departameinto of Nari o located in south-
western Colombia . . .............. 25

2. Colombia, South Am-erica, lshcrwing location
of the Departamento of iari;o ...... ... 38

3. Cross-section of Nilr o, Colombiastudy area
showing locations of profiles 1-6 relative
to Volcano Galeras ...... ... ..... 41

4. Pumice beds 1 km north of Ecuadorian border . . 42

5. Typical vegetation of sub-alpine grassland
near profile 6 .................. 50

6. Color photographs of profiles I-4 rcpresent-
ing altitudinal sequence of 2,00- 2,730 a . 58

7. Color photographs of profiles 5 and 6 rep-
reo-Ceinci attitudinal sequence of 3,120 -
3,510 m and ancillary profiles 8 and 9 at
3,020 and 3,060 m ........... ... 59

B. X-ray diffraction patterns of Ig-satura Led
air dry, K-saturatcd 100C, end K-saturated
500C treated coarse clay of horizon. A
pro file 7 demonstrating identification of
netaalloysi te, hydrated hilloysite, il i e,
and vermiculite by changc.f in d-spccings . 80

9. X-ray diffraction patterns of rg-saLtrated
coarse clay (0.2 2 ),) and fine clay
( <0.2 u) Fractions of profile I ......... 82

10. X-ray I ffraci on patterns of Hig-saturaled'
coarse clay (0.2 2 pi) and fine clay
(<0.2 p) fractions of profile 2 .. . . ... 83

11. X-ray diffraction patterns of ig-saturated
coarse clay (0.2 2 u) and fine clay
(<0.2 ;j) fractions of profile 3 ............. 4

2. Y ri iIr .io n frns alii, d
c .' c!y (0),2 ? :) anOi tini clay
(C O. I) 1r.r ': ,r 's oF pro file .

i X- y i frac ti i r s of I- sa Lurated
cc i c cli. (0.2 2 ji) and Iine clay
(< 0. j ) i' ctiors of profit le 5 .

14. X-r i fr iio pc.ati ternsi o r l-satur ted
coarai clay (0,2 2 j) and Fine clay
(< 0.2 i ) fractions of profit le 6 . . .

16. X-ray di ffraction patterns of Hg-saturated
coarse clay (0.2 2p) and fine clay
(< 0.2 ji) fractions of profile 7 . . . .

16. X-ray diffraction patterns of Mgl-saturated
coarse clay (0.2 2 u) and fine clay
(< 0.2 tj) fractions of profile 8 . . . . .

17. X-ray diffraction patterns of Mg-saturaLed
coarle clay (0.2 2 ii) and fine clay
(< 0.2 j) fractions of profile 9 . . . .

18. X-ray diffraction patterns of Mg-saturated
coarse clay (0.2 2;1) and fine clay
(< 0.2 j) fractions of profile 10 . . .

19. Infrared spectra of coarse clay (0.2 2P)
nrd fine clay (<0,2 )) Fractions of
profit 1 . . . . . . . . .

20. Infrared spectra of coarse clay (0.2 2 p)
Ind fire clay (<0.2 F) fractions of
profit . . . . . . . . .

21. Infrac d spectra of coarse clay (0.2 2 p)
and fine clay (< 0.2 1) fractions of
profile 3 . . . .

22. Irfrared spectra of coarse clay (0.2 2 )1)
and fine clay (<0.2 .i) fractions of
profile 4 . . . . .

23. Infrared spectra of coarse clay (0.2 2 i)
and fine clay (< 0.2 ) fractions of
profile 5 . . . . .

24. Infrared spectra of coarse clay (0.2 2 1)
and fine clay (< 0.2 y) fractions of
pro ile 6 . . . .

25. Infrared spectra of coarse clay (0.2 2 i)
and fine clay ( <0.2 y) fractions of
profile 7 .. ......... ......... 10)

26. Infrared spectra of coarse clay (0.2 2 j)
and fi-e clay (< 0.2 y) fractions of
profile S . . . . . .... . .. .. 110

27. Infrared spectra of coarse clay (0.2 2 u)
and fine clay (< 0.2 j) fractions of
profile 9 .......... ........ ... III

28. Infrared spectra of coarse clay (0.2 2 )
and fine clay (< 0.2 p) fractions of
profile I0 ......... ....... 112

29. Infrared spectra of coarse clay (0.2 2 p)
and fine clay (<0.2 i) fractions of ki, 119
buried argillic horizon; Ward's halloysite
No. 12; and KBr blank ............... 113

30. Differential Lhermal patterns of coarse clay
(0.2 2 Ji) fraction of selected horizons
from profiles I and 10 . . . . . . . 118

31. Photomicrograph of All horizon, profile 3
showing tio fecal pellets set against matrix
of interconnected ortho-craze planes end
orthovughs . . . . ... .. ..... 120

32. Photomicrograph of All horizon, profile 3
showing organan on andesite lithorelict
and subrounded blocky secondary structural
unit set against matrix of interconnected
orthlovughs ........ ......... . ..... 120

33. Photomicrograph of All horizon,profile 3
showing isotic plastic fabric . . . . ... 121

34. Pihoomicrograph of A12 horizon, profile 3
showing feldspar grain wii th weathering
fractures perpendicular to twins . . . 121

35. Fhotcmicrograph of [BJ horizon, profit le 3
showing weakly silasepic and undulic
plastic fabric ............ ...... 122

36. Photomlicrograph of [B] horizon, profile 3
showing hornblende grain with thin,
siliceous isotropic border . . . . . .. 1. 22

37. icro h *f 'i( h ri; i lI 3
i 1 e io Ii (h tit c)
noulo aIind *v, r a mial lor on.. . . . . . 123

38. Phot"ic i-ir ,h of C .horizon, przfilo 3
sho-ing h;i nb g rain wi h t'in,
iiscontinLcii s si i!.n (h'I atan) set
':; i!st i:m trix or inri rcorncted ortCovlJghs . . 123

aernac, o Oi)ss a ition PrsKn-:nd to the
Grad .te Coul :i ci f die Univ-rsity of Flurid: in partial Fu l il e.t
Sthe -,ire 'as for t' gree of 1caor of Phiilso'phy



Frank Gilbert Calhoui, Jr.

August, 197!

Chairman: Dr. Victor WI Carlisle
Major Departrnet: Soil Science

Ten soi! profiles from t're Adean highlands of Departimeno 'ariarno,

Colombia, were investigated by use of chemical, physical, mireralogical,

and micromorpholog9ical techniques in order to elucidate pedocenetlic

processes in soals derived from andesitic volcanic ash anId similarr

parent materials. Six Andosol profiles were selected iin an iltitudinal

sequence from 2,000 m to 3,510 m as representative of maxir'.,; ranog of

weathering severity experienced by aeolian p;roc'lastic :.aterlais ia the

altipliro cf 'i-in'Ro. Four additional soil profiles, 4ev lopirg in

pa:-ent materials with varying degrees of ash iinfl-ence, provided si.pple-

mentary pcdoga'r t'c ;nforn!ation. All of tHe-soils ,.;e re infienced by

annual rainfai? range of 500-2, 00 nn and an annual isothermal tei-:per-

ature range of 6C to !SC ..hih was closely relaLd to li

The folio'iljig parameters jwer dter, si ned on :.el ected horizons oi:

each soil: (a) particle size distribution, (b) buik density, (c) fifteen

atmosphere wa ter 'on-tet, (d) pH in water anJ t K;CI. (e) exchi;rqgeah!e

cautions, (f) cation exchange capacity, (g) organic carbon, (h) free iron

oxides, (i) X-ra, diiffraction of fine clay (< 0.2:') and coarse clay

(0.2 2p), (j) infrared analysis of Fine clay and coarse clay, (k) NaF

test for .'lop 'ah cid (!) liih anJi heavy nine' 'l contents "f the sand


,.0 I,0 p ) fr ion', Ili, action, pr apre'I f i, ,1 'h.c EoJ

*ori s of A o- .I p 'fils 1- (2,z 0t 3,510 ;) < d -I cri' cd in

Th 'O'. :.,ntioned an lys,; of Anrkl'ol p o fi 1 -6 r 11e ld the

I i,1 trI S itI inc sin l, v I t io : (I) i cre Ie 5 ir 5c11T1n-e--

'I i K ,iJ Ai; or A a ic carbon; olcanic glass, 1py'O xnes, alrd <,ndesi Lt

Ii ihorl icts in the j0 500 p said fraction; al lophane ni <2 )1 clay

I'rct ion; id isotropi sm of plasiic fabrics in subsurface horizons;

(2) decraasns in FH values; exchdii;nable Ca and Na; free i ron oxides;

quartz, feldspar arl; hornblende in 50 500 p sand fraction; hydrated

hal oysite in the < 2 ) clay fraction; sesquioxidic nodules; ondJ bire-

fri (eiince of plastic fabrics in subsurface horizons.

Vary ng amounts of hydrated hal loys i te, niLahs loysi te, vermicul i ic,

cristoba i Le, gibbsi te, feldspar, and quartz were identi filed by infrared

and X-ray analyses of the coarse clay fraction of Andosol profiles 1--6.

Hydrated hal loysi 'e, nsta'il1 losi te, and possibly allophane A ware

identified in the fine clay fraction.

Andosol surface horizons generally had isotic plasnmic fabrics while

subsurface ho i zons exhibited :wekly unduli c and silasepic pl5sm;c

fabrics which were replaced by isotic plasmic fabrics at higher eleva-

tions. The turri "Andic" vwas proposed as a rodifier for this sequence

of obhrlc types.

A clay mi neral wea hearing sequence or a llophane -- metaha loysi te--)

hydra.ld halloysite was postulated. The maturation of this sequence

increased with profi le depth and wi Lh incre-sii;r a!ti tude. CoIrse clay

varmiculi te was hypothersi ed as be;og tr'Icsi ent sr insui;-r cient time has

elapsed for it's tran forniation Lo other 2:1 phyllosi icaces. it was

Further postulated that vermicul i te pedogenesis occurred rather rapidly/

with some modification from a primary mineral source.



Op:or iuni ties h;,.e been al i o r're ora genetic i nv l ig3 tions

in the tropics invol ving soils deri ',J froli na teria ls in an initial

stoie of eri'athering. Research in temperate regions such as New Zealaid

and Japan has disclosed that unique weather ln products occurs during

Lte early stages or primary mineral cdgradation in ?/yroclastic materials.

-3ils dIrivd from such coniniinuted acol in pyroclastic rat, rials (us'sl ly

.'desicic) have been referred to as Andosols. Andosols present very

real mngndgernt call enges due to their aberrant ph;'sical, chemi cal,

-id mi n1ralogical property is. These aberrations are in large pert due

LO their prcence of a lophone w-;hich is a primitive Veqthering prod rt of

volcanic glass.

There is reason to believe that the genesis of Lropical Andosols

is to some degree thermal dependent. Wright (1961+) noted a "thermal

progression" in r:orphology of low latitude Andosols in Souch America.

There twas ev'ey reason to assume that morphological changes were related

to chemical and r;,ineralonicai di fferences; however, laboratory data w ere

unavailable for substantiion of this potential rsiationship. This

investigation provides such analytical data. it also may be considered

as an addiLion to the general reservoir of kno'.lcedge irgarding the nature

and properties of tropical soils.

Colombia is ore io the worlds' largest coffee exporters. Thiis C op

is grown primarily ard most efficiently on volcanic ash soils of the

highlands. Additionally these soils hold an important role in small


grain and .esetable production fo.- local c, su*.ption. The populace of

Colo-bia is concentrated in the Andcan highlands. iL's eco,',> is

still primarily agrarian ard population expansion i ll continue to

apply considerable pressure on soil resources, especially the com-

paratively fertile Andosols.

A dire need evolves, then, for an accurate inventory and character-

ization of Aniosols, especially in the tropics they most cormonly

occur. Genetic interpretation of these data can add to understanding

of Andosols with concomitant implications to their efficient management.

It would behoove the temperate climate oriented pedologists to partici-

pate in and contribute to these programs in order to broaden their scope

and understanding of soils derived front a wide spectrum of parent

materials. Selection of study sites in close proximity to areas of

historical volcanism would facilitate pedological investigation of

soi s dominated by easily wedatherable primary minerals. With this in

mind,arrangements ;eare mad vith the Insti uto Geografico "Agustin

Codazzi" in Bogota, Colo;mbia, to participate in their program of char-

acterizing and classic ying the Andosols of Colombia.

The intent and objective of the study presented herein was to

elucidate pedogenetic processes occurring in selected Colombian Andosols.

This .qas expedi ted by use of chemical, physical, mineralogical, and

heretofore seldom sedri micromorpholojical paramete-s.

L I Fi RAr u ;: REV I

-Ifini ti o

During sun ,ary and Lcchnical discussions of a "Meeeting on Lhe

Cliasification and Correlation of Soils From Volcanic Ash" held in

Tokyo, Japan ("noyir.ous, 19c5), the following dcini tion was proposed

for Andosols: "Mincral coi Is in which the active fraction is dominated

by a.:or'phous aeriall (rin;mum about 50%). These so Is hlave a hiyh

-orptivc capacity, 3 relatively thick friable dark "A" hori Zon, are

high in organic mi:tei r, have a low bulk density aid a ioi' stickiness.

They :3y have a (F) horizon and no' show sig.iificant clay r.oven:er.t.

These soils occur under humid and sub-hu mid conditions."

With the advent of a new so;l classification system in the Uni ted

States (U. S. Soil Survey Staff, 1970), an important si:border referred

to as ''Andcpt'" was developed which, according to Fiach (!969), en-

compassed ,any of Lhe soils that had beni called And'osois. Andepts

are described ai "i'ore or less freely drained inceptisois that have

Iow blk dr-sity and apprecciabl 1s.ounts of allophan -~h a high

ec,'chia:n capCa i y, o;t t ha i-v rie st'ly pyroclastic materials in the

sil t, sandI, Sa d jr- ei fr actionss"


According -,t tthe ., Si .So i S!,rvey Staff in Soil Ta;xo-omy (i9i0),

Andepts occur in or near mountains with active volcanoes from the

equal or co high lacitudes Andosols, therefore, occur ,.: .ly t ',o..h--

l-jt Lt= Pacific ,inj of volcanism '.4 inr volc'i c rc 1io of C 0 ELsi

Indies and Africa.

Luna (19s59) reported that in Colombia t:ypicil Andisols occur

most coi,,noily on gentle, slightly u:duidling topography betwen I 300 m

and 3,000 m abve sda level in the And.-s i;ounta;.' rnii j. Luai fu-her

stated that studies on volcanic ash '!is and volcanism in Sgnerai

were very l iri ted in Colombia. The i;pl icaeion was that expanded

programs in these areas would be qu; e desirable.

Var-ca (1963) completed a general study of soils in the Pasto-Rio

Mayo sector in Departamento Narino, southern Colombia. lie reportedd

chemical, physical and morphological properties for several profiless

collected fro: six irajor soil areas in the sector. The do!,i nant 0roup

of the six vaere soils derived from andes:itic volcanic aish. Vare

suggested that a large group of soils derived from vcicanic tuff had

"hardpans' cemented by carbonates and/or calcareous clay. Iield obser-

vation of these soils by the author aftiriro. thai the "hardpan" is

actually a duripan as ihe primaiy cerenting agent .as silica.

