Central Java Fieldtrip (instructed By Peter Lunt) 6z4v5h

© IPA, 2006 - IPA Field Trip to Central Java, 1998

Oct - Nov. 1998

IPA Field trip to Central Java

IPA field trip to Central Java,

Trip Guides Peter Lunt (Coparex Blora b .v.) Richard Netherwood (Schlumberger)

0.Frank Huffman (Archeological Research Laboratory, University of Texas at Austin)

Page: 1

Disc Contents

Search

CONTENTS Introduction

.

Location Summaries

.

Karagsambung Section

.

Baturagung Escarpment & Jiwo Hills

.

Sangiran Dome .

Addendum: Sangiran Dome A Palaeontologist's View of Java Man . 6 pages An Anthropologist's View of Java Man .15 pages by 0. Frank Huffman

-

P A Field trip to Central Java

Oct - Nov. 1998

IntroductEon The ob/ecthes d tbe t*

-

This trip was originally planned to visit the three classic Eocene sites in Central Java two of which have never been visited by previous IPA field trips. These sites were first uplifted in Miocene times so consequently we can take this opportunity to study the stratigraphic expression of this regional Miocene unconformity. Also the classicJava man sites can be visited, and we have a knowledgable guide in FrankHuffman. Frank has considerable experience in the oil industry and is currently trying to use industry subsurface data sets, including seismic, to better define the palaeogeography of Java during the Pleistocene,

As everyone is aware, this trip was first planned for May 1998. In principle it should have been , possible to run it to the same schedule in October. However the weather has not been kind and an unusually damp dry season has been followed by a very prompt start to the rainy season, so the trip W'%%-$ * will be run with a different itinerary. For the record this Field Guide is being mostly left in its original i&

, 5

We think that during the course of this field trip it will become apparent that the Eocene sections in Central Java have considerable stratigraphic similarities, a consistency that is not apparent fiom the sparse published literature on the sections. This stratigraphy, is quite different h m the well established Eocene stratigraphyof the offshore easternJava area (Bransden & Mathews, P A 1992) and apparently also western Java (Schiller, 1991). It should not be surprising that there was a varied palaeogeographyand complex patterns of sedimentation in Eocene times as it was in the mid Eocene that a completely new regional tectonic episodebegan. This was the initiation of many of the modern sedimentarybasins in SoutheastAsia. A commontheme observed during this trip is the transgression of basement by polymict conglomerates (composed of reworked basement lithologies)pas into shallow marine and then deep marine sediments. The sedimentswe see in the Middle Eocene outcrops commonly contain volcanic rnakrial, indicatingthat volcanism was initiated long before the Oligocene. However within Oligocene times a new influence affected the sites we will visit, and it is the reason why most of them are exposed today. This was the initiation of the "OldAndesite" volcanic arc, which intrudes, uplifts, and then coversthe areaswith thick volcaniclastic deposits. It is presumed that the original Eocene volcanic arc was located elsewhere and it was only in the Oligocene that it moved to the mappedb'Old-Andesite"locations. The originalarc position may have been to the south, migrating north in Oligocene times, just as it migrated north again between the activity of the Old-Andesites and the development of the modern arc. As a consequence of visiting the Eocene sites it is possible to see some outcrops relevant to Early and Middle Miocene stratigraphic development, especdy the events at the termination of Old-Andesiteactivity.

Page: 2

P A Field hip to Central Java

Oct - Nov, 1998

Page: 3

Oct - Nov. 1998

P A Field trip to Central Java

The use of the Global Positioning System (GPS) has revolutionised navigation in field geology.Now it is possible to record a outcrop or sample location to within about 50 meters, so that re-location is easy for anyone. This can become pinpoint re-location with the aid of a photograph, sketch or a good description. The following latitudes and longitudesare the locations recorded at key sites visited on this field trip, as well as some road junctions important for access. The data is given in decimal degreesto avoid confusion over different numerical bases, and not in UTM coordinates as this requires consistency and declarationof the chosen spheroid.

Karangsambung area (locations from south to north) River section of Late Eocene clastics The Jatibungkus Lst. -western tip, nearest road The Jatibungkus limestone center, base River section of Middle Eocene clastics Scaly clay location (ongoing coprolite deposition) On road next to the M. Eocene Nummulites limestone Near the road the Sanggrahan Sst & conglomerate The S a n g g r a b sandstone in the river Bridge over reported pre-Tertiary basalts and "pillow lavas" Turn off road to the "marble & schist" quarry The marble and schist quarry "Serpentinite" in road cut

-

-

1 0 9 . 689200°E, 1 0 9 . 682180°E, 1 0 9 . 687320°E, 1 0 9 . 685680°E, 1 0 9 . 676440°E, 1 0 9 . 6711709E, 1 0 9 . 670490°E, 1 0 9 . 667830°E, 1 0 9 . 667620°E, 109.675570°E, 109.674290°E, 109.691350°E,

Nanggulan area Kenteng X-roads Turn off main road to go west of Pendul Turn of E-Wroad to go north to Kalisongo Turn of E-W road to go north to Girimulyo Junction and parking at Girimulyo Kalisongo small cross-roads near center of Eocene outcrop Axinea beds N. Kalisongo (outcrops at heart of anticline) Center of anticline on road from Pendul D j o g d j h r t a e beds in north (in river valley) Top of section near Jetis village Upper W d e d volcanics in road cut Old Andesite breccias in road cut

1 1 0 . 210164'E, 1 1 0 . 213380°E, 110.204574°E, 110.187960°E, 110.188020°E, 110.196700°E, 110.199670°E, 110.200460°E, 110.200600°E, 110.186560°E, 110.184980°E, 110.177100°E,

Jiwo Hills Turn off road on track to Pendul (Nummulites boulders by road) Pendul location Mid Miocene Lst overlying Eocene, east of Pendul Slope site of Gunung Cakaran Top of Gunung Cakaran Turn off to reach access path to G. Cakaran

-

110.673003°E, 110.670450°E, 1 1 0 . 677300°E, 1 1 0 . 622660°E, 1 1 0 . 624130°E, 1 1 0 . 621910°E,

Oyo Widoro section Confluence of the two rivers 110.548135°E, Sangiran Dome,site of extinct mud volcano and Eocene / basement and other ejecta. 1 1 0 . 837140°E, At Sangiran, please do not sample if you do not have to. Ten years ago I saw a 2 meter plus boulder of Eocene limestone. This has now been reduced to small amounts of rubble by collectors who probably don't know a Pellatispira from a pisang.

Page: 4

IPA Field trip to Central lava

Oct - Nov. 1998

The dating of the onset of new basin development in SoutheastAsia is important in establishinggeological models, but it is difficult, as the first transgressivesediments are &void of marine fossils. The study of terrestrial microfossils (pollen and spores)rarely gives accurate absolute ages (although studies of changes in these fossil floras can yield much data on intra-basinal correlation and environments).In most of the new Sundaland basins the first marine sedimentsare dated as Middle or Late Eocene. Reports of older sediments are rare and one such example is the section at Karangsambung (however at this locality we can show not only that the oldest marine beds are Middle Eocene, but also explain fiom a historical perspective why early workers thought there may have been Paleocene at this location; see page 24). About the only reasonable data indicating marine sediments of Early Eocene or older Tertiary may come fiom the far east of the Java Sea region (Amoco's early 80's "L" .' series wells) but this still needs substantiation.These wells could be acceptable exceptionsto the general trend as *-- 5 $ i they are on the far southeast margin of Sundaland.

-4

2@

the^ are geological scenarios of Sundaland basins opening in geographically linear sequencesor at differ-

ent times throughthe Eocene, but it is difficultto prove these models in the absence of good age dating or good regional seismic.Around the southern and eastern margin of Sundaland it appears that many new basins were mitiat& and became the site of new, highly active sedimentarysystems within the diree to four million year period that is the later part of the Middle Eocene. The plate tectonic events that led to the new basins and sedimentary conditionsin Sundaland are reasonably well known and dated. These are the northwards acceleration of Australia fkomAntarctica, the collision of India with the Asian plate and the resulting Himalayan orogeny, and the shift in motion of the Pacific Plate and its resulting break-up into several sub-plates. Of these events the first and last can be dated by magnetostratigraphic and radiometricmethods with somepaision as about 45 MYFJP (mid Middle Eocene). The onset ofthe Himalayan orogeny, which may have triggered the other events after the long drift north of India was arrested, is harder to date as it involves uplift and non-marine sedimentation. However the sialic derived S$' of the worlds oceans has been increasing steadily since the start of monsoonal erosion of the Himalayas, and this first seemsto increase at about 47 MYBP (see figures on pages 27 and 29).

The preliminary field work for this trip can confirm that there was volcanic activity in the Central Java region in Middle Eocene times. This activitywas close, but not at, the areas we visit and is most likely some way to the south, in what is now offshore south Java. The "Old Andesite" volcanic arc is the oldest volcanic material mapped by the Dutch or the GRDC,and it is Oligoceneto mid Early Miocene in age. "Old Andesite" intrusive and extrusive volcanics are associatedwith the uplifted Eocene sites at Nanggulan and Jiwo. The "Old Andesite" phase came to an end at about 20 MYBP. After a period of relative quiescence, the modern arc appeared at about 12 MYEP, but in a position approximately 50 kilometers W e r north of the Old Andesite chain.

Page: 5

3x:GC" =, +-

.

;. -

,:. ; "-

a

IPA Field trip to Central Java

Oct - Nov. 1998

Page: 6

IPA Field trip to Central lava

Oct - NOV.1998

The following is a list of known Eocene outcrops in Java, the fmt four are visited duringthis trip

00 The Nanggulan site west of Yogvakarta. 00 The Jiwo Hills (Bayat area) sites east of Yogyakarta, which are mainly around Gunung Cakaran in the west and Gunung Pendull Gamping village in the east.

