NRC Publications Archive

Archives des publications du CNRC

The consolidation of peat: a literature review

MacFarlane, I. C.

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien
DOI ci-dessous.

Publisher’s version / Version de l'éditeur:

https://doi.org/10.4224/20331668
Technical Paper (National Research Council of Canada. Division of Building
Research), 1965-03

NRC Publications Archive Record / Notice des Archives des publications du CNRC :
https://nrc-publications.canada.ca/eng/view/object/?id=6e0fff9f-f5db-418e-908d-83e154c45786
https://publications-cnrc.canada.ca/fra/voir/objet/?id=6e0fff9f-f5db-418e-908d-83e154c45786

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
https://nrc-publications.canada.ca/eng/copyright
READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site
https://publications-cnrc.canada.ca/fra/droits
LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the
first page of the publication for their contact information.

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la
première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez
pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
 ctsTr/rcrsr

tltll
llllll
lllllltl
lltrllll
lllt
lllllll
rllllll
3 1809000469038
lltll
lll,ll
lllllrl
NATIoNAL RESEARCH COUNCIL

CANADA

DIVISION OT' BUILDING RESEARCH

THE CONSOLIDATION OT' PEAT

A LITERATURE REVIEW

by

I. C. MacFarlane
Ar'i,ql"Yr.ID

Technical Paper No. 195

of the
Division of Building Research

OTTAWA

March 1965
 THE CONSOLIDATION OF PEAT

A LITERATURE REVIEW

by

I. C. MacFarlane

An irnportant factor in the over-all stability of any structure

is the arnount of cornpression of the subsoil that will occur when the

weight of the structure is applied. For clays, a rational systern has

been developed for estimating the rnagnitude of the total and differential

settlernents as well as for the time reguired for a given arnount of

settlernent to occuri this systern is based on the Terzaghi theory of

consolidation (Taylor, L94Zl.

A characteristic feature of highly organic soils (or peats)

is their extrernely cornpressible nature; large settlernents can be

produced even under srnall loads. Consequently, peat is not a desirable

soil type on which to build roads or other structures" Until recently,

construction on peat has been lirnited and the chief interest in peat bogs

or rnuskeg areas has been with regard to their potential for reclarnation

for agricultural or forestry purposes, or in the rnining of the peat for

fuel or as rrpeat rnoss. rt within the past l0 years or so, however,

interest in peat as an engineering rnaterial has been substantially

increased as a result of two rnajor factors:

(I) The rapidly accelerated p a c e of northern developrnent

in Canada has required an increasing arnount of construction in regions

 )
-L-

of alrnost continuous rnuskeg cover. In the construction of roads in

such regions it is seldorn possible to avoid muskeg, and econornic

considerations often rnilitate against excavation of the peat and back-

filling with a rnore suitable rnaterial. The road, therefore, rnust be

I'floatedtr on top of the organic terrain. Consequently, the compression

characteristics of the underlying peat becorne a rnatter of sorne irnportance.

(Z) The second factor is related to rnore southerly regions

where urban expansion requires the extensive utilization of forrnerly

marginal land (such as peat bogs) for airfields, shopping centres,

factories, or warehouses. Here also the question of the cornpression

characteristics of peat becornes a rnatter of great irnportance.

Although the classical consolidation test has proven useful

in analysing settlernent characteristics of clays, the wide variation

between the fundarnental nature of clays and peats is sufficient to reguire

an independent evaluation of its use in assessing the cornpression

characteristics of peats. Several investigators in various countries

have cornpared settlernent predictions based on laboratory consolidation

tests on peats with observed behaviour in the field. In view of the

unusual nature of the rnaterial it is not surprising that the conclusions

of sorne investigators have been at rather rnarked variance with the

conclusions of others. One obstacle to the validity of cornparing results

of various investigators has been the lack of a rational (and international)

systern of classifying peats. A11 systerns of classification in current

 -3-

use (including the Radforth systern, which is used alrnost exclusively

in Ganada>i<)are qualitative rather than quantitative, and this is a serious

lirnitation to.investigations involving engineering characteristics of peat.

A particular area of controversy in the early literature was

centred around the relative contribution to the totaL settlernent of that

phase of consolidation known as I'prirnary consolidation" (associated

with dissipation of excess pore water pressure) and that known as

I'secondary colrrpressiont' (the plastic deforrnation phase). This

controversy developed largely as a result of the difficulty in estirnating

when the prirnary phase was cornpleted and when the secondary phase

began. The relevant literature on laboratory compression tests and

field settlernent observations of peat extends over the past 25 years.

About two thirds of the papers exarnined that refer to consolidation and

settlernent characteristics of peat, however, have been written in the

past 5 years. A trend towards sorne unanirnity of opinion is only now

being e stablished"

One of the earliest investigators to study the cornpression

characteristics of peat, Buisrnan (1936) considered consolidation of

peat to include a large rrsecondaryrr or ttsecularrt effect. He observed

that for laboratory peat sarnples of 2 crn (0.25 in. ) thickness the

settlernent - - log tirne curve becarne a straight line about one rninute

rrGuide to a Field Description

of Muskeg (based on the Radforth
Classification Systern)". National Research Council, Assoc. Cornrn.
on Soil and Snow Mechanics, Tech. Merno. 44, Ottawa, 1958.
 -4-

or a few rninutes after loading and rernained a straight line throughout

the test. The slope of this line increased for higher ternperatures.

Bui.srnan also noted that the settlernent -- log tirne diagrarn for a road ernbank-

rnent (observed over a ?-year period) was a straight line. He concluded

that laboratory consolidation is cornposed of a cornbination of rrdirect load

effectsrr and rrsecular tirne effects,rr the rnagnitude of the latter depending

upon the test procedure.

Another early investigator, Ringeling (1936) investigated

the settlernent characteristics of a peat layer 2 to 3 rnetres $-t to l0ft)

thick beneath a projected road. He determined both the rate of cornpression

under the fill load and the tirne at which consolidation would be cornplete.

Laboratory consolidation tests were carried out on several peat sarnples

and an experirnental road section was constructed to check actual

settlement against that predicted frorn test results. Settlernents of the

test fill did not agree with cornputed settlernents, being less for the

lower loads and greater for the higher loads than those predicted.

