Cainozoic Research, 3(1-2), pp.

189-219, December 2004

In: Donovan, S.K. (ed.). The Mid-Cainozoic White Limestone Group of Jamaica

Karst geomorphology of the White Limestone Group

David+J. Miller

Department ofGeography and Geology, University ofthe West Indies, Mona, Kingston 7, Jamaica; e-mail: dmiller@uwimona.edu.jm

Received 22 July 2002; revised version accepted 27 May 2003

The karst geomorphology of the White Limestone Group of Jamaica is reviewed. The lithological and structural characteristics of the

White Limestone Group are examined in relation to its karstification, specifically material properties such as purity, petrology, porosity
and mechanical strength, which have been regarded as important in the development of karst landforms. The structural setting and

palaeogeography during the deposition of the White Limestone, together with subsequent block faulting, are considered in relation

to the origin and evolution of the karst topography. The phenomenon of case-hardening is examined as are the dissolution
processes

and influences related to the of a soil cover. The karst features on the White Limestone Group are considered in two
presence

categories; small-scale dissolution sculpturing phenomena, collectively termed karren, and larger scale landforms. The latter can be

further divided into karst landform assemblages comprising doline, cockpit and tower karst. Cockpit karst is the most common

landform unit on the White Limestone Group, and occurs on the harder crystalline limestones, dolostones and where case-hardening

is important. Tower karst occurs on similar lithologies, but has a more restricted distribution. Doline karst is largely restricted to the

chalky Montpelier Formation. The

geomorphological characteristics of each landform type are described and their origin discussed.

Other karst landforms include poljes, ridge karst, glades and of fluviokarst, each of which described. The cave
a range are systems

and underground rivers are also considered. Karst morphometric studies and theories of landform evolution are reviewed.

KEY WORDS: White Limestone Group, Jamaica, karst terrain, geomorphology.

Introduction karstified than the underlying Yellow Limestone because

of its purity and as a result of the extensive development

The White Limestone Group is geographically the most of a secondary permeability along lines of structural

unit in Jamaica and weakness, leading directed dissolution of the

widespread lithostratigraphic crops to

out over 60-65% of the island. Much of the outcrop is limestones. Dissolution of the limestones along lines of

karstified, forming a
of karst landforms across the weakness has formed a classic karst landscape.
range

island (e.g., Sweeting, 1958; Versey, 1972). Indeed, The The palaeogeography and depositional setting of the

Cockpit Country (Figure 1) in northern Jamaica is famous White Limestone Group is examined by Mitchell (2004)
worldwide for its spectacular karst topography, it being and is considered to represent a tectonically quiet phase in

the type area for a tropical limestone terrain termed cock- Jamaica’s geological history. Essentially, the carbonates

karst karst is the laid down block and belt defined by

pit (kegelkarst). Cockpit most widespread were on a structure

landform type and occupies about 60% of all karst in two major fault systems, an east-west set and a northwest-

Jamaica (Day, 1979). It comprises steep-sided, enclosed southeast set (Draper, 1987). The rapid subsidence of the

depressions with side-slopes, forming depressions troughs led deposition of the Yellow Limestone
convex to Group
which are star-shaped or polygonal in plan. Other com- and, in the Middle Eocene, the deep-water chalks of the

mon karst landform assemblages are doline karst, where White Limestones, while eventual subsidence and sub-

the landscape is dominated by oval- or circular-shaped mergence of the blocks initiated shallow-water pure-

depressions; tower karst, forming isolated residual hills carbonate sedimentation. The karstification of the White

above alluvial limestone Limestone in the Miocene when renewed

rising an plain, planed or non- Group began
karstic and poljes, which large, structurally deformation uplifted the limestones. Surface erosion and
rocks; are

controlled depressions surrounded by a steep rim of lime- continued incision de-roofed the siliciclastics of the Yel-

stone
or, at least in part, non-carbonate rocks. There is low Limestones and the older Cretaceous rocks, to form a

also a wide range of fluviokarst features associated with series of inliers, producing surface streams which provided
the White Limestones. large volumes of water to the White Limestones via river

The White Limestone Group is more extensively sinks and which greatly modifiedthe karst from an essen-
-
190-

Figure 1. Locality Key: 1 =

Cockpit Country; 2 =
Natural Bridge, Riversdale; 3 =
Central Inlier; 4 =
John Crow Mountains; 5
map.

=
Hellshire Hills; 6 =

Kemps Hill; 7 = Brown’s Town; 8 =

Moneague; 9 =
Portland Ridge; 10 =
South Manchester plateau; 11 =

Cabarita River; 12 =

Nain; 13a =
Lower Black River Morass; 13b =

Upper Black River Morass; 14 =

Burnt Savannah; 15 =
Santa

Cruz; 16 =

Top Hill; 17 =
Ballards Valley; 18 =
Bull Savannah; 19 =
Barbecue Bottom; 20 =
Lluidas Vale; 21 =

Queen of Spain’s

Valley; 22 =
Duanvale Fault Zone; 23 =

Roaring River; 24 =
Martha Brae River; 25 =
Dallas Mountain; 26 =

Long Mountain; 27
=
Brazilletto Mountains; 28 =
Round Hill; 29 =
Great Goat Island; 30 =
Savannah La Mar; 31 =
Little London; 32 =
Discovery

Bay; 33 =
St Ann’s Bay; 34 =

Quashies River Cave; 35 =

Domock Head; 36 =
Bristol Cave; 37
=

Dry Harbour Mountains; 38

=

Windsor; 39 =

Troy.

Figure 2. General geomorphologicalmap of north-central Jamaica, showing the main karst landform assemblages (based on
Sweeting,

1958).
 -
191 -

tially autogenic to mixed allogenic-autogenic karst In the 1960s, a number of German

a (1958). geomorpholo-
system, particularly on the northern side of the island. Pfeffer
gists, notably (1967, 1969), also worked on the

Karstification of the limestones has continued through the karst of the White Limestones, in the north of
specifically
later and into the
Tertiary Quaternary. the island and across southern St Elizabeth.

Versey (1972) also reviewed the karst geomorphology

of the island from a geological perspective. Gardner et al.

Historical review
(1987) published a summary of Jamaican karst, chiefly
from the earlier reviews of and and
Sweeting Versey, a

Some of the earliest descriptions of the karst in Jamaica more recent review
summary was presented by Draper &

were related to early cave explorations, summarised in Fincham (1997). Other investigations are related to karst

Fincham (1997). The earliest descriptions of in landform evolution (e

caves
g., Landmann, 1990; Pfeffer, 1986,
detail those Edward who
were by Long (1774), published 1997) and to karst landform
development (e.g Smith et
.,

a three-volume history of Jamaica, which contains de- al.

,
1972).

scriptions of three caves. In his remarks on the geology of Karst studies the White Limestones
morphometric on

Jamaica, De la Beche (1827) described the Natural Bridge to date, been restricted localised
have, to relatively small,
at Riversdale, which was the first surface karst feature to areas
of Jamaica. These studies include the work of
Day
be documented (Miller & Donovan, Sawkins
1999). (1976, 1978a) and Brook & Hanson (1986, 1991).
(1869) provided detailed of the and
a account geology

hydrology of the island. The

Cockpit Country was men-

tioned in this volume for the first time, where the cockpits, Geological background
dolines and lightholes were described and attributed to

collapse of the limestones into pre-existing caverns. Some The karst of Jamaica provides a good example of both

thirty years after the publication of the memoir, a reas-

lithological and structural controls its
on development
sessment of the of Jamaica Hill
geology was produced by (Gardner et al., 1987), while the case-hardening of many

(1899) and, although there was little consideration of the of the White Limestones has also had a significance
karst, he attributed the origin of the influence
cockpits to solution, on the development of karst landforms.

not collapse. A few years later, Dane (1909, 1914) de-

scribed the cockpits and other karst features, and also and material
Lithological properties
favoured solution, predominantly along joints and fissures,
for the formation of the cockpits. His contribu- There is
principal a strong control on
karst landform
development
tion to the development of karst studies add de- related and
was to to lithology specifically to certain material

tailed observations of tropical karst in Jamaica Euro- of the limestones.

to properties Sweeting (1958) noted that

investigations which had hitherto been the basis for vertical relief in the form of both
pean development cockpit and

karst evolution theories (Fincham, 1997). Lehmann tower karst is confined to the of hard
areas crystalline
(1954) examined the karst in northern Jamaica and also White
Limestone, as well as to areas of rainfall
high
suggested solutional the of cock-
a origin to development totals. In areas where the White Limestone is more marly,
pits. or
where the rainfall amounts are relatively the
low,
In 1955, Marjorie Sweeting, karst geomorphologist landscape is dominated
a
by doline karst. Sweeting further

from the University of conducted two-month indicated

Oxford, a a link between lateral relief development and

programme of hydrological studies in Jamaica in associa-

lithology, as poljes are normally associated with the marly
tion with the Geological Survey Division. These studies which
limestones, promote extensive flooding and lateral
led to three publications the hydrogeology and karst
on planation. Urquhart (1958) also indicated that ‘cockpit
geomorphology of the limestones of the island, but karst’ is the hard
specifi- developed on pure, and fissile

cally the White Limestone areas (Sweeting, while ‘doline karst’ is those which
1956, 1957, limestones, present on

1958). Her 1958 publication was to become the seminal White & Dunn
are marly or impure. (1962) also suggested
work on the karstlands of Jamaica, and the first major the massiveness and
solubility, well-developed joints of

review of the limestone and of the White

geomorphology hydrology Limestone have been major factors in the

the island, including the first karst of karst

general geomor- development topography in Jamaica.

phological map of the northern part of Jamaica (Figure 2). Versey (1972) suggested that, in terms of the impor-
This stimulated a wider interest in Jamaican karst and of karst
tance lithological properties to
hydrology, three

established The broad

Cockpit Country as a geomorphological lithological divisions can be identified, each with

tropical terrain-type. She published later reviews and differing hydrological regimes. The first are
the
group
descriptions of Jamaican karst in her 1972 Karst chalks which,
book, Montpelier according to Versey, are char-

Landforms, where she also documented previously acterised rather than

unpub- by a primary, diagenetic, porosity.
lished work by other authors, notably Conrad Aub. Other with the of leads
This, together paucity fractures, to a

general descriptions of the karst of the island were pub- well-developed primary permeability, such that
ground-
lished in the 1950s Doerr & and
by Hoy (1957) Urquhart water movements occur through the body of the rock
-
192-

rather than via conduits (Versey, 1972; Gardner et al., The White Limestone Group formations are, for the

1987). Versey’s second rock division based on hydrologi- most part, extremely pure carbonates. Versey (1972) noted

cal characteristics is the recrystallized limestones and that the non-carbonate fraction seldom exceeds 2% and is

dolostones, representing the oldest formations within the generally less than 0.5%. Comer (1974) (Table 1) found

White Limestone Group. These rocks are characterised by the Middle Eocene to Lower Miocene White Limestone

no primary permeability and groundwater movements are Group to be on average 99.87 % pure calcium carbonate

entirely along fissures. Definite lines of groundwater flow and that the source
of the insoluble residue could have

have developed only along the largest fractures (Gardner been contaminationof the samples by bauxite or terra rosa

et al., 1987). Enlargement of joints and other lines of soil. Comer also examined the Middle and Upper Miocene

weakness is the result of directed dissolution and the carbonates of the Montpelier Formation (Table 2), which

development of secondary permeability. Versey (1972) contain several thin bentonitic interbeds and disseminated

characterised the rest of the White Limestone Group as a clays. These limestones were found to contain on average

‘rubbly limestone’, having a widespread nodular texture, 6.17% insoluble residue, chiefly oxides of Si, A1 and Fe,

with occasional hard bands. The variable lithologies in the form of bentonitic and montmorilloniteclays.

within Versey’s ‘rubbly limestones’ range from soft Day (1976) reported the results of acetic acid insoluble

which lime- residue experiments the limestones which indicate that

powdery textures are nearly impermeable, to on

the Browns Town, Walderston referred the

stone conglomerates which are highly permeable. (both now to

0/ % % % % % % %
/o

Insoluble Fe 2 Oj Si02 ai 2o 3 P2Os Ti0 2 CaO co 2

Residue

0.1275 0.0017 0.0275 0.0060 0.2072 0.0005 55.94 43.93

Table 1. chemical composition of Middle Eocene to Lower Miocene limestones of Jamaica (from Comer,
Summary

1974).

% % % % % % % %

Insoluble Fe 20 3 SiOz AhO, PzO s TiOz CaO COz

Residue

6.1715 0.47 3.179 0.975 0.069 0.0048 52.69 41.28

Table 2. chemical composition of Middle and Upper Miocene limestones of Jamaica (from Comer, 1974).
Summary

O/ O/ O/ Schmidt
/o /o /o

Formation Insoluble Dominant Texture Sparry Estimated Hammer

Residue Specification Calcite Porosity Hardness

Content Test

Montpelier 0.23 Fossiliferous Fine <10 <5 32.6

Micrite

Foraminiferal Fine <10 <10 30.4

Bonnygate
-

(herein, Montpelier) Micrite

Walderston 0.16 Biomicrite Fine-Medium 20-30 10-15 29.7

(herein, Moneague)

Brown’s Town 0.12 Biomicrite Fine-Medium 30-40 10-15 32.1

(herein,Moneague)

Somerset 0.08 Biosparite Medium 40-50 <10 33.1

Swanswick 0.09 Biosparite Fine-Medium 30-40 10-15 36.3

Claremont 0.03 - - “

(herein, Troy)

Dolomitic Medium 10 <10 42.3

Troy
Limestone

Table 3. material properties of formations of the White Limestone Group of Jamaica (from Day, 1976, 1979, 1982).
Summary
 -

193 -

Moneague Formation; Mitchell, 2004), Somerset and suiting geological structures have imparted significant
Claremont (now part of the Troy Formation) formations control the of karst landforms within the
on development
all contain less than one
considerably cent by weight White Limestone Group, least, the block and
per not
faulting
insoluble residue further examined
(Table 3). Day (1982) extensive east-west trending fault
systems, the anticlinal

the role of material in the of karst

properties development folding of the platform carbonates and the erosional

terrain in a number of localities within the Caribbeanand

breaching of this structure, together with
subsequent de-

central America, the basis of the

on purity, petrography roofing to form the Central Inlier. A detailed discussion of

and mechanical hardness of the limestones in relation to the structural and tectonic evolution of Jamaica is
geology
the karst landform type developed He also the
(Table 3). beyond scope of this paper and the reader is directed
described the White Limestone
as extremely pure; of the to recent reviews of the block and belt structures by Draper
samples tested, the Somerset Formation found to be Lewis
was (1987, 1998), & Draper (1990) and Robinson

the most
pure, with only 0.08% insoluble residue.
Day (1994). Essentially, Jamaica consists of three main struc-

examined the of the limestones and

petrographic nature tural blocks, separated by two northwest-trending graben
the samples he biomicrites micrites
analysed are mostly or structures (e.g., Draper, 1998). The blocks are composed
(Table 3), with low
a relatively percentage of sparry of Cretaceous
volcanics, siliciclastic
sedimentary rocks

calcite, though his samples of Swanswick and Somerset with minor and intrusive
limestones, granitoid rocks,
limestones classified The For-
were as biosparites. Troy capped by Tertiary limestones, and comprise, from east to

mation classified dolomitic limestone and dis-

was as a west, the Blue Mountain Block, the Clarendon Block and

played extensive recrystallisation. the rocks the Hanover Block

Texturally (Draper, 1998; Robinson, 1994). The

were found to be fine- to medium-grained, which is re- belts the blocks fault bounded and
separating are Tertiary
flected in the low estimated porosities of <15%. also in age. consist of the
Day They Wagwater belt of siliciclastic

provided a of results obtained from in situ the

summary sedimentary rocks, Montpelier-Newmarket belt com-

Schmidt Hammer testing and found the White Limestones

prising Tertiary limestones and the North Coast belt,
from hard the dolomi-
vary moderately to
very hard, with which abuts the northern
margin of the Clarendon Block
tic limestone of the Formation the
Troy being hardest, and consists of deeper water limestones.