00 Late Eocene ejecta from the Sangiran Dome mud volcano.The Eocene limestone boulders were over -+,-,.., 2 meters in size until a few years ago, but now are reduced to minor rubble. -.:- =? ,

+

-7

.00The Karangsambung sectionwhich is the lower part ofthe Lukulo River section, with perhaps the most g . .

.

<* s

+ -: + ,

L'

extensive Eocene section. Surprisinglypoorly documented.

?

,

' * _"r

O North of the Kasangsambung section, in the upper reaches of the same Lukulo River, is an extensive but .

.

*

I

poorly exposed and hard to reach site. This m a is southeast of Banjarnegara and southwest of Wonosobo. -; r $-+ar Almost no published data is available. Boulders of Middle Eocene Numrnulitic limestone are known =-%-. here.

0 Just north of the upper Lukulo River site, near Gunung Worawari, west of Wonosobo, is the type locality of the now defunct term "Bagelen beds" (see Marks 1957, and van Bemmelen 1949).Again little published data except the GRDC 1975Banjarnegaraquadrangle map reported on some samples from here. Outcrop is very poor. Boulders of Middle Eocene Nummulitic and Fasciolites 1Alveolina limestone and some beds of Eocene mudstone are known fkom here. 0 A location called Kalibongbong about 15 kms NW from G. Worawari (see fig 301 a & b of van Bemmelen 1949),however this sitewas not mappedby the GRDC. A further 15kms NW is the Sigugur Limestone type locality (Late Oligocene reefd limestone; see figure 301 in van Bemmelen) overlying a small area of poorly exposed Eocene. Access is difficult to both these sites. These two, plus the G. Worawari site above, are often lumped together as the North Seraju Mountains Eocene sites (the South Seraju Mountains sites are the north and south Karangsambung sites). O Near Nanggulan there are three other Eocene outcrops, which are smaller and also show the effects of metamorphism, which is absent at Nanggulan. About 10kms due east is a location marked as Gunung Wungkal on the Yogyakarta qwadrangle,GRDC geological map. This is not the type locality of the Wungkal Beds, which is a hill on the east side of the Bayat 1Jiwo Hills area (cf. fig 266 in van Bemmelen). The Yogyakarta quadranglelocation is the GodeanHills locationdescribed in van Bemmelen and probably the Gamping Hill of Gerth (1930; -gumping means limestone).Note that the limestone here has been mostly removed by quarrying and the remnants are difficult to find.

In the Old Andesite volcanic areas immediatelywest ofNanggulan, called the West Progo Mountains, are two small locations, onejust west ofBombudur (Gandul) and one southwestofNanggulan, not far h m Wates (site called Serrno) where Eocene can be found (see page 105 and figure 296 on p.597 of van Bemmelen where the West Progo Mountains are called the Menoreh Hills).

Page: 7

P A Field trip to Central lava

Oct - Nov. 1998

O In west Java there are three areas of outcrop quite close to each other, clustered around the fishing and resort town of Pelabuhan Ratu. To the west is Bayah which is described in van Bemmelen (spelt Bajah) and briefly in Keetly et al. (1997, IPA -Petroleum Systems). To the east is the Ciletuh Bay section well described in Schiller et al. (1991, IPA). A few smaller outcrops (eg. Gunung Walet) are reported in the Cimandiri (Tjirnandiri) river valley northeast of Ciletuh, and east of Pelabuhan Ratu (see map on page 618and sectionson page 619of van Bemmelen) but it is not certain if these sections are Eocene in age. Van Bemmelen suggested they were Eocene as they are composed of non-metamorphosed sandstones, carbonaceousshalesand coals,and are conformably suceeded by marly clays containing reticulate (=Early Oligocene) Nummulites.

Finally in a location southwest of Semarangworkers have noted Eocene fossils reworked into Late Oligocene or early Miocene beds as conglomeraticclasts (see van Bemmelen p 106).

Page: 8

-,c

LS

Oct

IPA Field hip to Central Java

- Nov. 2998

+ Stmt.position demd

a

Chalky Lm (Globlgrinr) Sandy forami,,ifd La Foraminifera1 Lst Giant Nwnmulites

Limestone

0Turbiditic facies 'UMudstcme UP-,

,cmi.ias+.ic sandstone

Page: 9

Page 9

P A Field trip to Central Java

Oct - Nov. 2998

The NonpguIan section swnbna4y

. .- &<-:

ex*>A'. -a

;^."-

The Nanggulan section was the first sequence of Eocene deposits recognised in the Indo-Pacific area. While the section is a key locality for Eocene biostratigraphy,particularly palynology, there is little publicly -- : availablemodem data on the sedimentologyor organic geochemistry of the sands or the lignitic coals found here. , g,. -.--*: In the vicinity of Nanggulan there are four outcrops of Eocene, three of which show the effects of metamorphism fiom "Old Andesite" intrusions. Nanggulan is the only outcrop that is unaffected. It is also the - . longest section, starting in the core of a gentle anticline and continuing for about 400 meters of vertical section, ,, entirely within the Middle Eocene. kgi? Tz

8

Geologicalmap of the area just weat of Yogyakarta. The Djonggranan and Sentolo limestones unconformably overly the rest of the (non-Quatetnary) sediments. The reefal Djonggrangan tending to occur over the "stubs" of the Gajah, Ijo and Menoreh (just to the north) volcanic cores. This grades laterally into the pelagic dominated Sentolo limestone. The southern hill at Godean does not have as much Eocene (Teon) as mapped, being dominated by Old Andesite volcanic breccias and tuffs which are extensively quarried for making into bricks and roofing tiles. The limestone that used to be hete has largely been removed for use in building.

Page: 10

-

P A Field trip to Central Java

Oct - Nov. 1998

The central, oldest, unit exposed is at least 30 meters thick and can be identified as the Axinea Beds of 0ppenoort.h& Gerth (1929). This is a quartz rich sand unit, often poorly cemented, highly fiiable with excellent visible porosity. At the base of the exposure the sediments are fine to very coarse grained, sometimes cross bedded and with small channels. These up into gradually finer sands, silty sands, and laminated sands and mudstones. Numerous lignite beds occur, up to 50 cm thick but usually 10 - 15 cm and laminated with sands. Some dolomitic cemented sand beds occur, as well as well cemented channel filling, gastropod coquina sands. The gastropods include Axinea dunkeri after which the unit is named. Localized trough cross-bedding and occasional ripple cross-bedding, plus the gastropods and coal layers suggestsa channelized, but otherwise low energy, littoral setting for thisunit. +q

The Axinea Beds up into bioclastic, often sandy mudstones. The fauna in the mudstone is dominated by thin-shelled bivalves, Nummulites, other benthic forms and platey codiacean algae. These are the ":Djogjakartae Beds, &r Nummulites djogakartae a Middle Eocene index larger foraminifera. The mudstones -. in this interval are often dark grey in color. Sands, when they occur, are matrix ed and not comparable-,,. with the reservoir quality sands in the underlying Axinea Beds, although they are apparently dominated by quartz ', :tgrains. -4 - $ r ** There is a considerablegap in exposure until the next unit is encountered.These are the Discocyclina Beds, .named afterthe Eocene index foraminifera which is p m t throughout. Typically the lithology is a silty or sandy '?+& mudstone with modest amounts of bioclastic material, locally with calcareousor dolomitic concretions.

% ; =

+

:z

%

7

5

-

"

-,

It is within this Discocyclina unit that the fmt tuffaceous beds occur. These appear as sandy mudstones with tu@ weathered crystallineor arkosic sands. Fine, white amygdaloidalmineralsmay also occur in the mudstones. On the limited section currently exposed the Discocyclim Beds appear to get progressively more sandy and coarser grained towards the top, although no true grain-ed sandstones have been recorded. Also upsection there is an increase in abundance of the larger foraminiferaDiscocyclina as well as an unidentified platy codiacean algae and traces of coal. All of this suggestsan overall regression. Overlying the Discocyclina Beds is a very different facies of very pale grey, slightly tuffaceous chalks. These are the Globigerinamark of Hartono (1969) or Seputih Member of Pwnamaninsih Siregar and Harsono Pringgoprawiro (1981). This thin unit, probably only a few tens of meters thick, contains abundant planktonic foraminifera and represents a flood or drowning out sequence. Above the Seputih Member is a unit that has so fhryielded only sparse nannofossik h m a sample near its base. This unit has no name, and is composed of thinly bedded, dark brown-grey mudstones with repeated, centimeter to decimeter thick, fine sands. These sands are composed of weathered feldspar with no quarb content. The thickness of this unit is uncertain but it appears to be at least several tens of meters. The sample that was dated from the basal part (by P.T. Corelab Indonesia) was reported as Middle Eocene, zones NP 15 NP 16 based on the overlapping ranges of Sphenolithus furcatolithoides, Reticulofenestra umbilica and Cribmcentrum reticulaturn.These same species are used to date other samples lower in the section as Middle Eocene.

-

Some comparison can be made between these upper bedded volcaniclasticsat Nanggulan and the clastic sediments at Jiwo (Pendul) section (Middle Eocene, zone NP16- 17), as well as the more fossilifemus turbidites at Karangsambung (Middle Eocene, within zones NP 14-16). Lithologically this unit appears to be a distal volcaniclastic sequence that might be regarded as a unit preceding the Old Andesite b i a s . However if it is truly Middle Eocene in age then somewhere between this and

Page: 1 1

IPA Field trip to Central Java

Oct - Nov. 1998

Map of the Nanggulan area. From new mapping work. Red dots are sample locations. Contours are at 25 meter intervals. Only roads close to the Eocene sediments are shown.