Ringeling also rneasured pore water pressures beneath the test fill and

noted that irnrnediately following loading of the peat, about 75 per cent

of the load is taken by the pore water. He pointed out that despite the

Iow perrneability of the peat, pore water pressures decreased practically

irnrnediately.

Van Mierlo and Den Breeje (1948) also indicated that the

so-called trsecularrr effect is responsible

for an appreciable part of the
 -5-

total settlernent of. peat. This secular effect had to be taken into

account in calculating the rate of settlernent of a sand fill on peat and

clay in the Spangen Polder. Laboratory compression tests on the peat

showed a straight line relationship between settlernent and log tirne.

The average coefficient of perrneability of the peat tested. was I x t0-5"rrrr/"...

Ug$-(1948) reported on an investigation into an ernbankrnent

slip failure on peat. The water content of the peat varied frorn 800 to

1000 per cent, the specific gravity, 1.2 to 1.5. Laboratory consolidation

tests on 2-in.(5.08 crn) thick sarnples were carried out; the samples never

fully consolidated although all loads were left on for a week. The

predicted settlernent -- log tirne curve assurned a straight line, but

the values were on the low side. Actual settlernent of the peat under

the embankrnent was 4ft (I. ?Z rnl in Z years, about double the predicted

value.

kllgy (1950) carried out one-dirnensional (4. 25 in. (I0.8 crn)

diarneter sarnple) as well as triaxial consolidation tests on peat, which

had an average moisture content of. 7zB per cent, an average organic

content of 83 per cent, and an initial void ratio range of.4.5 to t0.3.

The triaxial apparatus was arranged with a scale -size sand drain

installed in the centre to observe its effect on the rate of consolidation.

Colley reported that the sand drain increased the rate of consolidation

by 12 tirnes, in contrast to the observations of various other investigators

who found that vertical sand drains do not increase the rate of consolidation
 .6-

in peat. col1ey measured the perrneability of the peat, which averaged

o.3ft/day (1.059 x I0-4 crn/secl, in the vertical direction and in the

horizontal direction it ranged frorn 0.2 to 1.3 tt/d,ay (7.06 x l0-5 to

- 4-
4.59 x IO crn/ sec).

Thornpson and Palrner (1951) reported on a fairly cornprehensive

series of consolidation tests on a black fibrous peat ranging in natural

rnoisture content frorn 239 to 341 per cent, dry weight range of. 14.5 to

Z0.8Ib/cutt(Z3Z.3to 33?..?kg/c:urnl, specific gravity range of 1.80 to

2.04, and initial void ratio range of 5. I to 6.7. Consolidation specirnens

were 2in. (5.08 crn) in diarneter and,l/2 and I in.(l .27 and.Z.54 crnl

thick. It was not possible to draw an e-1og p curve for these specirnens to

deterrnine the preconsolidation load. The authors could see no line of

dernarcation between prirnary and secondary consolidation in laboratory

tests and postulated that the prirnary phase, if it exists, is apparently

'W'hen
completed in less than one rninute. observed settlernents were

plotted against 1og tirne, an approxirnately straight line was produced.

Laboratory specirnens also showed no tendency to deviate frorn a straight

Line" This line extended frorn one rninute to approxirnately one day,

changing to another and steeper straight line frorn one day until the end

of the test" Thompson and Palrner observed that the rate of consolidation

of peat is independent of the distance of flow of the pore water and

consequently concluded that it is futile to atternpt to stabilize peat with

vertical sand drains.

 -7-

In his pioneer work on the engineering properties of peat,

Hanrahan (l9SZ, 19541 carried out various physical and rnechanical

tests on peat with the following general characteristics: water content

range - zB0 to r45o per cent, specific gravity - r.z to 1.7, dry density -

4.0 to 7.SIb/cuft (64.1 to lzo.lkdcu rn) and initial void ratio - 9 to 25.

Hanrahan perforrned a nurnber of srnall - and large-scale consolidation

tests and observed that the curve of settlernent plotted against tirne bore

little resernblance to the theoretical curve. He attributed the discrepancy

to various phenornena, including the abnorrnally large decrease in

perrneability that accornpanied the application of load, the d.ecreasing

coefficient of cornpressibility, and thixotropy. High surface activity

and adsorption properties of peat also contributed to appreciable secondary

consolidation. The cornpressibility of peat was observed to be temperature

sensitive, agreeing with Buisrnants (1936) observations. As a result of

cornparative srnall-and large-scale consolidation tests, Flanrahan concluded

that accurate forecasts for settlernent can be rnade by assurning that the

rnagnitude of settlernent is directly proportional to thickness and that the

tirne of settlernent varies as the square of the thickness, for a considerable

period after the cornpletion of the excess pore water pressure phase.

Shea (I955) reported that the use of consolidation tests for

the prediction of settlernent of levees constructed on peat had been only

partially successful" The arnount of settlernent could be deterrnined, with

fair accuracy, but the predicted length of tirne required for it to take
 -8-

place was not even approxirnately correct. Consolidation tests indicated

I5 to 20 years for 90 per cent settlernent, whereas the actual settlernent

took place in a few days or weeks. Frorn settlernent plates installed

beneath the levees on peat, Shea observed that the total settlernent carne

very close to the predicted 50 per cent of peat thickness, but that

practically 100 per cent of this settlernent occurred during construction.

This latter observation is at variance with the findings of rnost other

inve stigators.

Cook (19561 presented a cornpilation of test data to show that

a relationship exists between the rnoisture content and the |tcoefficient

of cornpressibilityrr of a wide range of soils frorn pure peat to organic

clay silts in the Vancouver area. He also shows that there is a relation-

ship between the rnoisture content and the specific gravity, initial void

ratio, and submerged weight of these soils. He suggests that these sirnple

relationships perrnit the calculation of.expected settlernents without the

trouble of lengthy consolidation tests.

Lewis (1956) concluded frorn his consolidation tests on peat

(1 in. (2.54 crn) thick sarnples, average water content 580 per cent) that

100 per cent prirnary consolidation was about cornplete within the first

I0 rninutes. Much of the consolidation was observed to be of a secondary

nature , a fact that was attributed to plastic deforrnation, or creep, with

the rate of rnovernent a function of rnanv factors such as the nature of

the rnaterial, the rnagnitude of the applied stress and the ternperature.
 -9-

Although it was not possible to calculate rates of settlement, frorn

results of Iong-terrn consolidation tests Lewis estirnated the ultirnate

value of settlernent to be expected in the field" This agreed closely

with subsequent observations.