recrystallisation having destroyed the original limestone The Lower Cretaceous to in

Palaeogene rocks, exposed
texture and resulted in increased cementation
strength. several inliers, represent parts of an
island arc system

Day (1982) attempted to correlate purity, petrographic In the Palaeocene Jamaica rifted
(Draper, 1998). along
characteristics and mechanical with terrain
strength type, major northwest-southeast-trending faults, when the

and found that the most striking correlation is between block and of the island
present graben structure was

terrain type and hardness of the rock demon- formed blocks of

types as
(Draper, 1998) producing different
strated by the Schmidt where ‘tower karst’ tends After cessation of the
Hammer, height. arc system, from the late

to develop on the hardest rocks, and ‘doline karst’ the

on Early Eocene to the Late Miocene, differential subsidence

softest. Day concluded that the role of occurred and

lithological varia- transgression led to the accumulation of

tion, and particularly hardness, as a measure of mechani- shallow carbonate sediments the main structural
water on

cal strength is deserving of attention. The correla-

greater blocks, although the Montpelier-Newmarket and North

tion between and terrain

purity, petrography type was less Coast belts underwent
greater subsidence, allowing for

striking for the White Limestone samples, but this is due carbonate sedimentation
deeper water
(Draper, 1998). As
to the fact that all are extremely pure and mostly micritic, well as controlling the
lithology of the limestones, the

and is also reflection of the small size.

a sample More tectonic activity uplifted and
exposed the island during the

recently, however, Pfeffer (1986) indicated that cockpit later Cainozoic. The uplift from the Late Miocene is most

karst only the limestones and that the

occurs on ‘pure’ likely the result of northeast-southwest which
shortening
cockpit hills the limestones.
are steeper on purest accompanied east-west, left-lateral transcurrent motions,
Limestones with marl tend to form less
layers steep resid- which can be seen as three large, east-west trending fault

ual hills and dolines, while the limestones support that

impure systems cut through the island
(Lewis & Draper,
doline karst only. In of the structural influence
1990; Draper, 1998). terms

The relatively low of the limestones is

porosity impor- on karst
topography, the Duanvale and the South Coast

tant, as directed, rather than diffuse, dissolution occurs fault systems are the most important, and the Plantain

along lines of weakness to Garden-Rio Minho

develop a secondary perme- system is less so.

ability along joints and fissures, the

leading to develop- On the Clarendon Block, the structural
response to the

ment of karst terrain through enhanced rates of dissolution northeast the

shortening accompanying transcurrent
where joints are more common. tectonics the scale anticlinal
was large folding of the

platform carbonates form

(Draper, 1998) to a major
westnorthwest-eastsoutheast-trending anticline. The tilting
Geological structure of the limbs of the anticline also provided the hydraulic

gradient necessary for much of the horizontal cave devel-

The overall tectonic evolution of the island and the re- in the of the island
opment centre Devel-
(Draper, 1998).
-
194-

of the anticline followed by a general south- parallel to flow direction and they tend to impart a strong
opment was

ward tilting of the block. The erosional breaching of the control, acting as lines of preferential groundwater move-

anticline and subsequent de-roofing, to

expose the Creta- ment (Versey, 1972). The Miocene to recent tectonic

ceous rocks of the Central Inlier, began about 9 Myr ago activity also produced many faults, which form numerous

lineaments satellite and

in the Late Miocene (Draper, 1998). According to Draper, scarps, seen as strong on imagery

the of the Clarendon Block, the aerial photographs (Draper, 1998).

at margins early Tertiary
faults reactivated The of the karst landforms in the White Lime-
northwest-southeast-trending were as origin
thrust faults. stone Group are closely related to the character and atti-

In the east of the island, most of the northeast short- tude of the limestones and their tectonic evolution (Ver-

has been accommodated by thrusting along 1972; Gardner et ah, 1987). One of the main controls
ening a sey,

northwest-southeast directed axis in the Wagwater and on the overall physiography of the limestones is their

Yallahs fault systems, which resulted in spectacular uplift block-faulted structure. This episode of block faulting

(Draper, 1998). This has resulted in the erosion of the occurred in the late Tertiary, and is more pronounced and

the north. the

overlying White Limestone and the only remnants are the younger in the south than it is in Basically,
northeast dipping beds of the John Crow Mountains karst features are superimposed on the block-faulted

These limestones form structural elements. According and

(Draper, 1998). a pronounced to Versey (1972)

escarpment on the eastern slopes of the upper Rio Grande Gardner et al. (1987), karst erosional
processes
have

Valley. accentuated the structures in some areas. In other situa-

The renewed in the Miocene exposed the tions much of the original has been masked by
uplift car- structure

bonate the blocks to karstification and deposition of alluvium, particularly in the interior valleys
platforms on ero-

sion. As far the karst is the most and poljes, which have been structurally depressed, rather
as concerned, important
block is the Clarendon Block where two series of faults than being the result of erosion.

each have the has

strong topographic expression (Versey, 1972). Hydrologically, faulting generated drainage

The east-west faults occur at the northern margin of the barriers and created distinct catchments within the karst.

block and especially form the Duanvale fault system, Some of the streams have eroded to base level and this has

which exerts strong geomorphological and hydrological resulted in a major modification of earlier flow which was

controls karst Similar subterranean (Versey, Gardner ah, 1987).

on development. east-west trending entirely 1972; et

faults to the south of the block, near to the coast, also have A major event was the erosional breaching and subsequent

a profound effect on karst landforms. Across the Claren- de-roofing of the anticlinal structure which now
is exposed

don the faults less perpendicular in the Central Inlier, which fundamentally altered the
Block, are more or to

those its and trend between northeast- hydrological regime of the karst the Clarendon Block.
along margins, on

southwest and northwest-southeast. Most of the major Surface drainage from the Cretaceous rocks was directed

faults have clear topographic expression across

the White onto and beneath the karst plateaus surrounding them.

the north of the karst modified from

Limestones. According to
Versey (1972), to Therefore, was an essentially

the Clarendon in the North Coast belt, faulting autogenic system, where karstification was dominated by
Block,

gives to folding in the deeper water carbonates of the diffuse autogenic recharge by vertical percolation of
way

Montpelier chalks, which display clear dips, and faults rainwater through the limestones, to a mixed autogenic-
lose their clear topographic expression. The general allogenic system, where additional large volumes of water

absence of faults in the Montpelier Formation leads to low passed through the limestones as concentrated allogenic
from the Cental Inlier.
secondary permeability, which in effect acts as a barrier to recharge, especially Versey (1972)
the northerly groundwater flow. It is also possible that suggested that the sediments contained within this allo-

of is caused reversal of the water would also have abraded the limestones before
impedance drainage by a dip or genic
the of the karst base (Gardner ah, 1987). being deposited in interior valleys or carried the sea.
by upfaulting et to

The other facies of the White Limestones rarely display At the more detailed scale, Wadge & Draper (1977a,

clear dips and when they do the angle of inclination is b) investigated the tectonic control on cave development

generally small (15°), steeper dips indicating nearby in the limestones. Brook & Hanson (1991), in a quantita-
The had the effect of tive analysis of cockpit and doline karst Brown’s
faulting (Versey, 1972). faulting near

the of the limestones, Town, parish of St Ann, showed that northwest- and
generally increasing permeability
where the limestones incoherent’, northeast-trending fractures exert a major influence on
though were ‘initially so

the faulting decreases the permeability as evidenced by the topographic development in the cockpit terrains and that

occurrence of fault-line ridges which have resisted disso- east-west-trending fractures are relatively more important
lution in the doline
(Versey, 1972). areas.

Within the fault blocks of the Clarendon Block, the

north-south trending faults are roughly perpendicular to

groundwater flow and do not affect it significantly. How- Case hardening

ever, where the regional groundwater flow is north or

south through the limestones, the north-south faults are Calcareous weathering crusts commonly occur on residual
 -
195-

limestone hills and in environments.

outcrops tropical A 2 m thick caprock has been observed on some quarry
Their significance for karst first in the faces in
was appreciated Moneague Formation tower karst at Kemps Hill,
Caribbean where the known
region, phenomena is as case southern Clarendon, but more
work needs to be done on

hardening (Ford & Williams, 1989). Most of the work on the importance of case hardening in karst development in

in the Caribbean
case
hardening comes from Puerto Rico Jamaica.

(Monroe, 1966, 1968; Ireland, 1979) and Cuba (Panos &

Stelcl, 1968). Both Monroe (1966) and Sweeting (1972)

interpreted the hard caprock to be due to Dissolution

secondary depo- processes and controls
sition of calcium carbonate from
resulting evaporation
soon after torrentialrain showers dissolve small No
amounts systematic work has been done on dissolution proc-
of limestone. Monroe (1966) considered case hardened esses and controls limestone dissolution in Jamaica.
on

surfaces to be thicker windward due to Two groups of controls limestone

on exposed slopes, on weathering can be

pronounced wetting and though

more
drying, Ireland generally identified, these being environmental and geo-

(1979) indicated that it is much uniform,

more with the logical controls. In terms of lithology and structure, the

of induration
zone closely following topography. Com- main controls have already been discussed and occur at a

monly, case hardening in Puerto Rico is 1 to 2 thick, variety of scales. Environmental controls fall into
m spatial
but it may be least 5 in Ivano- four broad
at m extent (Ireland, 1979). categories, related to soil conditions, hydrology,
vich & Ireland (1984) considered for limestones
that, topography and climate. A major control on limestone

consisting of than 50% the dominant and dissolution in

more fossils, case weathering general is the
presence or

hardening process is precipitation into solution cavities. In absence of a

soil cover and the nature of the soil, espe-
rocks with less than 20% fossils, the main process is it relates
cially as to
acidity, soil carbon dioxide produc-
aggrading neomorphism, a recrystallisation process re- tion and content, and organic content. Sweeting (1958)
sulting in the progressive increase of In both indicated under
microspar. that, tropical rainforest conditions, it

the is caused by
processes, precipitation carbon dioxide would be expected that soils would have a high production
degassing and it leads to a significant reduction in poros- and concentration of carbon dioxide. disso-
Accordingly,
ity of the limestones. lution of the limestones would be rapid, which
helps to

Although there has been no systematic work explain the intense of karst
completed development landforms and

hardening in Jamaica, it is Pfeffer may also account, in part, for the of wide-
on case common. occurrence

(1969) reported three distinct variations of cockpit karst in

spread terra rosa soils and bauxite deposits, which are

one small area in southwest Jamaica in relation to the commonly associated with the White Limestone
(Zans,
of surface crusts. One type of conical hill is
presence 1951a; Sweeting, 1958).
completely formed in firm second type is
limestone, a Measurements of the levels of soil carbon dioxide were

encrusted with external stalactites and travertine made the

covering by University of Bristol Karst Hydrology Expe-
the rock surface and on overhangs, while a third type is dition to Jamaica in 1967 and
(Smith, 1968, 1969a)
covered with weathering residuals and limestone
a crust reported by Nicholson & Nicholson (1969). Carbon diox-

overlying decomposed limestone The limestone ide levels in soils limestone varied from 0.6
particles. over to 1.9%

crusts, travertine and stalactites reported by Pfeffer (1969) with the highest concentrations in soils under
occurring
were taken to represent reprecipitated calcium carbonate, natural woodland with and humus
a high clay content

which at least in part forms case-hardened &

a caprock on (Nicholson Nicholson, 1969). It was found that the

some of the residual hills. On a smaller scale, McFarlane lowest concentrations under cultivated soil
occur bare,
(1980) reported the of
occurrence
razor-sharp, case- with low humus content in the surface
horizon, but the

hardened karren features on the karst landforms of the carbon dioxide than for soils
content was generally higher

Newport Formation (now referred to the Moneague For- in result of the

temperate regions as a
greater levels of
mation; Mitchell, 2004) in the Hellshire Hills (Figure 1). biological activity in higher temperatures (Nicholson &

As indicated by Ford & Williams (1989), case

hardening Nicholson, 1969). Day (1976) analysed the soil carbon

is especially important in the of karst dioxide in

development on content slightly to moderately acidic, red-brown

porous, mechanically weak and diagenetically bauxitic loams associated with in

young doline-type depressions
limestones, where it gives the rock more resistance to the Brown’s Town and also found that levels
area were

erosion, and reduces its and

porosity permeability. generally higher than those associated with temperate
The case-hardening effect is normal the
throughout conditions. Soil carbon dioxide content was higher when

White Limestone but it is noticeable the soil

Group, especially on was approaching saturation (Day, 1976). This

those units, such as the Moneague Formation, which would be due the fact that soil carbon dioxide diffusion
are to

soft, chalky to granular micrites when fresh. it the surface is

Thus, to
considerably reduced with increasing soil

would that be important

seem case
hardening may more in moisture content. Day (1976) also found that carbon

the development of karst the and dioxide levels increased in the base of the
topography on rubbly depressions and
chalky limestone formations, rather than the older recrys- with he
increasing depth, though concluded that the

tallised limestones, of which have been dolomitised.

some importance of a soil cover in facilitating limestone disso-
-
196-

lution, as it relates to carbon dioxide content, in Jamaica vial fans, steepheads and natural bridges. The formation

and development of the surface karst phenomena is closely

requires further investigation.
the role of soil in associated with the of the White Limestone
Trudgill (1977) investigated a cover hydrogeology
limestone in small in the and cave systems and underground rivers are
weathering a study Troy area on Group,
the southern of Cockpit Rates of important karst phenomena across the island. Karst
boundary Country.
obtained by placing pre-weighed lime- speleogenesis and the history of speleological work on the
weathering were

stone tablets within both red and organic soils, which are island is summarised in Fincham (1997); herein, research

rivers is reviewed
moderately acidic and with low carbonate content, and at on the cave systems and underground as

the soil to bedrock interface. Soil carbon dioxide content it relates to the development of the karst as a whole.

was also measured at the two White Limestone localities Localities referred to in the text
appear in Figure 1.
and was
found to be lower in the organic soils, especially

at the soil-rock interface, resulting from rapid diffusion to

the surface in the open leaf litter organic soil compared to Dissolution sculpturing
Measured
the more compact mineral soil (Trudgill, 1977).

erosion under both soil types where systematic work has been undertaken the small-
rates were greater than No on

soil cover was absent, and this is related to leaching and scale solution sculpturing of the limestones of the White

the acid nature of the soils, rather than soil carbon dioxide Limestone Group. White & Dunn (1962) noted that the

also surmised that the karst landforms covered with extensive

content (Trudgill, 1977). Trudgill major are karren,
could markedly affect the devel- and classified the solution sculpturing cockpit hills and
presence of soil
a cover on

of dissected terrain on the White Limestone, as towers as a type of spitzkarren. McFarlane (1980) reported
opment
erosion rates would be greatest where pockets of red and a small descriptive study of karren morphology at three

organic soils leading to differential rates of dissolu- localities within the White Limestone Group at Troy (Troy
occur,

tion. Portland and Hellshire Hills

Formation), Ridge
Based on a considerationof effective runoff figures and (Moneague Formation). The features are predominantly of

the values obtained for the hardness of springs resurging a linear nature in the Troy area, which contrasts markedly

from the White Limestones Smith al. with the sub-circular forms within the Portland
(Smith, 1969b), et more

(1972) suggested that the overall erosion rate of the White Ridge and Hellshire Hills. It was further suggested that the

Limestone area lies between 40 mm and 70 mm per 1,000 karren features in the Troy area are
better developed and

Smith this variation is almost have wider of morphologies. McFarlane (1980)

years. According to (1970), a range

due variation in effective runoff, which that the differences be dependant on

entirely to spatial postulated may

variation in rainfall factors.

is controlled largely by amounts. climatological and, more particularly, lithological

More recently, Draper & Fincham (1997) noted the

of several types of karren, especially solution

presence

pits, micro-pits and Hortonian runnels, which are wide-

Geomorphology

spread on exposed surfaces on many different kinds of

The karst geomorphology of the White Limestone Group limestone across the island. Other runnel or rill-like

can
be examined at several spatial scales. On the small- features in the form of rinnenkarren and fluted scallops

scale, there is a
of fine solution sculpturing present are also present (Draper & Fincham, 1997). Draper and
range

limestones which is the of Fincham further indicated the of spitzkarren

on many surfaces, product occurrence

dissolution of the limestones by slightly acidified rain- and cleft-like grikes on the thicker bedded and well-

water, and subsequent surface vertical trickle and flow jointed limestones of the Troy/Claremont Formation,

under the influence of these small- while honey-comb weathering features and littoral karren
gravity. Collectively,
scale features are referred to as karren and a wide variety forms occur on the White Limestone exposed near to the

of such features date coast.

are present, though to
very little

systematic work has been undertaken on the phenomena. The most common solution sculpturing features iden-

Work completed to date on karren features is summarised tified on the south coast limestones, specifically the

there is wide range of karst be weather-

below. At the larger scale, a Moneague Formation, appear to honey-comb

features, from cockpit and doline karst, both largely the ing phenomena (Figure 3) and rain pits. Although spitz-
of vertically-directed dissolution, and karren is hydrodynamically controlled linear
product tower karst, present,
tend be less The author agrees
which is mainly related to laterally-directed dissolution, to sculpturing to common.

ridge karst and poljes, which have a strong structural with the earlier observations of McFarlane (1980) that

control. Poljes also induce horizontal water movements at rillenkarren may be more prevalent on the older Troy

or near to
the regional groundwater table. There is also a Formation, but more systematic work needs to be com-

of fluviokarst, including dry and blind valleys, pleted in this regard.

range

ephemeral limestone gullies and valleys with relict allu-

 -

197-

Figure 3. Honeycomb solution weathering on ridge karst in the Moneague Formation, southern parish of St Elizabeth.