Wright (1964) csated that, even thiuigh Ar.,dsols occur under a wide

range of clin3atic co;ldi iions f.'cm lo,- to high lati :-des, the jross

i;or-hology of the soil pr.ofil es remains rather ul;iform. -!-' listed

several properties as co;rmon to Andcools throighout t'ce korld:

(a) deep profiles o.:'tn with distinct deposit;oncl stratification,

(b) thick dark surface horizonss high in negli;gily hiodegradable

humic compounds,

(c) pi i r y I i ish br' ;u'':oil colors ~id. "' i ari"

(hixi x o n) ,

(d) low b il, densi ty i hinjh w.;.r--hoding and retention c paci ty,

(c) t I jctural ,a ilcgat:on aJn I ck of cuans on pod surfaces,


(f) lack oF I tickiress dor plastic Ly v.hern roist.

Wright also Ji cussed, in dtai l, properties of both high and low

latitude Ardosols. He indicated that at low- latitudes there is a

thcriIal scqrl: nc 4i lth altitude which imparts only faint morphological

trends to dte soil profiles. The following differences with increasing

terpern ture wrire noted:

(a) horizoa.lion progressed from AC to A1-A3-C to A-(B)-C as

n'oi s u r increased,

(b) topsoil colors changed from black to dark brown,

(c) weak qr-i:ll ar structure progressed to moderate blocky in the

surface h'ori'ons,

(d) subsoil colors became redder along with some plasticity develop-

ment, and

(a) organic riatter content decreased.


Very high natural moisture contents (measured on a weight basis)

cnd associated low dry bulk densities are comnion features of volcanic

ash soils (Swindale, 1969). Both total water and 15-har water contents

are high, however, and whpe nreasured on a volume bdsis, the values for

available water are not markedly different from other soils. When

these soi s are dried t,) i5 atmospheres or beyond, the shrinkage be-

comes increasingly irreversible. Gradwell and Birrell (1954) dis-


cove Id that if soi I samples ha,,e n pr'viou !y ai r-dr ied th re .as

a ,rarl' d dec rase in liquid 1 i it a; pl .as ic limit values as ihe

ol lophane content increased.

AnJosols, in general, have a very high total porosity. Ydioa-ki

et a,. (1963) reporLed a figure of 0S/ for Kanto loam, a predominant

soil around Tok;,o. Swindale (1969) stated thaL associated wiith this

high porosity is a high saturated permeability, a though these soils

are notoriously di fficult to rowet after they are dried because the

organic substances adhering to particle surfaces are generally hydro-

phob;c and the contact angles are large. Youngberg and Dyrness (1964)

published a range of 63-85% pore spaces for some Oregon Andosols.

According to Swindale (1969), the particle size range in Andosols

is quite often very large. He further stated that particle size

distribution data jre difficult to obtain due to poor dispersion

characteristics of these soils.


In general, pH values of Andosols rang. above 5.0 (hi lir et al.,

1968, and Flach, 1964).

Cation exchange capacity values for these soils are quite high,

Birrell and Gradw'l! (1956) have shown that this parameter varies with

concentration of leaching solution, nature of ions in solution, and

volux of wLatr content of the washing alcohol. G;rrell (1964) and

F'inik (196) poiitid out thla conventional CEC determination may be

subject to considerable I rror when applied to soils containing amor-

phous mineral cciloids and ali such determinations iust be interpreted

wi th caution.

Base saturation covers a wide range of values although pH values

S : i c, l ( ), i i t ibu ed ;o pi dep i -

Ano uu! s are kn 0.: io fix lirge quai ti 'ies of hosp'.ice, tolybJare,

ardI /th r i ili l rly sLruct 'rd nions. S~ uiin rs (1 59) a:d Mi l ler ot

.H. (1930) avi sho'n, for ;i i, Z.. li;ndl and C ,sto Rican 's i Is, respec-

tively, Lhat jphosplha is bound to a lumini, Blasco (1969) reported

d(t L for upla d soils in DeparLa, anLo ilri o, Coloh.bia, and claimed

Ltht inert plhospiaous represented about 52' oF tiha total P with rather

low values for other P fractions.

Andosols invariably contain relatively largyc counts of organic

maLLer in Lheir surface horizons. Kosaka et al. (1962) and Aominne

ardJ ;obaYashi (lS:64-, 196Lb) demons rated the negligible biodegrad-

ability of organic materials in these soils.


Eesoail (1969b) ipesented an extensive review of literature on

the sineralogy of volcanic ash soils. The sand mineralogyy of these

soils is composed mainly of prir'ary minerals such as ferron'acnesiums

(o livines, pyroxenes, and amphibolbs), quartz, foldspars, nmagretite,

and various silicate comply xes. The silt fraction may be constituted

of primary and secondary minerals. The secondary minerals, Ospecially

in high allophan.e oils, may be aggregates of cla/ occuriin9 as

psudosororphs of silt. ibbsite and hydrated iron oxides frequently

tend Lo accumulate in the fine silt fraction (5-2 -).

eesonin (I>6Sdb) wvnt into considerable detail regarding the

occurlircec and format io of various secondary minerals in the clay

fraction of volcanic ash soils. Allophane and halloysite were the

most prevalent species. Weathering sequences involving these two

members were especially interesting and will be discussed in greater

detail later. Under special environmental conditions, such secondary

mineral species as kaolinite, gibbsite, bohenite, imogoiite, montmo-

rillonite, hissingerite, vermiculite, chlorite, sericite, palagonite,

amorphous silica, aiiorphous alumina, iron oxides, and amorphous and

crystalline titanium oxides have been identified in the clay fraction

of Andosols.


There is a dearth of published information on the n:cromorphological

properties of Andosols. Frei (1964) described the microfabric of a

black Andean or "Paramo" soil and a brunizem soil located in the

central Andes of Ecuador. The black Andean soil was derived from

andesitic volcanic ash and was characterized by having highly stable

crumbs which formed a spongy and very porous matrix. He speculated

that black cutans on ped surfaces in the lower epipedon resulted front

precipi cation of dispersed organic matter migrating downwards by

percolation. Below the epipedon in the B/C horizon, organans occurred

on all the primary sand-sized minerals. Frei also suggested that

perhaps colloidal silica acted as a completing and mobilizing agent

for organic col'oids with subsequent precipitation on mineral surfaces

at lower depths in the profile.

llejia (1971) concluded from fabric analyses of a profile of the

Corzo Leries (an Andosol on t.e "'abana de Bogota") that tiese data

provided valid additional criteria for the classification of ash-

derived soils. 'e outlined several unique characteristics the most

important of which wa:. the presence of an isotic plasmic fabric (see

"I ;r, r I I, f'or di scussio Lhi: t!)e ,' niL .ri lCno ogy osn

nric an lys is) i n tl A v; r hol i > z i"< id [ 0 t iL !, ic 1,1iisliC

fabric in i ho dccc-z;t .h ri/ons. This feature along wi ih characteristic

void p:5toris, i:t i niincr'.loI y n,/ d 'n range ii: nt, iand s ;d mil'ralogy

u, : ALed J i ni ,in r'ibrric rypo pi culiar io Andosols. rI "ji prr osCd

h n ic ., rica A.dicc" (A idic F b hric) For this Fabric class.

iaej ( r' intained, hiov v r, that rigorous statistical studies ovc.r a

wide range of Andosols iws needed before precise definition and adop-

tion of this fabric tpe could be realized.

Luna (1969a) mentioned the existence of a fabric formed by a

plasma of porous aspect and of isotropic character in a Typic DysLrandept

of Dcpartarento Antioquia, Colombia. The fabric type was Lentacively

designated as Amorphic.

A ophane

i ntrodnction

As mentioned earlier,Andosols are characterized by low bulk

densities and low- base saturations in addition to high organic carbon

couiltnts and high phosphate aid water retention capacities. These

properties are to a considerable degree duc to high quantities of

-almohane (a disordered hydrous alumino-silicaLe) in the clay fraction.

The unique cualitiis of allop :-ne in contrast to the more common

orde-red alumino-silicates present special managrmen problems involvir.g

these soils.

Due to the widespread distribution of volcanic ash soils and a

well-developea science of pedology in Japan and aew Zealand, consider-

able research on allophane has been undertaken iin these two countries

in the past 20 years. This review will outline and discuss ajor

portion of this research in addition ao accompli; sh enLs in ot-:r r.eas

of the world. Use of the tern 'allophane" will be restricted to

pedogeneLic disordered hydrous u luino-si icaces derived frow volcanic

ash, generally of an andusitic type.

Structure ard Comrposition

Ross and Kerr (1934) were probably t:., first workers to present

significant data on allophane. They reported for geologic allophane:

(a) variable chemical composition, (b) variable index of rafr-ction,

(c) no definite crystal structure (lack of sharp X-ray pov;.-dr peaks),

and (d) no regular molecular arrangement as interpreted From infrared

spectroscopy. Ross and Kerr referred to al lophane as an amorphous

alumino-si 1 icate and i ttle more was ascertained as to i ts structure

until the 1950's when Japanese and New Zealand ipe'ologists began to

explore more closely the weathering sequences of volcanic ash clays.

Advances in structural knowledge of allIopTer-e came haout through the

increased use of techniqu-s iuch as X-r/y an:d electron diffracromitry,

differentia l 'rrral anaysis, iinrned spectroscopy, and electron

mi croscopy.

Throughout the !i 3 s and ear ly j1,6's,,seaer3 investigators

disco-/ered that al ophane in -.,,.ic:anic a-h soils could exist in a

structural continue iirl bel ow the "rcl i ci.i cr s talli e puri ty" of hal oy-

i te or ka li:ite. Fields (i935) ir-nicated. a weathering sequence of

i creasingg crystal l ni ty: allophane B --al loph3re A .-methal loy-

site. Yoishnaga e s Aoc inna (1'62a) ca;i,..d to have dlscovcrad a more

crystalline alc mlia-r;ch species wi ih fibrous morphology which they

called "imogclite." Later Acmine and Miyauchi (1965) through the use

of 1iy d I OL ol :Ar le to

iIo a 0 !.o l i lite:" A J 3. "le'golite 3"C 'T s sup--

p A 0d to r "!" -! a "913 crystelli,,, stage LhU "i liic AI"

y 'a (l'";) ';-, n to r', acir se tp, sp-ci ici .y of al Icphn,; struc-

e '; h, infrar,, a:- ion with dI oleri.. (0 o) ex-

i. f hy). ../ glo up (?;l - O 0) at roo.i ii;r; i a ura e o OOC.

This : ','s one a c:,iparative basis with ian tmorillonile anr. hydrated

halloysite. ;I four that all the OH groups in could be

replaced by OD,F'hile vury ;minor and parltal reply acemen ts resulted for

montLinri I lon ie and hal oysite. Wada concluded that a sharp line

exists between structural OH groups and absorbed 20 in aillphane.

Uada (1967) dhen sur.marized chemical, infrared spectroscopy, and

rorphological data for soi l illophane and suggested two end-inembes

.ith chemical composition 2SiO2 e Al203 3H20 (allopha,,e) and Si02,A120

H2 0 ("i 3r-ol ita") along w i:h a structural scheme for the two end-

e:;,bers and inter:edia'e coiipositions. HIe proposed a structure for

allophaone wi th a 1:1 Si/AI ratio composed of 3 si lica tetra!.edral chain

and an alumina octahldial chain sharing a corner of the respective

tetrahedron anj octc hedron. Addition of another alumiina octaliedral

chain to the silica .etraheidral chain results in formation of alIlophane

with a 1:2 Si/Al ratio ("i rogolit':), Wada reported that noncrystalline

and poorly crystai lna X-ray patterns oi alilopha-.e with a Si/Al ratio

1:1 and 1:2 could he interpreted in t"iam of the differences in the

case wi:.h ihich the respective unit chains are aligned.

App3arintly there can be considerable substitution of Al for Si in

tetrahedral co-ordination in alnophanae which deelnitely segregates it

from halloysito ai d k'aol iicc. iada's (1367) scheme claimed to predict


the trinsfo. :ation of Al co-ordination from 6 to 5 upon dehydration

and further to 4 upon cehydroxylation.

Russell Cet a. (1969) presented supporting evidence for the

uniqueness of "ii:ogole i e' khich they believed .,arrdnced i ts cilas;i iica-

tion as a distinct mineral species. They interpreted electron diffrac-

tion patterns in terms of repeat units 8.4A par-alll and 2A perpcndiic-

ular to the fiber axis. They proposed an alternate structure to that

of Wlada's (1967) for "imogolite" consisting of discontinucus distorted

chains of Ai-0 octahedra linked through isolated Si20 groups. They

contended that the existence of the Si207 groups was supported by an

infrared absorption band near 930 cm-". They reported composi tions

of 3 SiO2 2A 20 3 0 5P20 and SiO 2 Al203 3H2C for "imoyolite"

and al lophane,respeccively, somewhat different from /ada's formulae.

X-Ray Diffraction Data

The presence or absence of X-ray diffraction peaks for allophane

would, of course, be highly dependent upon its stage of crystal inity

between partially weathered volcanic glass and hydrated lalloysite.

X-ray data have been presented in numerous papers generally as a oub-

ordinate analysis.

Separation of the clay fraction (<2p) into s;,al ler size ranges

has usually facilitated diffraction in/estiga tins of allophane.

Yoshirnga and Aomine (1362a) showed that the 2-1 pi fraction of a Uemura

soil -rhibited seve-al distinct di fraction peaks indicatingg the

p esence of quartz, c'is obalite, fcldspar, and gibbsi e. They were

able to isolate "imogoliite" and allophane using-di fferential dispersion

properties. in the 0.1 ;r fraction, allophine displayed no diffra:cton

peaks or bands (X-ray amorphous) while "imogolite" gave distinct dif-

fraction bands at 4.5, 7.6, and 5.5A.

' b 1 'icrr ,opy'
i (i1 ) 1i d 1 t l .r ,: ,-c : s 1 alloe n ge

Srili ly s I a eJ fluffy a5reg wi th a rc',rdJ d r d! l,"r i-iearac-:e.

Yos'i r giy and Ao-ie (1962;i ) Jscrib-d ";V ry o glo'>>i's"' ile

Miy, : (1969) 'oh!i .v d 7.r.y "fibi us ': uters" whichh uI i ild in retro-

spect probably be ;ni fid 'as "ia5i oie." Yohiinap and Aomrine

(1i62') observed thread-- ike particles of relatively uniform size which

Lhey tenta-ivoly identified as "irogoli e.." Wiadd (1967) sunmmar;zed

several observations by ird ica ti n that a fairly good corre la ion

existed bel'ween iworphology and developing t of structural order. He

stated thit "imogolitc" developed smooth, curved and often branched

threads \with diameter 50-200A and extending as long as s'eeral microns.

W;th an increase in the Si/Al ratio, the thread-like tendency gradually

diminished and minute particles (50 to 500A) in aggregated masses or

thin film s baca-me dominant. The fibrous morphology of "itnogolite"

strongly suggested one dim'nsiona! regularity in structural organiza-


D fferential ThermT l Analy sis

Yoshinaga and Aomine (1962a) public shed thermal data on several

soil allophanes. They revealed no marked therm il reactions wit Lh the

exception of a strong cndotherm between 10-170C adsorbedd H20) and a

sharp exotherm bet^eenr 910-915C (fort,:aTion of gamma-aluminsa). A

sluggish exot.herm was nole-J just after the endotherm which was assumed

to be due to tightly bouod peroxide-resistant organic matter.