Brawner (1958) noted that for field observations of a fill

on rnuskegr graphs of settlernent versus log tirne (in days) were essentially

straight lines for the period following construction. He concluded that

the settlernent was principally secondary over this period. The rate

was very uniforrn, ranging frorn r.2to 7.4f.t per log cycle of tirne. The

rate and rnagnitude of settlernent were observed to be aknost identical

for sections of the road with and without sand drains, so that the sand

drains did not accelerate the rate of the settlerrlent to any great extent.

It was concluded frorn this that the rate of consolidation of the peat was

independent of the length of the drainage path. Inlater reports, however,

Brawner (19S9, 1961) suggested that the consolidation of peat rnay be

largely prirnary rather than secondary. Piezometers installed beneath

a roadway fill on DEI rnuskeg (average water content of the peat = 1396 per

cent, specific gravity = 1.61, initial void ratio = 24.8) indicated that

prirnary consolidation of peat extends well into what is norrnally considered

to be the secondary phase of consolidation, with the transition taking

place at about ?5 days. Brawner considered that the rnagnitude of the

settlernent varies as the thickness of the peat layer, and that the rate

of settlernent varies as the square of the thickness. Subsequent investi-

gations, however, necessitated a revision of these conclusions.

 -I0-

Hillis and Brawner (1p61) pointed out that sorne of the rnisunderstanding

regarding the process of settlernent within a peat layer had been caused

by the fact that significant pore pressures had been observed in peat for

a long tirne after settlernent had becorne directly proportional to 1og tirne.

They expressed doubt as to the validity of these pore pressure rneasurernents

because of inadequacies of the piezorneters used; and noted that for both

field and laboratory observations the settlernent of rnost peats continues

as a straight line on a serrri-Iog plot after a relatively short tirne. The

initial process is cornpared to the usual prirnary consolidation associated

with inorganic clays where the rate of settlernent is a function of the decrease

in pore water pressure. For peat this is qualified, however, by heterogeneityo

the presence of gas in the peat, and the rernarkable decrease in perrneability

with increase in stress" AJter the pore pressures have dissipated, and

probably for sorne tirne before, the secondary colrlpression phase of

settlernent occurs, The rate is independent of drainage conditions and

thus theoretically independent of the thickness of the peat layer. Hillis

and Brawner suggested that the prirnary phase of consolidation in the field

can be calculated frorn the laboratory void ratio-log pressure curve, and

recornrrr.ended that each load incrernent be perrnitted to rernain in the sarnple

for only about 30 rninutes. Long-terrn tests can be run to define the

secondary stage" The secondary consolidation is deterrnined frorn the

forrnula: S - C_-_"H. log t-/t, where C is the coefficient of secondary

SCC SCC Z' I SCC

compression and is the slope of the straight line portion of the e-1og t plot,
 -lt-

H is the thickness of the layer (after prirnary settlernent is cornpleted),

t, the tir4e of prirnary settlernent, and tZ - tt the tirne over which it is

desired to calculate secondary settlernent. The total settlernent at a

given tirne can be obtained by adding the settlernents for the prirnary and

secondary phases. Hillis and Brawner stated that sand drains do not have

a beneficial effect in peats.

Lea (1958) pointed out that for peat soils, where the

coefficient of consolidation is not a constant but rnay vary as rnuch as

100 tirnes in the comrnon range of pressures, sorne rnodification rnust be

rnade to the theoretical analysis of the non-steady flow condition. He

suggested that the basic principles of larninar flow would be applicable

and that studies of prirnary consolidation of peat should be based on these

general principles rather than on the rnathernatics of the Terzaglni theory.

Lea considered that primary consolidation is a rnuch more irnportant factor

in peat consolidation than is generally believed. The rate of volurne change

of peat under load is affected by a variation of strength with tirne and a

gradual crushing of cells and slipping of fibres. Lea and Brawner (Lg59I

carried.'.this thinking a step further and advocated a fresh approach to the

analysis of peat consolidation: to assurne that consolidation is prirnary and

that Darcyts Law applies" They suggested that calculations of settlernent

in peat can be deterrnined by direct proportionality between laboratory

tests and field predictions. They used the square rule to estirnate the

tirne rate of settlernent and reported reasonable agreernent between the

 -tz-

rates of settlernent in a test section and laboratory tests.

In a rnore recent paper Lea and Brawner (1963) report,

however, that additional inforrnation does not support the view that the

tirne to 100 per cent prirnary consolidation is proportional to the square

of the thickness; rather than the exponential rrirt is closer to the power 1"5

than it is to ? as for the classical theory. They report that both Iabotatory

and field tirne curves usually show the characteristic ttStr shape on serni-

log paper and indicate that it is usually not difficult to establish 100 per cent

prirnary consolidation. Secondary settlernent follows a straight line on a

serni-log plot of tirne settlernent. Records are assembled to show that

laboratory tests give a reliable estirnate of rnagnitude of settlernent. The

rnagnitude of secondary cornpression (t"".) was colrrputed frorn the forrnula

already referred to (Hillis and Brawner, 1961): t""" = C"...H.1og tZ/tl

The rate of secondary cornpression was found to be rnuch greater in the

field than in the laboratory by a factor of up to 5. Values ot C"u. were

observed by the authors to depend to a substantial extent upon the load

history of the deposit and the rnagnitude of the load. Lea and Brawner

also point out that rebound can be quite a significant rnatter in the rernoval

of surcr'-rge - frorn l5 to 35 per cent of the settlement upon full rernoval

of load. They found also that, in general, the field rebound is about double

the laboratory rebound. They suggest that this rnay be a result of elastic

rebound plus gas expanding and corning out of solution.

 -13-

In their cornprehensive report on a study of vertical sand

drains, Moran et al.(1958) considered in sorne detail the consolidation

characteristics of organic soils, including peat. They indicated that

the effects on consolidation of gas in the pore spaces (caused by

decornposition of organic rnatter) are quite significant: (l) when load

is applied, cornpression of the gas rnay cause a delay in the response of

pore pressures to a change in boundary water pressures; (2) when

loading has been cornpleted, expansion of the gas will decrease the

rate of consolidation. The effect of gas on tirne-settlernent curves is

sirnilar, therefore, to the effect of secondary compression, and care

must be exercised to distinguish between these two effects" In the

authors! wiew, alrnost all of the prirnary consolidation under a particular

load is cornpleted before an appreciable arnount of secondary cornpression

takes place, so that practically no excess pore pressures exist during

secondary cornpression. The rnagnitude of secondary cornpression,

cornpared to the rnagnitude of prirnary consolidation, varies with soil

type, state of stress and ternperature and is rnaxirnurn for highly

organic soils. Relative proportions of the two phases of settlernent

are said by the authors to be approxirnately independent of the rnagnitude

of effective stress for stresses in excess of the preconsolidation value.