Figure 4. Cockpit karst in northwest Cockpit Country, parish of Trelawny, showing well-developedpreferred orientations due to

faulting and joint-patterns (photograph courtesy of Mr Jack Tyndale-Biscoe).

-
198 -

Doline karst to contain semi- permanent ponds which fluctuate ac-

cording to supply of runoff.

Doline karst is an important landscape type on

the Mont- Day (1976) dye-traced the water sinking from the

pelier Formation and is also associated with areas of enclosed depressions to the dry and ephemeral valley
low annual rainfall The traversing the Montpelier Formation limestones
relatively (Sweeting, 1958). systems

of doline karst the White Limestone the north of the The marked water
largest area on Group to study area. com-

altitudes of 250 100 which

occurs in the north central part of the island (Figure 2), a monly reappeared at m to
m,

few kilometres inland from the coast, especially where it gives a

vertical height of underground flow in the order of

is underlain the soft, chalky Montpelier 150 Lateral distances of flow ranged from 0.5 km to
by relatively m.

limestone (Draper & Fincham, 1997). Sweeting (1958) 5.0 km and Day estimated travel times of between 0.25 km

suggested that lithological and climatic conditions are to 0.5 km per day, which supports the idea that flow

for the development of doline karst along the through conduits rather than by diffuse percolation.
responsible occurs

north coast, though she did indicate that dolines occur

in In most cases, the water sinking from the depressions
the Moneague area, despite a fairly high annual rainfall. contributed to the flow in one valley only, but the main

According to Sweeting, the dolines on the Montpelier sources of flow to the valleys are not the depressions under

Formation shallow and have but the karstified limestones of the plateau to the
are usually an open v- or study,

funnel-shaped cross section. Similar shallow karst depres- south of the study area (Day, 1976).
sions in the around Brown’s Town, of St indicated that it is to infer the
occur area parish Day’s study not possible
Ann in the north of the Harbour Mountains dominant in the formation and subsequent devel-
Dry (Day, process

carried morphometric and hydrological opment of enclosed depressions. He suggested that small
1976). Day out a

of the enclosed for which he careful scale collapse provide the initial form, which is
study depressions was may

not to use the term doline, though he asserted that, ‘Other subsequently modified by systematic solutional activity.

workers might describe these simple depressions by the The enclosed depressions generally show a regular distri-

term doline or sink’. The depressions examined by Day bution which could suggest solutional activity, as collapse

are laterally extensive and shallow, and some

individual events
may produce a more random distribution pattern,

depressions occur
within larger, more complex depres- while the control of structure on depression long axes

sions, though their complexity cannot be distinguished by orientation could also be suggestive of solution, though
size alone. The depressions range in diameter from <100 there is no reason why subsurface solution should not

m to over
1 km, with a mean diameter of 219 m. Their result in preferred orientation of collapse features (Day,

depth range is also variable, with a mean depression depth 1976). In a later paper, Day (1984) attempted to predict

of 18.4 m; although two of the largest depressions in the the location of surface collapse within these karst depres-

area of study are >30 m deep, many of the depressions are sions.

relatively shallow. Day found that the depressions are Draper & Fincham (1997) indicate that doline karst

on the
characterised by low angle slopes within to tilted plateau in the southwest part
a range of 2° also occurs areas

12°. The found be 12.5 per km". of the island which are underlain by the relatively soft,
depression density was to

Day also found a strong

correlation between depth and rubbly and mechanically weak Newport Formation (now

which considered unlikely to have devel- referred to the Moneague Formation; Mitchell, 2004).
diameter, was

chance and which relate the internal and Fincham, the lithology is
oped by must to slope According to Draper too

development of the depressions. Many of the depressions weak to support the steeper and more pronounced residual

show tendency towards orientation in hills associated with cockpit karst. Dolines and sinkholes,
very strong
a an

with other weak orientations in circular sub-circular in plan form and varying in di-
east-west direction, an to

northwest-southeast direction. Day from few metres to 1 km,

approximately (1976) ameter a over are common

the direction with the south Manchester plateau and westwards into
interpreted east-west as corresponding across

the eastward extension of the Duanvale fault zone,

which southern St. Elizabeth. There is a wide spectrum of forms,

orientation of between 71° from hollows and

crosses the study area with an ranging saucer-shaped to funnels, cones

and 80°, while a secondary group of faults are orientated cylindrical pits, and they may occur as isolated depres-
between 310° and 320°. Stream channel networks within sions the terrain in clusters. Within this
or pock-mark
also and the threefold subdivision of doline karst be related
the depressions were analysed depressions area, a can

ranked in relation to the network order. This was com- to the characteristics of their associated positive relief

with size and it found that depres- features.

pared depression was

sion area increases with increasing channel order. It was The first doline karst category is typified by circular to

from 100 in
found that the channels, which are commonly up to 2 m sub-circular depressions, ranging m to 400 m

and 4 contain surface runoff only times of but around 150 m to 250 m,
deep m wide, at diameter, more commonly
and/or intense rainfall. found that which conical to bowl-shaped, to 30 m to
40
prolonged Day water are up m

down the channels ponded in the lowest parts of deep, though usually shallower at around 20 m to 25
flowing m,

removed subsurface with depth to width ratios of <1.0. The intervening resid-
the depressions, before being by
found ual hills
drainage at an outlet. Some of the depressions were form small cones, <25 m high, with similar
 -
199-

diameters the and in

to depressions occasionally occurring extend to over 150 m deep and have a diameter of
up-
beaded arrangements. wards of 1 km. The associated residual hills are broadly
The second type of doline karst is the conical in
common across shape, though some are elongated, 30-130 m

southern Manchester plateau and consists of enclosed and 1 km in diameter. In the

high up
to
larger cockpits,
depressions associated with elongated hills. The dimen- there is a marked break of slope at the base of the hill-

sions of the relief elements similar the doline karst while in the smaller the floor and sides
are to slope, depressions
described previously, but the hills 300 long, into another
are up to m grade one (Sweeting, 1972). Although distin-

and commonly orientated northwest-southeast and north- between the and hills of cockpit karst
guishing depressions
south, probably in response to major joints and other is since
essentially meaningless, they are not
separate
structural controls. landforms, but integral components of the surface pitting,
A third type of doline landscape is characterised by individual cockpits described surrounded by
are as being
enclosed depressions with subdued <10 three
hills, generally m or more residual hills of a similar elevation to the

to 15 high, where the sinkholes dominate the of the When the

m landscape. depths depressions. depressions are

Generally the in clusters. Elsewhere delimited the basis of their

depressions occur on topographic divides, the

across southern Manchester and southern St residual hills

Elizabeth, surrounding and ridges connecting the

isolated dolines be identified. Some are shallow constitute cellular network

can large, depressions a (Day, 1979)
features, 1 km to 2 km in while others termed karst
diameter, are polygonal (after Williams, 1971, 1972a). The

narrow and deep, with a depth to widthratio of 1.2 2.0, which

to depression side-slopes are convex, gives the overall

commensurate with collapse origin and possibly linked Che-

a landscape a star-shaped or polygonal plan, though
to well developed cave systems. noweth & that there is
Day (2001) suggested a range of
It is likely that a number of processes are responsible from
cockpit morphologies complex star-shaped patterns
for the formation of the dolines across the south Man- to circular forms. The with
simple cockpits are contiguous
chester Plateau and southern St. Elizabeth. The
morphol- a
clearly identifiable col or divide between each depression
of the shallower, and
ogy clustered, cone- bowl-shaped (Barker & Miller, 1995; Miller, 1998), forming corridors

dolines suggests solutional where solution is and between hills that

a origin, passages connect adjacent cockpits,
focussed along lines of weakness and where the vertical though other divides form which less
saddles, are pro-

permeability of the limestones is

spatially variable. The nounced notches between adjacent hills (Chenoweth &

steeper-sided, more isolated dolines probably the result Canter estimated

are Day, 2001). (1987) a frequency of 15
2
of collapse, though many of the depressions likely
are to cockpits per km in the area around Quickstep, with an

be of compound origin. estimated 20 cockpits per knT in the of

a
centre Cockpit

Country, totalling about 4,000-4,500 depressions.

The bounding slopes of the cockpit depressions are

Cockpit karst indicated

extremely irregular, although Sweeting (1958)

they average about 30°-40°, consisting of chemically

Cockpit karst (Figure 4) consists of a succession of cone- weathered and honeycombed blocks and scree. Where

like hills and intervening enclosed bedrock is the sides

depressions (Sweeting, exposed are steeper and form cliffs

1958). Cockpit karst in Jamaica has also been termed and precipices (Sweeting, 1972). Aub (1964a, b, 1974a, b)
kegelkarst, gerichteter karst examined in detail the of different slope elements
(directed karst) (e.g., array

Lehmann, 1954) and cone karst (Sweeting, 1958, 1972). associated with the and six
cockpits recognised types of

Versey (1972) considered the term kegelkarst inappro- These defined Aub
as hillslope (Figure 5). were by (1964a in

priate due to the fact that the Jamaican cockpits modi- staircase small
are Sweeting, 1972) as; slopes, comprising
fied of the and vertical
by an over-deepening depressions. Cockpit ledges steps, the latter being up to 2-3 m high
karst is restricted to the fissured and and where
hard, recrystallized bedding planes are conspicuous; broken cliffs,
White Limestone and, previously indicated, it does of higher
as not consisting steps and less uniform ledges, where

appear on the chalky Montpelier limestones, nor does it the steps intersected widened
are by joints; steep even

occur on White Limestones which associated with of limestone covered with loose talus
are slopes honeycombed
bauxite deposits, as on the Manchester plateau and of varying
(e.g., blocks; major cliffs, height, but undercut by

Sweeting, 1958, 1972). The overall landscape is highly horizontal notches

a to depths of 2-3 m and often associated

irregular combination of positive and negative relief with cliffs of similar but without undercut-
springs; height
elements of & and
roughly equal prominence (Chenoweth ting; scree slopes covered with small limestone frag-
Day, 2001; Day, 1979, 1982). This type of karst landscape ments.

is the most widespread in Jamaica Chenoweth and Day (2001) identifiedtwo major slope
(Sweeting, 1958),

covering some 60% of all karst terrain their cover, talus and soil
(Day, 1979), types according to covered

Cockpit Country being the type locality for cockpit karst slopes. The former are the more common and it was

(e.g., Sweeting, 1958, 1972; Versey, 1972) (Figure 2). suggested that tree-tipping and bedrock disruption as a

The cockpit depressions Sweet- result of windthrow

are deep according to
may play an important role in the

ing, with depth of 90-120 and maintenance of the talus

an average m, though they may development slopes.
-200-

Figure 5. cross-section of cockpit karst, showing the main elements of relief (from Aub, 1964a in I
Stylised Sweeting, 1972). Key:
=

glade floor, with soil and projecting rock; 2 =

vertical shaft, joint-guided sinkhole; 3 =
cliff over sinkhole, with gentle solution

scallops and other karren phenomena; 4 =

scree slope; 5 =
undercut notch at base of major cliff and associated with softer

limestones; 6 =
high cliff with undercut base, with scalloped or pitted surface, often with a case-hardened surface; 7 =

steep even

slope; 8 =
hill summitwith projecting rock and loose fragments, some summits are planar, others more conical; 9 =
pinnacles sum-

mit surface, dissected into pinnacle and cleft ‘castles’ (may be isolated or in groups and up
to 10 m
high); 10
=
cliffwithout under-

base, rising from a bench in solid limestone; 11 staircase of steps 3 high and ledges wide;
cut =
slope, consisting up
to m
up
to 5 m

of the undercut and the covered with large limestone blocks; 12 covered with soil and boul-
many ledges are slopes =
cockpit floor,
ders; 13 sinkhole, vertical shaft with rubble floor and solid limestone walls, indicating the cockpit is 14
collapse feature;
=
not a

cockpit floor, with some joint solution and abundant rubble; 15 =

broken cliffs, with honeycombed walls, narrow
bedding plane
ledges and solution-controlledjoint plane openings; 16 cleft topography, with 17 small
pinnacle-and large boulders; cockpit,
= =

with soil covered slopes and surface drainage channels; 18 =

sinkhole, impenetrable sink among boulders at the lowest end of the

drainage channels.

Chenoweth and also identified two basic hilltop 1977b). Aub (1964a) also indicated the cockpits he studied
Day
in the in the around show apparent fault and that the
morphologies cockpit karst area no or joint guide mor-

Windsor; some are dome shaped and rocky, while other phology of the slopes and cones is variable.

hill summits are flatter in plan view. Smith et al. (1972) also stressed that the cockpit areas

Normally, the cockpit depressions and conical hills are show considerable variation in morphology and display

more or less symmetrical, but some asymmetry occurs. irregular slope forms, though they have a mean slope
Two explanations have been presented to account for the angle of about 30°. According to Sweeting (1958), the

The first is related the of the summits of the hills tend reach level, which she
asymmetry. to dip to an even

limestones, that is, when it exceeds more than a few interpreted as a structural surface, or part of a more exten-

degrees, the cockpits become steeper on the up-dip side. sive peneplain surface, dissected by cockpits of variable

The second is differential dissolution rates in response to depth, representing one-cycle and multi-cycle landforms.

exposure to the oncoming trade winds, as occurs near the However, Smith et al. (1972) concluded there is no corre-

north coast where north facing slopes are less steep lation in the elevation of hill summits or depression floors

(Sweeting, 1958). Similar explanations have been pre- except where the depressions have eroded down to the

sented for mogote asymmetry in Puerto Rico (e.g, Day, underlying Yellow Limestone.