Thar.,i l a:s;alsis of "irfcogoiite" by Yoshinaga ar. Aomine (i962b)

displayed the early endothenr and late ex.otherm of allophane, cnly

they occurred at temperatures 20 to 40C higher along with a more intense

exo thbrm. In cJdition, "hia olite" exhibited -, endotherra betL,;en

425-435C, indicating driydroxylation.

Miyazaw!a (1Yi2) reported a sharp e.xother!a c 850C for clay From

a hufnifed all9ophlnic soil of Japan. This appeared to be a somewhat

lowi nucleation temperature for allophane.

Miyiuchi a:.nd Ao.mine (1966) indicated that DTA curves varied con-

siderably with 'exchangeable cations, especially for the high temper-

ature exotherm. They ranked the cation effect of this exotherm from

highest temperature to lowest in the following order:

Ca> K > > g > Na

Campbell et al. (1968) revealed that particle size, pH, and organic

matter content :had considerable effect on the thermal analysis of

allophane. They found that the area of the low temperature endotherm

and the height .nd temperature of the high temperature exotherm in-

creased w;ih decreasing particle size. With increasing pH, here was a

considerable reduction in height and sharpness of the exotherm. Removal

of organic matter had the same effect. The authors concluded that con-

siderable changes in the high temperature exothermic peak of allophane

could be produced by treatments often used in the separation of

amorphous colloids from soils.

I n f red Spectrosc.o-

Infrared absorption spectra maxira For clay minerals have been

generally identified as follows (Grim, 1968):

F-eg e Lgm ...i(J Absorptior due to:

3,500 3,700 -4 Cil groups
1,300 1,90 6-10 -1adsorbed 1120
400 1,150 8-25 lattice vibrations

Several investigators such as Lai and S-Cindale (1969), Yoshinaga

A (la i, 196?T ), i V ti-'c Os '- d F. ik (i' ) reF ried

r, -. 1 '. o;- i :i r. .J n ': xi Iy 3 c "1, D, 0 c i 1-,

J 930 :-"1 d< e .o hydro,/yl rou s, ab o- ,J H.,0, amd Si--0 or '5E -0-

AI -nd vibr ion r re pecti vely

U ( f (19 ')w I ",LOe lo o-cor I i i bU-ic dif !rtnc o bEc-w.en

i"ir te" and aiu' p ',; in Si 0 s trcthi n band. Allow phae' a i ith

Si/AI ratios of 1:1 to 1:1.3 displayed broad absorption naxima at 1,0!0

and 9':5 cml1 while "i aogoli Le" (Si/Al = 1/1) had rather w.l el l-d fi ned

absorption maxiri at 990, 955, and 925 cm-1.

Lai and Swindale (1969) reported that for four Andosols of Hawaii

and one from Japan the main infrared absorption band shirl'ed froia

1,100 to 900 cm-1 with increasing v.'aight ratio of A20 3/A1203 + Si0O2

This apparently reflected a change in state of Al coordination. Cation

exchange capaci ty and infrared data together suggested that soa

stabilized AliV 0 Al bonds occur in allophane minerals.

DchyI.r tion and Rhl'dration

loc,'gral dehydration curves of aIlophare hive generally been

ob-served to be qui te smooth (Girn, 1968, Yoshinaga and Aomine, 1962a,

1962b); however, Yoshinaga and Aomine (1962b) reported an inflection

for "inogolite" indicating rapid dehydration bot';cen 275-350C which

was attributed to structural H20.

Schalsa e a!. (1365) reported that air-drying of Chilean volcanic

ash soils produced sigaificanr Irreversible changes in CEC, soluble Fe,

and P. Air drying has been ecpo lted by many workers to cause formation

of pseudomorp s of sand and sil t which became difficult to disperse.

This gives sore indication of the rehydration properties of allophane,

Surface Area

Aomine and Yoshinaija (1955) '-ported that echylene glycol ritenL;on

was high for their allophane samples; retenLion by unheated samples p 'as

si milar to that of montmori lloi e, but heated materials gae valie-s

higher than those of heated mnontmori l on!it.

Orchiston (1959) indicated that heating of allophane considerably

increased the area available to the water molecule at mono-layer cover-

age. He further stated, from a comparison of the surface areas avail-

able to water vapor and nitrogen, that some allophanes may possess the

equivalent of internal surface.

Clay contents of soils, containing varying amounts of allophane,

were calculated from measurements of surface areas of separated frac-

tions using adsorption of nitrogen, acetic acid, and water-vapor as

methods of measurement by Birrell (1966). Values for nitrogen adsorp-

tion ranged from 29 to 320 m2/g of surface area. Since these clays were

not pure allophane, the above data should be interpreted with care.

Optical Properties

Allophane is optically isotropic with an index of refraction close

to that of glass. Discrete clay units are unobservable in thin sections;

however, due to the anisotropic properties of layer silicates, their

overall orientation relationships can be observed and interpreted pedo-

genetically (see Brever, 1964).

Luna (1969a) reported mainly isotropic clay in thin sections for

a Colombian Andosol; however, Luna (1970) indicated thb.t as allophane

weathered to a relatively nore crystalline state, it also gradually

tool; on some anisotropic properties.

Steen and Fryxell (1965) stated that weathering has had little


rdi il .r i. lfl on 'hi val,-.s of !'9c i. 'l-x of 1 f r .;c )f pumice

I I S :,:' 1 1, ;, j, ard G I 3c i ri Fa k.

,)s Jr i Kcrr 1934) oi 0) rvcd a vir i ble i -icx of refr.actonc

( .';72 1- i .' ) ior al lophn

I on E:. c i d emotiorn

fhi ion .xch.ino i oprt ies of allohane p-i:e Andosols in a

very unique nanig','- t position. lon exchange capacities of allophane

are highly pH-depcndent.

Ao-1ine and Yoshinaga (1955) reported a CEC (at pi. 7.0) range of

23-51 mraq/100 g oven-dried clay from three Japanese Andosols; however,

Girrell and Gradwell (1956) published data indicating that conventional

CEC methods were not necessarily appropriate for allophane.

'.ada and Ataka (1958) then shed considerable light on the CEC

and AEC pH-dependency of allophin e. The CEC and AEC values were

n terni ned with Hil, CI solutions varying in concern ration and pH.

Two adsorption mechanisms were discerned, based on; pH and equi ibrium

concentration. They found that the higher the pH of the solution,

the more cations adsorbed vi th the opposi te true for anions; also that

tahe higher the concentration of I Cl, the greater was the apparent


Aoriine and Jackson (1959) claimed to be :ble to use "cation

-xchlage capacity i delta value" (ACEC (: EC pH 10.5 CEC pH 3.5) to

calculate allophanc content of Andosols.

Fieides and Schofield (1960) reported on two allofhanic soil

cl-ys ,*ich had a net positive charge below pH 5.5 but acquired con-

siderable negative charge between pi-1 6 and 7. They postulated that in

allophane, because of its origin From glasses and disordered feldspars,

octaiedral posi> -ns .ere icrc'ing and its ch)arce tic var bile

charge was atli 'bca ad to the J'-iavicr of speci ic sites origin Liig

from bro'-en boiLs of Al in ittrahedral co-ordination. In l!ht of

Wlada's (1967) observations, octa Idral posi tions inere ,-ot r ccssarily

lackingg", esp: cially in "imolgolite." It could alIo be that h'ie a .tnit

of tetraharaal alu.ii nu.m, as less than that inferred by FieldIs and


Houng et aL. (S566) published data indicating that the pH-dependent

charges of allcphane were more likely due to dissociation of surface

hydroxyl groups associated with silicon atoms (silanol groups), and

that permanent charges were probably due to tetrahedrally co-ordinated

aluminum in the a!i.mino-silica network. They further sugay5sed that

the amount of exchangeable bases should be expressed in terms of per-

centage of the permanent charges, since the pH-dependent charges were

considered not to take part in the exchange reaction under normal soil

condi tions.

Aomine and Egashira (1968) reported flocculation of allophane

occurred 'ith a relatively snall arrount of divalent anions such as

sulfate as against ironovalenl anions. Their results indicated that

allophane w!as positively charged at the prevail iing pH of Andosois,

unlike the usual crystalline soil colloids.

Eschena and Gessa (1S63) showed negiti e charge density of

allophanic clays calculated from permanent cation exchange capacity

(CECp) as lower than that of bentonite and kaolinite. Heating For

2 hours at 400C ruduced pil-dependent exchange capacity (CECv) and

specific surface, but had no effect on CECp. Tan (i~5's) useJ CECv as

a criterion for classifying the Andoso!s of Indonesia.


I ( 1 ) v ,ti j). r ic f)01 :,ho h,' wi 1 h 1 1oph nre.

;'; ''.0, O,,i. . ospi- )2 '" ] :ci J r 1 1h r r .pidly

i a lloph :- ilo r i n r," )l'b i ,: a ; hi e pIH 7.0, the

ri',- ction c i t- b: I is il lustraicd !h i portsnce of

p',., ph ia il 1on in lade. aly -icid An:ioso' s,

Adosois typically have thick dark A horizons hii.h iae hii l in

organic matter. Org--.ic mrtier accu;iuleatlin takes place rapidly after

new additions or volcanic ash and is exceptionally resistant to bic-

logical deccon,;,osition. A series of .Jpanese paplars in the i,-rly and

mid 1960'5s concerned with the relationship bet.c.en a lophian and

d-cor positionally resistant organic matter of Andosols.

Early paper by Kosaka (19f3), Adachi (-i963), and Tan (1i9l +)f!

I ndoo sia vcr- concurred vwi th humi ficat ion s cci's and hur:ic-fulv c

acid conLIents 'rcm a soi clasi Fication viewpoint. A series of aj.b-

licaLions then fol 'ou.ed on specific allophan-organic reactions.

Aoi ino and Ko'byushi (196!! 1965 ,)) found .hat a:lophane inhibited

the activi ty of protease, cC- cnd -a,rylases, and cellulase enzymes.

Aliophaaie sere -d to adsorb enzymes at both positive and negative charge

si tes.

Polyphncrols '.re found to undergo rapid oxidative changes in the

preseaice -f a; lophane according to Kymia and Kawaguchi (1964). Non-

bi:loi!ical c~taly'ic oxidatoin appeared to be responsible for the

changes. These polyphencls could be cdsorbed by ailopiane and calaly-

tically e,;d!ied to give dark-colored, highly acidic polymerized

produciLs similar to humic substances, regardless of the acid environ-

ment and low base status of the soiI. Tannin-rich qguercus sspcies are

ofien aound anong the vegetation on Andoso!s.

Kobo d Fujisawa (196) c]ai,'ed that ai lophane showed preferen-

tial adsorption for different types of hod;ic acids. Toki;L o, r end annoo

(1i65a, 1S65b) reported that in Japan, Andosols ha -e u 1 ic/

fulvic acid ratios than older Andosols indicating that the concentra-

tion of hunic acids, the alteration of fulvic acids to ht:nic acids,

and the carbon content of the so is increased wi th developsren of

pedogenetic processes, Based on their analytical data, they c!aii.iod

that most of the hurus jas bound with allophane and sesculoxi'des,

especially aluminum. Humic acids were characterized by high darbor

contents and high C/H, C/N ratios while fulvic acids had low carbon

contents and C/H ratios.

According to Tan (1966), Al and/or Fe will migrate to hori-

zons depending on cho humic/fulvic ratio, In Latosols, A! appeared to

be mobilized due to the large amounts of fulvic acids pree-.t whi le Al

and Fe were both robile in Andosols due to the presence of both fulvic

and humic acids.

Pedoggentic ,ea l then g

Under well-draired, i.oist conditions, the weathering of volcanic

ash may lead to formation of i3iy different secondary minerals. Besoain

(1969o) claimed that under humid conditions accompanied by good drainage

conditions a mineralogical sequence front allophane to knolinitic type

r!inerals alnmot invariably occurred ji ih Li:e. The incidence of allo-

phan- and/or kaolinite type clay ,,i. :- a s in arid envi ronmeii ts i s rare.

i'ieldes (1955) stated that andesitic volcanic ash developed in

time, che folliowinq series: Allophane (6, A3, A) --- halloyi tc/

me tahal ioysi te --' aoi i ni te.

3esoain (1969b) suggested that the inclusion of kaolinite was a


logical bu' LIeo. Lical ai sibilily si; ce e, a d soils con-

h l loys i e.

Fi eI :s (l 5) o tlii d A b ril strucLur-l col ti u rii r rhe

.Iove-r ti ...... 1h riirg e .e;'a : ( ) allop hnn> 0g in which .r ';oI

;ilica ,i3 di crc tnd cl y partitcl s ul ra fine, (L) Allophaie A in

whi h alumii'li and i I iica v 'wre randomly colii')ined and particle si ze was

greater, and (c) nctaihalloysi e in which alumina and silica are coiabirned

in a kaolin structure.

ilay (1560), on the island of St. Vincent, 8.W.I., reported that

andesitic ash originally consisting largely of glass and the minerals

anoirhite labradorite, hypersthene, augite, and olivine had altered

to hydrated halloysile, allophane, and hydrated ferric oxide in 4,000


For Ha'.:aiin Andosols, Swindnle and Sheri an (1964) indicated that

in the early stages glass weathers to yellow, brown, and orange

palagoniteC (Ca, Mg, and K containing amorphous alumi no-silicate) due

probably to hydroxylation and physical breakdown. They said that glass

contained much strain energy; consequently expansion due to hydroxyla-

tion increased the strain, and physical comminution occurred as a

strain-relieving mechanism. Some removal of alkali and alkaiine-earth

caLions undoubtedly occurred simultaneously with hydroxylation.

Ol.nrasa (1964) outllined what he conceived as the weathering

s'.a:ue1nco of volcanic ash in Japan: (a) leaching caused immediate

neutrality of acidic new ash, (b) large amount of bases are ieieased

to a great depth and the ash becomes alkaline, (c) freed bases are

i .h ,d quickly along with Si due to the alkaline or basic condi tions,

S h S i s s a ,d --iLl- s occt (d) and Al are w-e-k y i'5incd

,n t C._ b a-inr in" but later 3a1 i;: c j;-aduaily forms, ia d (e) .ilopine

gradually crystal ies to I-yl'ra d halloysite ind, in soie cases,

gihbsi re.

Fields (10G6) co: silered structural randomness in hydros al: Liino-

silicates as being highly significant in the formation of ailop't ne.

He outlined four possible mechanisms of al !ophane formation in soils:

(a) weathering of basic silicate minerals dissolution of discrete

Si tetrahedra occurs under local alkaline condi tions with subsequent

reprecipitation with Si and Al ions under more acid pH after leaching

of basic cations, (b) weathering of glasses direct formation from

glass, (c) weathering of feldspars the ability of feldspars to aiter

to sericite apparently depends on the proportions and ordering of Al

and Si in 4 coordination. Thus in suitably ordered fe!dp.-rs,

single chains wi th Al -- Si order identical with feluspar ca.: detach

from (100) and (010) and polymerize to produce mica sheets.