The rnagnitude of settlernents resulting frorn secondary cornpression

was calculated frorn the forrnula: 0"""" = Co.Hr' log ,r""/tn, which

isessentia1lythesarneasthatusedby@(lgot}and

Lea and BTawner (1963). Co is the coefficient of secondary colnpression

 -14-

and is equal to the amount of secondary compression per unit thickness

of soil occurring in one cycle incrernent of tirne. The rate of secondary

compression is independent of the perrneabitity of the soil and of the

coefficient of consolidation as is the rate per unit thickness at which

secondary compression occurs also independent of thickness. Laboratory

consolidation tests to deterrnine the effect of preloading on second.ary

compression indicated that although preloading rnay not cornpletely

elirninate settlernents due to secondary corrrpression, they reduce thern

to very srnall values.

Root (1958) considered that settlernent analyses of

peat by the conventional rnethods developed for clays are not always

reliable. He questioned whether prirnary consoLidation can be identified

in the consolidation tests of peats. lt there is prirnary consolidation

he believed that it takes place during the very early stage of the test

and is cornpleted in I to 10 rninutes. The initial perrneability of peat

before loading is relatively high but dirninishes rapidly as the peat is

compressed. Root suggested that at some tirne subsequent to cornpletion

of the initial rrprirnary consolidationtr the perrne ability of peat rnay

decrease to a point where t},e rate of consolidation is influenced. or

controlled by the dissipation of pore pressure. In the field, the rate

of settlernent greatly dirninishes a short tirne after loading of the fill

is discontinued, thereafter following a straight line on a serni-log plot.

Root reports that the slope of the line varies rnarkedly even between

points where fill loads are cornparable and the rnoisture content and
 -I5-

thickness of the peat layer appear to be the sarne. Root also pointed

out that the rate of settlernent is not proportional to the thickness of

the peat layer. Long-terrn consolidation, which is proportional to

log tirne, he considered due to possible plastic deforrnation or slow

structural failure of peat fibres. It was found that sand drains did

not accelerate settlernent in peat, but did effect an increase in its

shear strength during construction.

Tressider (1958) pointed out that the first phase in the

consolidation process in peat is governed by the rate at which the water

can escape frorn and through it" This is predorninantly a hydrodynarnic

phase and the tirne of consolidation varies as the sguare of the length

of the drainage path. The tirne involved for this first stage varies, but

he noted that it is often very lengthy and rnay continue for rnany years

in deep peats. Tressider stated that the second phase continues at a

rate independent of the drainage process and thickness of the peat

layer" This occurs sirnultaneously with prirnary consolidation, but

continues long after prirnary consolidation has ceased. Laboratory

sarnples were observed to cornpress slowly for years under load. He

pointed out that the perrneability of peat decreases significantly during

the consolidation process; and also stated that there is very little

rebound in peat when consolidation is complete and the load removed.

Arrderson and Hernstock (1959) observed that Ior a test

fill on FI rnuskeg (with peat ranging frorn non-woody fine fibrous to

 -r6-

amorphous-granular) initial or prirnary consolidation takes place within

a rnafter of days. This was indicated by the observed excess pressures

being dissipated at an 8-ft depth in 3 to 4 days under a load of 0, Ztott/sqft

(0. t9S kg/"g crn). Settlernent, in addition to this initial arnount, continued

at a rate proportional to log tirne, as was confirrned by laboratory

consolidation tests. Anderson and Hernstock considered that this secondary

compression is independent of drainage, but influenced by the soil type

and by the state of stress. Settlernent for the first several weeks under

the loading conditions noted above was 0.IIft per ft of depth per log

cycle of tirne. On the other hand, field observations of settlernent gauges

under oil tanks on ABI rnuskeg indicated a rate of secondary consolidation

of 0.0I?ft per ft of d.epth per log cycle of tirne, for a load of 0.15 tons/ftz
.)
(0. 15 kg/crn').

Zegarra (I959) reported on consolidation tests carried. out

on a very fibrous tropical peat (water content range T19 - 795 per cent,

specific gravity 2.r, organic rnatter ?7.8 per cent). He pointed out

that unless the vegetal fibres were broken during loading, the peat would

regain its original volurne after rernoval of cornpressive load if no

restraint were placed on the rebound" This was not too ewident frorn the

e - log P curves; the discrepancy was attributed to side friction.

Settlement -- log tirne curves did not produce the usual straight-line

portion (for tirnes up to 1000 rnin) but all the curves are slightly convex

down for all load incrernents except the final one of 3.5 tons/ttz(E.+zt<g/crnzl
 - 1 7_

when the curve has a slight tenddncy to flatten.

Keene (1960) found that for Connecticut upland and tidal

peats settlernent estirnates based on consolidation tests agreed quite

well with field observations. Preconsolidation load obtained by the

Casagrande rnethod gave close agreement with the caleulated overburden

Ioad for norrnally consolidated peat, and settlernent calculations frorn the

virgin cornpression portion of the e-1og P curve were confirrned by

observed settlements. Rates of consolidation presented a nrore difficult

problern. Keene noted frorn laboratory and field observations that when

prirnary consolidation is alrnost cornplete, secondary consolidation

proceeds as a straight line on a settlernent -- log.tirne graph. The slope

of the secondary line gives a settlernent in one cycle of tirne of about 10

to 2O per cent of the prirnary settlernent. Keene pointed out that it is

often difficult to deterrnine frorn the data at what point the prirnary

settlement is about cornpleted and the secondary settlernent has begun.

Carrying out field and laboratory tests on Scottish peats

(average water content = 800 per cent), @!s (1960, 196I) found that

vertical sand drains did not effect a significant increase in rate of

settlernent, although they did increase the rate of excess pofe water

pressure dissipation. Laboratory tests also indicated that the rate of

dissipation of excess pore water pressure was affected by the length of

the drainage path, conforrning to the Terzaghi square law. He noted that

excePt for the first incrernent of load, the rates of settlernent of the
 -I8-

laboratory sarnples one rninute after loading were largely independent

of the length of the drainage path. Lake observed that for both srnall-

scale field tests and laboratory tests the settlenrent -- log tirne curves were

not linear - contrary to the observations of rnost other investigators.