Page: 12

Oct - Nov. 1998

P A Field trip to Central Java

Samples, and nannofossil zonation.

400m

Q

E

0

o

NP15-16, Middle Eocene (poor floras)

Pa,

0

NP16-18, Middle or Late Eocene

.'L?

B?

0 Q 08

NP15-16, Middle Eocene NP15-16, Middle Eocene

80

NPl5-16, Middle Eocene

Q

NP15-16. Middle Eocene

Q

0

00

300m

8

E; k

P

h.

0

200m

-

Tuffaceous beds Coal 1 lignite I-Discocyclina . N djogjakartae M Planktonics i - Axinea gastropods /

1OOm

p 2

B B

90 8

NPlS-16, Middle Eocene

8 0 @

Indeterminate

0

base of exposurc

Simplified stratigraphy of the Nanggulan area.

Page:13

IPA Field trip to Central Java

Oct - Nov. 2998

Above: Thin section through a dolomitized unit in the Axinea Beds, lower part of the Nanggulan section (Middle Eocene). Note the quantity of fine quartz grains and some lithic !hgments. On the left is a foraminifera and at the top a mollusc. Below: Thin section through the conglomeratic beds at the center of the anticline, part of the Axinea Beds. In this very coarse grained sediment there is a higher proportion of lithic fragments, compared to medium or fine grained sands which are dominantly quartz. Scale in both images is a pencil lead 0.5 mm in diameter.

Page: 14

Oct - Nov. 2998

P A Field trip to Central Java

Thin section through a sandy marl from the Discocyclina Beds, middle part of the Nanggulan section (Middle Eocene). Note the quantity of volcanic material and lack of quartz grains. For scale; on the bottom right is a pencil lead 0.5 mm in diameter.

the younger breccias are beds including Late Eocene Pellatispira limestones. The evidence for this limestone is fiwntwo locations near by. Firstly such limestones are known h m boulders in the basal conglomerate ofthe mid Early Miocene Sentolo Fm. some 2%kmsto the south, near the village of Karanganyer. Here are "fairly abundant boulders of limestone full of Pellatispira" van Bernmelen p. 105. A similar limestone is known fiom the Godean Hills about 8 kilometers east of'Nanggulan, however this outcrop has been quarried for building material, reportedly for repairs to the Kraton at Yogyakarta, and the limestone is now hard to find.

-

The basal Sentolo boulders are assumed to have been reworked during the Early Miocene after cessation of Old Andesite activity, when the volcanoes were eroded to their cores along with the flanking Eocene sediments. The Godean limestones are stratigraphically isolated, presumably occurring below the Old Andesite volcanic deposits that occur in the surroundinghills, but at least they underline the occurrence of Late Eocene carbonate deposition in the Nanggulan area. A second attempt to c o n f m and more widely date the bedded volcaniclastics yielded no nannoflora or other microfossils, which is not very surprising considering the lithology and the sparse flora in the firstsample. This upper sbratigraphyat Nanggulan requires furtherwork.

Both the Karangsarnbung turbidites and the bedded mudstones and tufFaceous siltstones at Jiwo are followed by Late Eocene carbonates(Jatibungkus Limestone) or larger forarn mads (Discocyclinamark at Pendul).

Page: 15

Oct - Nov. 1998

P A Field trip to Central lava

+

The Karangsambung location is part of a wider area often referred to as the Lukulo River section or the South Seraju Mountains. The Karangsambung locality is the main area of Eocene outcrops in the south, while immediately to the north there are extensive outcrops of basement. We will start the day on the edge of the , basement complex and move upsection through the overlying Eocene sediments. -*

k .

g L

:--

" C -

5

Basement in the region is composed of crystalline schists, serpentinites, metasediments including low g*. grade marble, cherty red limestones, and locally volcanic rocks including pillows lavas of a pre-Tertiary age. :z.,:*.* Early Cretaceous Orbitolina larger forarns are recorded here, as are Cretaceous radiolaria from the cherty *?: limestones, and the GRDC has located Late Cretaceous plankton such as Globotruncana, Heternhelix and ":q -Q Rugoglobigerina. The pre-Tertiary outcrops are called the Lukulo complex, in which individual lithologic units ; , <* are very unevenly distributed,with each ridge or river visited having a different lithology. The whole unit is interpreted as a melange and the overall structural lineation is WSW to ENE a trend thought to parallel the orientation ofthe SE Sundaland margin in Late Cretaceoustimes. e

fi

-

Overlying this complex are Middle to Late Eocene sediments called the Karangsambung Formation (Sukendar 1974).The Karangsambung Formation is sometimes cited as an olistostromedeposit but there is little evidence suggestingthe whole sequence is allochthonous, althoughthe Late Eocene Jatibungkus limestone may be a large, &rived block (anolistolith). The overall stratigraphiattern is similarto other sites such as Jiwo and the Nanggulan section (although basement is not exposed in the latter). Overlying the basement are fluvial to shallow marine polymict conglomeratesthat up througha bransgressive sequenceto volcaniclastic turbidites, all withinthe Middle Eocene. Also within the Middle Eocene are relatively clastic-starved periods(local flooding events) with either Nummulites limestone or chalks.

748 O I I - ~ y r n #n)I~rn e As long ago as van Bernmelen (1949) and Harloff (1929,1933) the Eocene sediments at Karangsambung were recognised as tectonically complex; "Theeocene strata have served as a lubricant between the basement complex and the overlying Neogene" (van Bernmelen 1949,p. 603 and see cross-section on page 18). In the field this tectonic complexity is seen as very variable dip and many small faults in the brittle mudstones, especially in the central area of the Eocene outcrops where Harloff and later mappers interpreted a major thrust fault. The "scaly clays" reported by other workers have been visited at their type locality but do not appear to haveanunusualfizbrc. TheyaremtainlymuchmarebrittlethanthemoreusualtyencountcredNeogenemdstones, and consequently have a conchoidal to splintery fkture rather than the moist, plastic breakage of younger clays. This brittle character is common to nearly all the Eocene claystones we will see on the trip

The first reports known to me that discuss olistostromes in this area are both h m 1975;Paltrinieri suggests "The palaeontological examination shows more over that these sediments [Jatibungkus Lst.] have been deposited in shallow marine water, an environment quite diffeerent from the rest of the [Karangsambung] formation. The hypothesis that an olistostmmal mixture, which was occurred during the Late Eocene time, jilling troughlike depression, is the most probable.".

Page: 16

L~

/ _!

W

MmC. Te OuG.

Mid-

I 'F~ratB r c c c ~ aHor ' or WaturandaFm ; ~ O . + i y ~ T u f f or

-

O

1-

Kilometers

1

Fault, observed (a) mfcred (b)

1 Schist & others

'

Penosogan Fm

or

2

D~abare1 gabbro dror~te

"Second Breccm'. or Semoor F m in sandv facie5 to west

'Tblrd Marl Tuff or Rawakelc Fm

t - I 'Second 'Mart Tuff

LATE

Oct - Nov. 2998

P A Field trip to Central lava

In the same year @PAproceedings) Bolliger and Ruiter state "In the geographic center of Java (Lok Ulo, Banjarnegara area) an interesting melange of shallow and deep marine deposits is present, ranging in age from Upper Cretaceous (Cenomanian / Turonian), over Paleocene to Upper Eocene. Most probably we are dealing here with an olistotromal mixture, which was emplaced into a trough during the Late Eocene."

The former hypothesis above is a valid suggestion for the origin ofthe very localised Jatibungkus limestone as an olistolith.The * second statement may be responsible for starting the idea that the -% whole Karangsambungformation is an olistotrome*.However it is a bald, over-simplified hypothesis. Mapped distributions show the Cretaceousrecords to be within the basement melange complex, .:., and there is a considerable missing section of about 20 Ma before the appearance of Mid Eocene sediments (the Paleocene reported by Sukendar Asikin in 1973 unpublished thesis was shown to be -, Eocene by Bauman in 1974, see also page 24). -. +.-;

-

t L

r

I

+

:

-

-

-j.-%&

In a 1976IPA paper Paltrinieri and others returned to the theme of an olistotrome, citing the Bolliger and Ruiter paper in of this concept (the papragraph quoted above).No new data was presented for this interpretation, and the basal conglomerates, sands and Middle Eocene nummulitic limestonewere not found or considered by these workers. Since this time there have been no widely circulated papers on the section, and the olistotrome model has been unchallenged The Karangsambung area is a window of Eocene surrounded by Early Miocene volcanic breccias.Thesehard breccias (Waturanda Fm.) form a high ridge around the soft Eocene inlier. The precise stratigraphic relationship between the Waturanda and the Eocene sediments is difficult to establish as it would appear (cf. section on the left)that in the north it overlies a thinner Eocene section than to the south. However the east - west axis of folding of the Waturanda, which produces the elongate Karangsambung ink,also coincides with the area of disrupted bedding in the soft Eocene. It is therefore possible that the thrust slightly precedes the deposition of the Waturanda and causes some repetition of Eocene section. Whatever the exact relationship, it is important to note that the bottom and top of the Eocene section dips in a similar fashion to the overall trend of the Karangsambung monocline and it is in the central, axial zone that the erratic bedding is observed. It is also important to note that the stratigraphic succession, fiom Middle Eocene conglomerate on basement to Late Eocene turbidites, is comparable to other

* An olistrostrome is defined as "a sedimentary deposit consisiting of a chaotic mass of intimately mixed heterogenous materials (such as blocks and muds) that accumulate as a semifluid body by submarine gravity sliding or slumping of unconsolidated sediments."