1978b; Monroe, 1968) and a more thorough discussion of Most of the depressions ‘drain’ towards a deep vertical

residual hill asymmetry is presented in the section on shaft in their lowest part eroded in solid limestone. In

tower karst. The cockpits and residual hills are often some areas the shafts may have a number of narrow

in lines faults entrances which unite at a relatively shallow depth (Smith

arranged following or jointing (Sweeting,

1958) (Figure 4) and the cockpit karst often occurs

in the et al., 1972). Smith et al. indicated that it is possible to

form of winding sinuous chains or ridges separated by gain entry to some of the shafts, though others are blocked

glades (Sweeting, 1972) (Figure 6), though elsewhere no

by debris or too narrow for exploration, contending that it

structural be identified & is for the shafts to be 30

guide can (Wadge Draper, rare explored beyond m
depth.
 -201 -

Figure 6. Cockpit karst in the north of Cockpit Country, of of dissolution widened glades
parish Trelawny, showing development
and pocket valleys (photograph courtesy of Mr Jack Tyndale-Biscoe).

sinkholes that connect with deeper horizontal caves.

However, Baker et al. (1986) explored several of these to

depths of 80 m, without intersecting the water table, and

Canter (1987), in an attempt to estimate the frequency of

pits, indicated they reach up to 70-80 m deep, but without

significant lateral passage development with most shafts

bottoming out in narrow joints or

debris chokes.

further delimitedcockpit karst

Sweeting (1956, 1958)
into two variants which she termed ‘true and
cockpit’

‘degraded cockpit’. According to Sweeting, cockpits

become degraded when due ‘slack-
deepening ceases to a

ening’ of solution processes. Degraded cockpits are

shal-

lower with slumped and gentle side slopes, the overall

reliefof the landscape becomes more ‘subdued and roll-

ing’, and there is a greater dissociation between surface

and groundwater circulation (Sweeting, 1958). The basic

morphological difference between the two relates to the

accumulation of debris within the depression. The bases of

‘cockpits’ contain little superficial material (Figure 7) and

a near vertical shaft is evident, whereas in the ‘degraded’

form the bases of the depressions bauxitic
are occupied by

material, commonly up
to 10 m deep (Day, 1979). De-

graded cockpit karst exhibits a wide variety of forms, in

relation to the degree of accumulation of ‘soil’, from a

nearly flat bauxitic plain with occasional protuberances of

limestone, through a vermiform pattern of limestone

ridges separated by bauxite infills, where

Tower in
to cockpits only
Figure 7. Troy Formation showing basal undermin-
the central portions of the depressions are filled with
of Lluidas Vale, parish of St Catherine.
ing, west

bauxite (Smith et ai, 1972).

Versey (1972), as indicated earlier, questioned the use

Versey (1972) indicated that many cockpits are floored of the describe karst and
term kegelkarst to cockpit con-

with boulders, while others have solid floors with vertical sidered the be
term to more aptly applied to depressions
-202 -

which have undergone ‘overdeepening’, where the

not underground drainage. Day also concluded that the de-

side-slopes are relatively gentle and where the depressions graded cockpit forms are currently experiencing the

filled with bauxite. main terrain

are Versey proposed two greater solutional attenuation as dissolution is focussed

types of the karst uplands, cockpit karst and kegelkarst. He beneath the superficial deposits. However, Aub, in a

considered that ‘degraded cockpit karst’ represents a personal communication to Versey (1972), considered that

karst which has suffered either

cockpit uplift or freeing of an important difference between cockpit karst and karst

its drainage. areas elsewhere on the island that there has

was never

A number of contrasting observations have been made been a bauxite cover in the former and dissolution has a

about the hydrological characteristics of the karst. effect than in where the
cockpit more powerful areas depressions
Zans (1951b) noted instances in which the depressions infilled with bauxite, thus dissolution and
are are inhibiting
partially flooded following heavy rainfall and similar impeding deepening. It is clear that more work needs to be

observations have been made by Sweeting (1958), who done the between the bauxite and
on relationship cover

noted temporary ponded water, though frequent flooding karstification of the limestones, but, as indicated in an

of cockpits only when extend level earlier work

occurs they near to a section, subsequent to that of Aub has shown

close underground circulation. that in the

to water Versey (1962) most cases presence of a soil cover enhances

described the of after dissolution

flooding cockpits heavy rainfall, (Trudgill, 1977).
where water issues from basal sinks which other times A number of
at explanations have been given for the

drain the cockpits, and Aub noted the presence of the but
(1974a, b) origin cockpits, they basically fall into either

of small channels central with solution-related or

leading to a depression collapse-related categories. The origi-
temporary surface drainage after rainfall. Aub nal for the
heavy collapse explanation cockpits was presented by
(1964b) found the receive
depressions 13-15% more Sawkins (1869), where in the first detailed
descriptions of

precipitation than the hill summits and that The the ascribed the
suggested Cockpit Country, depressions were to

vegetation rafting of rain water into the cockpits by sinking of the of the
water through cavernous structure

throughfall led to dissolution being greater there. Smith which forced its way and
et limestones, through subsequently
al. (1972) noted that the presence of near-vertical shafts in removed beds of shale and sand of the
underlying ‘Trap-
the bottom of cockpits would suggest surface runoff, but Series’. The cavities formed and
pean were unsupported

they found no evidence of recent flow in the shafts and collapsed form the Some
no to cockpit depression. thirty
surface runoff, other than in few locations where thin
a a
years later, a lengthy reassessment of the geology of

soil is
cover present, when small channels develop and Jamaica was published by Hill (1899), who, although there

become locally active. also found well little interest in the

They no developed was descriptive karst, ascribed the

channels in the degraded cockpits and evidence of of dissolution rather than

no origin cockpit depressions to

flooding. In contrast, Day (1977, 1979) noted occasional collapse. Danes (1909, 1914) was
the first worker to

basal ponds in ‘degraded cockpits’ and small channels the that the
cut widely adopt theory cockpit depressions
into the superficial bauxite but extending of the limestones
cover, not onto originated predominantly by dissolution
limestone bedrock, where vertical fissures and other lines of weakness, he did
water disappears by along though
infiltration. Similarly, in the slow vertical concede that
‘cockpits’, they may be further deepened by localised

percolation is the dominant hydrological of which would help further

process, no collapse caves to
enlarge
surface flow occurred, either dispersed in surface them. evidence such the distribution and
or Using as regular

channels, the shafts inactive and relict linear arrangement of the cockpits, with their
are essentially together

(Day, 1979). dissociation from subsurface groundwater circulation,

Surface drainage channels similar to those in New Sweeting also concluded that the
(1958) cockpits are

Guinea cockpits (Williams, 1971, 1972a, b) virtually formed by dissolution with subsequent
are enlargement by
absent in the Jamaican cockpits and are confined to the collapse along fissures. that the
Versey (1962) suggested
superficial cover in the ‘degraded’ forms. According to physiography of The Cockpit Country is closely related to

the shafts
Day (1979), once acted as drains at the foci of the hydraulic and abrasive action of floodwaters in under-

surface channels under conditions of either lower and that solutional and
secon- ground drainage systems, me-

dary permeability and/or a wetter climatic It is chanical action of confined floodwaters

regime. was responsible

highly likely that the vertical shafts developed due for the enlargement of with
to cockpits, along collapse.
surface drainage and, according to Gardner et al. (1987), Versey also noted that the where
deepest cockpits occur

the vertical shafts in Jamaica

may have developed because underground water movement is close to the surface, and

of the greater relief available, due he further refined this

to more recent tectonics, theory by suggesting that a rise in

or because the limestones dense than in Puerto levels rainfall brings circulation
are more water during groundwater

Rico, where shafts seldom found. It could also be the close the bottom of the
are to cockpits, eroding them, and that

case that shafts in Puerto Rico are obscured by debris. then the and
subsequent collapse deepens depressions
The hydrology of the is dominated by the soil
cockpits now removes cover (Versey, 1972). According to

slow and diffuse percolation, which is the Versey, ifthe cockpits have developed
only contempo- over a long period
connection between the and of time, then the of erosion has matched the
rary cockpit depressions rate rate of
 -203 -

uplift. In areas where uplift has been more rapid, water size and shape of the intervening residual hills (Figure 8).
circulation becomes be then dissolu- The three types with conical
too deep to effective, are: cockpits hills; cockpit
tion takes the main erosive process and soil depressions with small hills up 50 high,
over as a cover
elongate to m

in the to the of in north-south

develops depressions leading development aligned a general direction, probably along

‘degraded cockpit karst’. faults and major joints; and with elon-
depressions larger
Other workers have suggested that dissolution alone is
gate hills in a beaded arrangement, grading into ridges. A

responsible for cockpit formation.Aub

(1964b) suggested recent
survey of ‘cockpit karst’ terrain in the south of the

there is very little evidence of any widespread collapse island, in of Westmoreland, southern St Elizabeth
or parts
mechanical action, and that there is no correlation of and Manchester (Sir William Halcrow & Partners,
1998),

cockpit shape or
size with height above the underground classified the cockpit landscape based on
the or
presence

drainage. Aub further indicated that the depressions absence of an integrated surface drainage pattern across

receive more throughfall than the hill summits and sug- the terrain, on the size and shape of the intervening hills

gests that solution could be greater in the depressions and on the extent of the enclosed depressions, especially
which makes them self-perpetuating features. Aub also whether or not they contain a residual soil infill. In this

link the development of of karst to the east of the Cabarita

attempted to ‘degraded cockpits’ study, an area cockpit
and the of bauxite, where he considered that of is associated with
occurrence River, parish Westmoreland, ephem-
the shallower cockpits the result of greater sediment eral and
are drainage lines, comprising gullied slopes dry,

deposition within them, and they associated with short valley often interior
are parts systems, draining to larger
of the Central Inlier that were breached first. He further valleys and poljes. Most of these lines
drainage are proba-

suggested that if this were

the case
then the ‘degraded bly inactive and the landscape is partly relict terrain.

cockpits’ be development from ‘normal’ cock- indicated

may not a Versey (1972), as earlier, suggested that the

pits, but have a different history. Smith et al. (1972) and term ‘kegelkarst’ is an inappropriate term for the over-

Day (1979) also supported a dissolution origin for the deepened cockpits within The Cockpit Country heartland,

development of cockpit karst. and that it is more appropriate to use the term for the

Two geomorphological aspects of the cockpits which ‘degraded’ cockpits around its margins. It is evident that

have been largely overlooked are the importance of me- the terminology used needs to be cleared and to help
up,

chanical breakdown to their development and modifica- overcome terminological confusion some workers have

tion, and possible fluvial inheritance. Many of the cockpit turned to morphometric analysis to develop a more objec-

slopes are strewn with angular, block- and boulder-sized tive classification of the terrain (e.g., Day, 1979, 1982).
talus and which be explained by simple This and other the of
scree cannot morphometric approaches to study
dissolution and the importance of mechanical cockpit and other karst terrain
processes, types in Jamaica will be
breakdown actively modifying the landscape has been discussed in a later section.

understated. These talus-strewn slopes may be the result

of bedrock disruption by the windthrow of trees (Che-

noweth & Day, 2001), while dissolution along joints and Tower karst

could increase the of the

bedding planes susceptibility
cockpit slopes to rockfall. Some of the cockpit depressions Lehmann (1954) described the geomorphology of tower

have a beaded arrangement, others are linear in nature, karst (Tumi karst or Mogote karst) in Jamaica, which

while the karst often residual hills scattered

cockpit landscape occurs as winding comprises a
landscape of across a

chains ridges separated by sinuous channels and flat plain Some of the residual
or glades, relatively (Figures 9, 10).

particularly the margins of The Cockpit Country. The isolated and others in groups.
at towers are occur They occur

geomorphology suggests possible fluvial inheritance to the White but karst is

a on crystalline Limestones, tower

the cockpit landscape, at least in parts. Aub’s much less widespread than cockpit karst
some (Sweeting, 1958)

(1964b) suggestion that ‘true’ and ‘degraded’ cockpits (Figure 2). According to Sweeting (1958), towers are

have a different evolutionary history also requires further steep-sided hills which slope to 60°-90°, rising up to
up

investigation. 100-150 m above a flat alluvial plain. The towers are sub-

From the foregoing, it is clear that cockpit karst has a conical, though many have flattened tops and appear

wide variety of form and several interchangeable tabular in profile. The broken
terms tower slopes are frequently
have been used to describe the landscape, including ‘cock- and devoid of a soil except for isolated pockets.
cover,

pits’, ‘kegelkarst’ and ‘cone karst’, while ‘true’ cockpit Many tower bases are undermined and display undercut

karst and ‘degraded’ cockpit karst were additionally notches associated with well-developed foot-caves and

introduced differentiate between cockpit depressions A of

to springs (Figure 7). well-developed ring sinkholes,
devoid of, with soil and debris Pfeffer which often become flooded in the
or a cover. (1969, wet season, commonly

1984) identified distinct variants of cockpit (Kegelkarst) occurs

around the base of the undermined and over-

in southwest Jamaica and, more recently, Miller (1998) steepened towers. Indeed, some of the towers are
situated

mapped three principal classes of cockpit karst in The in small enclosed depressions, similar to those described

Cockpit Country between Windsor and based the alluvial in Belize 1975, 1976).
Troy, on on plains (McDonald
-204-

Figure 8. Geomorphological map of the area of Cockpit Country between Windsor and Troy, parish of Trelawny.
 -205 -

Figure 9. Tower karst in MoneagueFormation with basal pedestal in plane limestone, south Manchester Plateau, parish of Man-

chester.

According to
Sweeting (1958), two particular geologi-
cal conditions favour tower karst development. They occur

near to
the base of the White Limestone at the junction of

the Yellow Limestone and within the crystalline facies of

the White Limestone Group in close association with the

marly Montpelier Formation. They are also associated

with rapid spring-head recession, which gives rise to the

flat-floored, steep-sided and steep-headed ‘pocket valleys’

which accompany tower karst on the northern margins of

Cockpit Country (Sweeting, 1958). Sweeting (1972)

further identified two main types of tower karst, which

both occur within the White Limestone Group. One type

of tower occurs as a visible remnant of limestone sur-

rounded by alluvium, beneath which is a planed limestone.

The other tower type is developed on a plain of non-

limestone rocks, as occurs where towers within the White

Limestone form outliers on the much less pure Yellow

Limestone, though they may also occur where the White

Limestone is in faulted contact with older Cretaceous

rocks, especially in the Central Inlier. & Fincham

Draper

(1997) indicated that isolated towers of White Limestone

formed by erosion down to the less soluble Yellow Lime-

stone or older Cretaceous rocks are best developed around

Maroon Town, on the western edge of The Cockpit Coun-

try, while the other tower type, with residual hills pro-

truding through an alluvial blanket in a glade, polje or

plain is well developed in the Cave Valley area and on the

western edge of Lluidas Vale. Draper & Fincham (1997)

10. Case hardened tower karst in MoneagueForma- have observed many examples of shelter caves at the base
Figure
with solutional pockets and anastomoses,
tion, south Manchester Plateau, parish of Manchester. of the towers,
-206-

suggesting solutional undercutting at the base of the east-west

asymmetry, where the west facing slope (60°) is

towers. about 25° Pfeffer

steeper. suggested aspect related mois-
In addition to undercut towers with notches, Sweeting for the asymmetry, where
ture changes as an explanation

(1958) identified a second tower type based morphol- rainfall evaporates the
on more quickly on east-facing slope
which she to be karst’ and with it corrosive action is Pfeffer also noted
ogy, interpreted ‘degraded tower impeded.