For a 3-in. (7.6zcrn) diarneter sarnple, 0.5 itt. (1.2?crn) thick, settlernents

that took place after pore water pressures returned to norrnal varied frorn

39 per cent of the total after the first load incrernent to 8I per cent of

the total after the final load increment.

Frorn his research on Japanese peats (water content range

300 to 1000 per cent; organic rnatter content range 50 to 95 per cent; dry

density range 5.0 to 15.5lbs/cuft (80.1 to Z48.3kglcucrn), Miyakawa(1960)

found that frorn a fairly early stage settlernent becarne a straight line with

relation to log tirne, both in laboratory tests and in the field. He suggested

that the arnount of prirnary settlernent could be predicted frorn the ordinary

consolidation test with a 24hour loading period, which contrasts with the

30 minute loading period recornrnended by Hillis an{B1qw4er (1961).

Mayakawa calculated the rnagnitude of total settlernent frorn the equation:

6 = a * b log ttt/to), where ttarr is the settlernent of the prirnary phase, rtbrr

is the rate of settlernent of the secondary phase for one cycle of log tirne,

rrttr is the elapsed

tirne, and rttott is the tirne for the prirnary phase. The

coefficient of secondary consolidation, c", is equal tob/H, where ilI{il is

the thickness of the cornpressible 1ayer. He found that C" increased with

stress for loads less than the prgconsolidation load, after which it showed
 -19-

no direct relationship. Miyakawa also jrdicated that C" can be shown

to vary directly with the rnoisture content of the peat. He doubted that

sand drains would be effective in accelerating settlernent in peats except

in the prirnary stage of consolidation.

Adarns (196r) conclud.ed frorn earry tests on woody, fine

fibrous peats (water content range frorn 375 to 430 per cent; specific

gravity range frorn I.62 to l. ?0; ash content range frorn IZ. Z to ZZ. 5 per

cent) that the consolidation of peat is predorninantly due to the expulsion

of water under excess hydrostatic pressure. A large proportion of the

settlernent takes place very rapidly when the rnaterial is relatively pervious.

with tirne the perrneability is reduced with a corresponding reduction in

the rate of settlement. He reported that an increase in load resulted in

a significant reduction in voids volurne and in perrneability, changing

frorn approxirnately I x l o-r tt/rnin (5. r x r o-z

at zero road. to
"^/sec)
I x ' 1 0 - 6 t t / r n i n ( 5 . I x I O - 7. o y ' " e c ) a t Z . 8 J k g 1 l c r n } .
Adarns considered

that the consolidation occurring under excess pore pressure was extended

to relatively long-terrn cornpression by this large red.uction in perrneability.

He estineated that approxirnately 90 per cent of the consolidation was due

to expulsion of water under excess hydrostatic pressure. rn a rnore

recent paper, however, Adarns (19631 reported that for peat sarnples

8 in. (20.32 crn) in diarneter and 3.g in. (9"65 crn)

thickthere was an

initial settlernent of relatively large rnagnitude which occurred in a very

short period of time (5 rnin ). The long-terrn settrernent occurred

 -20 -

relatively slowly and was linear with tog time. Excess pore pressures

were dissipated during the initial consolidation period, although a

residual excess pressure of low rnagnitude rernained and showed only

a slight reduction with tirne. On the basis of these laboratory tests and

the results of field instrumentation, Adarns has concluded that the

consolidation of peat occurs in two separate stages: an initial or irnrnediate

stage and a long-terrn stage, which continues indefiniteLy at a slow rate.

Available data indicate that the rnagnitude of the initial settlernent and

the rate of long-terrn consolidation bear a predictable relationship to

the applied load and the thickness of the peat. Adams suggests further

that the initial consolidation is due to the expulsion of the pore water in

the peat rnass, and that the long-terrn consolidation is due to the expulsion

of the water in the solid rnatter. This is essentially the same conclusion

as that reached by Evgenbv (1961).

Barber (tg6t), in an effort to deterrnine the cause of

secondary consolidation, carried out a rather extensive series of

laboratory tests on various organic materials, including peat, using ?-in.

(5.08 crn) diarneter, 0.5-in. (I.27 crn) thick sarnples. Tests on desiccated,

air dry and wet rnaterials indicated that the seat of secondary consolidation

is in the hygroscopic rnoisture. Barber noted that for a specified loading

incrernent for a peat sarnple, the typical S-shaped prirnary consolidation

phase was essentially cornplete in 0.1 day. Linear extension of the

secondary consolidation beyond the l-d"y record consid.erably under-estirnated

 .zI-

the consolidation at I00 days. Barber suggested that secondary

consolidation rnay also produce an s-shaped curve sornewhat like that

of pri:nary consolidation, but extending over rnore log cycles of tirne.

He showed that an increase in load tends to accelerate the rate of

sepondary consolidation and to reduce the magnitude of the long-terrn

settlernent.

Evgenlev (196i) concluded frorn an analysis of nurnerous

consolidation tests that deforrnation of peat under load consists basically

of two phases: (I) deforrnation due to squeezing out of water frorn the

open macropores and (Z) deforrnation resulting frorn the squeezing out

of water frorn the closed pores. In the first phase, the rnovement of

water follows Darcyts Law and compression proceeds very quickly even

und.er srnall loads. The arnount of consolidation varies considerably,

depending uPon the loa<iing history of the peat and the disturbance of the

sarnple. Evgentev found that deforrnation in this phase arnounted to

75to !0 per cent of the total settlernent. The arnount of deforrnation

in the second phase is believed to depend basically upon the degree of

decornposition of the peat and varies within rather narrow lirnits for a

given type. Evgenlev presented tabulated data for the arnount of deforrnation

to be expected in the second phase for various peat types and loads.

He suggested that for calculations not requiring great accuracy the

second phase can be ignored and the settlernent of the first deterrnined by

a relatively rapid rnethod involving application of load incrernents every

 -zz-

24 hours rather than at the end of consolidation. Evgenlev observed

that when the load was rernoved frorn peat sarnples they rebounded

alrnost exactly tp the deforrnation stage of the beginning of the second

phase, frorn which he concluded that the second phase is cornpletely

elastic deforrnation.