Page: 18

IPA Field hip to Central Java

Oct - Nov. 1998

Totopn Fm. (also called "Clay Breccia" and "First Marl Tuff': breccia fragments, fine grained calcarenite and grey tuffaceous mark The beds are quite rich in planktonic foraminifera. Faunas indicate a Late Oligocene age and possibly a shallower environment than the underlying sediments.

Upper clastics: "bluish grey tuffs, to argillaceousmarls intercalated b) bluish tuffaceous sandstones". Globigerinatheca seminvolutu indicates the zone P 15 1 to basal P16, Globomtaliu cerroazulensis forms continue above them indicating Late Eocene sediments. Nannofossils indicate NP17-NP18, Late Eocene. Jati Bungkus Limestone: containing rare broken corals, common red algae, abundant larger foraminifera and bivalve remains. Locally with quartz grains, especially near the base where limestone conglomerate also occur. Forams include abundant Nummulites (inc. N nuttalli), Discocyclina and unconfirmed reports of Pellatispira species indicating Letter Stage Tb,Late Eocene in age. Paltrinieri et al. interpret the bed as an allochthonous unit transported into a deep marine setting. Conglomeratic to sandy base. Limestone conglomerate and also sandy limestone with non-calcareous clasts to 1 cm, mostly quartz and minor lithic fiagrnents (metasediments). Quartz is often polycrystalline.

Lower clastics: "Clay ironstones and sandy marls" rich in planktonic foraminifera and also with radiolaria (abundance unspecified). Truncatulinoidesmhri consistently present and indicates zone P 14, top of the Middle Eocene. Nannofossils indicate zone NP 15 or NP 16. Bedding and sedimentary structures indicate turbidite facies with finc sands of abundant feldspar minerals, minor volcanic glass and very rare quartz. Localised outcrops of deep marine mudstone, descibed as a "scaly clay" by some authors. Dated as zone PI 2- 13of foraminifera, NP15- 16 on nannofossils. Limestone: with abundant giant Nummulites indicative of the Middle Eocene. Basal sands and conglomerates with carbonaceous horizons.

Page: 19

P A Field trip to Central Java

Oct - Nov. 1998

sections in the region and there are no stratigraphically displaced beds. Based on this data it is simpler to interpret the Karangsambung formation as a normal stratigraphicdeposit with, as van Bernrnelen put it "TheEocene strata [serving] as a lubricant between the basement complex and the overlying Neogene".The structural developmentof the Karangsambunganticline would therefore be duringor after Waturanda deposition.

T4e dnt&@lc

rwmdu

A review of the volcanic arc history shows that Karangsambung would have been in a melange setting in Cretaceous times. The Paleocene and early Eocene are unknown but are considered to be a period dominated by erosion, as suggested by a significant d t yjump h m mid-Eocene to late Cretaceous sedimentselsewhere in east Java. It was only with the northwards acceleration of the Australian Plate at about 45 MYBP (mid Middle Eocene)that the southern margin of Sundalandwould have been changed to include a volcanic arc and back arc complex

The oldest beds of the Karangsambungformation are in the north of the outcrop area, although the with the underlying basement is not exposed (Prasetyadipers comm).In one location, identified as Sanggrahan, there are sands and conglomeratesnext to the road and more extensively in the south bank of the river about a hundred meters to the west. These clastics are well bedded polymict conglomerates, sandstonesand mudstones which are locally carbonaceous. Beds are typically 10to 20 centimeters and internal features are rare. No crossbeds or ripup clasts were seen although simple burrowing trace fossils are common. Sands are mostly quartzose with somemarked lithicmaterial.Carbonaceoushorizons or lambationsare common, sometimes as argillaceous or sandy "C081S" to a centimeterin thickness. Shortly above this are Nummulitic limestones, although again no is seen. These limestones contain typical very large Nummulites (N.javanus) species indicative of the Middle Eocene. Outcrop is poor until the location of the "Scaly Clays" just east of Karangsambungvillage. As mentioned above these clays are not especially outstandingalthough they contain a good Middle Eocene (zones P12 to P 13), deep marine planktonic and benthic assemblage. To the south,generally upsection, outcrop again disappears underneath padi, however in stream sections there are small exposures of bluish grey or brown clastics. These are thinly bedded turbidites, with milky white fine sand I silt beds a few centimeters thick composed wholly of feldspathicgrains, and with traces of volcanic glass. The mudstones are moderately rich in planktonic foraminifera of Middle Eocene age and indicative of a deep marine setting. Scours are sometimes seen on the base of the sand beds, as are de-watering flame structures and rip-up clasts. Less common are slightly graded sandy beds and irregular laminations that could be ripple cross-bedding. The turbidite sedimentsweather easily and both large and small blocks slump into the river sections. There are also many small faults, and dip, while mostly steep, varies in direction considerably. The thrust fault interpreted by Harloff (see earlier figure) is probably responsible for much of this disturbance.Near the stratigraphic top and the southern limit of the Karangsambung Beds are two E-W elongate hills on the east side of the road. These are composed of the Jatibungkus limestone. The limestone units dip south and have an abrupt and apparently conformable with the underlying turbidites. This has led workers in the past to suggest the limestones are displaced (slumped I olistostrome) units.The fact that the limestones have such a very limited lateral extent and are both under- and overlain by similardeep marine clastics sthis idea. However there must be a stratigraphiccomponent to this lithologicalboundmy as the underlying mudstones are Middle Eocene while the limestone are Late Eocene. Also there is a significant sediment.contrast associatedwith the base, with the lowest part of the Jatibungkus Limestone being coarsely conglomeratic (rounded limestone pebbles boulders) and even locally a coarsesandy limestonehorizon with abundant quartz material; but no feldspars or

-

Page: 20

.

-

rr'

>:,

,

_. Y

$3*' "5 -.

5

; - .:?.

-

-&%,

IPA Field trip to Central Java

Oct - Nov. 2998

volcanics. This contrasts with the pre-limestone clastics which exclusivelyfeldspathic,with no quartz grains in the sands. The majority of the JatibungkusLimestone is a larger foraminiferal and coralline algal calcarenite.The coralline algae often forms large rodoliths, sometimesoriginating on other bioclastic fhgments such as coral fhven@.

Earlier work on the Jatibungkus limestonehas reportedthe age index larger finamhiferalgenus Pellatispira but extensivesampling by me has yet to tum up a single individual. The limestonecan be dated as Late rather than Middle Eocene on other faunas components, chiefly the species of Nummulites present. Above the Jatibungkus limestone is a small areawith more outcropping bluish grey turbidite mudstones and fine sandstonesof Late Eocene to basal Oligoceneage (Oligocenebased on the faunas collected & examined by Paltrinieri, 1975& 76), but exposure is greatly reduced approachingthe slope of the Waturanda Hills.

,

The following notes describe the Oligocene through Miocene succession south and southwest of the Karangsambung section. This stratigraphy compareswith the Baturagung hills sectionwe will see tomormw. No stops are planned in the volcaniclastic section today. The top of the Karangsambung Formation is not exposed. It may be a hiatus and unconformity, or a condensed section. Paltrinieri (1975) studied the next outcropping unit which he called the Clay Breccia formation. He could date samples as old as P21 and deep marine above the non-exposure, contrasting with P 18 below. Considering the difficulty of identifjhg the effectively single P 19-20 zone, and the deep marine nature of the samples above and below the sample gap, it is easiest to interpret a continuous but condensed deep marine clastics section in mid-Oligocene times here.

The Clay Breccia formation of Paltrinieri is the Totogan Formation of Sukendar, and this latter name is used on GRDC maps. Both are probably equivalent to at least the lower part of the First Marl Tuff of Harloff. Paltrinieri dated the section as old as Late Oligocene from planktonic faunas. Harloff had already noted planktonic globigerinidsand radiolaria in the lower part, as well asthe larger foraminiferaMiogyprina and Flosculinella globulosa indicating an Early Miocene age (note that in 1933the genus Miogypsina would probably have includedthe Oligo-Miocene Miogypinoides form). The Totogan Formation is therefore a transitional unit grading h m tuffaceous deep marine mudstones in the lower part, into mudstones with breccia fmgments, tu&eous mark and tuff8ceous calcarenites,locally with possibly derived shallow marine limestone lenses. We know from radiometric dating (Bellon et al1989) that the period from the end of the Karangsambung Fm. and the time represented by the depositionof the Totogan Fm. saw the begining of intrusive igneous activity in the South Seraju Mts. region. Intrusive dykes in the area are intra Late Eocene, Early Oligocene and latest Oligocene. [39.9 *1.99,37.&1.88 & 26.5k1.33 MYBP respectively]. These radiometric ages and the ages of the subsequent extrusive igneous sediments record the activity of the "Old Andesite" volcanism in this area, which was similar to what was happening in the Jiwo Hills. The Totogan formation interfingers with the base ofthe Waturanda Formation, or "First Breccia Horizon" of Harloff. This formation is composed of sandstones,conglomerates,and breccias of andesite. Small limestone

Page: 2 1

a

.<

Oct - Nov. 1998

P A Field trip to Central Java

N8 &

older Stratigraphiccolumnfor Neogene section southwest of Karangsambung,

through the formations overlying the Eocene - Oligocene sediments. Note that the Waturanda Formation is much thicker than shown here, but is notfossili/erous. The key p o i n t h m this data is the clear intra Middle Miocene unconformity. This is a regional unconformity that correlates with the radiometric dates for the start of renewed volcanism in the Java area, along the axis of the modern arc. Data re-interpreted3om Kadar 1986.