(Figure 2). These are towers where the lower slopes dis- that the west-facing slopes have a conspicuously thicker

play slumped sides, and the sharp break of slope at the case-hardened caprock.
undercut between the tower base and alluvial floor karst in Puerto
disap- Similarly, east-west tower
asymmetry

resulting from a decrease in limestone dissolution Rico has been linked to the dominantwind direction and
pears,

at the foot of the tower due cessation of of the resultant differential dissolution and reprecipitation
to a flooding of

alluvial area. calcium carbonate. Gentler slopes on the eastern sides of

Some towers show little evidence of and have been related the increased
undermining, towers to
exposure to

contain colluviated footslopes or of blocky lime- rainfall and dissolution while the
aprons (Thorp, 1934), asym-
talus and of 25°-30°. These has also been linked
stone
scree, sloping upwards metry to variation in the extent of

the edge of Lluidas Vale case-hardening

are seen on western at
higher (Monroe, 1966, 1968). In this case, the

altitudes than the current level of alluvial cover within the rim of indurated is thicker the side and
caprock on eastern

vale. They be similar to towers at interfluve locations the thin case-hardened surface the tends
may on opposing slope
in Belize, where McDonald (1979) showed that the the
mor- to spall-off slope, leaving a vertical cliff and over-

phogenesis of located interfluvial is in

towers on plains hanging slopes. Day (1978b), a study of the mogotes on

controlled by the progressive lowering of the alluvial the Limestone in northern Puerto
Aymamon Rico, showed

surface by erosion due to overland flow, and the that

tower- tower asymmetry is not as simple as previously sug-
base landforms revealed gentle, colluviated bare the steepest do the
are or gested, as slopes not always occur on

bedrock footslopes. As the interfluve plains eroded and western side of the but found at different
are towers, are

lowered, bedrock slopes formed around the locations around the towers, that
are towers including many
towers

with a slope angle approaching 20°-60°, which rise north-south asymmetry. linked the asym-
to displayed a Day
meet the base of cliffs about 10-50 m above the surround- concentrations of sinks in locations
metry to particular

ing plain. The towers on the interfluve plains therefore where became undermined by both corrosion
are tower slopes
not the result of tower undermining and lateral planation, and mechanical action. The is also the
geological dip to

but are formed by the vertical dissection of the plain north in this and it might be that
area expected water

sediments. would be in that direction,

movement leading to a con-

The most common tower karst in the White centration of sinks the steeper southern slopes. Day
on

Limestones involves residual hills from either considered that the of the
protruding general dip strata
may also

a planed carbonate surface veneered by alluvium or from contribute to the asymmetry, though this is <5°.
generally
alluviated surface burying limestone. In the of the Bull
an a pinnacled Many case towers near Savannah, parish of St

of the towers have been case-hardened and two the author considers the
morpho- Elizabeth, Jamaica, present

logical types of tower associated with an alluvial infill asymmetry is related to the dip of the limestones, which

from allogenic streams be recognised, namely circular about 10°-15° the and southeast. the
can dip to east Thus,
and elongate (asymmetric) towers. Both types of tower steeper slope facet is the side of the and
are on up-dip tower

formed by laterally directed solution at or near to the water

suggests a strong structural control in the morphogenesis

table, coupled with aggressive floodwaters from allogenic of the towers.

rivers. Symmetrical (circular) towers commonly to Small

are up towers, from 5 to 20 m high and 100 m or less

100 m elevation and 1 km in diameter with steep, almost in with and

diameter, together asymmetric elongate forms
vertical slopes and with associated sinks and depressions similar the hills of Puerto Rico,
to Pepino are common on

around their bases. They may occur singly or in small the plateau surfaces and slopes of southern Manchester

clusters. Asymmetrical tend be and southern St

or elongate towers to larger Elizabeth, developed chiefly on case

and be 1 km with width ratio in hardened Formation limestones.

can over long a length to Moneague They were

excess of 2.0. Asymmetric towers occur where described Pfeffer where small isolated
one slope by (1967), towers

facet exceeds the opposite aspect by more than 20°. There are associated with corrosion surfaces and what he termed

is no general trend in terms of the orientation of elongate ruined

Kuppenruinen (which literally means hilltops),
towers, but where they occur in clusters across the south which are basically small towers
up to 8 m high and <100

of the island, their broad arrangement is northwest- in diameter The in this

m (Figure 11). towers area are

southeast and north-south. Elongate the unusual in that associated with

towers occur on they are not low-lying

slopes to the west and north of Alligator Pond and around alluvial plains close base level. Pfeffer
to (1975, 1997),
Bull Savannah, with where the relief
an east-west asymmetry, through analysis near Mandeville and Alligator
west-facing slope is steeper than the slope on the eastern Pond, considered that the karst in that area had been

side. These tower-like hills were described by Pfeffer uplifted during the Quaternary, when they were lifted well

(1967) being only about 10-12 high, with striking above the table. The mode of for-
as m
regional groundwater
 -207-

mation of these towers is still uncertain, as some are aligned roughly along northwest-southeast and northeast-

oversteepened with footcaves, indicating active rela- southwest in response and

or axes, probably to major joint

tively recent undermining, whereas other towers, in the fault trends. A second set of predominantly east-west

same area, rise above small limestone pedestals (Figures trending ridges displaces the north-south ones, in the area

9, 10). The towers are probably the product of vertically between Top Hill, Ballards Valley and Bull Savannah,

directed dissolution rather than corrosional of St Elizabeth. The of the limestone

processes parish development
planation at or near the water table. They are therefore ridges appears to be related to the prominent faults, where

similar in origin to towers in northern Puerto Rico, where the faulting leads to a marked resistance of the limestone

there is a concentrationof water within the coversands at around the fault The mechanism of
to weathering. exact

the base of the towers and vertically directed solution their formation requires detailed
more investigation,
controlled by the presence of depressions and sinkholes, although Versey (1972) attributed this resistance to disso-

concentrating erosion at the tower base (Miotke, 1973). lution as a result of reduced permeability around the fault.

Some of the towers on the plateau surfaces and slopes

above local base level in southern St Elizabeth seem to be

aligned and are separated by corridors where the topogra- Glades

phy is less marked. The tower alignments correspond to

major joint directions. They have evolved in similar Glades first described elon-
may a were by Sweeting (1958) as

fashion to the giant grikeland labyrinth karst in the and similar

or gated enlarged cockpit depressions, to uvalas,
Nahanni region of northern Canada & where individual cockpits extend lines of
(Brook Ford, by growth along

1978), where joint controlled grikes and and coalesce form

are opened up by jointing faulting, to more complex
dissolution to form enclosed depressions with forms well-defined tectonic line. Glades and
steep, along a

residual towers to 50 high, emerging from devel- tend

up m a cockpits to
merge into one another (Sweeting, 1972).

oping, karst plain. The corridors between the indicated that Barbecue Bottom in
uneven Sweeting (Figure 12),

aligned towers
may be similar morphogenetically to the the northeastern
part of Cockpit Country, is a typical glade

zanjones of Puerto Rico (Monroe, 1968). formed in such northnortheast-

a way along a

southsouthwest-trending fault line. According to

Sweeting

(1972), glades are

sinuous depressions with concave and

Ridge karst angular contours, and are sided. The floors of

steep glades
also tend to consist of a
series of shallow basins being
A number of limestone ridges have been recently reported separated by low divides which are
much shallower than

across southern St Elizabeth and to a lesser extent on the the passes leading out of the glades. Sweeting also indi-

Manchester plateau, in the Moneague Formation cated the contrasting drainage of cockpit depressions and

limestones (Sir William Halcrow & Partners, in that of tends be toward the
1998). They glades, drainage cockpits to

initially described by Pfeffer who noted

were (1967), a centre, whereas drainage in channels on the floor of the

range of narrow ridge-like horsts and what he termed glade often disappears into caves and sinkholes at the

Kalkrippen (limestone ribs), commonly associated with margin of the Glades have also been described
glade. as

karst corrosion surfaces, to the south of Nain, southern St broad between residual hills that be
areas can individually
Elizabeth (Figure 11). Versey (1972) also noted the occur- classified as either such
compound depressions, as uvalas,
rence of ‘fault-line ridges’ associated with low- dry and underdrained valleys (Chenoweth &
or as Day,

permeability faults that have resisted erosion, though he 2001). Accordingly, their cross-profile flat,
may be con-

did not indicate their precise location. The karst limestone vex, concave or undulating, and in plan view their
shapes

ridges in the parish of St Elizabeth are up to 4 km long, are either linear or sinuous. According to Chenoweth and

but 0.5 2 km, and from <50 300 have thick soils and when left undis-
more commonly to to m Day, many glades

wide, with heights ranging from <5 40 Some turbed they and dense
to m. are support large trees vegetation.

topped by limestone exposures in the form of rock walls.

They occur across a variety of geomorphic settings on Poljes

plateau surfaces, slopes, broad depressions and on valley

floors. They particularly occur in topographically low- Poljes are large, flat-floored, enclosed in karst
depressions

lying settings to the south of Burnt Savannah, parish of St terrain, commonly with ephemeral or perennial streams

Elizabeth and to the east of Santa Cruz, where their surface &
they disap- flowing across (Ford Williams, 1989).

pear northwards beneath the swamps of the upper Black Poljes have three
diagnostic criteria (Gams, 1978): a flat

River morass. They are also associated with a

of floor in solid rock or unconsolidated sediments such as
range

other karst landforms, including dolines, elongate hills, alluvium; closed basin bordered rim of steep
a by a mar-

large uvala-like enclosed depressions, dry valleys and ginal slopes, though the steep slope be restricted to
may
karst. side of the with break of be-
tower one polje, an abrupt slope
of the ridges broadly symmetrical, though the floor and and karstic with
Many are tween sides; drainage
others asymmetrical their Most stream the of the
are across axes. are sinks, especially at margins polje.
-208 -

St
Figure 11. General geomorphological map of southwest Manchester and southeast Elizabeth (redrawn from Pfeffer, 1967). Key:

30-50
1 = karst marginal plain -
alluviated karst plain (Karstrandebene);
2 = cone karst (Kegelkarst): (a) residual hills up to m

tower-like residual hills, to 8-12 elevation; 3 corrosion surface (Korrosions-

=
relief gentle slopes; (b) up m
elevation, high on

horsts and limestone ribs (Kalkrippen); (b) escarpments and landslides; 5 =

karst
flache); 4 =

(a) small steps, narrow ridge-like

plateau (Karstebene).
 -209-

Figure 12. Glade Barbecue Bottom, northern

at Cockpit Country, parish of Trelawny.

Figure 13. Polje at Lluidas Vale, of St Catherine, displaying gently undulating floor with
parish subsidence doline and steep-sided
rim in
part due to faulting.

The poljes or ‘interior valleys’ of the island have been have

(1972). Specific poljes also been studied in more

described in general by Sweeting (1958) and detail, especially The of

Versey Queen Spain’s Valley (Pfeffer,
-210-

1986) and Lluidas Vale (Landmann, 1985, 1990) (Figure Lateral planation at the margins of the polje has resulted

indicated that in Ja- in the dissolution of and their within

13). Sweeting (1958) poljes occur cockpits integration
maica where lithological, structural or relief conditions the polje (Landmann, 1985). According to Landmann

flood collect and where is of the base level has lead the
permit water to rapid drainage (1985), a lowering to polje

conspicuous the becoming dissected, forming channels and sink-

impeded. Accordingly, they are along more

northern margin of Cockpit Country, where lateral water holes. At present the permanent water table is at some

movement is promoted by the Montpelier Formation depth beneath the polje surface and it is more likely that

chalks, and on
the northern side of the Duanvale fault local perched water tables within the alluvium account for

(Sweeting, 1958). Sweeting (1958) also noted lateral after rainfall. Landmann (1990)
zone poljes planation heavy
to be common
where the White Limestone Group comes further developed his theory of polje formation at Lluidas

into contact with the older Yellow Limestone Group, Vale in relation to changing climatic regimes. He attrib-

where flood water can

be ponded due to drainage uted the landforms around Lluidas Vale to different cli-
poor

through the older rocks. Sweeting (1958) further sug- matic regimes, in that intensive weathering conditions in

the the island the formed and karst

gested that, as poljes across occur at a
range ‘Neogene’ etch-plains cone sur-

of elevations, local water table fluctuations were more rounding much of the Vale, while alluvial fans and

their than variations. dolines in the Vale drier climatic conditions and
important to development regional represent
She also noted the occurrence of drained poljes where a lower intensity of weathering in the Quaternary. Ac-

flooding is absent infrequently due to changing cording to Landmann the polje is the result of
or occurs (1990),

hydrological conditions and marginal poljes (border lateral expansion of the limestone at or near
the water

poljes) where they are not completely surrounded by White table and began as ‘the meandering canyon of a river’.

Limestone. Pfeffer (1986) described the Queen of Spain’s Valley

According to Versey (1972), all of the poljes are as a large polje with an extensive plain, about 90-100 m

structurally controlled in their entirety and there is no elevation, surrounded by higher relief. To the south, the

evidence to an erosional origin. They are down- valley is bordered by steep limestone hills which rise
suggest
faulted on at least one side and the character of the fault- abruptly to form cockpit karst; to the north, east and west,

ing tends to impede drainage, which forces groundwater the polje gives way to limestone hills and dolines. Ac-

to the surface during high water stages. The drainage may cording to Pfeffer (1986), the polje floor is dissected by
such the reversal of which contain shallow and
be impeded by a structural control, as dolines, may ponds, many

dip, by the upfaulting of the karst base (Gardner et ah, karst springs rise on
the polje floor near to the southern

1987), or by an abrupt decrease in limestone permeability border to flow northwards as

the Roaring River and Mar-

(Versey, 1972). Some poljes are to alluviation tha Brae River systems. Pfeffer (1986) interpreted the
prone

during flood events, especially in low-lying areas, while in landforms of the Queen of Spain’s Valley and the sur-

others the alluvium is being dissected. In some poljes, the rounding karst landscapes as the result of different relief

up-slope part of the valley is periodically flooded and generations, possibly influenced by tectonic uplift and

subject to deposition of alluvium, while the down-slope climate change. He considered that the polje and the

area carrying floodwaters to sinkholes is being dissected surrounding cockpit karst are Tertiary landforms, and that

(Versey, 1972). The poljes are generally linear, with the subsequent Quaternary uplift and climate change led to

long axes perpendicular to overall direction of groundwa- the development of the dolines, dry valleys and subdued

flow (Gardner ah, described limestone hills forming the northern border of the polje,
ter et 1987). Versey (1972) a

of the best known the island including the similar to that of the evolution of Lluidas
some poljes on interpretation
Nassau and the interior valleys of northern Vale given by Landmann(1990).
Valley

Trelawny. He considered that faulting and folding within In Jamaica, the three basic types of polje outlined by

the Nassau Valley forms a barrier to drainage and forces Ford & Williams (1989) can be identified, these being

groundwater to the surface, which causes periodic flooding border, structural and base-level poljes. Border poljes
and leads to the deposition of alluvium. In describing the receive allogenic drainage from a non-carbonate area and

valleys of northern Trelawny, Versey (1972) indicated that are therefore not bordered on all sides by limestone. The

the structural barrier which impedes the northerly flow of valley floor contains a number of allogenic streams which

groundwater, leading to a series of springs, is a combina- bring alluvium onto the polje and eventually disappear at

tion of block of Yellow Limestone beneath stream sinks A border

an upfaulted (ponors). typical polje occurs at

the and the of facies within the White Lluidas Vale. Structural poljes dominatedby geologi-
valleys change are

Limestone from the crystalline and rubbly limestone to the cal control and associated with graben or fault angle

Montpelier Formation chalks. depressions formed by block faulting. Structural poljes are

According to Landmann (1985), lateral planation was common across the White Limestone. Base-level poljes are

for the of the polje at Lluidas to the north of Cockpit Country, though here
responsible development common

Vale. The process takes place at the border of alluvial fans they are also associated with faulting. Base-level poljes are

the floor of the depression and in small in water table dominated where the
on areas a
regional epiphreatic zone

neighbouring cone karst area to the southwest of the Vale. reaches the surface. They are typically located on the
 -211 -

outflow side of the karst drainage, which partly explains are associated with residual tower karst above
emerging
their occurrence to the north of Cockpit Country. the alluvial cover.