(1961), in describing the perforrnance of

!!!g\}gbg53ggb.

ernbankrnents constructed on rnuskeg, stated that the rnagnitudes of

prirnary settlernent were found to be sornewhat less than had been

predicted on the basis of laboratory consolidation tests. He reported

that the trend on the tirne-settlernent curve for field conditions indicated

that for the rnuskeg under consideration (depths up to 8ft (Z.44rn)) ttre

rnajor portion of the secondary colrrpression would take place in the

first 3 or 4 years after cornpletion of the ernbankrnent.

Rip1ev and Leonoff (I96I) reported significant differences

between the observed behaviour of peat and the predicted behaviour

cornputed by conventional theoretical analyses based on field and laboratory

test results (arnorphous granular peat with woody fine fibres; average

water content = 1000 per cent). The data frorn long-terrn laboratory

consolidation tests (both one dirnensional and triaxial) indicated that

long-terrn settlernents occurring during the period norrnally attributed

to secondary cornpression rnay arnount to 50 per cent of the total

settlernent. Their field settlernent rneasurernents revealed a high rate

of continuing long-terrn settlernents. Settlernents of raft sections of

 -23 -

an ernbankrnent where no fill load had been placed for 3 -l/Z years

were observed to be continuing on a straight line relationship at a

relatively steep slope.

Goodrnan and Lee (1962) considered that the classical

consolidation theory predicts the rnagnitude of expected settlernents of

peat rather well, although the rate of consolidation in theory and practice

do not agree. Their conclusions were based on experience with peaty

soils in the Syracuse, New York area (water content range 282 to 403 per

cent; initial void ratio rarlge 5.73 to 5.48; specific gravity l.5l to l.62;

organic content 42.5 to 75.2 per cent). In contrast with rnost other

investigators, Goodrnan and Lee believed that relatively rninor effects

would result frorn secondary corrrpression, since their observations

indicated that most of the settlernent of structures on peat appeared

to have taken place shortly after construction.

schroeder and wilson (1962) carried out an extensive series

of consolidation tests on partly disturbed peat sarnples with initial void

ratios varying frorn 40 to 10. The initial heights of the specimens

ranged frorn 0.5 in. to 25 in. (1.2T to 63.5 crn), and were consolidated

under loads of 0. 5 to 20 psi (0.035 to r.406xg/.rnzl. They observed.

that the settlement -- Iog tirne curves for peat had characteristics

sirnilar to clay consolidation curves for the first incrernent only. For

succeeding load incrernents the tirne graphs were either straight lines

or slightly curved downwards at the beginning. Points of zero pore

 -24-

pressure occurred at various positions on the straight-line portion

of the tirne-settlernent curves owing to differences in drainage paths,

but were observed to have no effect on the cuaves thernselves. Schroeder

and Wilson concluded frorn their test results that gecondary consolidation

and, consequently, plastic or viscous tirne relationships are rnore

irnportant for peats than for clays, at least for a load incrernent ratio

of unity. A second series of tests was carried out using sarnples taken

frorn one large undisturbed specirnen and all sarnples having the same

drainage path. These sarnples were exarnined and analysed on the

basis of rheological principles. From this they concluded that

consolidation characteristics of peat are governed by: (I) variation in

the rnaterial; (21 drainage conditions; (3) load incrernent ratios; (4) range

of void ratios; (5) stress and strain history; and (6) viscous or plastic

resistance to flow. In a more recent report, l4rilson (1963) described

a series of large-scale consolidation tests (75.* (9.8 in.) diarneter by

75 crn(29.5 in,) high) carried out on disturbed arrorphous granular

peat to rneasure the pore water pressures during consolidation. These

tests showed that the pore water pressure, irnmediately after loading

was considerably less than the theoretical value, and that it then increased

to a value slightly less than the theoretical value during the early part

of the test. Wilson cornpared. the theoretical and experirnental U/T curve and

reported that the plastic structural deforrnations took place at a rapid

rate in the early stages of consolidation as a result of a decrease in

Pore water pressure with consequent increase in effective strees.

 _25_

Tests conducted with identical loads on sarnples of different heights

showed that the tirnesfor the dissipation of excess Pore water pressures

conforrned to the Terzaghi theory relatjng length of drainage path to

tirne. Due to structgral deforrnations, however, the rate of settlernent

or change in void ratio did not conJorrn to the theory.

Wahls (I962 ) reported on an experimental study of the

prirnary and secondary consolidation characteristics of a calcareous

organic silt, the results of which rnay be applied generally to sorne

peats. Both undisturbed and rernoulded sarnples wele used. Frorn

his experirnental results he concluded that the coefficient of secondary

compression C^, is dependent on the void ratio (and, consequently,

on the total pressure) and independent of the magnitude of the pressure

incrernent and pressure-incrernent ratio. The coefficient of secondary

cornpression, Co, represents the rnaxirnurn rate, with respect to log

tirne, at which secondary compression occurs during a pressure

incrernent. During prirnary consolidation, secondary colnpression is .'

occurring at a rate tlnat is not linear with respect to the logarithrn of

tirne. The tirne required for the rate of secondary colnpression to

approach Co is directly related to the tirne required for cornpletion

of prirnary consolidation.

As the pressure-incrernent ratio is reduced, the shape

of the corrlpression-logarithrn of tirne curve differs rnore radically

from the theoretical curve for prirnary consolidation.

 -26-

For sufficiently srnall pressure-incrernent ratios (np/po < t/! tne

tirne at which prirnary consolidation is completed cannot be found by

the conventional Casagrande rnethod. In such instances, the rna-:<irnurn

slope of the curve during prirnary consolidation is less than the final

secondary slope, and thus there is no inflection point on the curve.

As the pressure-increment ratio is reduced, the tirne required for the

cornpletion of prirnary consolidation, as rneasured by the coefficient

'Wahls
of consolidation C.r, increases. concluded, therefore, that:

(a) investigations of secondary compression effects are sirnplified

if srnall pressure-incrernent ratios are used, and

(b) the use of laboratory tests with Ae/eo = 1.0 in the analysis

of practical problerns for which the field pressure-incrernent

ratio is rnuch less than one rnay lead to unconservative

conclusions. In such cases the secondary effects will have

greater significance in the field than in the laboratory.