-

Page: 22

Oct - Nov. 1998

IPA Field trip to Central Java

lenses contain corals, Lithothamnium red algae and rare Lepidocyclina indicating a shallow marine setting (unless derived fiom nearby shallow enviroments). The Waturanda Formation forms the hills that surround the Karangsambung area and it is estimated to be up to a thousand meters thick. The Totogan and Waturanda formations are seen on the GRDC geology map to cover much of the northeastern Karangsambung area in places nearly overlying Middle Eocene sediments. This change in depositional geometry is probably due to the intra Oligocene uplift associated with intrusion and the Old Andesite arc development. It is possible that the condensed Totogan formationdoes laterally develop into an angular unconfdty. Conformably overlyingthe Waturanda Formation is the "Second Marl Tuff" or Penomgan Formation. This unit is dominated by white tufbceousmudstonesand mads with tufbxous sandstonesand bedding showing turbidite features. This transgression to white mads may correlate with the transition from Kebo-Butak and Nglanggran breccias to off white, planktonic foram bearing Sambipitu Oyo mark in the area south of the Jiwo Hills. Both the Penosogan Beds and Sambipitu Fm. have turbidite characteristics,and contain in their mid sections the base Middle Miocene Orbulina datum (Kadar 1986). The top of the Penosogan Fm., studied by Kadar about 15 kilometers west of the Karangsambung section, contains mid Middle Miocene markers (zone N 11-N 12)before the transition to the Sempor Formation.

-

The Sempor Formation is a sandy lateral equivalent to the "Second Breccia Horizon" of Harloff (1933), the base of which was recognised long ago as an important unconformity. Kadar reports boulders of reworked rock. Harloff recognised the onset of strong volcanic activity resulting in deposition of basaltic lahar breccias, lava flows and dikes of olivine basalt. The SemporFormation is dated by Kadar as old as zone N15, in the early part of the Late Miocene, therefore quantifjlingthe extent of the basal unconformity as being a period h m mid Middle Miocene to early Late Miocene. The data of Kadar (1986, figure 2 1) is important as it clearly shows Globorotaliafohsi lobata assemblages at the top of the Penosogan being overlain within 25 meters by N. acostaenk assemblages in the new, sandy Sempor h i e s . This indicatesthat the Penosogan to Sempor transition is very close to the regional mid Middle Miocene unconformity (the 12 m p event and onset of the modem arc development).

In summary, the Karangsambungsection shows mid Eocene transgression over a basement complex, with a very fast rate of subsidence that only allows relatively thin basal conglomerates1sands and shallow marine limestones to develop. These are followed by deep marine mudstones and thick, distal, tuffaceous turbidites. The basal sands are composed of locally sourced quartz and lithic material. The later deep marine turbidites contain very different sands, being wholly tuffaceow / feldspathic in composition. These tuiTaceoussands are deposited in sedimentsas old as [wellwithin] the Middle Eocene but it is only in the latest Eocene and basal Oligocenethat radiometric evidenceof in-situvolcanics are found in the area (inhusivedikes,just to the north in Lukulo). Other radiometric ages of intrusives in the Lukulo area and the transition to proximal breccias (upper Totogan and Waturanda formations) indicate that the Eocene Karangsambung formation was deposited in a rapidly subsiding back-arc rift, but in latest Eocene through Oligocene times the rate of sedimentationslowed considerably, the area underwent uplift, and the immediate (Lukulo) area became the site of active volcanism.

Page: 23

P A Field trip to Central Java

Oct - Nov. 1998

The Jatibunglcus limestone that occurs within the Eocene turbidites may be a large allochthonous block or olistolith, but the base almost certainly represents a stratigraphic event as it includes the Middle to Late Eocene boundary, and a b a d conglomerateto the limestone contains locally sourced quartz and lithic material. The Early Miocene Waturanda breccias are probably equivalent to the Nglanggran agglomerateswe shall see in the Baturagung Escarpment, and the overlyingdeep marine t u f i of the Penosogan Fm equivalen to the Sembiptu - Oyo formation. This transition from lahars, breccias and other eruptive products to transgressive marine tuffaceow mads probably marks the end of the Old Andesite arc activity in the area, dated as within the middle of the Early Miocene (c.20 m ~ ) . After the mid Middle Miocene unconforrnitythe new volcanic arc is slightly north of thisarea as seen in the :=$$+distribution of radiometric dates younger than 12 m~ (see figure on page 32). Volcanic breccias occur in the -+?qz , area north and northeast of Karangsambung (Second Breccia Tuffs etc). d

-

-+&=

h them k h u e n e

8t

cY~un~m~~)~anj?

The Karangsambung section is one of severalareas in Southeast Asia that have been dated as Paleocene, although much of the literature availablein the last 20 years only details h e planktonic foraminifd data refirting this, and dating it as Eocene. The exact details ofwhy some early workers thought the section covered an older age are not yet known to me, but one possible reason is to draw an analogy with other sections in SoutheastAsia such as the sedimentson the Mangalihat Peninsula, east Kalimantanand the Pinugay limestonenear Manila in the Phhppines. Early workers at the former, Kalimantan, location noted a foraminifera in the basal Taballar Fm. they described as Nummulites nuttalli, a distinctive and widespread form that was later split out into a separate genus Ranikothalia, after Ranikot in Pakistan. This genus is identified by an extremely inflated marginal cord. At Ranikot and in the Middle East it is an indicator of Paleocene sediments. In Kalimantan the "Nummulites nuttalli occurs in association with other non-Ranikothalia nummulitids such as N. bagelensis, as well as Biplanispira, indicating a Late Eocene age. In the latter, Philippine, location a more recent paper (1970) illustrated a species it named Ranibthalia bermudezi and used to date the host limestone as Paleocene, even though it occurred with traces of mid to Late Eocene forms such as Astemcyclina and close to better dated Middle to Late Eocene carbonates. "

A form virtually identical to Nummulites nuttalli can be seen in tremendous abundance in the Jatibungkus limestone, identifiable with only a hand lens. Thanksto planktonic biostratigraphy we know the sediments below and above this are Middle and Late Eocene respectively. What appears to be happening is that the Nummulites group first appeared in the Paleocene and by Late Paleocene times some distinct, specialised forms such as Ranikothalia had evolved. Through the Early Eocene these were replaced by other Nummulites forms with increasingly large sizes, increased differentiation between the haploid generation (half a set of chromosomes, as in gametes) and diploid generations (twinned chromosomesas in sexuallyreproducing animals), and increasingly complex chamber shapesand cunsequently complex sutures between chambers. By Middle Eocene times Nummulites achieved the giant sizes distinctive of this period, as we shall see at the base of the Karangsambung section and at Jiwo. At about the end of the Middle Eocene there was a mass extinction of larger forms and the giant, complex Nummulites forms disappeared. The ancestral Palaeonummulitestypes survived, and in the Late Eocene a new radiation of Nummulites occurred with one of the first forms having a repeated appearance of the inflated marginal cord that defines

Page: 24

.., J

b

Oct - Nov. 1998

IPA Field trip to Central Java

Ranikothalia. This is an example of iterative or repeated evolution, a phenomenon fi-equently seen in the fo-

Stratigraphic distribution of the Nummulitidae and probable evolutionary relationships

(.:septa

meandrine to sub-reticulate

(

-

--

I

'Involute forms somdmcs r c f d to (a , supcr-genus concept o f )Nummuliter Much like P a l ~ n u m n n r l i t a ccx walls tend to be +,thinner, usually

anstgt3"' *u,rm

t

p a

-

.-

_-

--

P A Field trip to Central lava

Oct - Nov. 2998

The division, naming and spellingof the Tertiary is a little conbing but this reflectsan inhistory,so the next few pages are devoted to an explanation ofthe Tertiary, rather than just presenting a dull stratigraphic column. The dull stratigraphiccolumn is on the next page. The term Tertiary survives from an eighteenth century attempt at stratigraphic subdivision (Primary, Secondary, Tertiary, and Quaternary),of which only the last two survived.In the nineteenth century the term Tertiary was applicable to the sediments that overlay the [Cretaceous] chalks in the Paris Basin and southern England. In 184 1 Phillips defined the Cainozoic(or Cenozoic) as sedimentsdeposited after the end-Cretaceous mass extinction (Cainomic is from the Greek; kaiios = new and zoon = life). Just prior to the definition of the Cainozoic, Charles Lyell (1 833) divided the Tertiary of Europe into four parts, namely Eocene, Miocene, Older Pliocene and Newer Pliocene (revised in 1939 into Pliocene and Pleistocene respectively). As with the Cainozoic these were biostratigraphy based units, as reflected in the Greek roots of the names, viz.: Eocene = "dawn of the new", Miocene = "less new", Pliocene = "more new", Pleistocene = "most new". Lyell defined his Eocene as strata with less than 5% of mollusc fossils still living, the Miocene as strata with about 17% extant forms, and the Pliocene as strata with 35-50% extant forms. The Paleocene ("ancient dawn of the new" stage, see below) and Oligocene ("lesser new" stage) were added in 1874 and 1854 respectively as workers discovered significant sections of sedimentsat the base of the Eocene and between Lyells' Eocene and Miocene. By the end of the 19th Century European stratigraphy, at the Seriesor Epoch level, was virtually complete. Work was then concentrated on correlatingthese biostratigraphic epochs with the lithostratigraphic stages defined on various stratotypesacross Europe. In 1902 E. Haug intraducedthe term Numrnulitique (or Nummulitic) for the Palaeogene (Palaeogeneitself introduced in 1853, along with Neogene), This termNummulitique survived for some fitly years and is directly comparable to the more diversely fossiliferousIndonesian Tatiary a through d. Many of the beds we shall see on the course of this field trip are distinctly "Nummulitic".