Incipient (weakly developed) karst terrain Limestone and other fluvio-karst

valley systems

There number of massifs within the and

are a upstanding Dry ephemeral limestone valley systems and lime-
White Limestone the southern
Group, mostly across stone gullies are quite common across the White Lime-

sections of Jamaica, where karstification is restricted to stone the but very few have
Group outcrops over island,
small-scale dissolution sculpturing, about 1 2 been described in detail.
to m deep, Allogenic streams, blind valleys,

extending to a maximum of about 5 and

m, accompanied by steepheads pocket valleys also occur, the latter having
the development of a case-hardened surface. also been described by
They are Sweeting (1958) from around the mar-

often covered by limestone talus and

scree, and may have gins of northern Cockpit Country and were further briefly
limestone debris fans their lower
on slopes associated with examined by Miller (1998). Gullies and
gorges are com-

dry valley systems and gullies. Uncommon sinkholes, mon on the limestone outcrops to the south of the
island,
collapsed dolines and surface lines and
ephemeral drainage are are an expression of ephemeral drainage or
may
in

associated with the incipient karst surfaces of these lime- some cases be relicts of in the
wetter periods recent
geo-
stone massifs. Typical locations include the Dallas Moun- Gullies drain the southern
logical past. part of the south

tain, Long Mountain, Hellshire Hills, Brazilletto Moun- Manchester plateau, especially in the Robert’s Run-

tains, Portland Ridge, Kemps Hill, Round Hill and Great Sixteen Mile Some of the the south-
Gully area. gullies on

Goat Island. Although they have

poorly developed surface ern end of the Manchester
plateau are associated with

karst, some of the massifs have well developed alluvial fans that have been
cave slightly displaced by faulting,

systems, such as the Jackson’s Bay Great Cave Port- that the
on suggesting faulting is a relatively recent occur-

land Ridge (e.g, Fincham, The surface karst

1997). poor rence, though no detailed work has been undertaken to

development on these massifs could be a factor of the date.

duration that have been sub-aerial weath-

they exposed to
Day (1985) examined the limestone
valley systems
ering with been in rela-
processes, most having uplifted along the north coast, particularly where they are well

tively recent geological time. developed between Discovery Bay and St Ann’s Bay
where seventeen major valley systems extend up to 7.5 km

inland. Streams in these

valleys range from permanent to
Karst surfaces of low relief (buried karst) times of intense
mostly dry, except at or prolonged rain-

fall, while others show alternate and

wet dry reaches. The

Both buried and exposed karst surfaces of low relief are headwaters of the most extensive lo-
valley systems are

extensive in the upper and lower morasses

of the Black cated in the so-called Browns Town-Walderston forma-

River. This area was described by Pfeffer (1967) karst tions the
as a (now Moneague Formation; Mitchell, 2004),
marginal plain (Karststrandebene) (5), which is
basically though much of their valley length is across the Mont-

alluviated karst plain. Karst surfaces also

an are exposed pelier Limestone, and they are associated with a series of

above the blanket of alluvium in the CabaritaRiver imme- fans of the

gravel Quaternary age near to coast. Day
diately west of where the north-south
Savannah-la-Mar, investigated the geomorphology of the valleys and exam-

alignment of the exposed surfaces

may be a response to a ined their morphometric properties, including drainage
structural control. Karst surfaces of low relief also occur density, valley order and orientation. Two distinct valley
around Little London, parish of Westmoreland, where orientation concentrations
they were shown, one virtually
are associated with a thin alluvial cover and normal the present coastline
numerous to (north-south), and a sec-

enclosed water-filled ond which

depressions. parallels the coast
(east-west). Day compared
Many of the karst surfaces
may be a
legacy of a former the valley systems with similar dry valleys developed on

polje surface, now covered by alluvium and wetlands, such the raised reefs in Barbados (Fermor, 1972) where the

as the lower of the Black River 1964;

morass (Grontmij, drainage pattern was attributed in part to the effect of

Pfeffer, 1967). An alternative explanation is that raised reef morphology

they on drainage extension across the

represent partially buried base level corrosion reef surface runoff under
plains, during wetter conditions than

while allogenic rivers the former polje

flowing across
present. Day indicated that the Barbados model can only
aided in the planation be invoked for the
process by periodically introducing valley sections across the raised reefs

large volumes of floodwaters. Corrosional and that the

aggressive east-west valley orientations are structurally
planation may have also been
encouraged by the sandy controlled due to east-west
faulting in the Duanvale Fault

bedload of the rivers. In addition, overbank Zone. Control the north-south

deposition on
valley trends is less

would stimulate biochemical the

activity on floodplain clearly explained, but Day considered that
they may either
and, through the production of soil carbon dioxide, would be joint controlled in similar way in Puerto
a to zanjones
enhance soil water impacts. Many of the karstic surfaces Rico or be related to north-south faults.
-212-

Smith & Atkinson described the between the Montpelier Beds (Montpelier Forma-
(1976) landscape contact

the Limestone fossil fluvio- tion) and the crystalline White Limestones. Sweeting
developed on Montpelier as a

network. (1958) further indicated that fairly

karst with a well-integrated dry valley Day most cave
passages
are

(1985) considered the valley systems to be fossil in that long, with tunnel-like cross-sections and undulating floors

certain of their components may be attributed to past

due to the common occurrence
of roof-fall debris (break-
conditions and that they are undergoing little or no
devel- down), though for the most part gently inclined passages

under most conditions, though during predominate similar to the generally shallow gradients of
opment present

prolonged or
intense rainfall they are fully active, so it underground water flow.

would be incorrect to describe them as totally fossil. The According to Draper & Wadge (1997), most of the

extensive fans their lower reaches also in the shallow limestones

gravel on were explorable caves occur water

taken to indicate greater transport ability in the past. Day which are commonly crystallised to at least 40% sparite.

(1985) examined the sources of runoffwithin the valleys, There is also evidence to suggest that the sparite-rich

which surface runoff after limestones resistant solution than the micritic
comprise two components, are more to

rainfall and underground drainage limestones with low sparite content and tend to produce
water supplied by

sources, the latter providing the baseflow for the perma- cave passages
that are
taller than they are wide and are

nent streams. Some of the baseflow is supplied from more strongly joint controlled (Wadge & Draper, 1977a).

enclosed depressions and sinkholes via springs Wadge & Draper (1977a) investigated the size and
adjacent
within the valley 1976, 1978a, 1984), shape of cave passage cross-sections in several formations
courses (Day,

though much is derived from the limestone and non- of the White Limestone Group. They found that most

carbonate areas to
the south of the valleys. The dominant caves of great length occur in the oldest Troy Formation

formative mechanism for the valley systems was thought (and in the Chapelton Formation of the Yellow Limestone

to be fluvial erosion rather than karst solution (Day, Group), associated with allogenic rivers flowing off the

1985). Cretaceous inliers to form stream sinks. Thus, the large

Miller & Donovan (1999) examined the Natural cave passage lengths encountered in these formations may

of St which be under rather than lithologic control,

Bridge at Riversdale, parish Catherine, occurs stratigraphic
in a small gorge cut into limestones of the Somerset and related to water flow routes. Wadge & Draper (1977a)

formations. The 20 22 section of further indicated that cave cross-sections within

Troy bridge spans a to m passage

the Rio Doro channel and is 17 with distinct the Formation typically rectangular in shape,
m high a Troy are

limestone bed forming the roof span. The natural bridge indicating a strong bedding-plane and joint control. The

was inferred to be the result of partial roof collapse of an biggest cave passages occur in the Swanswick Formation,
river system, while river capture also typified by huge, single-passage dip-tubes formed along
underground may

have contributed to its formation through the work of a major phreatic routes, possibly related to the massive

more aggressive lower base-level capturing stream. Miller bedding and lack of jointing within this formation (Wadge

& Donovan further suggested that initial collapse probably & Draper, 1977a).
formed karst windows that were subsequently connected The tectonic control on cave development was also

by more complete collapse to form the gorge. investigated by Wadge & Draper (1977b) who found little

evidence of caves along major fault planes, though some

caves
be along small subsidiary faults. A few river
may

of steep gradient do alongside major fault

Cave systems and underground rivers caves occur

systems (e.g., the Bristol Cave-Quashies River Cave

A summary of cave exploration and the development of system) (Waltham & Smart, 1975). Major faults tend to

studies within Jamaica presented by produce large collapse features especially where they
speleological was

Fincham (1997). The first detailed speleological studies of intersect cave passages at high angles, while strike-

cave systems within the White Limestone Group were passages and dip-tubes tend to develop where drainage is

strike The
published in the 1950s (e.g., Zans, 1951b, 1953, 1954a, b, parallel to (Wadge & Draper, 1977b). most

1958a, b, 1959). White & Dunn (1962) indicated that the important cave type is the fairly straight, large river

to be with little simple in plan and flowing down dip. According

largest caves appear single cave passages passage,

or no branching. Brown & Ford (1973) investigated the to Wadge & Draper (1977b), although phreatic ‘lift’

caves and associated groundwater patterns in northern passages would be expected to occur in caves where the

is less than the of the

Jamaica, and the hydrological link between the Quashies hydraulic gradient dip beds, only a

River sink and the Domock Head rising. few exist, as they may have become fragmented and

Sweeting (1958) suggested that most known caves

in disconnected by rapid surface lowering, such that re-

the White Limestone Group are located in two main maining passages only occur within inter-depression hills.

locations: first the between the horizontal in of

stratigraphic at contact Although some passages occur areas

lower formationsof the crystalline White Limestones and cockpit topography, these are essentially fossil features,

the top beds of the older Yellow Limestones Forma- later erosion of the karst surface &
(Troy cut by (Draper Wadge,

tion/Chapelton Formation); and, secondly, at the zone of 1997).

 -213-

In terms of vertical cave development, large number and considered a

a compound origin was likely.
of ‘cockpit’ shafts known within the White Limestone
are Day (1979) used vector analysis to examine the distri-

Group, and many have been explored, described and bution and orientation of
planar surfaces within several

surveyed (Fincham, 1997). Most cockpit shafts are re- sampling areas across the Caribbean, including localities

garded to be solutional in origin, though block- and boul- the White

on Limestone, to develop a surface roughness
der-chokes within
many shafts may indicate some col- index in order to classify landforms on a more
objective
lapse. There are
also a significant number of caves with basis. Day calculated the surface roughness values of the

both vertical and sub-horizontal

passages, which seem to ten sites to develop a
threefold classification of the land-

be concentrated in specific of steep

areas hydraulic gradi- scape, the three types being roughly equated with doline

ents and where faulting is prevalent (Wadge & Draper, (type I), cockpit (type II) and tower (type III) morpholo-
1977b). Many caves are abandoned vadose systems de- additional
gies. Day (1982) developed an general classifi-

scending in a series of shafts. Wadge & Draper (1977b) cation index from Caribbean-wide karst terrain, which

suggested that the development of these steep vadose different of theoretical

analysed degrees roughness and

systems may have been related to the presence of thin, simulated overall karst landscapes. The negative and

impermeable bentonite the

clay layers near top of the positive reliefelements of the landscape were analysed to

White Limestone which would have

Group, provided develop a ratio of vertical to horizontal
components. On
sufficient concentration of surface for the basis of the ratio of their vertical
drainage to account
component compared
the vadose development. Elsewhere, with their horizontal
steep cave many dip- component, twelve types of landform
tube phreatic modified under unit could be
systems are now being recognised within the three classification

vadose conditions.
types, giving 36 possible landform unit combinations. The

scheme index of terrain type

provides an upon which

comparisons be made
can (Day, 1978), and was correlated

Karst bedrock such

morphometry against properties, as purity, porosity, pe-

trology and mechanical in order ascertain the

strength, to

There are many quantitative morphometric techniques most influential properties on

karst terrain
type.
available for karst terrain but few have
research, only a Brook & Hanson (1986, 1991) used morphometric
been used in research the karst of the
on phenomena analysis to assess how successfully karst terrain could be

White Limestone Group. The first work

published using modelled, to reveal organised patterns in the terrain and

morphometric analysis the karst of Jamaica that the

on was by to assess
potential of
morphometry in distinguishing
Day (1976), who considered that such karst
an approach pro- types. They used double Fourier series analysis to

vided the basis for a more objective classification of karst uncover significant patterns in the spatial variability of

terrain, while the correlation between form and doline and karst in northern
process, cockpit Jamaica, and pro-

and the controls on karst development, also greatly vided correlation data between landform
are
type, and bedrock
aided by a morphometric approach. Specific morphometric and Brook
structure lithology. and Hanson measured

analysis of karst terrain on the White Limestones includes fracture trace, and orientations in the
depression lengths
that of Brook & Hanson
Day (1976), (1986, 1991), Han- Brown’s Town area, parish of St Ann. The data were

son (1985) and Draper et al. (1998), while a review of

grouped into ten degree orientation classes and the
length
morphometric techniques used in karst studies, with weighted values were converted to percentages. The

examples of work from the Caribbean dominant directions concurred with the directions of the
as a whole, was

presented by MacGillivray (1997). Morphometric analysis more important frequency pairs of the Fourier analysis,
of karst on the White Limestones has also been while the of the
incorpo- wavelengths frequency pairs coincided

rated into Caribbean-wide research, specifically in the with the fracture and depression densities analysed from

classification of karst terrain types (e.g, Day aerial concluded that

1979, 1982). photographs. They local structures

In morphometric study of depressions in the Browns had the formation of

a a strong influence on depressions in

Town area of St Ann, Day (1976) used randomly located the there also difference in the
area, was a significant
profiles to measure depth, diameter and internal slope density of depressions related to two bedrock
types, which
angles of the features, from which number of function of
a morpho- was a folding and faulting. However, the

metric ratios developed. The facilitated results could be

were analysis a not used to determine the reason for

correlation between shape and distribution of the variance in in similar bedrock.

depres- depression depth
sions with factors influencing their morphology. It al.
was Draper et (1998) applied a nearest neighbour
thought that the analysis, by quantifying the spatial distri-
analysis method, used in structural geology to measure

bution of dolines, would allow for inference

a preliminary strain, to the distributionsof sinks and summits of
polygo-
as to their formation, collapse might be be nal karst in
as expected to Cockpit Country. Results showed the distribu-

more randomly distributed than solution dolines. How- tion of sinks is while that of the
essentially random,
Day concluded that it was impossible to infer from summits is uniform with
ever, more nearest neighbour spacings
morphometric data alone whether surface of 125-250
collapse or m, with the longer nearest neighbour distribu-

solution was the dominant

process in depression formation tions generally trending to the northeast. Draper et al.
-214-

suggested that the polygonal karst develops from origi- which occur in the southern part of the vale in the same

nally randomly distributed sinks, and that White Limestone facies the
depression as surrounding cockpit karst,

growth ultimately produces uniformly distributed sum- developed on an etch-plain which was uplifted at the end

mits. of the Pliocene the of the

or beginning Quaternary during

a drier climate, as the residual soils associated with the

dolines are not as intensely weathered. The alluvial fans

Landform evolution and the floor of the vale simi-

gravel terraces seen on are

larly the product of a drier climate in the Quaternary,

the karst of the White Limestones though
Early publications on Landmann (1990) suggested that several phases of

interpreted the development of landforms in relation to the alluvial fan formation occurred as a result of humid and

petrographic differences between the pure, hard crystalline dry cycles, which were repeated at least eight times during
limestones and the less in that the
pure, marly varieties, Quaternary.

cockpits and towers tend to occur on the pure limestones Thus, according to the German school of
geomor-

and dolines the less limestones (e.g., Sweeting in Jamaica and the wider
on pure phologists working Caribbean,

1958, 1972). More recent studies by German geomor- the evolution of limestone landforms can
be explained not

phologists have suggested that palaeoclimatic change may only by geological controls, but also by contrasts in the

have had important influence the development of intensity of karstification the and
an on during Tertiary Quater-
karst landforms in Jamaica 1986, 1997; Land- result of palaeoclimatic and
(Pfeffer, nary as a palaeoecologic