As the observed behaviour of a soil exhibiting considerable

secondary colrrpression cannot be explained by classical consolidation

theory, W'ahls proposed a new rnathernatical rnodel for the consolidation

process. This rnodel is used to explain the effect of pressure-incrernent

ratio and to analyse tirne-cornpression curves when secondary cotrrpression

predorninate s.

SUMMARY

Frorn this study of the literature, certain conclusions rnay

 -?7-

be drawn on which there is fairly general (although not necessarily

universal) agreernent. These are as follows:

(l) In general, field and laboratory tirne-settlernent curves for

peat on a serni-log plot show initially the characteristic |tSrt

shaped curve, although it is not as clearly pronounced as

for inorganic soirs. rt is usually diJficult to define any crear

dernarcation between the end of the prirnary consolidation

phase and the beginning of the secondary colnpression phase.

(21 In the laboratory prirnary consolidation occurs very rapidly,

generarly within a few rninutes, and rnay constitute as little

as 50 per cent of the total settlernent.

(3) The rnagnitude of settlernent in the prirnary consoridation

phase varies directly as the thickness of the peat; conseqluently,

this phase of field settlernent can be predicted frorn laboratory

te sts.

(4) The tirne rate of the prirnary consolidation phase of peat

compression is also proportional to the thickness. It is variously

reported to be directly proportional to the thickness, proportional

to the thiclqress to a power of 1.5, and proportional to the thick-

ness to a power of 2. This variation rnay be a function of the

peat type; further investigation of this possibility is required.

(5) cornplete consolidation of peat is reached only after a 1ong

period of tirne - several rnonths even for srnall specirnens -

 -zB-

due to secondary effects.

(6) Secondary corrrpression is considered to be of a viscolls or

plastic nature, its rnagnitude affected by ternperature, the

nature of the peat and the state of stress. It is independent

of drainage conditions. Secondary cornpression is unusually

high for peats and rnay account for as rnuch as one half of

the total settlernent. The order of rnagnitude of field settle-

rnents due to secondary compression can be calculated frorn

long -terrn consolidation tests.

(7) Secondary colnpression follows a straight line on a serni-1og

plot of tirne-settlernent for both laboratory and field curves.

The slope of this line, the coefficient of secondary cornpres-

sion, depends upon the load history of the peat and is influenced

by the rnagnitude of the applied load"

(8) The coefficient of perrneability of norrnally consolidated

peat is initially very high, but it dirninishes rapidly as the

peat is cornpressed and is dependent upon the rnagnitude and

duration of the load. The coefficient of perrneability is

greater in the horizontal plane than in the vertical plane, with

an anisotropy ratio of about 3 to 6.

(e) If the load is fully rernoved during or following the secondary

corrpression phase, the rebound rnay be a very significant

factor to consider.
 _z g _

(l0) Gas (caused by decornposition of organic rnatter) in the

pores rnay be a factor co'tributing to the difficulty in

interpreting tirne -settlerne nt characteristics for peat.

(l l) Vertical sand drains do not accelerate settlernent of peati

they rnay effect an increase in the shear strength, however,

during construction.
 -30-

REFERENCES

Adarns, J.I. Laboratory cornpression tests on peat. Proq, Seventh

Muskeg Research conference, lB and 1t April 1961. National

Research Council, Canada, Assoc. Cornrnittee on Soil and

Snow Mechanics, Technical Merno. 7I, P.36-54, 196l-

Adarns, J.I. A cornparison of field and laboratory consolidation

rrreasurefirents in peat. Proc., Ninth Muskeg Research

Conference, Z0 and ZL May 1963. National. Research Council,

Canada, Assoc. Cornrnittee on Soil and Snow Mechanics,

Technical Merno. BI, p. I I7 -135, 1964.

Anderson, K.O. and R"A. Hernstock. Relating sonre engineering

properties of rnuskeg to some problerns of field construction.

Proc., Fifth Muskeg Research Conference, 4 March 1959.

National Research 3ouncil, Canada, Assoc. Cornrnittee on

Soil and Snow Mechanics, Technical Merno. 61, P. 16-25, 1959.

Barber, E.S. Notes on secondary consolidation. Proc., Highway

Research Board, Vol. 40, p"663-675, W'ashington D.C. 1961.

Brawner, C. O. The rnuskeg problern in B" C. highway construction.

Proc., Fourth Muskeg Research Conference, Ll March 1958.

National Research Council, Canada, Assoc. Cornrnittee on

Soil and Snow Mechanics, Technical Merno" 54, P.45-53, 1958.

Brawner, C. O. The principle of preconsolidation in highway construction

over Muskeg. Roads and Engineering Construction, 97:)z 99'

100, L0Z, I04, Septernber 1959.

 . 3 1-

Brawner, C. O. The practical application of preconsolidation in

highway construction over rnuskeg. Proc., Sixth Muskeg

Research Conference, ZO and 2l April f 960. National

Research Council, Canada, Assoc. Cornrnittee on Soil

and Snow Mechanics, Technical Merno. 67, p.l3-3I, 1961"

Buisrnan, A. S. K. Results of long-duration settlernent tests. Proc. ,

First International Conference on Soil Mechanics and

Foundation Engineering, Vol" I, p. I03-106, Cambridge,

Mass., 1936"

Colley, B.E. Construction of highways over peat and muck areas.

Arnerican Highways, Zgil: 3-6 January 1950.

Cook, P.M. Consolidation characteristics of organic soils.

Proc., Ninth Canadian Soil Mechanics Conference, l5 and t6

Decernber 1955. National Research Council, Canada,

Assoc. Cornrnittee on Soil and Snow Mechanics, Technical

Merno. 4I, p.BZ-87, 1956.

Evgenrev, r. E. The accelerated m.ethod of obtaining peat soil

cornpression characteristics. (In Russian) Torfyanaya

Prornyshlennost (Peat Industry) No. I, p.25,26, Lg6t.

Goodman, L.J. and C.N. Lee, Laboratory and field. data on

engineering characteristics of sorne peat soils. Proc.,

Eighth Muskeg Research Conference, IZ and IB May 1962.

National Research council, canada, Assoc. cornrnittee on

soil and snow Mechanics, Technical Merno.'r4, p.l0?-l zg,1962.

 -3e*

Hanrahan, E. T. The rnechanical properties of peat with special

reference to road construction. Bull., I.C.E. of lreland,

78:52 L79-2I5, April L952.