-

Note that the British convention is to spell "Palaeogene" with the "a" and "Paleocene" without. This is because the former is "Ancient Race" (and consequently ''New Race" for the Neogene), but the latter is "ancient dawn of the new" - or p a l m + eos + kainos. Thiscombination drops both the "0" and the "ae" of "pal-ae-o"on combination, as each would be followed by another vowel, leaving the word as "Paleocene". The Greek word for ancient is ~ahartoawhich becomes "Palaeos" in English; -the Greek "aa" became 'a?' and then "ae". This diphthong was used for a vowel with a sound between "a" and "e", as in Caesar or aegis. In some cases, such as the beginning of words, this uncommon inflection is not kept, so words such as "ether" or "edifice", which have mots in the Greek "at" and Latin "a?, were modified to have a single "e". The American have taken this a step further to replace all diphthongs "ae" with "em.(Note that if the American preference is adopted, fossil names such as Palaeonumrnulites do not get changed, as these are formally defined and fixed upon publication; the only changes allowed are when s u f i e s are altered to matching the gender of the speciesto the gender ofwhichevergenus is prefered).

Correlation curve of seawater strontium isotope ratios

anhvard

~ t t e Stages r 'ollen & Spores

P A Field h i p to Central Java

Oct - NOV.1998

The Dutch were apparently not happy with these exotic, sometimes overlapping names and the ensuing grammatical picky points. They must have wanted a simpler, logical schemeto divide the Tertiary. In Europe they had little choice as the Brits had got there h t (ed by the French, another nation renowned for tolerance to eccentricity)and had set up the accepted system of stratigraphicbenchmarks. In Indonesia however they could start with a blank stratigraphiage, and so the Letter Stages were born, Tertiary A to H, or whatever, seems a fine idea, except, as in the case of the Oligocene in Europe, you find a &on that had previously been omitted. Compound thiswith the inevitable subdivisionsand you end up with what we now have in Indonesia; - a system no more logicalthat the European epochs, except that it has the slight advantagethe "Ta" is immediately recognisable as ol&r than "Te".

Subdivisions of the Tertiary in Southeast Asia In Indonesia the Tertiary is proportionally a far greater part of the sedimentaryrecord, so while British and French palaeontologists made inroads into the understanding of trilobites, ammonites and other very ancient creatures, the Dutch palaeontologistsconcentratedon m o b s and vertebrates (see the second half of this guide for some notes on the latter). Professor K. Martin (1 852- 1944)was particularIy outstanding in the field of molluscs. His work was able to follow that of Lye11 in Europe in utilising the ratios of extant to extinct faunas to indicate relative ages. Two major contributionsto Tertiary stratigraphycame fiom the work of Martin.Firstly he showed that the mollusc fiunasof the East Indies developed separately h m those of Europe and secondly, as a result of this, the stage names of Europe (lithosttatigraphictype sections such as "Aquitanian" or "Messinian") could not be correlated with confidence to the Indonesian region.

Martin studied a number of Javanese marine sections in detail and gave ratios of extant to extinct molluscs and predicted the stratigraphic order of these strata. Mapping work, and studies at about the same time on the other regionally significant fossil p u p , the larger foraminifera, edthe mollusc stratigraphy,but there was still little idea on how it Minto the European stratigraphiccolumn. [Foraminifera,by theway, were at this time included in the class Cephalopoda, part of the phylum Molluscs, but they were not used by Martin in his scheme]. By the 1920scompilations of work on larger foraminiferabegan to make this fossil group the premier tool for biostratigraphy in Indonesia. They have a natural advantage in that they are more abundant than molluscs, and secondly a scheme was developed that utilised assemblage zones rather than percentages of extant forms. To date sections with molluscs required extensive knowledge of both living and fossil species. The larger foraminifera assemblage zones could be identified by the presence of a few key tam. In 1927 Van der Vlerk and Umbgrove published the Letter Classification of the Indonesian Tertiary, based on larger foraminifera.This scheme subdivided the Tertiary into seven parts. Six parts were labeled "a" through "f' (e.g. Tertiary "a", or "Ta" for short), and a seventh section of non-orbitoidal Tertiary between Tf and the Quaternary was noted. This scheme was immediately successfbl and was rapidly adopted as a workable biostratigraphic scheme for the Southeast Asian region, even if the correlation with the European epochs and stages was not adequately known.

In 1931 Leupold and Van &r Vlerk published a detailed revision of the Letter Classification. This new version had sixteen pre- Quaternary divisions or subdivisionsto replace the previous seven. The primary divi-

Page: 28

Oct - Nov. 1998

P A Field trip to Central lava

sions were a through to h and were unfortunately called "stages". Subdivisions, such as Tal ,Ta2 etc., were called zones. Their new Tg and Th units were not defined on larger foraminifera and are not actually detailed in the text of the paper (their accompanying figure noted the percentages of extant molluscs for the Tg, Thl and Th2 divisions). This 193 1 paper marked the peak of the Letter Classification.Over the next fifteen years the scheme was periodically modified, with a tendency to lump the 1931"zones" rather than split them. By 1950Van der Vlerk (1949, with Rutten in Van Bemmelen, 1949;and 1950)had reduced the number of pre-Quaternary divisionsto eight. Tertiary a and b as well as Te 1to Te4 had been reduced to single units after criticism from Tan (1939) and others, as were Ti2 and Tf3 and Tg and Th.

-- Y ‘ --y'. F>

The Letter Classificationwas largely unchanged &r this until 1970when GeofEey Adams undertook a detailed review of the scheme and the ranges of the fossils covered by it, in light of new data from planktonic ssatigraphy.This 1970 paper can be considered as lxg.umngthemodem era of larger foraminif' biosmtgmphy.

A summary of stratigraphic Series and Letter Stages, and some notes on factors that are possibly linked to larger foraminifend diversity

LETTER STAGES SERIES (or Epochs) n7

Sr / '"sr

0 7090

Blue line: Strontium isotope, curve retlecting weathering p of Himalayas (Sr87 is 3 derived from sialic rocks)

70e5

,,,,,

c

0 7076

8

3 Black line: Sea level curves from Vail et al. * 977), Haq a1 (1 987). Scale 8 at right is modern day s l . at (1

et

0

Tectonolstratlgraphlc sequences in Indonesian area

.t. Cmt.

- m. Eo

'"ega-quen-

m. Eoc. m. 01ig.

L t 0110 to Lt. Mioc. moqasequence wldespnmd carbonatm

Oct - Nov. 2998

P A Field trip to Central Java

The concept behind this day's geologicalexcursion is to drive firstthrough the products of an Old Andesite volcano, heading up-section until post-volcanic sediments(eventually limestones) are encountered. Then to turn around and head north to the oldest rocks in the area, where a basement and mid Eocene complex is intruded and metamorphosed by crystallinevolcanic rocks, in what may have been the center of the Old Andesite volcanic cone. This central area is eroded and capped by limestones that correlates to the youngest of the limestones seen to the south.

-

The Eocene sediments at Jiwo are very similar to the conglomerate sandstone - limestone - clastics (?turbidite) sequence seen at Karangsambung and, to a lesser degree, at Nanggulan.

JIWO HILLS or Bavat area

BATURACUNG ESCARPMENT

Simplified section across the western end of the Southern Mountains, Note the non-conformable change from volcaniclastic dominated beds to carbonates.

One of the reasons that we are able see Eocene or basement sections in Central Java is that some locations were uplifted by the formation of the "Old-Andesite" volcanic arc. We can therefore examine some important regional aspects of early to mid Miocene Javanese stratigraphy while we are in the area. We know that the Old Andesite volcanic arc was most active in Oligocene through mid Early Miocene times, with compilationsof radiometric ages of volcanic rocks fiom Java forming a cluster h m almost 40 MYBP up until about 20 MYBP. From 20 to about 12 MYBP there is a gap in volcanic record, both in radiometric ages,

Page: 30

P A Field trip to Central Java

Oct - Nov. 1998

in Javanese sedimentarydeposits, and in the occurrenceof h e tu& offihore in the IndianOcean DSDP material. At 12 m p the modem volcanic arc was initiated, although not along the same axis as the Old-Andesite arc.

Immediately west of Nanggulan a complex of three, eroded, Old Andesite volcanoes form the present West Progo Mountains (dated as 29.6,25.4 & 22.6 WBP). Late Oligocene through Early Miocene volcanic breccias and basalts unconformably overly the Eocene outcrops in the Nanggulan section. These Oligo-Miocene volcanics are deeply eroded and unconfonnably covered by a shallow marine limestone as old as late Early Miocene age (Djonggranan Lst., Letter Stage Tfl). This limestone is depositedright over the intrusive core (the "hearth" of van Bemmelen) suggesting rapid erosion followed by at least partial transgression all within the mid Early Miocene. Away fiom the volcanoes of the Progo Mts. the Djonggrangan limestone grades into the Sentolo formation which is a mixed planktonic chaky marl, with sandstonesthat includereworked volcanic and even Eocene rnateral (clasts of PelIatispira limestone). At Karangsarnbung the deep marine distal tubidites of Late Eocene age are apparently uplifted and succeeded by widespread Oligo-Miocene volcanic sediments. The thick Waturanda Fm. is a coarse, probably proximal, volcanic breccia which is overlain by later Early Miocene clastics (Penosogan Fm.), probably the erosional products of the recently extinct volcanic cone.

ills and Baturagung Escarpment there is Early Oligocene onset of volcanism, uplift,and In the Jiwo H growth of a new volcanic cone. Volcaniclastic sedimentsapparently culminate with the thickNglanggran breccia, similar in character and maybe cornlatable with the breccias around Nanggulan and the Waturanda Fm. The later Early Miocene Sambipitu beds of the Baturagung Escarpment are probably equivalentto the Penosogan Fm. in south Karangsarnbung and the Sentolo beds south ofNanggulan, all ofwhich are later Early Miocene, deep marine sedimentsfrom erosion of a recently extinct volcanic cone. The Sambipitu and Sentolo Beds both grade up into transgressive carbonates. These transgressivecarbonatesare seen unconformably covering parts of the Jiwo Hills Eocene and volcanic outcrops. An interesting point of sedhnentology is that the clastics that were shed offthe eroding Old-Andesitearc contain significant amounts of quartz. This mineral is completely absent from the underlying Oligo-Miocene volcaniclasticsand is also extremely rare in the new phase of volcaniclasticdeposits that are "switched-on" at about 12 MYBP. Within the later Early and base Middle Miocene sediments quartz o h n composes up to 30% of the sand grains in a sample. This quartz must have been sourced from the eroding arc to the south as the nearest continental source was Kalimantan to the north. This northern source developed its own sedimentary system (Tuban-Ngrayong)that only reached the northernmost edge of the present onshore Java area.