1990). Pfeffer
mann, (1997) agreed with many of the early changes. Some of their interpretations are highly conten-

observations of Sweeting, in that dominated by tious and both the and residual
areas speculative, as depressions

impure or thinly-bedded limestones with interbedded and the and soils

hills, weathering products they contain,

marly layers always develop doline karst, and that the cannot be treated both
as separate landforms; they are an

purer, more massive limestones underlie areas of residual integral part of the overall pitting of the surface. Their

hills with cockpits. However, some areas have dolines interpretations also rely strongly the that the
on premise

developed in bedrock which would otherwise show cock- bauxites residues of the karstification. It is that
are likely

pits and, according to Pfeffer, cannot be explained by geological controls combined with climatic change and

Sweeting’s (1958) interpretation. climate variations have both had a profound influence on

Pfeffer that the relief the of karst landforms the White Lime-
(1997) argued original was a development on

peneplain, the main line of evidence for this being the stone Group. Of equal importance to karst evolution would

similar elevation of the residual limestone and that have been the erosional of the anticlinal
hills, breaching struc-

this be dated from the The kar- and form the Central Inlier.
can Oligocene-Miocene. ture subsequent de-roofing to

stification of the peneplain in the Miocene-Plioceneled to This non-carbonate rocks and led the
exposed to develop-
the development of cockpit karst with poljes, and that of surface streams which carried volumes of
ment large
karstification may have been triggered by sea-level fall the limestones. and subterra-
or water to Subsequent sinking
local uplift. According to Pfeffer (1969), analysis of relief flow the White Limestones modified their
nean through
in southwest Jamaica shows that Tertiary cockpit karst hydrological characteristics from essentially autogenic
was uplifted during the Quaternary. Pfeffer (1969, 1997) systems dominatedby vertical percolation routes through
also argued that the bauxites found in the large cockpits the limestone, to a
mixed allogenic- autogenic system with

are the weathering residue of Tertiary relief formation, not only vertical percolation, but also horizontal water

and have a higher rate of quartz dissolution and de- movement, forming fairly rapid flow paths.

silicification than the red-earth and red loams associated An earlier theory on the evolution of karst landforms

with the residual hills, which are considered to be the on the White Limestones also stressed this link to the

result of soil formation and

present younger karstification. bauxite deposits, where Smith et al. (1972) suggested that,
Pfeffer this as evidence for ‘the origin of the karst landforms is connected
(1997) suggested higher intimately

weathering intensities and different morphodynamic with that of the bauxite deposits’. Smith et al. considered

during the Miocene-Pliocene which formed that the bauxite the of limestones
processes originated by weathering

cockpit karst, compared to the Quaternary karstification. in the cockpits, from which the residue
weathering was

which led to a new ‘relief generation’ as a result of a transported via underground drainage to the ‘degraded’

change in palaeoecological conditions. where it

cockpit, was deposited by floodwaters, a process
Similar interpretations of the landform evolution of which is active. A mathematical model
no longer was

Lluidas Vale were made by Landmann (1990), who constructed in which that the White Lime-
con- they suggested
sidered the oldest reliefunit in the Lluidas Vale area is an stone was
first exposed to erosion by uplift in Late Mio-

etch-plain which was formed by intense weathering con- cene times and the development of karst landforms began
ditions in the The etch-plain subject with surface with shallow
Neogene. was to an undulating depressions. In

intense karstification the late Miocene late the model, soil

during to an incomplete cover then developed in the

Pliocene, which led to the development of cockpit karst depressions leading to differential rates of dissolution, the

and a strongly-weathered residual bauxitic soil. Dolines, latter being concentrated beneath the soil cover. Smith et
 -215-

al. then contended that significant slopes is the host of the

more develop mation; Mitchell, 2004) most extensive

due to preferential dissolution, and they become bauxite while the

as steeper deposits, Browns Town-Walderston

their soil cover is lost by mechanical erosion. Thereafter, Formation is the

(also Moneague Formation) principal
the slopes cease to become the erosion rate is host of bauxite in the north with
steeper, as coast belt, important
constant over the entire length of the slope. Smith al. the Formation
et reserves on Troy (Lyew-Ayee & Stewart,

assumed in their model that this ‘equilibrium time’ is The residual of the has
1982). theory origin of bauxite
reached when the is 30°, after which time the been
slope angle questioned and, although the deposits are mainly
slope will retain its form. confined underlain by the shallow carbonate
to areas water

Smith al. also into the mathematical units of the White Limestone
et
incorporated Group (Robinson, 1994), the

model the development of the of these

cockpit shafts, by supposing relationship deposits to the bedrock has been the

that while the slope remains soil all the of

covered, drainage subject numerous
papers (e.g., Wright, 1977; Lyew-
leaves at its base, through fissures in the lead-
limestone, Ayee, 1982).

ing to the formation of shaft. The depths of the shafts Zans also
a are (1958c; see Chubb, 1963) proposed an

also controlled by the presence of soil and, alternative alluvial theory for the of bauxites which
a cover once origin
this is the shaft and erode
removed, slope at the same rate, is closely associated with the groundwater hydrogeology
due to the fact that drainage is no longer concentrated of the White Limestone. Zans theorised that alumina-rich

down the shafts, but seeps directly into joints and fissures argillaceous debris was eroded from Cretaceous tuffs in

in the bare limestone. Smith

This, according to et al., the Central Inlier and fluvially transported by surface

accounts for the fact that shafts are rarely greater than 60 which sink when the White Lime-
streams flowing onto

m deep and the final depth of the shaft will be reached stone. The fine debris continued be
to transported by
when the surrounding slopes reach of 30°. Based
an angle underground drainage, and was carried to the surface and

on their mathematical model, Smith al. maintain that deposited by karst the floors of karst
et a waters over depres-
shaft of some kind will form at the base of the slope sions of various sizes. Zans’ the first
as a According to theory,
natural
consequence of the presence of a cockpit. They of these
muddy sediments were laid down while the island

also modelled final and radii of and still

depths shafts, ‘equi- was being uplifted and continued uplift elevated the

librium times’ in relation to cockpit development. The alluvial deposits high above the where lateri-
water table,
development of degraded karst in relation to bauxite sation converted them to bauxite. In the alluvial theory,
deposits also modelled Smith al. and for the of bauxites
was by et (1972) a uplift accounts occurrence on the

true ‘equilibrium landform’ towards which all slopes in limestone plateau surfaces high above the water table.

degraded cockpit will tend equated. It has also been that

was proposed much, if not most, of the
Smith et al. (1972) to show that their of bauxite is derived from the
attempted superficial cover
weathering
mathematical model of landform development produced of ashfall volcanic in
deposits produced by activity the

landform shapes which are not dissimilar real karst in the late Ceno-
to region surrounding Jamaica, especially
landforms on the White Limestone. Based their model of which still ash beds in the
on zoic, some are preserved as

of landform Smith al. contended that the

evolution, et
deep water limestones of the Montpelier Formation

equilibrium time for the production of cockpits is about Comer

(Comer, 1974, 1976; et al, 1980; Comer & Jack-

0.6 million
years, while the total time required to produce son, 2004). The ash-fall theory is also supported by other

an equilibrium bauxite deposit, or fully degraded karst, is workers Comer

(e.g., Lyew-Ayee et al, 1989). (1974)
0.6 million years plus 2.7 million years, or 3.3 million indicated that the of bentonitic
relatively high percentage
for economic and 3.7 million
years an average deposit clay within the Montpelier chalks, the chemical composi-
years for a non-economic deposit, similar to the 3.5 mil- tion of the
clay (including the high alumina and low silica

lion that Smith estimated the time and the

years (1970) as re- contents), stratigraphic position above the host

quired to produce the total known bauxite reserves. limestone all suggest that these rocks the
are more likely
The times presented by Smith al. for the of material for the bauxites than the
et (1972) source
parent rocks
evolution of both the cockpit landforms constrained in the residual
are by proposed theory. According to Comer, the

assumptions in the model, least the of of volcanic ash

not generation timing deposition coincided with the

erosion rates, which constructed using emergence of the and much of the
mean were water island, so ash was

hardness values and runoff rates data.

using present day deposited subaerially and concentrated in pre-existing
The model also that the bauxite karst
presupposes deposits depressions. Subsequent elevation of the ash deposits
originated the insoluble residue of limestone from the and intense
as leaching of ash and ash-bearing carbonates
cockpit which is the subterra- caused the formation of bauxite. Comer
areas, transported through eventually con-

nean drainage system and deposited by floodwaters to cluded that the exclusive association of bauxite with

form areas of degraded karst. In addition to Smith et al. karstified limestone was related to both the and
drainage
(1972), other workers who favoured a residual origin for the dissected of the limestones.
nature According to

the bauxites include Ahmad al. Hose

et (1966), (1963), Comer, the effect of karst development was to trap the
Clarke (1966) and Sinclair The
(1966, 1967, 1976). parent ash-fall material in karst depressions which pre-

Newport Formation (now referred the For- vented erosion, while the
to Moneague high relief topography of the
-216-

pre-existing karst resulted in well-drained slopes where Bauxite soils of Jamaica. Soil Science Society of America

Proceedings 30, 719-722.

vertical drainage through the limestones provided the

for the subaerial Aub, C.F. 1964a. Limestone Scenery in Jamaica. Cambridge
optimum conditions leaching processes
of Cambridge) (unpublished MA thesis).
for bauxite formation.
(University
required
Aub, C.F. 1964b. Karst problems in Jamaica, with particular
reference to the cockpit problem. Abstracts, Symposium on

Karst Phenomena, 20th InternationalGeographical Union,

Conclusions
Seattle, Washington, USA.

Aub, C.F. 1974a. The nature of cockpits and other depressions

This has to review our current under-
study attempted in the karst of Jamaica. Proceedings of the 5th International

standing of the karst geomorphology of the White Lime-

1969, M15/1-7.
Speleological Congress, Stuttgart,
stone Group and to provide a summary of the work com- Aub, C.F. 1974b. Some observations on the karst morphology of

pleted to date. It has also attempted to identify the gaps in Jamaica. Proceedings of the 5th International Speleological

Congress, Stuttgart, 1969, Ml6/1-7.

our knowledge and to indicate areas of future research. A

detailed of each of the main karst landforms Baker, L.L., Devine, E.A. & DiTonto, M.A. 1986. Jamaica, the
description
the White Limestone has been pre-
1985 expedition of the NSS Jamaica Cockpits Project. Na-
occurring on Group
tional Speleological Society News 44 (1), 4-15.
sented. The most extensive karst landform type is cockpit
Barker, D. & Miller, D.J. 1995. Farming on the fringe: small-
karst, followed doline and karst. Poljes also
by tower are
scale agriculture on the edge of Cockpit Country. In: Barker,
landform elements and have strong
important most a
D, & McGregor, D.F.M. (eds). Environment and Develop-
structural related block faulting.
control, being closely to
the Caribbean: 271-292.
ment in
Geographical Perspectives,
Both and structural controls important in
lithological are
Kingston (The Press, University of the West Indies).
the of karst landforms within the White
development Beche, H.T. de la 1827. Remarks on the geology of Jamaica.

Limestone and the de-roofing of the Central inlier

Group, Transactions
of the Geological Society London (2) 2, 143-

altered the karst hydrology of the 194.

fundamentally
& Ford, D.C. 1978. The origin of labyrinth and
limestones and greatly modified the overall karst land- Brook, G.A.

Climatic also have had tower karst and the climatic conditions necessary for their
scape. change may a profound
influence karst of the White Limestone development. Nature 275, 493-496.
on development
G.A. & M. 1986. Quantitative analysis of
Brook, Hanson,
Group, as
it may be argued that some karst phenomena are
Brown’s Town, Jamaica and
depression morphology near

essentially hydrologically relic and represent karstification

bedrock factors influencing cockpit or doline development.
during wetter periods than experienced at present, specifi-
the 9th International Speleological Con-
Proceedings of
cally the vertical shafts associated with cockpit karst.
124-127.
gress, Spain 1,
of the karst has
Detailed geomorphological mapping Brook, G.A. & Hanson, M. 1991. Double Fourier series analysis
been restricted to a few small localities and more system- of cockpit and doline karst Brown’s Town, Jamaica.
near

atic mapping of the karst of the White Limestone Group 37-54.

Physical Geography 12,
needs to be undertaken. In addition, only a
few morpho- Brown, M.C. & Ford, D.C. 1973. Caves and groundwater
in a West In-
metric studies of the karst have been undertaken in Ja- patterns tropical karst environment, Jamaica,

dies. American Journal ofScience 273, 622-633.

and these have been limited to small isolated areas
maica,
Canter, R. 1987. Cockpits and pits. National Speleological
of karst for quantitative analysis and classification of the

Society News 45, 53.

landforms. A combination of geomorphological mapping
Chenoweth, M.S. & Day, M.J. 2001. Developing a GIS for the
and a more systematic morphometric analysis of the karst
Jamaican Cockpit Country. In: Geotechnical and Environ-
would increase base and under-
greatly our knowledge
mental Applications ofKarst Geology and Hydrology, Pro-
of the karst the White Limestone
standing landscapes on
Sink-
ceedings of the 8th Multidisciplinary Conference On

Group. holes and the Engineering and Environmental Impacts of

Karsts, Louisville, 1-4 Balkema (Lisse).

Kentucky, April.
Chubb, L.J. 1963. Bauxite genesis in Jamaica. Economic Geol-

58, 286-289.
Acknowledgements ogy

Clarke, O.M. Jr. 1966. The formationof bauxite on karst topog-

thanks Dr Simon F. Mitchell for his in raphy in Eufaula district, Alabama and Jamaica, West Indies.
Many to help

1, and Mr Leonard Notice who Economic Geology 61, 903-916.

compiling Figure to com-

Comer, J.B. 1974. Genesis of Jamaican bauxite. Economic

pleted the remaining figures and scanned the photographs.
Geology 69, 1251-1264.
I thank Mick Day and Donald McFarlane for their con-

Genesis of Jamaican bauxite Eco-

Comer, J.B. 1976. -
a reply.
structive review comments.
nomic Geology 71, 822-823.

Comer, J.B. & Jackson, T.A. 2004. Miocene bentonites in the

White Limestone Group, Jamaica. In: Donovan, S.K. (ed.).

References White Limestone of Jamaica.
The Mid-Cainozoic Group
Cainozoic Research 3, 31-37.

R.L. & A. H. 1966. Genesis, miner- C.W. & McDowell, F.W. 1980. Fission-
Ahmad, N., Jones, Beavers, Comer, J.B., Naeser,

and related of West Indian soils. Part I; track of zircon from Jamaican bauxite. Economic Geol-
alogy properties ages
 -217-

ogy 75, 117-121. Gardner, T.W., Back, W„ Bullard, T.F., Hare, P.W., Kesel,

Danes, J.V. 1909. Studien im Karst- R.H., Lowe, D.R., Menges, C.M., Mora, S.C., Pazzaglia,
Geomorphologische
Gebiete, Jamaikas. 9eme Congres International de F.J., Sasowsky, I.D., Troester, J.W. & Wells, S.G. 1987.

Geographic, Geneve 1908 2, 178-182. Central America and the Caribbean. In: Graf, W.L. (ed.).
Danes, J.V. 1914. Karststudien in Jamaica. Sitzungsberichte der Geomorphic Systems ofNorth America, Centennial Special

Bohmischen Gesellschaft der Wissenschaften, Volume 2, p.343-402. Boulder (Geological Society of Amer-
koniglichen
mathematisch-naturwissenschaftliche Klasse 20, 1-72. ica).