Feldspars in ,which A! and Si have disorda,-d arra-gements such trans-

formations are not possible, so that sericitisation cannot occur and

a!umino-si licate weatheling products of such feldsparts appear as

allophane. Fields claimed that this process occurs during initial

weathering of feldspars from iNew Zealand volcanic ash, and (d) grinding

of minerals.

Kanno et al. (1968) claimed that aliophanization of glasses

proceeds through desilication, hydration, and replacement of Si-O-Si

li-ikaes by Si-O0-Ai linkages.

a3iN 3 et al. (1368) recognil.'ed a weathering sequence from ;luophane

to halloysite i ia Coioibian Andosol, while Luna (1969a) also recogni,-:d

alloohane A ar-d 8 In Colombia.


3r-main (1 ', ) (U .co pr a '- ,r, .parl p a 2 a "dolil'"

C'iil ; :l' ; ic ,--. A idools caLS-in

"i! o.ol i '" 3 I :I, n i p arc, .pci clay r A cLi on r ,IJ ih L its

occorr cce i f ro! I / I'ltei to i the :i pn -,nt oi Fi i : re 2o ordered

Utili gI i ; n ied in this litera ure review, V ii

l01 ph< (19/1) repoiLed ihat im bers of a recent amorphous a!uiiio-,

silicate w\orkhlop in Japan agreed to the Foolloini tentative defini tion

of ill ophane:

"Allcphanes are rm i:L:s of a series of naura lly occurring minerals

which are hydroja -l.i]:,'' i lic-ites of widely varying chemical con-

position, characterized by short range order, by the presence of :i-0--Al

bonds, and by a differential thermal analysis curve displaying a low

temperature endotLherm and a high temperature exot'ierm vli th no inter-

med Lte endo Lthrr."


Site SLi- action

Soil profile s ies were selected by investigitic,, of -ivi hle

iarial photographs in Bogota and an initial land recoirai sanca of the

Narino area about one month before actual fe!ld work co,:e-ccj, It

was decided, after observing ;Nari ,u Ai.dosous occurring ':iaely bet'lcn

2,000 m and 3,500 m in al titude, that meaningful pedogen:tic data

Lould be obtained by profile selection in an a!t tidinal sequence.

Profi le locations in the Departuento of Narino are shown in Fiy.

i. Profiles !io. 1-5 were selected at altitudes between 2,000 m and

3,200 m within a 25 iki radius of Volcano Galera (Pasto) in order to

ai low tihi assunprtion oi ash-proi ncc uni formity. Profi le 6, represent-

r, ..he higl.est altitude investigated at 3,510 m, ,'as near Vc!cano

Cjti!al about 80 kra southwest of Galerss, SiSusti tuition of tiis si e

--s necessary because chick ash Jeposits do not occur above 3,500 n

on the flan'ks of ial eras. An assu'mptior was made that ash nincralogy

,.'as noi o\e,- dissinmlar in 1hea Cumbal region due Lo predominance of

-.jsi te tholrougihcut t.eo uplards of Marno. These six piofi ls ,.sre

r-cside-ed Lo Ieprc-senr app,-o:.'iiately the maxi u,! ringe cr weat a :iering

s',e i-ty exe-ri ,i-ced by loessial pyroclastic deposits in tie a;tiplano

or lr-no. Factors supporting this ssurrption will be discussed :n a

later section.

Profit i 7 was SrTpLJ 57 kic, nr, c::heas t Of ; iers in the vicinity

o, Volcano Dona Juare at 2,G003 a for comparison purposes with profile 1,



V.G T, l s, M6 1
Io o," Caho 1e

Fij. 1.De partame to ofr Nriio I]ocatd ip sout!.este rn
'Col biAe, Smil] inst numbers Adic-ate loca-
tions of prof; es 1-10.
tions of pro. Fles 1-10.


i 3-'ofi 83 s'pled in a lJarg valley I(Sbana de Gunchcal: )

pro i 6 ; : e pirpos cf cr.r; n ininI the : n lu nce ci ..jlcinic

activist on all u lo- -l u rin's s il ;iLhi n th s udy ar .a, Pro miles

3 and 10, altihoioc'i .ro'hlc o ic-illy similar ] o Andosols, W:epre Found by

h!F field 'est to be lo:, in all phane. Obviously, these soils nr rited

further ir. t i on,

Field Proc-dJres

All soil profiles :a.-r sampled 7rcm road cut'. Descr options ,ere

a.ado and samples cotalned dfter rriosal of app;oxirately 50 to 75 cm

of soil iate'ial] paralla to the profile face. Profile descriptions

for all i tes were mrade in accordance wi h criteria described ii, t:ha

Sol; Survey iMarlual (Soil Survey Staff, 1951).

3u k saimpes (approxi.;:ael one-ha3 f gal I ) '.*"-e co l i-::ted from

each horizon ad replaced in plastic bags. Undi iturbd ccre sa:; les,

epproxiia Le ly 8 cm in diamicicer, .:sere col.l ?c'd fior.I s'eicted hSori ons

of most ioi ls for ,ulk density mieaj-ji events.

N-atrai ly oriented clods were col lected f o.l ; ele) t major lo'-"

zons of prof'Ies i through 6 for Cthn Jsetion analysis. These speci-

nc!Ts ',,er- placed in labellpd ne'tl Lins .i ;h ',I c ri ietln.aion mr'rk-d o0

the containers. Sjlops wr re determined by use. o. an Abey level.

tPrttcle Siz Dilst.ibuJtiolu

Partici C si.e l e.te rination.Ca .as b a .odli fiid hydi-c.etet :eth'od

adcpei fr, :m ouyoucos (1926). Organic atteta, ,J:hcn pr-es. t, wtcs

p* ,c 0ed prior t o rialys' s usii;, h/ydr-9n per; d Thf. d i

3 -7. u p-IK pL '- i I i i c ilfc!r lat i on

'j t U. S. D ,- r t .; ,,1 2 ,- i, 1, cl. s i' ca lilonl
( I I l i r C it.

,ii DIe ity

T io s a ,dificn tio o, f : o a r l outlined

y 31 k ( 19 '). S i c coar s s ci! ton con t' n L s vi tir al ] noi-

eisLcnIt in h' i!,':ino Andosols, the core rrii'.od was col-siderilcd to be

a sai ai act or :.' i: iq ic for bulk densi ty du; rmi nation.

Fif oitn-lo sOsre ict-or Conteit

This dcteri.lj ira tio in .ws nade according to the r tr-iod ouc'iined by

the U. S. S.liniLy Laboratory Staff (1951).

Chemic a M methods

. '.~. g:j ,':_i tat cis t

A soi -I:ol tion ratio of !:1 in distil ld water and, in 'os0 cases,

I N ,CI -as i.a srAd by a ecklan Mot;del G ipl meter.

Cat.ion nExa' ."ntr P C. '

This paF-ar' ir is dctrni ined after the Nrthed of Bower et al.

(i:, 2) usingj I l ven ,;it niul, ,s.eti:te at pH- /.0 as leeching ac enr,

Thies ccr,,,!nts ve-e da 'tr'.i-ined '.pa -ait ely using the EDTiA method

oi: darros,, aind iri'pso (i962). Tenl giram of air-dry soil was leacihd

with 0i 0 in of ',IAc. c at ptH 7,0 as described in Pech et iL. (19t7).

An aliquot of the N l,OAc leachate used for C--Hg determination

was .- ayzed for petass i I: aing a Coleman Jr. Spectrophoc e.- a

770 m)l.

Ex chan e ble SodL -im

An aliu ot or the NH4OAc Ilachate used ror Ca-Mg d'torf ,innation

vcas analyzed f sr sodii'l using a Col.: ,ian Jr. Spec trophotoaiTiecr at

590 miJ.

Organic Carbon

This was de ermined by the Walkley aid Black (1934)) method.

Elxchageabhe Al'.rmi num

Exchangeable aluminum was determined by ili fluoride tiLration

method of Yuan (1959).

Free Iron Oxides

Free Fe203 was analyz.J d by a technique reported by Ching ~a al.

(1i9 3).

Minerulonical Methods

Disrersicn and Fractionation of Clav

Portions of the so l samp es in plastic bS: .ra .'einoved a. d

allowed to air-dry, -hile the remainder was i-ai stained at epproxi atiely

Field r oist re. This .-as ado.e, because An:dusols, upco drying, oItun

form sand andn sil I pse,'domophs o f c!ay particles (Birre i, i964). This

oulid decrease tremindoansiy the a61ount of ci iy that might be dispersed

( right, 19 6LS ).

Duplicate 5 cj9 saples of soi I fromin each horr'zoi. ,'ere piaccr in

130 ml poiyi thylene centrifuge tu aes A; r--dry surface sa.:p leas r;t.'

i l I l.p i b blh ouE ,. t o ro i'g t:c r :d to r etain

'ici1nt ii ,s ri err e Aible i and At' ;' t io r,

Iu U', i, o !,zi ors '. k ac !< t? ppr I" y ) icd .0i :> ure si nce

in most c Is s s i :. irJjcr ibe in t e 28 ;d C h ori ns, Organic

a. t ,U 1 i ; I 'or sur face' 'or i ,.ois by ;i (r th [ t oI L vk' I i ch

)d in, ,s (19 u) t i'S o j codiul- hypochiloi it e solutionl adj ,st d ,. pH y.5

by 1.0 N HCi I i; iately prior to use.

T1n m ili liitc s of NTOC solution (mini r um 6% uvai labie chlorine)

wa!s ad1d'd to each noi [ ia1le in the polyethyle.,e Lubes. This fwas

repeated fou.- to five times, hating the suspension in a boi ling water

baih [ r 15 minutes during each treat, !ent. The supernatant Ic-

canted after centrifugation. Surface horizons treated in this manner

t.are qui te v.ll dispersed after completion of ianOCI oxidation.

Consicdrable amounts of aluminum oxalate and fertic oxalaite com-

plexes in peroxide extracts of soil clays have been reported by Fr.-nler

an!d Miltchell (1963), especially in soils containing a higqi amount of

nallphanic colloids. Since Andosois contain large quantitri s of

1 loph.ane, whhich is quite v-! cablee to common pretreat-;unts such as

organic matter oxidation and free ilroi oxide removal, it would ba

cdvantosrous to :.i i, i~iz possible pretreaLimenL -efr ccts. Therefore,

iron oxides rcc.-,' not -enmo,/ed ard NaOCI rather than Hi22 was used oa an

oxidiz'ing agent for orin c matter.

Subsurfo.ic s-.iples .cere di:sl:ersed with 2% Nl.2CO3. Approximately

5 9 of roist 0sil P;as placed in 100 ,1 polyethylene centrifuge tubes,

Twent mi ii i tit ers of 2^, Na2CO3 solution was then added to the soil

samples ard the s,'spcnsion ',as stirred for 2 minutes. The less than

2 p frtaction was rtl!on scpara ed by centri fugi ng as described by

thittig ( 9J5) o. :ently, the less t 2 2 p 'etion :s ,-rated

into 2 0.2 ~ and less than 0.? 9 i frac ,. s by centri-. i 'uj i.

X-K- Di ffraction

Clay suspeqsions representing the 2 0.2 pj nd less i0a:. 0.2 i

fractions placed on cermic tiies after the ,.:thoui r.', ;:;. and

Diamond (1956) as modified with the use of a suction apparatus de-

scribed by Rich (1969). The initial treatment pr;or to irradiation

was saturation by r-agnssiLun (I T.SCC2). The tile with plated seriple

was subsequently irradiated, glycerol so!vated, irradiated, washed \iLh

methanol and water, saturated with potassium (I N KCI), irradifaed,

heated to 50CC tor Ii hours,and then irradiated once more. In some

cases, the sample was hated to 100C for 12 hours after KCI satura-

tion and irradiated prior co 500C treatment.

Obtaining and storing the aforementioned oriented clay specimens

presented somn problems especially for the fine clay ( <0. 2) samples.

There was a te.defncy for the fine clay to be dra'.:n through the cel,-.iic

tiies under vacuum, This was alleviated by concentrat:n9 and flcccuiat-

ing the suspension .i th 1 TJ MgCi2 prior to applying it ro the tile.

hininmum water washirgs, usually equivalent to volume of llgC1, used

vuere necessary; otherwise, the problem arose again, Crackin g -;nd

fiaking of the fine clay speciiens of the tiles, even in a 100C glycerol

ai.,,splphere, Is:o created difficulties s. Thi: was ce,rnered in ,ost cases

by Lho-ough satur ation ith a 30% jiycerol solution dcra'.i through the

clay under sucticn,

Potassium-saturated samples ;:e re irradiated irmediately a ter

treatment with I N KCI. Anticipl.ting the cracking a-d flaking bronighc

on by dehydration and dehydroxylation of allophane at temperatures

v, r , d I i r o .' J d'y ik ..! c .ile

1 c. Wi h a very feI ,:C i ns f' or '0s Clay

of l 1 cCs for to rse cl y (0.2 2 ), i't iipos-

t u L i t t I for irradia. icn witho t losing all of the

Sle ii, i i'ic l iy 3si ci r i.cvenlid by i thr of '.o rietiods

'liic' s c orl : iually :ll. Double stick Lransp i
i: i Ald in < glass slide and LIhen pressad onto the dry, loose clay

filkc s lying on 'he ceramic tile. The glass slid' could then be

placed in the X-Irny isam path. The second rcethod wms som0iihat io:0i

ti e consuiiing. Ic involved transferring the clay From the tile to

a, agate mortar and pestle. The sample ,as then gently ground in a

srcall .niousit ovf wi'er Intil miost of the aggregates were coirpie'ily

broken. At thi s Lir,: the suspension was transferred witih dropper

to a glass slide end placed in an oven at 100 for about 2 horse.

Aftar drying. the clay was irradiated.

X-ray d i'fraclegrams '.?re obtained utilizing a Ceneral Electric

XRD-700 insLruci nt. Radiation was detected with a proportional co.ntlr.

i19-snturated tpcci':ens were X-rayed from 3 to 320 (2 0). IMg-saturaLcd-

glycciol-solvatad, 1,-sa tu.-ated, 1OOC and 500C treated specimens were

irradiated frocn 3 co 150 (2 f). Instrument conditions and settings

were as follows:

iadiation: Copper K alpha.

Fi ter: Ni cke I

Potential: 25 KV.

Current: 25 rpa.

Gonioaeter scan speed: 20 (2 0) per minuLe.

St.ip-chart drive speed: 30 s per hour.

i.ulse h.t selector: L =

Scale e:'pans7on: 600.

,ro oF Fset: 20.

Plot: Linear.

'oC-ii '/i -1 li L: U.' .

Serene El its: io.

C'ot:ir high voltage: 2.25 KV

A,|ii fiber ratemetor: Time constant 2.5 seconds, range

2 5 ;mal fier gain 16 low pps.