In the present day Southern Mountains there was apparently some uplift at the time the Old Andesite volcanism stopped,as earlier submarine volcaniclastic deposits are eroded prior to mid Miocene transgression by carbonates. This uplift is not well dated or understood, but it is probably not the regional block faulting movement known in the area, as this faulting offsets the mid Miocene carbonates and older rocks, and is most likely post mid Miocene. Within the early Middle Miocene most of the SouthernMountains area became the site of reefal or platform limestone deposition. This was also about the time that the Jiwo Hills eroded core was covered by carbonate

Page:3 1

, J

--r

F--

O

r

through Middle Miocene carbonates in low lying areas

1

J _-

7

,/

Early and later Miuene volcanics

i

loo 200

Generalised lithology distribution in the Southern Mountains, plus locations of dated igneous rocks, and sites visited on this trip.

!undifferentiated

,

Bawean Island

I

Mostly Oligocene & basal Early Miocene volcanics (Old Andesites) forming high topography Red = Intrusive rocks

JAVA SEA

Karimunjawa Islands

(post 12 MYBP) 'SC ~ R A N G E- souiherlv Old Andesite volcanism-2

RED symbols - more northerIy modern an:

P A Field trip to Central Java

Oct

- Nm.1998

sedimentation.Further west, around Karangsarnbung, deeper marine clastic sedimentationcontinued until an unconfonnity in mid Middle Miocenetimes, which is the event associatedwith the start of a new volcanic arc and sediment source.

AGE

FORMATION

Middle Miocene

nin. 250

Lab Oligocene

Lab Eocene Mlddie Eocene

Stratigraphy of the Jiwo Hills and Southern Mountains (modified from Budianta et al., 1986)

Page: 33

P A Field hip to Central lava

Oct - Nov. 2998

Driving south fiom Yogyakarta a range of hills is visible. These are mostly composed of a thick unit of resistant volcanic breccia, the Nglanggran Formation. Before reaching these hills we will drive over the Kebo Beds and the Butak Beds, which outcrop near the afternoon Jiwo Hills sites. As the road starts to climb there is a place we can stop to examine the third volcaniclastic unit, the Semilir Beds.Whilst the Kebo and Butak beds are Late Oligocene marine conglomerates,sandstones and claystones pas into more regularly bedded, turbiditic, sandstonesand claystones (at least 1,000 to 2,000 meters thick) the Semilir Beds are distinctly off-white to pale grey, light-weight, tuffaceous or pumaceous claystones, . gr-=* locally with basal Miocene marine microfossils. The Semilir beds range fiom 100 to about 1,000meters in <$s thickness. --. $s- -'. . The most resistive unit at the top of the hills is the Nglanggran Beds, which are poorly stratifiedagglom- > **.- ,:~2 erates and breccias. A.

v

+ :

'=

Although most authors consider the transition from Nglanggran to Sembipitu Beds to be gmhtional and not particularly remarkable, I consider this the prime candiate for a change from active volcanic deposition to erosion ofthe volcanic islands. Datingthe base of the Sambipitu is difficult as many samplesare non-fossilifmus but Kadar notes sporadic diverse planktonic assemblages of very broadly 'mid' Early Miocene (intra Early Miocene zonations are weak due to the scarcity of index species). This facies change is the closest to the mid Early Miocene unconformity seen in the shallow section. The Sembipitu Beds are turbidic mudstones and sandstones locally containing some quartz, a feature not known from older sediments. However there is no known systematic data on sand mineralogy through this section.

In the section we shall visit t h m are mudstones, sandstones,glauconitic sandstones, reports of channel like features, and slumps. One bed, which may be visible if water is low enough, is a sandstone rich in planktonic foraminiferaand deep marine in origin but with ripple cross-beds, indicating traction currents. Up-section the beds become more calcareous and less sandy until the Oyo Beds of some authors are reached. The base of this sandy facies is close to base of the Middle Miocene as indicated by Orbulina. Above this are mads and globigerina sands, often with abundant bioturbation. Still within the early part of the Middle Miocene the section es up into limestones and marls with coralline material and larger foraminiferathe dominant component of calcarenite beds.This is the Wonosari Formation. At this point the Oyo River changes course and starts to expose older sedimentsagain.

North of the Baturagung escarpment are a series of low hills where microdiorite igneous intrusives, and ?pillow lava extrusives of Late Ohgocene age penetrate and overly basement and Eocene strata. In many places the Eocene sediments show the effects of metamorphism. Above this is an unconformity and Middle Miocene limestone has covered the eroded stump of the old volcano. The absence of a volcanic cone of Oligocene or Early Miocene age suggestserosion prior to Middle Miocene transgression. The most likely time for such erosion is in mid Early Miocene times, from about 20 MYBP when the Old Andesite arc became less, or

Page: 34

IPA Field trip to Central lava

Oct - Nov. 1998

Page: 35

IPA Field trip to Central Java

Oct - Nov. 1998

wholly in-active, but before base Middle Miocene times when clean, clastic h e carbonates began to transgress the Baturagmg - Bayat areas. Radiometric data on intrusive volcanic rock &omthe Jiwo Hills gives ages of 24.25 1.21 MWP and 33.15 k1.66 ~ y s (Bellon p et al, 1989);base Miocene and mid Oligocene respectively.

*

Our first stop in this area is the location mund Gunung Pendul and the village of the same name. At the tunoff from the main road the soil in the path heading north can be seen to contain abundant loose Discocyclinu

specimens. This larger foraminifera1genus is a general Eocene indicator. Also near this road junction there are boulders ofNurnrnulitic limestone very similarto those seen at Karangsambung and characteristicof the Middle Eocene. The boulders have been carried out fiom the location we will walk to. Several publicationsand reports mention Pellatispira in Jiwo, some specificallyin the Pendul area. In spite of much searchmg I have yet to find any of these Late Eocene diagnostic fossils. I have found some excellent, robust Assilina specimens, which are the Middle Eocene ancestor to Pellatispira, and am currently of the opinion that, as with the Jatibungkus limestone, the identificationsof Pellatispira might be incorrect. As we approach the Pendul location some lavas with concentric onion-skin type weathering can be seen.

These are better exposed a few kilomekm to the west on a stop not planned for this trip.However even the small examples seen near Pendul are similar to pillow lavas, although as yet no mudstones layers have been seen between 'bpillows"and a broader pattern of interlockingpillows has not been recognised in such small outcrops. Whatever the settingthese extrusive lavas suggest that the Eocene sediments had only a thin sedimentary cover before igneous activity intruded the strata. The existenceelsewhere in the Jiwo Hills of moderately coarse crystalline igneous material indicates that a substantialcover quickly grew over the site so that slow cooling of molten rock was possible later on. This cover, presumably a volcanic cone, has since been lost through erosion.

Figure 32. Above: Thin section through a sample of limestone from the Pendul location of the Jiwo Hills or Bayat area. Most limestone contain giant Nummulites specimens but the above view illwtrates the smaller genus Assilina, a Middle Eocene marker, in a quartz sandy carbonate facies.

Page: 36

-'*

-

IPA Field trip to Central Java

Oct - Nov. 1998

The main Pendul location is composed of brittle mudstones described by previous workers as "alternations of bluish-grey marls, calcarenite intercalations of tuff8ceous sandstones". These rocks, usually assigned to the Gamping Formation, are actually mudstones, only locally c a l w u s , and usually with a distinctivebluish-grey colour. They contain regular thin beds (a few centimeters, rarely decimeter size) of non-calcareous, white sands or silts of feldspar / tuffaceous material. These are very similar to the splintery, brown to blue-grey beds seen at the top of the Nanggulan section and in the Karangsambung section.An unusual fof at least one mudstone beds is the occurrence of coarse "dropstone" like clasts within them. These clasts have been observed as up to 35 crns in s k , and are either basement metasedimentsor Nummulitic limestone clasts hlly embedded in the bluish grey mudstone. A mystery that has been slightly deflated by the gradual removal of the boulders ofNummuliticlimestone is that of their origin. As with the Eocene boulders at Sangiran, about ten years ago there were several sizable boulders to 2 meters. Now boulders to 50 crns are frequent, but not particularly common. A search of the section at Pendul has uncovered no strata of limestone, and one presumes that such a hard ltihology would stand out. This leaves open the possibility that the boulders might all be clasts within the mudstone. The giant Nummulites is N. javanus and is a typical Middle Eocene form. A few planktonic foraminifera have been seen in thin sections of the silty mudstone and while some samples are barren of nannofossilsa few rare forms have indicated a broad NP16-NP22 zonal assignment (Mid Eocene through Early Oligocene). The with the Discocyclina bearing beds is not seen, and could be faulted.