Day, M.J. 1976. The morphology and hydrology of some Jamai- Grontmij, W. 1964. Black River Morasses Reclamation Project.

karst depressions. Earth Processes 1, 111-129. Report to the Government ofJamaica, Ministry ofAgricul-
can Surface
M.J. 1977. Surface hydrology within polygonal karst 291
Day, ture and Lands, pp.

in northern Jamaica. Proceedings In- M.

depressions of the 7th Hanson, 1985. Quantitative Analysis ofDoline and Cockpit
ternational Speleological Congress, Sheffield, 143-146. Karst near Brown's Town, north Jamaica. Georgia (De-
Day, M.J. 1978a. The Morphology of Tropical HumidKarst with partment of Geography, University of Georgia) (unpublished
to the Caribbean and Central America, MA thesis).
particular reference
611 Oxford (University of Oxford) D Phil Hill, R.T. 1899. The geology and physical geography of Jamaica:
pp. (unpublished
thesis). a
study of a
type of Antillean development. Bulletin of the
Day, M.J. 1978b. and distribution of residual Museum Harvard 34, 1-256.
Morphology of Comparative Zoology,
limestone hills (mogotes) in the karst of Puerto Rico. Geo- Hose, H.R. 1963. Jamaica-type bauxites developed in

limestones. Economic
logical Society ofAmerica Bulletin 89, 426-432.
Geology 58, 62-69.
Day, M.J. 1979. The hydrology of polygonal karst in Ireland, P. 1979. variations of‘case harden-
depressions Geomorphological
northern Jamaica. in Puerto Rico.
Zeitschriftfur Geomorphologie, Supple- ing’ Zeitschrift fur Geomorphologie, Sup-
ment-Band 32, 25-34. plement-Band 32, 9-20.

Day, M.J. 1982. The influence of some

material properties on Ivanovich, M. & Ireland, P. 1984. Measurements of uranium

the development of tropical karst terrain. Transactions ofthe series disequilibrium in the case hardened Aymamon lime-

British Cave Research Association 9, 27-37. stone in Puerto Rico. Zeitschrift fur Geomorphologie 28,

Day, M.J. 1984. Predicting the location of surface collapse 305-319.

within karst depressions: a

Jamaican example. In: Beck, F.B. Landmann, M. 1985. Poljeboden und Kegelkarst im Lluidas

(ed). Sinkholes. Proceedings of the 1st Multidisciplinary Vale, Jamaica. Annales de la Societe geologique de Belgique
147-151.
Conference, Orlando, Florida, 1984, 108, 175-183.

Day, M.J. 1985. Limestone in north central Landmann, M. 1990. Relief genesis in the Lluidas Vale of
valley systems area

Jamaica. Caribbean Geography 2, 16-32. Central Jamaica. Zeitschrift fur Geomorphologie, Supple-
Doerr, A.H. & Hoy, D.R. 1957. Karst of Cuba, ment-Band 77, 67-80.
landscapes
Puerto Rico and Jamaica. 178-187. Lehmann, H. 1954. Der tropische Kegel-karst auf den Grossen
Scientific Monthly 85,
Draper, G. 1987. A revised tectonic model for the evolution of Antillen. Erdkunde 8, 130-139.

Jamaica. In: Ahmad, R. of Lewis, J.F. & Draper, G. 1990. and tectonic evolution
(ed). Proceedings a Workshop on Geology
the Status of Jamaican Geology. Journal
of the Geological of the north Caribbean margin. In: Dengo, G. & Case, J.E.

Society of Jamaica, Special Issue 10, 120-150. (eds). The Geology ofNorth America, Volume H, The Car-

Draper, G. 1998. Geological and tectonic evolution of Jamaica. ibbean Region, 77-140. Boulder (Geological Society of

Contributions to
Geology, UWI, Mona 3, 3-9. America).

Draper, G. & A.G. 1997. of Jamaican E. 1774. The History

Fincham, Geomorphology Long, of Jamaica, or General Surveys of
limestones. In: Fincham, A.G. Jamaica The the Ancient and Modern State
Underground: of that Island. London (T.

Caves, Sinkholes and Underground Rivers ofthe Island, 29- Lowndes).

33. Kingston (The Press, University of the West Indies). Lyew-Ayee, A. (ed.). 1982. Proceedings of Bauxite Symposium

Draper, G., Lindenberg, M. & Miller, D.J. 1998. Evaluation of V. Journal ofthe Geological Society of Jamaica, Special Is-
a new, rapid nearest neighbour analysis of locations of sinks sue 1, 1-328.

and summits of tropical karst in the Jamaican Lyew-Ayee, P., Scroggins, E.A., Kronberg, B.I., Doherty, &
polygonal W.

Contributions Vander Voet, A. 1989. Genesis of Jamaican bauxites:

Cockpit Country. to Geology, UWI, Mona 3, a geo-
th
79. chemical model. In: Marusic, R. (ed.). Proceedings, 6 In-

Draper, G. & Wadge, G. 1997. The geology of Jamaican caves. ternational Congress of ICSOBA, Pacos de Caldas, 1989.

In: Fincham, A.G., Jamaica The Caves, Sink- Academic Yougoslave des Sciences et des Arts, Travaux-
Underground:
holes and Rivers
Underground of the Island, 23-28. King- ICSOBA 19, 193-200.

ston (The Press, University of the West Indies). Lyew-Ayee, P. & Stewart, R. 1982. Stratigraphic and composi-

Fermor, J. 1972. The dry valleys of Barbados: a critical view of tional correlation between bauxites and their limestone hosts

their pattern and origin. Transactions of the Institute of in Jamaica. In: Lyew-Ayee, A. (ed.). Proceedings of Bauxite

21st- 1982.
British Geographers 57, 153-165. Symposium V, 24th June, Journal of the Geo-

Fincham, A.G. 1997. Jamaica Underground: The Caves, Sink- logical Society of Jamaica, Special Issue 7, 19-37.

holes and Underground Rivers of the Island, xv + 447 pp. MacGillivray, C.M.I. 1997. A review of quantitative techniques

Kingston (The Press, University of the West Indies). for measuring karst terrain, with examples from the Carib-

Ford, D.C. & Williams, P.W. 1989. Karst and bean. Caribbean Geography 8, 81-95.
Geomorphology
Hydrology, xv + 601 London (Unwin Hyman). McDonald, R.C. 1975. Observations on hillslope erosion in
pp.

Gams, I. 1978. The polje; the problem of its definition. tower karst topography of Belize.
Geological Society of
Zeitschrift fur Geomorphologie 22, 170- 181. America Bulletin 86, 255-256.
-218-

McDonald, R.C. 1976. Hillslope base depressions in tower karst Longmans, Green & Co.).

topography of Belize. Zeitschrift fur Geomorphologie, Sup- Sinclair, I.G.L. 1966. Estimation of the thickness of limestone

plement-Band 26, 98-103. required to the Jamaican bauxites

produce by the residual

McDonald, R.C. 1979. Tower karst geomorphology in Belize. process. Journal ofthe Geological Society of Jamaica 8, 24-

Zeitschrift fur Geomorphologie, Supplement-Band 32, 35- 31.

45. Sinclair, I.G.L. 1967. Bauxite genesis in Jamaica: evidence

new

McFarlane, D.A. 1980. Liverpool University expedition to from trace element distribution. Economic 62.
Geology
Jamaica. Transactions of the British Cave Research Asso- Sinclair, I.G.L. 1976. Genesis of Jamaica bauxite. Economic

ciation 7, 150-168. 821-822.

Geology 71,
Miller, D.J. 1998. Invasion of the cockpits; of Partners. 1998.
patterns en- Sir William Halcrow & South Coast Sustainable

croachment into the wet limestone rainforest of Cockpit Development Study. Technical Report 4, Geology and Natu-

Country, Jamaica. In: Barker, D. & McGregor, D.F.M. (eds). ral Hazards, 115 pp. Halcrow (Kingston).

Resource Sustainability and Caribbean Development, 373- Smith, D.L 1968. University of Bristol Karst
Hydrology Expedi-
389. Kingston (The Press, University of the West Indies). tion to Jamaica; preliminary report.
Proceedings ofthe Brit-

Miller, D.J. & Donovan, S.K. 1999. Geomorphology of the ish Association
Speleological 6 (43-44), 31, 32.

Natural Bridge at Riversdale, parish of St. Catherine, Ja- Smith, D.L (ed.). 1969a. Limestone geomorphology: in
a
study
maica. CaribbeanJournal ofScience 35, 112-122. Jamaica.
Proceedings of the British Speleological Associa-

Miotke, F.D. 1973. The subsidence of the surface between tion 6 (43-44), 85-166.

Puerto Caves 1-
mogotes in Rico east Arecibo. and Karst 15, Smith, D.L 1969b. The solutional erosion of limestone in the

12. area around Maldon and Maroon Town, St. James, Jamaica.

Mitchell, S.F. 2004. Lithostratigraphy and palaeogeography of Proceedings ofthe British Speleological Association 6 (43-
the White Limestone Group. In: Donovan, S.K. (ed.). The 44), 120-135.
Mid-Cainozoic White Limestone Group of Jamaica. Caino- Smith, D.L 1970. The residual for the for the forma-
hypothesis
zoic Research 3, 5-29. tion of Jamaican bauxite. Journal
of the Geological Society
Monroe, W.H. 1966. Formation of tropical karst topography by of Jamaica 11, 3-12.

limestone solution and reprecipitation. Caribbean Journal Smith, D.L & Atkinson, T.C. 1976. Process, landforms and

ofScience 6, 1-7. climate in limestone regions. In: Derbyshire, E. (ed). Geo-

Monroe, W.H. 1968. The karst features of northern Puerto Rico, and Climate, 387-409. London
morphology (Wiley).
National Speleological Society Bulletin 30, 75-86. Smith, D.L, Drew, D.P. & Atkinson, T.C. 1972. Hypotheses of

Nicholson, F.H. & Nicholson, H.M. 1969. A new method of karst landform development in Jamaica. Transactions of the
measuring soil carbon dioxide for limestone solution studies, Cave Research Group of Great Britain 14 (2), 159-173.

with results from Jamaica and the United Kingdom. In: Sweeting, M.M. 1956. observations in parts of the
Hydrological
Smith, D.l. (ed.). Limestone Geomorphology, a Study in Ja- White Limestone areas in Jamaica, B.W.I., July-September,
maica. Proceedings ofthe British Speleological Association 1955. Bulletin ofthe Geological Survey of Jamaica 2, 27 pp.

6(43-44), 136-148.
Sweeting, M.M. 1957. Notes on the caves of Jamaica. Transac-

Panes, V. & Stelcl, O. 1968. Physiographic and geologic control tions of the Cave Research Group of Great Britain 5, 1-11.
in the development of Cuban
mogotes. Zeitschrift fur Geo- Sweeting, M.M. 1958. The karstlands of Jamaica. Geographical

morphologie 12, 117-173. Journal 124, 184-199.

Pfeffer, K.-H. 1967. Neue Beobachtungen im Kegelkarst von Sweeting, M.M. 1972. Karst Landforms, xvi + 362 London
pp.

Jamaica, Tagungsbericht und wissenschafliche Abhandlun- (MacMillan).

Deutscher GeographentagBad J. 1934. The asymmetry of the Hills of
gen, Godesberg. Thorp, Pepino Puerto

Pfeffer, K.-H. 1969. Kalkkrusten und Kegelkarst. Erdkunde 23, Rico in relation to the trade winds. Journal ofGeology 42,

230-236.
537-545.

Pfeffer, K.-H. 1975. Zur Genese Trudgill,

von Oberfldchenformen in S.T. 1977. The role of a soil cover in limestone weath-

Gebieten mitflachlagernden Carbonatgesteinen. Weisbaden. ering, Cockpit Country, Jamaica. Proceedings of the 7th In-

Pfeffer, K.-H. 1984. Relationen zwischen Formenschatz und 404.

ternationalSpeleological Congress, Sheffield,
Gestein im Suwesten Jamaikas. Die Hohle 35, 247-252. Urquhart, A.W. 1958. The Landforms of the Cockpit Country
Pfeffer, K.-H. 1986. Queen of Spain’s Valley, Maroon Town, and its Borderlands, 87 Berkeley (University of Califor-
pp.

Jamaica: cross-section of different types of tropical karst. nia

a at
Berkeley) (unpublished MS thesis).
In: Paterson, K. & Sweeting M.M. (eds.). New Directions in Versey, H.R. 1962. The character of the White
hydrological
Karst. Proceedings of the Anglo-French Symposium, 1983, Limestone Formation of Jamaica. In: Weaver, J. (ed.).
349-362. Transactions ofthe 2nd Caribbean Geological Conference,
Pfeffer, K.-H. 1997. Paleoclimate and tropical karst in the West Puerto 59-68.
Mayagiiez, Rico, 4-9 January, 1959, May-
Indies. Zeitschrift fur Geomorphologie, Supplement-Band of Puerto Rico).
aguez (University
108, 5-13. Versey, H.R. 1972. Karst of Jamaica. In: Herak, M. &
String-
Robinson, E. 1994. Jamaica. In: Donovan, S.K. & Jackson, T.A. field, V.T. (eds). Karst: Important Karst Regions of the

(eds.). Caribbean Geology: An Introduction, 111-127. King- Northern Hemisphere, 445-466. Amsterdam (Elsevier).

ston
(University of the West Indies Publishers Association). G. & Draper, G. 1977a. Influence of lithology Jamai-
Wadge, on

Sawkins, J.G. 1869. the Geology of Part

Reports on Jamaica, or can cave morphology. Proceedings ofthe 7th International

II of the West Indian Survey (with Contributions from G.P. Speleological Congress, Sheffield, 414-416.

Wall, Lucas Barrett, Arthur Lennox & C.B. Brown). Mem- Wadge, G. & Draper, G. 1977b. Tectonic control of speleogene-
oirs of the Geological Survey, 340 London (HMSO & sis in Jamaica. the 7th International
pp. Proceedings of
 -219-

416-419. 61.
Speleological Congress, Sheffield,
Waltham, A.C. & Smart, P.L. 1975. Caves of Jamaica. Bulletin Zans, V.A. 1951b. On the karst hydrology of Jamaica. Interna-

Association 10,25-31, tional Association de Hydrologic Scientiflque 2, 267-279.

of the British Cave Research

White, W.B. & Dunn, J.R. 1962. Notes on the caves of Jamaica. Zans, V.A. 1953. The geology of Mosely Hall Cave, Jamaica.

National Speleological Society Bulletin 24, 9-24. Natural History Notes, Jamaica 58, 156-163.

V.A. 1954a. The St. Clair Cave in St. Catherine. Natural

Williams, P.W. 1971, Illustrating morphometric analysis of karst Zans,
New Jamaica 68, 137- 141.
with examples from Guinea. Zeitschrift fur Geomor- History Notes,
V.A. 1954b. Caves and sinkholes. In: Handbook of Ja-
phologie 15, 40-61. Zans,

of
Williams, P.W. 1972a. Morphometric analysis of polygonal karst maica for 1954, 16-17. Kingston (Government Jamaica,

in New Guinea. Geological Society ofAmerica Bulletin 83, Government Printing Office).

761-796. Zans, V.A. 1958a. A note on Dunn’s Hole and its cave. Geo-

Williams, P.W. 1972b. The analysis of spatial characteristics of notes 1, 95-99.

(ed.). V.A. 1958b. Caves and sinkholes. In: Handbook of Ja-

karst terrains. In: Chorley, R.J. Spatial Analysis in Zans,
of Jamaica, Govern-
Geomorphology, 136-163. London (Methuen). maica, 18-19. Kingston (Government

Wright, R.M. (ed.). 1977. Origin of Bauxite: Selected Papers ment Printing Office).
of bauxite. Geo-
reprintedfrom Geonotes and the Journal of the Geological Zans, V.A. 1958c. Recent views on the origin

Society of Jamaica, 67 pp. Kingston (Geological Society of notes 1, 123-132.

V.A. 1959. Caves and exploration in Jamaica.
Jamaica). Zans, cave

59-69.
Zans, V.A. 1951a. Economic geology and mineral resources of Geonotes 2,

Jamaica. Bulletin ofthe GeologicalSurvey of Jamaica 1,1-