TaLe--of angie: 30

i fferentTal Therrm'l Anilyvsi

Portion; of the stc,'ed fine clay and coarse clay suspension were

transferred to aluminum neighirg pans, dried at 'OC, and subsequently

gently pulverized in an agate mortar and pestle. A 100-mg portion of

the dried sample .as then transferred to on;e of the four sample positions

in the insert no d.:r of the Deltatherrm tnermoanalyzer where it was

gettly taped into place around a plati 'um t'ermocouple folloIrg

improved packing prroecdures as outlined 'y Re-eau and Carlisle (1970)

wherein the clay sample was "sandwirchei" between asbehcos which had

been fired at 1000C a'd passed through a h"frr--,n ill fitted with a fine

screen. The sample holder c.ntal'ing the four clay samples with alter-

nating reference sample posiLions was then ccrnected to a transistorized

circutl oeTeremi nations were made at 25" sensitivity using a heating

rate of 10C pur minute fro:c r,'oro temperature approximatelyy 23t1 to


nfrred! soc-trc" rtr Aa sjs

!n'f i bn : o I.,aL i c b, c' y 1 n rin is *s investigat.d 'i th

1 10 -c i rho.o ar. es -sturatred .;s were

S'OC 1.5 erloi ulv'rized u irn an agae -ortar and pestle.

A m'xture consisting of 1.0 rj of the ground racnjs um-saturated clay

and aJpioxirmatiely 'lCO ng of spn:troscni c-grad KBr was used to form a

itr.nparent Fellet. The pellet was formed using a Reckman K ,r die.

Aproxiieately 200 ig of the KBr-clay mixture was placed in a Wig L -

Bug electric mortar and pestle (i,:ae by Crescent Dental Mfg. Co., 1839-

45 South Pul;ki Road, Chicage Z3, 111.) to reduce the particle size

and thoroughly disperse the sample in KBr. The optimum grinding time

appeared to be app.-oxiiately 20 scconids. The Beckman KBr die consisted

of an anvil and a ram constructed in -uch a manner that when pressed

together by a hydraulic press, the sample is compressed into a disc by

the confined space. A partial vacuum was created in this space by a

vacuum pump connected to a hose coupling on the die which led to the

aforementioned confined space. This tended to remove entrapped moisture

and air .-h;ch othernvi-c coull result in a cloudy to opaque pellet.

After cleaning all parts of the die with acetone, it was assembled

arid the 200 mg of Krr-clay mixture was transferred from the Wig L -

Bug mortar and pestle to the die. The assembly was then placed in a

hydraulic press and the vacuum pump activated. After 1 minute, slight

pr.assure was applied to the ram for 3 minutes. Hydraulic pressure was

then c.-ased to a reading of 8,CO0 psi and Iaintained for 5 to 10

:min utes. The vacuum was then released after which the hydraulic pressure

;as gradually released.

The pellet was removed with tweaze rs and placed in a Beck!ian


Infrared sample holder. It was then introduced to the infrared beam

'or an-l yis. The scan speed was 15 minutes from 4,000 to 350 c'1

The instru.eiL was calibrated by scannigr. a st'edard polystyrene film

from 4,000 to 350 cm-1. A KBr blank and Ward's halloysite No. 12 were

scanned as reference samples.

Sodium Fluoride Test for Allophane

Fieldes and Perrott (1966) indicated that large quantities of

hydroxyls were released when allophana was treated with NaF. The

magnitude of release was great relative to other soil constituents.

The aforementioned authors outlined certain modifications of this

knowledge which allowed a positive field and laboratory test for allo-


This procedure was used for field work in Colombia by placing a

small portion of soil on phenclphthalein paper and treating it with

I N laF. The appearance of red color was accepted as an indication of

an appreciable allophane content. In the laboratory, pH values were

obtained by suspending 1 g of soil in 50 ml of 1 N NaF solution. The

suspension was stirred constantly for two minutes after which a read-

;ng was Immediately taken on a pll meter. If pHi values were greater

than 9.4 after t,'o minutes, the sample was assumed to have an allophane

content equal to or greater than 60% of the colloidal cc.mplex.

Luna (1569b) pointed out, however, that this test has been found

to be positive for certain Colombian soils in the presence of apprecia-

ble ha!loysite 'nd aibbsite.

Sand ili nra-logy

Sand mineralcgy \,as determined for the 500 50 j) fraction sepa-

ra I ro- L: I Ii ey ic.l3

::sity u ing q .

1r, 1 1 i ,, roin. ,inre or

C p- t. > r I i u- ,t OIpa uC

in e c'?o of h-vy minerals.


'1 sis. Li jht J he vy ii i ra1s

I-o r,.i (s. g. 2.9)., /evy inernl

re i'cntifi i d wi Lh a 1 lari i ig

: /,;r,.] ,l :n i Irobenzri nc (n 1: I .552).

bi is of speci. s pr' 'ent in 100

miciner;' were abulat d epraritely

Mljor hor ions from profiles no. 1 through no. 6 wcrc: selected

for ricromorpholcgicil investigations,

V r i- ical y-ori '.ntod undisturbed peds iwere i; loved froim thie me tal

count i; her's use. in the tie ld and p lced in si.aldl 'luminu!i con l.t iners

approximately 4 ci: in dianimeer and 8 cm high i-d allowed to air-dry

for a few' days. Dry peds remaining in alumin um containers vecre then

placed in a vacuum -Iesiccator, An impregnation solution consisiingj of

a mixturee of 50% L .nirac r-esn /4116, 4: rronor-er'Tc sty-one, and 1%

Lupa sol DDM hardener (Pettry, 1969) w-'s th-n periodically introduced

to oacn contaiolLr ov'/ir a period of: 4 to h-ours unidJ- vacuum. Sialiies

erre then reinmved from the vacuum and Dpl-3:e. in an initially cold oven

.i,.] hated sloiwi to 10CC for approximat ly i2 hours. After cooling,

Lhe Alulminum ccnolainer ;as pecled off and the samip!es wire ready for

se5i,: on i nr .

Erypccdtisn of Thina Sec'ion

Thle initial step r;uas to bioc'k our a sect.ioi: 2.5 x 2.0 c.-t btwcen

0., to 1.2 cm thick using a Felker Di-Met, ,:odel 1i -R sna' i th a diamond

jlede and .'ter as a coolant. The slab obtained in this manner was

then lap-ground on ore side using aVon Hulen Thin Section Grinder.

This waas done in three steps, s arting with! coarse 150-mesh silicon

carbiJe on the rough grinding 1ip using Water as a lubricant. This

operation uas then followed by lap grinding on the inLter,-ediate lap

with 320-mesh silicon carbide. Final grinding was done on a 15 cm

agate lap with #1200 optical finishing powder. The last two operations

used kerosene as a lubricating agent.

The smoothly ground surface was then cemented to a 2.5 A 7.5 cm

petrographic slide using an epoxy cementing agent. The excess saeple

was then removed to a thickness of about 90 i above the glass slide.

This was accomplished using a specially constructed slide holder vhich

allowed a parallel cut to be made with the diamond saw. Further reduc-

tion of thickness rias obtained by intermediate grind with 320-mesh

abrasive on the cast iron lap. This process was continued until bire-

fringent colors of second order blue for highest orientation of quartz

and feldspar were obtained. This indicated a thickirss of about '60 J.

Final grinding was necessary to reduce the specimen to about 30 p in

thickness. This was done on the agace lap with #1200 aluminum oxide

optical finishing ponder. At the desired thickness, the highest bire-

fringent colors of quartz and feldspar were light yellow or w.hite. The

slide was then thoroughly /washed, and a cover glass was mounted over

the thin section with epoxy cement (n = 1.58).

Anay' sg

Thin section slides were described using a petrographic microscope.

Photo icrographs were obtained wiith a 35 rnin camera attached to the

microscope. Thin section descriptions were made using the terminology

of Bre' (IS64).

S .i,'lI '- 1 i STUiY A FA

ILo ; ion

Thli Cap r e- nto of ik -rilo is o3 cated in th' c txtre:,e soi.ithl.,as torn

pFr of Cole rbi S< uth Arci ica (Fig. 2). Nari o is bourdd o1 t-he

ror:-h by t'e Dep5 rtla ,lo of Cauca, the Pacific Ocean on the ,,est, the

DeplrL't. ,,L o of Putom;yo on the east and Ecuador on ith south, it has

iian -r of 32,70' km2 (Banco de la Repub ica, 1959) and !ies fro: less

than I to slightly more than 20 nori' of the equator.

rE si/ cg'PJ.

Miari o is ch1 aci eri ze b by .levationsi ragi ng fror.i sea 1 vel a lng

its Pacific Ocen lshol%' i nei to n miaxili m of 1,830 im at the snow covered

su. ,. t of volcanoo Cil bc!. near Tu;iIerres i tih' Andean uplands. The

teinnina! ranges o: the Ai;ne dividai' in the northern part of Narir.o i -

to distinct t chains called "cordill!eras" rd extend almost i the

At!ntic Oce in (or iC'riihbcn Sea). This junction imparts a very ruggd,

mountainouss characLtr to the castin part of Lhe departamc.]to. The

An <'. n -i fhli hands .-r co ns i tu d by i tera!ti it y i toLrmontai e *,al ieys

and ridges. Probably the most striking feature of the uplands is the

--b.nldanc of volcenoe., alasco (I1;6) reported that 2, volcarces habe

Le. iJ '.ti fied in iNar;7o of which 5 are knov.n to be active (Cumbai,

Ch! l-s, G;]a res Az:ifral ~nd Dona Juara).

nt.i;::i;nLarne sirs of cons de'rable exCent iccur frequently in


0' s^s

Fig. 2.--Coiombia, South America, shoving location of
the Departanento of Nari- o.

ha 1 s v. t h

hills whichh

T-.pic-l of C- is a ]Iar I1' e r Tlque-rres

c;iy it i)lf la r. ea .r i.,c vao'o. ,rl (Chiles,

Az r'o j l) rF r 'I e w s -frn p ripl.i y,, of the Anres. ThIu

is 1ppo 'i; tely 3,000 m in at i Ltd and is occ(ujpid by

S fi t .strii. A 'i M i er ba sin eJi ts east of Volc.:no

U ti tuJ3 of 2,530 inm

st,.?r part of Narino is co-priscd of Andean piedmont foot-

;.rg ra her rapidly into the Pacific coastal plain,

Be rock and Surficial Geology

Lana (lSjna) indicated that Andean basement rock composition in

Narino cwas primarily andesite, basalt, and tuff of the Cenozoic.

V.rcla (1933) concluded that recent eruptive rocks due to volcanic

activi ty of the Cenozoic domindtrd a rmljo part of Nlarino.

Luna (i969a) stated that volcanism tu.dides and pyroclastic nmatrial

inventories wver quite limited in Colcnobia. He reported that rost

references were Lo soccific OruplionS .,iLn insubstantial details on

quantity and comrpsition of pyroclastic materialss A maximum age of

7,000 years by C 1 has been reported for an ash 1dposit in .he Departa-

mento of Toi ma.

lEipirical observaLions by the author in the Nariino area revealed

that the highJ! -n' have teen universally affected by aeolian and fluvial

l' rocl asti deposit ts; t'ocever, the influence of this factor on soi I

ldevelcpi.ment varied considerably in magnitude. volcanic ash 'ccumula-

tion ar:d di; tribhion in the Nari-o highlands has been r mandifested as a

'"p ch.:orl'" or discontinuous p.tern over the laodscape. S-:vecal basic hia. pro:'ab!l contributed to this pattern, De to .iic strongly/

dissected nature of the region,considerable ash has been eroded from

the steeper slopes. Differential deposition due to a-berant int r-

,ontane wind pattLrns has, u-doubtedly, been a supplsnaintal factor.

Additionally, it was very difficult to assess the territorial limit of

various ash deposits. Wrinh) (1964) indicated t'at mapping of ash-

derived soils is a probleri in peripheral areas at the ou:er territorial

confines of ash-mantled landscapes. Figure 3 shows the general distri -

bution of volcanic ash deposits associated with Volcano Galeras in

addition to locations of profiles 1-6.

Areas in the Mari-o higniands \rhich ;ere covered by thick ash

deposits characteristically exhibited smooth, rounded land f orm. which

generally conformed to pre-e"i tin surface configurations.

Nature of the pyroclastic deposits changed considerably from

northeast to southv.-est in the highlands. Deposits in the Pasto-C'.:-saco

region (Fig. 2) were composed of finely corTiinuted ardesitic ash ex-

hibiting little or no stratification. The character of these duposits

indicated gradual accretion of ash where eruptions were infrequent and

not of great magnitude. The contesting nature of pyrcclastic deposits

in tne Tuquerres Ipiales region (Fig. 2) was r,.anifested in stratified

beds of ash and lapilli. Lapillitic strata ,.era quite often pumiceous

in nature.

Thick wall-stratified and well-sorted fluviatile pumice beds were

observed '!ong the western walls of the previously mentioned Tuquerres

basin. Similar pumice beds were observed along drainage outlets to-

wards the Pacific lowlands and I k- north of the Ecuadorian border

(Fig. 4).

S Salers ,



a .f/r:

Oio-8 :ii j OF nIlAn'i,1 SLIBla

S-in 21 0 ci

dl?:itU (m ml

Fig, 3.--Cross-section of !ari n Colombia study -ea showing locations
of profiles I-6 relative to Volcano Caleras. Stippled araas
sho5: ;U >ei;zed disLribution of volcanic ash.


Fig. 4.--Pumice beds I km north of
Ecuadorian border.

S i ty oi i c i i o I p r ,, .ri o

t .'c. x::s" a ri l e rath r i i .i

,"J/0 I! oi s of !o!lcano tl rnIsi i stc red 0n n I r .j o annual

precipita lon of 700 n'm and a a verae te Q era tJr'; of 131 At the

lotonai -gricll Iur l station (2,/50 m), thl e i ver e annual rainf ia l and

.tmp-ratuce .,re 781 mm and 13C, respectively. A station near Consaca

(1 40 rim) in :he subtropical zone had an average annual precipitation

of 1,9,2 mn and temperature of 20C. Tumaco, on the Pacific coast,

reported 2,819 nm of rainfall per year and 26C annual temperature,

The co;,,lcx topography and wide altitudinal range of Narino

i.,p rt considierble variety to local climatic conditions. Pro;xi i y

:o i-e lqunor, however, dictates an isothermic temperature rcegi,-a

which is c!os-lyc related to elevation. Luna (1969a) stated thar. oi--

plarature is related to elevation in Colombia by a 6C dccrase for each

i ,CG m i ncr ,c-s, in altitude. Espinai and Montenegro (19e2) re?.o.' zed

five i3,ajor la-inperai:ure zones in Narino based on the aforel-'n tionc

Ther..l *riadient (Table 1). These zones ranged ''oil t;op;cil (> 2-C)

in ie FU r c fc lovlarids to sub-ioipine (3-SC) in t:-e Ar'-daln highlands.

lti tlin each tleriail region, subdivisions w re r ade on t'>c; basis of

p-ci pi :t tion di lfcrcr;cs. Precipitation is maxi,tus.i alorg sIopes F. .'i n9

;ie Pac;iic io',-lanids a.d tend. to decrease It th depch in i ite&:-. Laie

v!illeys. Varele (G93) reported a disi.;'ct drc/ s aon occutri rg; during

jl.iy, August, a-id Septeriter in the Pasto-fio 1:y1, s- :ior oF N.-roi 'I .