Hanrahan, E. T. An investigation of sorne physical properties of

peat. G6otechnique, 4z3z I0B-I23, Septernber 1954.

Hillis, s.F. and c.o. Brawner. The cornpressibility of peat with

reference to rnajor highway construction in British Colurnbia.

Proc., seventh Muskeg Research conference, rg and 19 April

196I. National Research Council, Canada, Assoc" Cornrnittee

on soil and snow Mechanics, Technicar Merno. zl, p. zo4-227,

I 961.

Keene, P. Discussion on rrA review of the engineering characteristics

of peatrr by r. c" MacFarlane, Arn. soc. civ. Engrs. , Journar

of the soil Mechanics and Foundations Division, g6:sMl; p.?l-72,

February 1960"

Lake. J.R. Pore pressure and settrernent rneasurements during

srnall-scale and laboratory experirnents to deterrnine the

effectiveness of vertical sand drains in peat. proc.,

Conference on Pore Pressure and Suction in Soils, British

National Society of the llternational Society of Soil Mechanics

and Forrndation Engineers, p. 103-10?, Butterworthss, Lond.on I960.

 _33_

Lake, J"R. Investigations of the problern of constructing roads

on peat in Scotland. Proc., Seventh Muskeg Research

Conference, 1B and l9 April 1961. National Research

Council, Canada, Assoc. Cornrnittee on Soil and Snow

Mechanics, Technical Merno. 71, p. 133-148, 1961.

Lea, N.D. Notes on the rnechanical properties of peat. Proc.,

Fourth Muskeg Research Conference, National Research

Council, Canada, Assoc. Cornrnittee on Soil and Snow

Mechanics, Technical Merno. 54, p.53-57, f 958.

Lea, N.D. and C.O. Brawner. Foundation and paverrrent design

for highways on peat. Proc., 40th Convention of the

Canadian Good Roads Association, p.406-424, 1959.

Lea, N.D. and C.O. Brawner. Highway design and construction

over peat deposits in the Lower Mainland region of

British Colurnbia. Highway Research Board, Research

Record I, Washington D"C. , 1963.

Lewis, W. A. The settlernent of the approach ernbankrnents to a

new road bridge at Lockford, Wsst Suffolk. G6otechnigue,

VoI. 6, No. 3, p. f 06-II4, Septernber 1956.

Mickleborough, B.W. Ernbankrnent construction in rnuskeg at

Prince Albert" Proc., Seventh Muskeg Research Conference,

18 and 1 t A p r i l f 961. National Research Council, Canada,

A s s o c . Cornrnittee on Soil and Snow Mechanics, Technical

Merno. 7I, p. L64-I85, I961.

 _ 34 _

Miyakawa, r- some aspects of road construction in peaty or rnarshy

areas in Hokkaido, with particurar reference to filling

rnethods' civil Engineering Research rnstitute, Hokkaido

Development Bureau, Sapporo, Japan, 1960.

M6a.n, Proctor, Mueser and Rutredge (consulting Engineers).

study of deep soil stabilization by verticar sand drains.

Bureau of Yards and Docks, Departrnent of the N"oy,

W - a s h i n g t o nD . C . , NOy gBBlZ, 195g.

Ringeling, J-c.N- Measuring groundwater pressures in a rayer of

peat caused by an rrnposed Load. proc.

, First rnternationar

conference on soil Mechanics and Foundation Engineering,

Carnbridge, Mass. V. 1, p. 106_ltl, L936.

Ripley, C'F' and C-E- Leonoff. Ernbankrnent settlernent behaviour

on deep peat. proc., seventh Muskeg Research conference,

l8 and 19 April 1961. National Research Council, Canada,

Assoc- cornrnittee on soil and snow Mechanics, Technical

Merno. ZI, p. IB5_204, i961.

Root, A. w. carifornia experience in construction of highways across

rnarsh deposits" I{ighway Research Board, Vlashington, D. C. ,

Bulletin 173, p" 46_64, 1958.

schroeder, J' and N. E. $rilson. The analysis of secondary consolidation

of peat. Proc-, Eighth Muskeg Research conference, r7 and 1B

May I962. Nationar Research council, canada, Associate

cornrnittee on soir and snow Mechanics, Technical Mern o.74,
p . 1 3 0- L 4 4 , 1 9 6 2 .
 -35-

Shea, P.H. Unusual foundation conditions in the Everglades.

Trans., Arner. Soc. Civ. Engrs., Vol. LZ}, p.92-102, 1955.

Taylor, D. W. Research on consolidation of clays. M.I. T., Departrnent

of Civil Sanitary Engineering, SeriaI 82, August 1942.

Thornpson, J.B. and L.A. Palrner. Report of consolidation tests

with peat. Syrnposiurn on Consolidation Testing of Soils,

ASTM Special Technical Publication No. L26, p.4-8, I9Sf .

Tresidder, J.O. A review of existing rnethods of road construction

over peat. D.S.I.R., Road Research Lab., Road Research

Technical Paper No. 40, H.M.S.O., London, 1958.

Van Mierlo, W.C. and H.I. Den Breeje. Deterrnination of expected

settlements of hydraulic fill in the Spanger Polder near

Rotterdaryl. Proc., Second Isternational Conference on Soil

Mechanics and Foundation Engineering, Rotterdarn, Vol. Z,

p.125-130, 1948.

W'ahls, H.E" Analysis of prirnary and secondary consolidation. Proc.,

Arner. Soc. Civ. Engrs., Journal of the Soil Mechanics and

Foundations Division, Vol. 88, SM6, p"2O7-23I, Decernber L962.

W'ard, W. H. A slip in a flood defence bank constructed on a peat bog.

Prof . , Second International Conference on Soil Mechanics

and Foundation Engineering, Rotterdarn, Vol. 2, p.L9-23, 1948.

 -3 6-

w'ilson, N. E. consolidation and flow characteristics of peat.

Proc., Ninth Muskeg Research conference, zl and zz May

1963, National Research council, canada, Assoc. cornrnittee

on soil and snow Mechanics, Technical Merno. BI, p. l50-160, 1964.

Zegarta, E. J. Discussion ontA review of the engineering characteristics

of peatrr by r.c. MacFarlane, Arner. soc. civ. Engrs., Journal

of the soil Mechanics and Foundations Division, vo1.g5: sM3,

p.89-P3, June 1959.