Table Cl i mate-vegeta 'on rel ci onsii ps in Do rt
Colo bia (Espina! and iontoneg'aro, 1352)


Temperature (C)

500 -- 2,000

9 SA 3,500 +

500 !,000 hf M
1,000 2,000 vhf M
> 2,000 pf M

500 1,000 df L'
1,000 2,000 hf LI
2,000 4,000 vhf LI
>4,000 pf LI

500 1,000 df S1
1,000 2,000 hf S1
2,000 4,000 vhf S1
> 4,000 pf S1

500 1,000
1,000 2,000
2,000 4,000
4,000 8,000
> 8,000

3,000 3,500

2,000 3,000

800 2,000

T 3 800

= grassland ("paramo")
= pluvial Forest
= very humid forest
= humid forest
= dry forest
= very Jry forest

SA = sub-alpine
M = montane
LM = lower montane
ST = sub-tropical
T = tropical


9 12

12 18

18 24


A3proxi mate
Rai nfa l I (mli) V g La i on A I ti _uiue (m)

i na| I i' ro (1962) us th s of .yo l ;, (;-I)

ipr Ci i ". C r c vir ft 'i'o i n C )cb i I.

U 1vill ia3 i/ i C 1 c uI C isp i ioa1 trr r ll-

So";h ,:., y i dd" i ;d 23 differ,
indi nons s io OD-prt i .onto Nri no (Table I). The Nlriro st,dy area

-rcom ,ssad 5 majorr v2jetation formations rangiiig from dry sub-tropical

forest to sub- pinr grassland. These pertinent formations ire discussed

below as outliin a by Espinal and Montenegro (1962).

ry: yLTripi ca I Forest

This Formition is found predominantly in intermontane valle-s and

is characterized by 18-24C annual temperature and 500 1,000 nmm annual

precipitation. Rainfall deficiency in these valleys is most likely due

to orographic effects which produce aberrant local wind patterns. The

inajor part of this formation occurs in mountainous terrain with roderate

to very steep slopes. The original vegetation has been almost eitt:rely

altered by man. The lone Narino plant species list published by Espinal

and :iontenrgro (1962) was enumerated in the Guaitara river valley for

this formation. They reported that the area uas covered with dense

clusters of a trce ca led Cocaa' (E rthroxylon sp.) associatLc with:

Crtoln. fr-l~ ie.s H.B.K,. (nosquero),
fa .a- 23:cl t, ? (tacishuo) ,
C~li and, a sp, (carbonero),
A clj.a f n--si t i l id. (pe ),
a." na- c lese s H.B.K, (vrenturosa),
lporbi a cj-aaca:eni 8Sos (Ichero),
antisa sp. (tuna) ,
C-'' lo_ reu. sp. ( c e i on),
TJ.c3P_, FJuss. (chi r lobi rlo),
aursra tc., -eiceca Fr. aed !1. t i arraco ,
-C_ s-l i L:ts : (ti ntilla), and
Ve rbenaceae (pendo).

In Nari o, areas of dry sub-tropicai forest are used for the

cultivation of coffee, suigr cane, plantain, a,:d in sore cases exiensi c

grasslands for cattle production.

Drr' Lower Montane Forest

In gener al this vegetation Formation has as climatic linl- s an

annual temperature between 12C to 18C and annual rainfall bt!e-en 500 mm

and 1,000 mm. Dry lower montane forests are generally encounteiud be-

tween altitudes of 2,000 m and 3,000 m in interimontane valleys and

basins of the Andes. The primary vegetation of this formation has been

completely altered and destroyed by man. Due probably to an ideal

climate, indigenous pre-Colombians populated a large part of these

areas. In establishing their agricultural systems they initiated the

demise of the natural plant communities of this ecosystem. Areas which

are left uncultivated are invaded by grasses and small trees or i!hrubs

such as Cordia sp.

Some common species identified in this formation are as follows:

Dodonaea viscosa,
Solanium maria natum,
Euhor bia orbiculata,
Opuntia sp.,
S particum jsceum,
Phyl lanthus salviaefoliu s HB.K.,
Bacchars sp,,
Croton sp., and
Prunus capuii.

Large tracts of this formation have been reforested with eucalyptus

(Eucalyptus lobhuus) and pi-. (Pinus radiate). This vegetation forma-

tion has traditioeall, supported a high population concentration in

the Andean highlands. Ory lower Ircntane forest alees are considered

to be some of the mosi productive regions in the country. Within these

areas are located large cattle ranches and commercial forests. Intcn-

si v, c iuc ti' of v 'at, corn, pott tes, and tree frui ts is -ncouiraced

in io /-G I r irrigation systems.

MI 1n t o' r o n r Fi, ost

I is v g,!t tioni fo irmation is of I ir i ted occirr'r ic in Col .' bi' .

fhe r,:'-ratur r hiJ al i tude raJge in this 'orma ion is sirii'ar to

that of the dry lover i,:ontane forest system; however, avcragq: annual

rainfall is higher ranging between 1,000 mm and 2,000 rim. Cliniat is

generally mild with gentle rains occurring throughout the year,

Occasional frosts occur at the higher elevations of this formation.

These forest formations occur on undulating to moderately steep moun-

tain slopes and intermontane valleys and basins.

In areas where the population has been established for several

centuries the natural vegetation has been considerably altered.

Some common species identified in this formation are as follo'-s:

< Crcus sp.,
Ced re I a
R-.pina sp.,
Ficus sp.,
eLin^annia sp. ,
Cro ton,
LPhy11anthus, and

In areas here the topography is flat to undulating, crops such

as potatoes are grown. IMore strongly sloping areas have generally been

esabli sherd as pastures composed mostly of kikuyu grass (Pennisetum

rlanderstinu) Pine and eucalyptus reforestation has occurred in many

miarginal agricultural areas.

iumid io!itarle Forus t

This formation occurs generally between elevations of 3,000 m and

3,500 m in the Andes although locally it ,ay extend in sowe cases up

to 4,00) m. It has been referred to as "sub-para ,o" by natives. The
/ A
Tuquerres basin in Harino is corposed almost entirely of this vegetative

formation. Average temperature generally fills between 9C and 120.

Precipi Lation ranges betv een 500 r" and 1,000 mm annually. Even though

rainfall is relatively low, the climate is quite humid due to low

thermal efficiency that is associated with considerable cloud cover

and high altitudes. This is reflected in low evapotranspiration rates.

Day temperatures are generally quite cool and nights are frequently

below OC with common frost and ice formation.

Humid montane forests occur on varied topography in the Andean

highlands. Considerable areas of the formation have been sacrificed to

cultivation;howeverremnant trees and on occasion complete con',unity

remnants have remained intact. A partial species list of a humid

montane forest community in the Departmento of Boyaca is presented as

fol lows:

Wei nmannia sp.,
Polylepi s p.,
Viburnium sp.,
,e ;ari s sp.,
Seneclo sp.,
Vaccinium sp.,
Rubus 5p.,
Orthrosanthus sp.,
Buddleia sp.,
Bona rea sp.,
Espletiva sp., and
Mi conia sp.

Population density is not great at this altitude althc9gh the

soils in tnis formation are quite fertile and have sustained cultiva-

tion for many years. Potatoes, wheat, barley, and vegetables are grown

on the flat and undulating areas while the sloping regions are generally

utilized for pasture.

01 *'1"t.i~ P cu pios th, hi

I;' t .-n l i S I i is :not axt

c- tur l i ,3or :c Col','biaJ st'

s iici ily ,id h e to th ir i

cl i i to. R i, iif i :y ranqe froS r 5i

an n al t~'nper-l t iur bet:,v.c n 3C a dd 9(

St!3cs., especially F stuca and C_1I

~ad shi ubs (Fig. 5) such as:

Di.plo t.ejhi u revolutum Blake,

S ciioa s ccinioides,
V -l rimn s arborea,
icL-i i a buxi folia,
P Jiyl/eji_s acea is,
S. .- sp. and
I sp,

gl st ?.l.vo,! , s in -he Ai dl s bef} ;

ensive and is of very little ijrl--

b-alpine ecosyste-s have i:':n in-

nacce ss) ility I-rd iniospl : bi

00 mm to 2,000 : jn annual ly with an

C. The veoytation is dominatO d by

Ingrostis, interspersed vwtth bushes

Fig. 5.--Typical vegetation of sub-alpine
grassland near profile 6. Broad
leaf plants are Esp el i sp.




ro nic Mlatter 0 Oxidi tion

Lavkulich and iens (13/0) p esented datia ccarli-y howing that

not only was NcOCI r.ore efficient in oxidizing organic matter !but ihat

the content of Si, Mn, Fe, and Al in extracts after treatments was

signi f'ic-ntly higher for H2 2. Comparative data for selected Andosol

horizons using H202 and NaOCI treatments (Table 2) were even more

driaaatic than those presented by Lavkulich and Wiens. H202 removed

five to ten times the amount of Si and ten to seventy times the r.:ount

of Al than that removed by NaOCl. The H202 treatment was the sarm as

th t reported as H202 I by Lavkulich and Wiens (1970).

NaCCl would appear to be less destructive of sesquioxides, silica

coatings, allophane, and crystalline clay minerals. It can be con-

cluded from these data that NaOCI was a more acceptable reagent than

H,20 for rapid routine organic matter destruction in Colombian Andocols

prior to nineralogical analyses.

lsPersc ion of Andosnl ClIs

iL would be aiJvan : jt us at this poinE Lo cx;
d;spersion properties of soi Is conai ning high qlluai i te s of al ophane.

Ai thoughh Lhe intent of this tiudy was not: ie thlod oriented, an interest-

irg point has a risen. -iistori ; iIy most soil mini erc logy analyses have

been performed on air-dry samples. As has been mentioned previously,


Table 2. Fe, Sioand Al amounts in ettrarts after :r: ant of certain
Andosol horizons by IaOCI or H202


Profile ?1o. Horizon Fe Si Al Fe Si A
Sppm -- -

3 A12 2 240 0 0.2 902 88
2 240 10 0 1,001 99

5 Al 5 230 30 36 1,738 1,705
6 270 30 35 1,914 1,760

5 AC 6 390 30 117 2,500 2,079
6 360 30 118 2,519 2,211

6 Al 64 330 170 164 4,025 6,105
73 350 220 116 3,630 5,830


h, nh-e "'-n enc cs in AnJoloi cly dispr'sicn due to ir-

Ir-....i c t. It is :, p l it 'i riors iive hocn

J reki nj-p th3 r s. A yriad of toch-

ni.,; J 1-'v b):?iL :i, l, ; th c; xyin3'x g o. O success (or faii Irc).

Ni'y e L ase .:iho
Dii i 1 i r .oc1 r c ac i J (see, for ex- pl e, 3irrell I : Fi eles, 1952)

his been used by cmr,/ workers Fields (1955) and Fieldes aind Taylor

(1361) dispersed Andosols, wi th varying success, by adjusting the pH

to 10.5 with H.1-. Kobo and Oba (1965) used a sonic vibration method

for disintegrating aggregates to facilitate mechanical analysis of

volcanic ash soils. They reported that in most cases the clay was well

dispersed by Calgon after sonic treatment but addition of dilute HCI was

still necessary for subsoils dominated by allophane.

In most cases mentioned above, dispersion was obtained by some-

w.t'hli drastic treatments with little consideration for allophane. Ahmad

and Prashad (1970) until recently had been impeded in their investiga-

tions of Andosols of the West Indies because they found it impossible

to disperse these soils even by the aforementioned "conventional"

methods. They discovered that zirconium nitrate (Zr(N03)4 5 51120)

solutions were very efficient soil dispersing agents, especially for

Andosols. They postulated that the high ionic charge, small atomic

radius, and low ionization potential of Zr apparently resulted in

saturation of the exchange sites and some tetrahedral substitution for

Al leading to a net positive charge on the exchange complex. Achmad and

'rashad felt that their data were the most reiiaBile so far quoted for

tnese soils. Unfortunately, zirconiume nitrate does not lend itself to

use ;s a dispersing ;cgent in mi rer.a og;cal i nvestig atios, Due most

i kely to its o', ioniza tion potential, it focis prec;pi tacc~ with most

of the accessory solutions such as i;gOc, r.qCl2, and acetone which are

used for satiratfon and washing of cl-e colloidal complex in cly miir r-

alogy studies.

It was observed by the author that seemingly good production of

clay (i.e. good dispersion) jas obtained front surface horizons using

NaOCI and from field moist subsurface horizons using 7i Na2CO Zir-

conium nitrate also seemed to provide adequate dispersion so it was

decided to verify qualitative observations using zirconium nitrate as

the best available standard. Sample dispersion with zirconium nitrate

was according to the method utilized by Ahmad and Prashad (1970).

Samples were centrifuged, clay decanted, flocculated accordingg to

Whittig, 1965), and weighed. Silt was separated from sand by wet

sieving through a 300-mesh screen and each separate was weighed. Data

in Tablc 3 clearly indicate that in most cases there was no striking

difference between dispersive quality of zirconium nitrate and either

NaCCl or Na2CO In 50% of the cases zirconium nitrate gave slightly

better clay production. Seventy-five percent of the time zirconium

nitrate dispersed samples, in which organic matter was removed by

11202, gave slightly better clay production than NaOCl. Three out of

the four horizons dispersed with Na2CO3 produced better than zirconiun


So it would appear that for Andosols of the humid subtropical to

sub-alpine uplands of southern Colombia the maintenance of field-mcist

condition assured adequate dispersion. it also sceeis that even though

thesee soils contain high quantities of allophane the principle of sodium

saturation of the exchange complex cad resultant development of strong

electrical repulsion forces remain valid.


Th,)ble 3. i'f, t: f trit Ln ad, disi s in t i Li cl s i /.e
I i : ic

Oi Dv i por-
Rrij 1 .,.ov l s i on

C-- 2 All H202 a
NaOCl b

C-2 2 A 2 H202 a

C-3 2 A3 H202 a
iaOCI b

C-4 2 [B] a

C-5 2 C a

C-15 3 C a

C-l7 5 AC H202 a

C-34 4 C a

- Organic matter removal unnecessary.

a Zirconium nitrate,

b Sodium hypochlorite.

c Sodium carbonate.

Pro i o
Lab, o. ;o.

Sind Silt Clay

39 '3 32
38 28 34

40 30 30
39 34 27

34 39 27
31 L6 23

37 40 23
35 39 26

29 42 29
29 40 3i

22 40 38
24 46 30

39 34 27
33 41 26

25 48 27
22 48 30

Detai led field descriptions for the 10 oil- elecued in this in-

vestigation are presented in Appendix Table 8. The discLs' on of

morphological and subsequent data wil relate initially to p-ofi ls 1-

which we're selected in an altitudinal sequence between 2,000 and

3,510 m. Profiles 7-10 will be treated individually as ancillary sites

providing additional pedogenetic information.

All soils were well drained and quite permeable. The altitudinal

sequence of profiles 1-6 exhibited definite morphological trends with

decreasing temperature. Horizon sequence at low and middle altitudes

(2,000 3,000 m) was AEB] C; however, above 3,000 m the carbic sub-

surface horizon was absent imparting an AC sequence to these soils.

Surface (A) horizon thickness tended to increase with increasing alti-

tude. Surface color was dark grayish-brown (IOYR 3/2) in profile 1,

but all subsequent surface horizons at greater elevations were black

(IOYR 2/1). The [B] horizon thickness tended to decrease with elevation

and was replaced by an AC transitional horizon in profile 5 (3,120 m).

Cambic horizon colors displayed maximum development (7.5YR 4/4) at

2,000 m. The EB] horizon color at 2,150 m (profile 2) was yellowish-

brown (I0YR 5/S);and with decreasing temperature,both value and chroma

of this horizon tended to decrease. The C horizon colors tended to

have similar hue and value but slightly lower chroma than the overlying

[5] horizon. Str.ucture or A horizons tended to be subcng'-lar blocky

IBrackets F] are used to indicate a caenb;c diagnostic subsurface
horizon. Right (i164) advocated the ses of this symbool in describing
Andoso s.

hb lo. 2,,U') i o I C. ul 1a ib ,v'- is I vatioi. C..,1 bic pr I horizon

S ; s ,are 1- ,'.i < blo y. a joriity of the soils, C

he I r1 r, c I s si ve zxce t at l Ow r a 1 tii tu the e,

.a o jLcrncy s.o r-s the lsl ian ular blo k 1 y typn. Notable .was thc

1.'k o i i-. t d clay iargill n -) on p-"1 urfce's. Oro nic coltin. i s

, a cor aily :otad on individual sand grain surfaces in A ior ;iois of

ill soils (Figures 6 and 7).

Pr of l 7 u3s a shallow Andosoi at 2,000 m in eIevation. The A

horizon was unusually dark (IOYR 2/1) for similar soils (i.e. piofi le I).

Profi le 8 was sampled and described in an alluvio--lacustrine region.

It appeared to be a young, well drained soil and showed little evidence

of profile development. The most striking feature of this profile was

a layer of white (IOYR 8/1) silt at 131 cm. It is possible that this

Iayer represented relatively unweathered fluvial volcanic ash deposits.

There was a pronounced change in depositional history below 152 cn.

Profile 9 had a thick (66 cm) black granular to blocky Al horizon.

Colors of the C horizon were redder in hue (5YR and 7.5YR) then was

typical for Andosols at comparable depths in the study area. The C2

horizon was probably developed from andesite. Although morphologically

similar to Andosols this profile exhibited negative NaF test results

for all ho,-izons. Profile 10 was similar to profile 9 but it was more

strongly dIveloped. The A horizon thickness was less (39 cm) and not

's dark (10YR 3/2). It was the only profile with an argi llc horizon.

Although aeolian pyroclastic materials have undoubtedly influenced Lhe

pedogenesis or this soil it has developed predominantly in andesitic

residuum. According to Wright (1964) the thick dark A horizons ex-

hibited by these soils have sone connection wi ih primi tive iwes.lesring

Profile 1

Profile 2

Profi l 4

1-4 repre-enting a i i i

P.-ofi le 3

Fio. 5--Co1or photographs of profiles
sequence of 2,C00 2,730 m.



le 5 ir 6

Frof i le 8 rufi ~e 9

F;'. 7,--Coior photographs of prcfili s 5 ard 6 reprcsrniin al tit ud i al
sequence o- 3,120 3,510 -. ad cirillary profiles 8 and 9
at 3,020 and 3,06.' Im.

products of volcanic glass. T- :' ,o Andcsol A Iorizons r~~: prol'- ly

zhe result of organo-metallic (.n"i,p;- nic) complexes encouraged by

occasional uniform depositions cf fresh volcanic ash. Increasing thick-

ness of A horizons with altitude was probably due to decreasing bio-

logical and/or oxidative decomposition due to decrease in temperature.

Proximity to volcanic ash source was most likely an additional thick-

ness-encouraging factor.

Genesis of the [B] horizons was somewhat more difficult to ascertain.

The soils are quite permeable but in their natural state do not have

well developed ped faces. Due to relatively uniform rainfall the soils

appear to remain moist throughout the year. Year-round moist condition

is encouraged also by high water retention capacities. These factors

tend to discourage desiccation fractures. Lack of layer-lattice sili-

cates whose plate-like morphology expedites orientation on available

surfaces is an additional feature contributing to the lack of B horizon


Overall development of profiles 1-6 seems to have been highly

temperature-dependent. Decreasing potential evapotranspiration with

increasing altitude may have discouraged profile development at higher


Chemical Properties

Soil Reaction

Soil pH measurements in water and I i KCI solution are recorded in

Table 4 with the exception that neutral salt pH determinations were

not made for profiles 9 end 10, In all horizons ji' profiles 1- p:

values hihber when .e-3sur-,d in afterr -tan wen .casured ;n 1 '

-1 ~ ~ ~ C 00' -to c 3u7Cr-mi~

40M0 ~r
O . .

--41440 -0404







-n Ln oD o o~
ccr 00cc cc\ .

-0 o- -t mt um-T

Ln i: LA tr ii Ln A

.-4 0 -- -r- io r
cr. L1j **^ (

0 0 r- o r--- m- n
\ o i'^o r,% d.o ur r0-- c -> -.J t n

-o o o o 00 00

o -: TNN -T -- -Ct(

'.o. ml i, .- .
S oo. ooo0 o o oom o
S'4- 00 0 '- 00000 4- 0000

_4zr ooocc- o-o-

rocin- occccoM oooo

cco 4O 00-0- 00-0








" _


1 CILi. 04.o C Lc.0CO 041 O 0 m oo. -0

0 0 d0 -cOdd --0 4 -O0dd
o -- . . 0 . . | . .
--o....... 4 i 0 -00-

.j- o0 <0-0' -0 c.- o0 o -- ,4 ', o r--. io

SAdd-000 -0000 4 000

un. )O i0)00 .- -t 0)-N

0 43 0 -- N

10 ^fjo40) ~t -0J0

0 cI U 0000 '1 00000 . 0 000

S -oo00 00000 0 000

I. I. . .1 m .

o -N20 )0 N 040) 0)00)
u0 I .m . 4 . .
_i> I r 0)0)040 0)0 0 -

0 -

01 04N1 -41 O --1.

S100 1 1 1 1 -s
~ ~ ~ ) \ m -j- h-c o oc r-- ~ l\ r~c

~J '


Or -'. '.-' t-j- N CO -'i
0--- 1 o r N- -. rr, r^, r- r- N o o r-.,o


0 o Ln --OO N 6-- 0 c0
CNI >0 0 -Zrr N N -^-^- m*2ccONo(

S,0 co- r --o ,io -co -. o-l

ooo Q.I -c-N- c o\o
II 0-0 0000 0010

ou I, . .-0 .N. . .

IU) --t o -20020 0 '.l G '.0~Nr'
S0 N N

I I' 4 o o -
(0- N- Nfl2 r- *- (N --Nt20

0 Ie '- 000 C- 0000 IIo 00000

000 4f-0NN

S-Nt 021N- r02 N-.Z-Nrc

Ic --N2 cm- O\cOc


-U -- 204) u\ oco-4 -

4- --(Ni--C- --<<<-2
' O~-- ~ C I- << << <- ce ^U O
(-1 0 -

I-- z[

a 2


m Lt-

u ^l






O ((

a, Lo 00
N-2dL 0 Oh-3

-0 L -- o a-

uo o o

- N (N c1
0000- 0

-a,--( -40


, 00 ao 00

0 IO i 0 0 n

-,l iti n. C li isl- (I. ) ,' ,- r i 'nlt neutral ,t pH 'i,. l s ir..present

i i ty in !i ) sdil d t:' ,fore l,:uld l a ,jctel .o in 'lr.

il rI fiction values r c iar, ab ly uni forn for profit les f-6 with

j i' of 01 5.1 to 6.0 in wai er and 4.2 to 5.3 in KCI. Even Lhough tle

i-; i i:deJ of vnriabi li ty hardly cnconipised one pH uni t, certain trends

or, ti noticeable. The pH values tended to increase with depth in tl-.u

ploFile and to decrease with increasing altitude.

r i':i c Carbon

Organic matter is formed in the biological decomposition of plant

ard anlial residues. Narifio Andosols had unusually high organic carbon

conients as shown in Table 4. Organic carbon contents of Al horizons

of profiles 1-6 ranged from a low of 2.9% at 2,000 m to a maximum of

oq.L in profile 3 at 2,440 m. Values tended to decrease with depth and

to increase for comparable depths below the Al horizon with increasing

ali tude.

Profile 7 exhibited lowest relative organic carbon contents of all

profiles investigated. Profiles 8-10 had similar A horizon organic

carbon contents;however,for profile 8 these values remained relatively

hish with depth.

.Lha eable Bases

Eci lr.neable calcium values geqera i' decre e -ed .ith depth in the

pronfil and ,i th increasing altitude for comparable depths in pilfi es

:-b. Treirds .e re mr're variable for profiles 7--0, i'xcha geable Ca in-

cri',:?Sd wirh depth for profi l-s ard 8; remained reF!mii-ely constant

For prcfile 9; and cec ea-.ed with Idepth to rch C horizonn for profile 10.

Corn isLent tr-ends '.re ;. l di -o avrd hy r chc 'y-,o abI e M s. dist,'ibu; tioas


(Table 4) for pio, iles 1-6. M.axium exchangeable Mg encountered

in profile 1 (?.,COO m). Exchargeable leMg increased with depth in

profile 7 but decreased with depth in profile 8 where maximum values

(6,1 9.1 meq/100 g) for all profit les investigated were encouniteed.

Exchangeable Mg values were variable with depth in profiles 9 and 10.

With the exception of profile 3 exchangeable K contents tended to

decrease with depth for profiles 1-6. There was a general tendency for

exchangeable K to decrease with increasing elevation for these profiles.

Maximum exchangeable K contents for all profiles investigated occurred

in profiles 8-10.

Exchangeable Na tended to decrease with increasing altitude for

profiles 1-6. No noticeable trends were observed within individual

profiles. There was some tendency for exchangeable Na to increase

with depth in profiles 7-10.

With the exception of profile 3 total exchangeable bases for

profiles 1-6 tended to decrease with depth in the profi le;although in

some cases such as profiles 2 and 5,there was a slight increase in sub-

surface horizons. Total bases increased with depth in profile 7 and

were maximum for all profiles investigated in profile 8, Total ex-

changeable base contents of profiles 9 and 10 were similar in magnitude

to those of profile I.

Exchanqeable Aluminum

Exchangeable aluminin values for profile 1-6 tended to increase

wi th increasing elevation and decrease with profile delth again wi th

the exception of profile 3 Exchangeable Al distribution was erratic

with depth in pr-oaile 7. Profiles 8 and 9 were quite le: in exchange-

able Al in comparison to all other profiles investigated. Profile 10

Si, i it d a r in fch Ie

r: r Ci' C??'iLy.

CI ion change cpaci ty vailucs w,;re va-ri l'e and did ot display

ti ..! i h profile depth or altitude. Profiles 7-10 did h1ive slightly

io.:r C vC values than those exhibited by profiles 1-6. CI.C values

-a.-ged from a low of 14.9 meq/100 g in a buried C horizon of profile

6 to a raximun of 65.7 meq/100 g in the Al horizon of profile 2.

Fr c.e 1ron Oxides

Free iron oxides tended to increase with depth in all profiles

and generally decreased with increasing altitude for profiles 1-6.

rMximum percent Fe203 for profile 10 coincided with the B2t horizon.

todinjiu Fluoride Test for Allophane

Sodium fluoride reaction values in Table 4 indicate that all

horizons of profiles 1-6 were dominated by allophane as evidenced hy

values greater than pH 9.4. These values ranged from the minimum of

p! 9,.4 in the Al and [Bjhorizons of profile I to a maximum of p'l 11.4

in the Al horizon of profile 3,

lanetic J nterre-ati;onlshsps of Chemical Preperties

The aik of picfi ;e development and disti'ibution with d
::-st chemical p rca.,.-ters indicate t',it profi les 1-6 were ycuihul

soils. The A horizGins of thesee s.ils -Aith their cark colors due to

high organic carbon contents provided visible evidence of profile cif-

ferntiuation. Acc!u:;ulation uf -:rganic carbon especially in the surface

horizons of profiles I-6 was most litcy encouraged by a high amount

ani of 1,.43 i: q/./100n n in n J

of biologic a i cc,'i ty and the presence of hi ghly rea' ,.i ve i a- 1 ic

-'rfaices in rth firn of allop:rns and oiher ar.orphous col lics. Tii

relative. ly Ic;' org ,!-ic carbon content of the Al horizon of prof!l I due mainly to t'e oxidation enhancement of hig'-er temp ie ure-s.

SOganic collo;d translocation was noted as evidenced by the rei Lively

high ioranic carbon contents in lower horizons.

Base saturation and pH values were generally unrelated for profiles

1-7. Interpretation of CEC and base saturation data was confounded by

the presence of allophane which has a highly pH-dependent exchange

capacity. Cation exchange capacities for these soils were determined

r;: h i-14HOAc at pH 7.0 and consequently were high due to the inclusion

of a considerable amount of variable charge. Cation exchange capacities

deaLrmined at lower pH's or at the prevailing field pH of the soil

generally give a more realistic picture of the actual charge participiat-

i:n in exchange reactions under natural conditions. The magnitude of

permanent charge would be expected to vary according to the degree of

tetrahedral substitution of aluminum for silica. The amount and crysi.-

aliin:ty of allophane present in addition to the presence of crystalline

colloid.- species with low pH dependent charges would considerably

influence cation exchange capacity measucremnts /t pH /,0. As wiold

be expected chi aforementioned factors vary from horizon to horizon

and fro., soil to soil. Co.-squeijntly base saturat:on value h-ve dubious

inter.prc active -orth Ju to a changing cation exchange capacity base.

Tfie cnt;-ibutioi of organic matter, especially in surface hori,.rs, to

confounding ilterpietation of these data cannot be ignored.

The distribution of free i on oxides with profile depth indicates

that Fa has bean relea. d from primary minerals and cranslocatcd by


percolatinig wat-r to lcwer horizons. Choluviation has most likely leen

r o, .e ri :, ech-i s of ir.n trr n oc i n in dd'Jition to

r-djctio" ,nd sub a jent lnecdhi .

/I lepha.t ccn n;i; as slo;n by N-iF Le t pH vai.s wer, "eater

at :.isher altitides for 1-6. The tendency for th':e values

to decrease with depth in the profile indicates that allcphane has

taken.on a more crystalline character and/or there has been an increase

in crystalline phyllosilicates. Either factor or the combination of

both i.p1lies a dccrc3sed capacity to release hydroxyls upon treatment

with sodium flucride solution.

Dhvsical Properties

Particle size distrib-'tion data for profiles 1-10 are presented in

Table 5. Several horizons in profiles 2, 3, 5, 6, and 7 refused to

disperse in sodium hexraetaphosphate. The. A12 and A3 horizons of

profile 2 and the AC horizon of profile 5 were dispersed with NaOCl.

The [B] and C horizons of profile 2 and the C horizon of profile 3 were

dispersed with N2CCO Clay distributions with depth for profiles 1-9

indicate no evidence of downward movement and accumauation in lower

horizons. A maximum of 483 clay (<2 p) was noted for the BI horizon

of profile 10 which did not coincide with the argiilic horizon as

determined in the f'led. Fine clay (< 0.2 ;j) conrtints are irosi likely

greater in the B2t horizon d';e to maximum deve!opiisnt of clay skins

(argillans) in this horizon. ii ller (1965) reported that the major

potion cf argillans in a Canif;id silt 3lan were composed of < 0.2 i

clay. This conclusion was confirmed by Caii-o''n (i68) on similar soils

in northeastern Ohio.

Tale 5. Phy ical properties of s lce, d Colt -ian soils

P, r ticl SiU a D;s : ut bu 1o;I
Sand Silt Ciay
2 .05 .05 .002 <.002 15-bar 3uik
Horizon Depth ------------- ,"-------------- H20 -osi iy

Profile I

Al 0-45 27
[8] 45-73 37
CI 73-121 25
C2 121-151 59
C3 151+ 15

All 0-41 45
A12 41-58 39
A3 59-70 31
Le] 70-96 35
C 96-160 29

All 0-50 *
A12 50-100
[C] 100-120
C 1201 27

Al 0-68 29
A3 68-88 28
183 83-115 26
S15+ 31

Ai 0-55 *
AC 55-69 33
C1 69-94 4G
C2 94-139 30
i C3 139+

39 34 25.) 0.93
38 25 31.1 0.80
44 31 30.7 0.72
21 20 0.90
32 53

Profile 2

30 25 29.0 0.44
34 27 26.2 0.71
46 231 29.4
39 26 38.2 0.49
40 31 34.7 0.64

Profile 3

33.2 0.44
32.7 0.54
42 312 33.0 0.64

Pro-i Ie 4

34 37 14.2 0.84
33 39 11.2
41 33 11.7 0.81
1,3 26 13.0 0.60

Pi ofi le .

30.9 0.49
41 261 22.5 0.49
30 30 11.9
34 36 22.3 0.45

2 .O5
,-i; : n De th ----.--


Al 0-50 *
A3 50-67 43
IIC 67+ 25




05 .002 <.002
-- --la~ ;-----------


Profi le


Piofile 8


Prof i le


Profi le 10



Sample wouldn't disperse nor determined.

1 Dispersed with NaOCI.

2 Field moist soil dispersed with Pa2dCO.,

I IAl b
I IClb
I I Alh



10. L


Bil k
O]nsi ty





A I 2







iulk densi y values for p ofilas 1-6 varied from a lo; of 0.44 /,:c

:.r the Ai Ihorizon of profile 2 and the Al horizon of profile 3 to

rs'i mum of 1.22 9/cc for the I rlb horizon of profile 6. 3ulk .nsit ies

,;are characteristically low. as uxp'-cted for soi is developed fron aolian

pyroclastic materials. 'he lo! e exception of the i lClb '-ori.-cn n n.ioned

above was due to a high sand'content consisting of vitric c~r:c'rs and

other coarse-textired pyroclastic components.

Fifteen-bar water contents were typically high for profiles 1-6

reflecting the voicanic origin of their parent materials. They ranged

from a low of 10.4% in the lidb horizon of profile 6 to a high of 38.2%

in B horizon of profile 2.

Low bulk density values were a reflection of the exceptionally

high porosity of Andosols. High 15-bar H20 contents were a manifesta-

tion of the affinity of allophane for water.

Sand Mi neralogv

The primary mineral content of the 50 500 }j sard fraction rep-

resents the original mineralogical nature of the volcanic ash parent

material as modified by pddcgenetic processes.

Li!ht Miinearals

The distribution of light minerals n the 1 500 pu fraction is

given in Table S. Quartz was a minor light mineral constituent in all

profiles investigaled. Feldspar jas the dcmineni species observed

generally followed in abLndance by volcanic glass. We3atheiing products

represented difficulty identifiable, higher, ai tcred minerals which may

have been derived from biolt ce and/or Feld.s ar (Luia, 970).

Quartz contents generally displayed erratic distribution with

T e 6. 1." ( O- I .) I .1 I of i l cL,! Colom'bian soil

tor i O : Pir

Al 12 /17
[]1 15 33
Cl 11 83
C2 7 55
C3 8 78

All S9 45
A12 5 60
A3 2 63
[B]1 19 60
C 16 64

All 4 52
AI2 3 66
C3] 7 30
C 4 68

Al 2 53
A3 3 40
[3] 3 57
C 10 75

AI tr 57
AC 3 64
Cl 1 47
C2 5 45
11C3 6 62

_-!. ;(,;L;1LIs IJQ 5t. q rivc. s . ..- .
'Wa th- Rock
Ic nic crri g frog- Other
311i products ,e its Siort-i mineral

Priofle I_

5 36
Z cr
Lr 6 tr
13 tr 20
10 4

Prori le 2

Profit e 3

Pro:i ie 4

Profi 1|

27 13
17 12
22 29
13 37
8 24

tr tr




Table 6. Cont;Inu:d

-He- v rnirn als iin O0 .. tr,.iLparent '_grains
Horn- horn- Hyper-
blende blended A.: ite st tene Zircon Biotice Altered Opaque

Profile 1

95 tr 5 tr tr 17
100 tr tr 3V
96 tr 2 tr 2 12
100 tr r r 12
85 4 tr 9 2 79

Profile 2

87 tr 5 8 25
87 5 8 tr 21
89 4 7 24
88 2 6 2 2 I1
88 6 tr 4 2 17

Prof I le 3

86 3 8 3 3
85 6 9 tr tr 14
66 2 10 19 tr 3 !4
47 cr 10 37 tr 6 22

Profile 4

71 2 7 16 tr 4 23
67 2 13 15 3 15
62 18 18 tr r 23
89 2 4 3 4 32

Profi e 5

87 tr 4 9 tr 12
76 8 16 tr li
64 17 19 tr tr 13
It 2 19 5 0
58 18 18 2 4 6

1!;1 Co.i IJ

"ri.1 O1 ri Cptr

Al 3 36
As 5 32
I IAlb 1 29
lICl 8 31
I IAlb 5 26

AI 5 68
A3 2 80
iIC 17 76

All 9 29
A12 6 38
A13 7 45
AI4 7 33
IIC tr 25

Al i tr 52
A12 55
A3 52
Cl 2 i4
C2 53

Ai ;9
AI2 30
81 tr 77

i;.3 ot de .ermined.

rr Trace.

-None present.

Ia in 100 or, ills

V ic<.c e ing frig- Other
I ass pr lcii c ,ts 5 t ii to minerals

Proii le 6

iPr- i le i .

Prefi le 3

Profi !e 9

P;ofile 10



tr tr
2 2



D0 rND

Table 6. ConLinued

Sr i.neral s in 100 LrnLpare2ntari n ____
Horn- horn- lHyer-
blonde blende Augite scihne Zircon bioti e Altered Opaque

Prof; e 6

42 24 32 2 8
44 tr 18 32 6 0
67 16 9 8 9
73 10 15 tr tr 2 16
44 tr 33 23 -

Profi e 7

87 tr 5 8 tr 4 tr i9
95 tr 3 tr tr 2 tr 36
93 3 2 2 tr tr 47

Profile 8

88 tr 7 tr 3 2 7
77 tr 11 3 6 3 3
85 12 tr 3 6
81 6 8 tr 5 7
80- 6 5 3

Profile 9

15 21 28 -6 9
53 20 27 tr tr 18
33 3 17 33 12 22
32 4 18 34 2 7 1I
16 tr 10 40 I

Profile 10

96 tr 4 -
78 tr 8 12 tr 2 10
86 2 5 7 2 tr 14
94 2 tr 2 14
97 tr tr 3 -
85 2 9 t r 54

I; h profit' 1- I1 pro file i h r, ihi s it,i r l sI eci'es

.1o . '' re ci ''ol ie lir'" n cy

ior qnIrtz c crL ,:h i n i r n j i a l de. QuIirt content

Oxs oe< r v L i 1 ilS 9 d cho upper hor i :o is of

I'o: ssi : ; d ppar s aid th plogiocl s series r. re not Ji Fforen-

Li, :d biut cor: i cd a-nd reported as fldspars. Feldspar ,as present

in grcjt sa ai 'Onts ,^nerally at lower altitudes for profit !es 1-6.

Fi ldspar disL ibutions within these profiles were ro.ostly erratic how-

cver in profiles 1-4 hhoween 2,600 m and 2,730 m there was sonle Lenden-

,y for feldspar to increase wirh depth. Feldspar content of profile 7

iwa shri lr to that of profile 1. Tle content of this species in

profile 8 paralleled that of profit c 6. Profile 9 exhibited uniform

fold- par distribution with the exception of a discontinuity at the Cl

hori zon.

Volcanic iass dncreased w\i th profile depth and increased with

incre-iinj altitude in profiles 1- Profile 6 was highest in class

co-ntnt but did not show trenJas ii th dJpl-h due to deposi tional strati-

fication and possibly t.o rcJ:ced weathering Tntensity at the 3,510 m

eieva tion.

jiy 4_Mi neral s

Distributloii o heavy mi; e rai in the 50 00L ;i fraction is shown

;n Table 6. I;-;o blonde w,;as t!-e dominant individual mineral species

for profiles 1-6. The pyroxene group, of which augite and hypersthene

.are rost co.mion species, doT;inated the A ,, As, nd il horizons of

profit e 6. Distribu lio of ih-;e i:e vy lircreol suite in profile 6

supports field identification of lithologic discontinuities. These

data suggest t!ha, the I Alb horizon was developed in n.aterials miner-

alogica ly simi lar to the A1 an-d s horizons. Ti- I !Alb and !I b

horizons probably represent pyroclastic depositions eitther from a dif-

ferent source or variation in roij n cl9-nabr com .Pos i ion from i:le share

soul Ce.

HornbT'nde tLcindeds to idecr; e i :h pro file I :ph ar:d ; crei sing

elevation in profils -6. rrofile cxhiibi ted lcwer hornblende con-

tents in contrast to profiles 7, 8, and 10. T-ne pyroxene group

dominated all horizons of profile 9 and hypersthene as an individual

species was essentially equal in quantity to hornblende in the A3, Cl,

and C2 horizons.

Zircon, oxyhornblende, and biotite eore observed occasionally and

only in minor quantities.

Opaque minerals were tabulated separately and were mainly rep-

reFented by magnetite and iron oxide nodules.

Genetic Relations of Sand Mineral Distributions

Abunaance of hornhblende, particularly in the heavy mineral fraction,

is indicative of pa-ent materials in their initial weathering cycle and

of youthful sofIs developinrq in these materials,. MInor quantities of

quartz in :he liqht: i-eral fraction provide additional evidence for

this conclusion. The tendency of hornblende to increase at the expense

of hypersthene and ugi te vi th decreasing elevation indicates that

pedoligic sic ithering has proccceded at a more rap;d pace at lower altitudes.

C ja. Mi ralo0

a ii ati I e es

l;l ::ir 'is be identified through Lhe diffraction of X-rays

Jue to a sivii lit-; between interplanar distance (d-spacing) end wave-

leogrh. The characteristic d-spacing of a specific clay minerall may

Ie czlclted a-,curding to Bregg's law (n = 2d sin 6) since X

(ua'l enr1th) and the anyle e are known. The d-spacing corresionds to

the pln:.e iperpendicu-ar to the c-axis of layer silicates.

T'ie d-spaciny of specific clay minerals will vary predictably ac-

ccrdinj to cation saturation, interlayer liquid (organic molecules)

sorbed on clay surfoccs and the solution from i.,\ch the suspended clay dried (Jackson, 19 ).

Hydrated halloysite was identified by first order spacing of

C1.OA which coill.-ped to 7.41 7.20A after heating a ll-saturated

soeclren to 100C for 24 hours. Metahal loysite was suggested after

Mg-s-turatioe by a /.25A flrst order peak. Vermicalite was concluded

to bo present by the !I-saturation 14A !SA spacing and a 10.0A -

10.2A spacing after hoatint to 00C. Illite was identified by the

presistence of a l.'. pek for all treatments. Figure 3 demonstrates

the technique utii zed to identify vermiculite, i lite, hydrated

Khlloysite, and metahliloysite. The diffractio pattern for the Mg-

saturacd corps cocr ey (0,2 2 r) from horizon A3 of profit e 7 exhibits

pea's at 14.2A, 9.93A, and 7.25A -hi h car be attrlibu'ec tc vermiculite,

Si te--hydrateoi :illoys ite, and r.:etahalloysite, respectively. Heac1ring

of this spec men to 100C for 24 hours re',,ea id J fi action peaks at

appr-oximately the same pic i ions. However it can be assume" Tiat the


Fig. S.--X-ray diffrdccion patterns of mg-saturated
;,i c dry, K-saturatedj,iOOC, ardi K-satuateJ
5CCC treated coarse clay of hcriz'n A3 in
profile 7 demonstrating identification of
!eLart- loysi te, hydrated halloysi t-, i l i te.
and vermiculite by changes in d-spacings.


So. n -h n .1 ,I ria d lloysitae -nd tfha

S* t i c e rn r a ir i rn L'. of c t ,l oy-

i ; i co fIor ihe I u C Iiffr:ction ni tern i figiire 8

S i 1r -. i ri 1' T i Yy ; cr. /, 25A ,-ik an; rccioju iiin ensi y of

S 10, C I 1 I, 1"k T s .ll i A. IA k.-- can e t t ri-Fbu o to illi t .

t I I I,, )!C JLti,'J t:he hai ysi q a l cy n n' a d caused

-. rti l ol ,ii <; of ve 'i- c li cj re to 10. '. A due to d s triuction of hydroxy-

'1 i.. t o rl ;. I A p rtion of the 10.2A peak can be assigned to

ill te ,'l7 ch ;-- .: i ,t 500C.

Non -.:r/s I ll i n si n pecies such as a I lophane l hicih are amoi-phous to

'X-ilys c-a ,ot be i ent i field due to absence of di ff actionn maxima,

Ch -ori te a.!d mon.. Lori l c.i te (s5, cti te) ..-re not i en L field in any of

ohe soi ls i ve iratcd. Peaks i, the 7A region were ino' assiq'ied to

koolinitr a Iu:c to :.hir broadness and 7.2A or greater position. It is

oi lin a :. ''Ler of opinion as Pto *.bhethri components in this region may

he iden'if I d as i. etthalloysite or poorly crystal zed kaolinite.

Obvio;']ly me abha loysi te-porly crystal lized kae l ini te-crystal 1 in

k;ir, i rite continil a exists which complicates assi nment of a species

Cr.-, cbal itoe nd gibbsite rwere ideciL 'id by 4.0"A and 4.33A

sp;:'cinjs, respectively. Spacings of 4.2A and 3.3.A iere assigned to

quartr:P. FI ,ld'.par r. as idontif:e by di ffr cti : peaks in the 3.0 A to

3.-5A region.

X-ray diffra.ctgra:'s of fira c:ay (<0.2 p) and coar:e clay (0.2 -

2 ,.) frI, rions foe- Fr Fi s -0 lre p ,re noted ;n Figs. 9-ri. Recog-

rizable cr', ta! ina species f, on X--ray diffract;cn p-ters are li sted

in Table / und d/n spci-g5s are rl:,-e:!td in Appendix "cbil 13 for the

ao-o' e. tior~, clay fracl.: n tnd pr; fi l s.

4 5 7 10

Fig. 9.--X-ray diffrac ion pattern of
iM-saturaTed coarse clay
(0.2-2i? anjd Fine clay
(<* ,.2,'-) fra-ctins of profile


S ,I

F -X6 i on ,
prof? / 2.

dill ,o Ii-S ROitlS
F3 4O--X-ia dF cion 13a7i

clay O,-.2pi) and fine c!.y
(< 2,1) fractios of
profile 2.

AL --- I .
4 5 7 10 17

Fig. il.--X-ray diffraction patterns
of !g-saturated coarse
clay (0.2-2jj) and Fine
clay (<0.2i) fractions of
profit l 3.