The second stop in this area is at the western end of the Jiwo Hills. In climbing Gunung Cakarnn we by someweathered phyllitic basement and then walk around the hill to see the polymict conglomerate and then some large boulders of black, metamorphosed sandstone and limestone. The sandstone of the Wungkul Formation is used by the locals as a knife sharpening stone. The limestone can be seen to have the same giant Nummulites we saw at Kamngsambung,just above similar polymict conglomerates and sandstones. Above the limestone outcrop is poor, however note that the weathered rock and soil, above where the boulders of limestone are, contains fragments of grey - brown, splintery mudstone which may be from the Garnping Beds. The core of the Gunung Cakaran is a diorite intusion, and at this location I have found crystals of quartz to several centimeters, as vein fill.

Page: 37

Oct - Nov. 1998

IPA Field trip to Central Java

The Sangitan Dome The Sangiran Dome is a unique feature in Javanese geology. It is set on the southern margin of an area of intense folding and faulting (the Kendeng Zone) where structures are dominantly east-west in orientation and elongation, yet it is a dome with, if anything, slightly northern elongation. At the center ofthe Dome are several .-.. - ~3~ small saline seeps and the remnantsof an extinct mud volcano. A-+,

* , Q I %

L

The Sangiran location was the second site where Homo erectus fossils were found in eastern Java, afkr the river section at Trinil, 50 kilometers to the east. It remains one of the most active sites for Plio-Pleistocene research and is also an excellent location to study fluvial and volcanic sedimentation. A wealth of detailed descriptions of the stratigraphy, palmntology, mdiometric dates and magnetostratigraphy has been compiled by a t Indonesian Japanese research team and published in a book by Watanabe and Kadar (1 985). A detailed guide book is available at the S a n g h Museum with a wide mnge of maps, figuresand faunallists that we will not attempt to reproduce here.

-

Most p d i n g though is the presence of blocks of metamorphic basement, Late Eocene limestone (blocks that used to be 2.7 meter high -now reduced to rubble by collectors), polyrnict conglomerate identical to the varieties seen at Nanggulan, Karangsambung and Jiwo 1Bayat, as well as Miocene limestone, Miocene and Pliocene mads. These have no local source. All these blocks are thought to have come h m the subsurfaceas a result of the mud volcano.

I N.W. Pagerejo

Sangiran Dome ..-..-

*-

*--.-

.--. ..

_.-.--_

-..?

S.E.

.- .-.-

-.

Lower lahar

I

Upper Kalibeng Fm.

-

NW SE profile through the center of the Sangiran Dome (von Koenigswald, 1940)

Page: 38

Oct - NOV.2998

P A Field trip to Central Java

The stratigraphic section at Sangiran

Fluvial chaMol conglomcrata. sand clay

Pucanaan Fm

-

Bluish gray manna cl sot lhm tuffbeds

I -

planktonic

--

iI I

MUSEUM location - is on hill composed 1 ; of Lower Lahar - - o 0.5 1.0 1.5 2.0 2.5' 3.0 rnilbns d yeor More present

1- -!

C

The Sangiran Museum contains hominid and mammal fossils.Just behind the Museum we will look at a site where the Pliocene marine "Kalibeng formation" (a rather loose term for sediments of latest Miocene through Pliocene ages in the Kendeng Zone) is overlain by marginal and non-marine Pleistocene beds. The is a scour surfacejust above some mollusc rich mudstones.

The Sangiran Dome is very young, - less than half a million years old, as is shown by the age dating of the Kabuh Fm. which was accumulated in a low area at about 0.7 m p . Its origin is not well understood and various causes have been proposed, such as a compressive feature from collapse of the old L a w volcanic cone, a

Page: 39

Kilometers \ ,

gA MURIAH VO(CAN0I \

plus location of mud volcanoes and other feature mentioned in the text Note that the Kendeng zone shows most uplift in the west

Location of the Kendeng Zone and the Sangiran Dome

Oct - Nov. 2998

P A Field trip to Central Java

diapiric shale flow, an incipient volcano, or due to a basement related fault. The regional unusual presence of the extinct mud volcano at the center suggestsbuoyant shale diiirism may be linked to the formation of this fature. Following the suggestion of Itihara in Wanatabe & Kadar (1985) the most likely explanation ofthe Sangiran Dome is that a basement involved fault, perhaps with a north south orientation, occurred first and produced a fold perpendicular to the ~gional trend. This hult acted as a focus for overpressured Early Miocene muds which could then rise and, along with over-thrusted, fractured rock from the fault plane, escape to the surface.

-

The s h t g t q h y ofthe SangiranDome showsthat marine conditions persisted in the area until the end of the Pliocene. There was then a rapid regression through brackish conditions to lacustrine muds which are the Black Clays outcropping over much of the dome. It is in the upper part of the Black Clays and the base of the overlying fluvio-deltaic clasticsthat the hominind fossils are found. This stratigraphy showsthat the area ofthe Sangiran Dome was a relative low featureup to, and for a little while after the level of the horninid fossils - fkom before 2 MYBP until at least 0.7 MYBP. This contrastswith datajust to the north, in the main outcrops of the Kendeng Zone, where long standing very deep marine conditionsaxe interrupted at mid Pliocene times (3.5 m p ) by uplift and the formation of Klitik reefal or epi-reefal carbonates on new highs. Regionally across eastern Java the Plio-Pleistoceneboundary is

Om-

500

-

lo00

-

1500

-

trxm

-

2500 m -

Old Andcsites (Early Miocene)

1

Basement Rocks (PWT~~~Y)

Schematic profile of the Sangiran Dome showing origin of the mud volcano (from Wanatabe & Kadar, 1985)

Page: 4 1

P A Field trip to Central Java

Oct - Nov. 1998

typified by the end of marine sedimentation,which could be due to eustacy, tectonic uplift, or a rapid increase in fluvial sedimentationdrowning out the marine influence.

-

A palaeogeogtxiphicreconstructionthereforewould have the Kendeng Zone uplifted to form an east west ridge of land or string of islands at 3.5 MYBP with some erosion and some areas then transgressed and acquiring small re&. South of here, including what is now the Sangiran area, was open sea. Close to the Plio-Pleistocene boundary was a second pulse of fault activity along the Kendeng Zone that was greatest in the west, near Semarang. There, Middle and Early Miocene beds are now to be seen, intensely faulted and with near vertical bedding. Between Semarangand Ngawi the degree of uplift, hulting and erosion diminishes, until virtually complete stratigraphicsections are present. East h m Ngawi the intense hulting style gradually changesto elongate h t3.5 MYBP Kendeng tectonism, and the folding. This folding was already present as a result of the events ofthe f base Pleistocene tectonism had only relatively minor additionaleffects in the east, other than general uplift and 4 some reactivation of folding. ,* FA-

g v . - .

.&

Thin section through a sample of Eocene limestone from Sangiran. The large fossil (about 4mm in diameter) is the Late Eocene index form Biplanispira absurda.

Page: 42

" .

.% ,

IPA Field trip to Central lava

Eugene Dubois and Me dr'scovery of Java Man Eugene Dubois (1858 - 1940), the discoverer of Java Man, appears to have been an ambitious scientist. In his mid-20s, while a physician and lecturer in anatomy at Amsterdam University he developed a taste for the hottest scientific topic of his day; the subject of evolution. With his expertise in anatomy he would naturally have been drawn into the most controversial side of the subject, - the debate over human evolution. It is generally assumed that this background motivated him to enlist in the "Dutch East Indies Army" as a surgeon. This would have allowed him to travel and search for a potentially enormous prize; the missing link between apes and man.

Java Man'

(Homo erectus)

Homo sayiens

He started his search in 1887, in Sumatra. This was already more than a decade after Charles Darwin had published a ground-breaking work (Descent of Man, 1871), which specifically dealt with the position of the human animal within his new theories on evolution. At that time little hard data was available as only two types of prehistoric hominids were known; namely Cro-Magnons and Neanderthals, - both of which were far too close to man to be a missing link to apes. Within the racist, Euro-centric climate of the day even these nearly modem remains were viewed with considerable bias by most people. For instance, after the 1856 discovery of the classic Neanderthals in , Professor Mayer of Bonn explained the man-like, but obviously quite different remains as being from a Mongolian Cossack with congenital rickets; -the reference to the Cossacks attempted to explain what the "Mongolian" bones were doing in western Europe. In "Descent of Man" Darwin suggested that pre-humans may have originated in the tropics where the evolutionary loss of body hair would be no disadvantage, and of course it is where apes currently live. For Dubois, being Dutch, the Indonesian region was his pre-eminent choice for tropical field work. AAer three years searching caves in the Padang Highlands of Sumatra and another year in the Kediri area caves of East Java Dubois had only found traces of Homo sapiens, with ages estimated at about 10,000 years old. In mid 1891 he changed his methods and began excavation with the aid of convict laborers. He started digging in a section of the Solo River at Trinil that had previously yielded elephant bones, and which were obviously much older than his existing finds. Increasing official recognition and

Oct

ZPA Field trip to Central ]nun

-

Nou.1998

Lower Kalibeng Fm. Upper Kalibeng Fm. Pucangan Fm.

Notopum Fm. River teraces

Pithecanthropus I locality Dubois monument and museum X

Vertebrate fossil locality

Line of section below is the faint north-south line marked on the right hand side of the map

Geological map and cross-section of the Trinil area (after Duyfjes, 1935, in Von Koenigswald, 1940

of his efforts was shown by access to the convict labor, and also by the ing of legislation banning the trade in fossil bones. Vertebrate fossils were sold as "Dragon Bones"; ingredients in Chinese remedies. As the town of Tuban was an opium smuggling center in the eighteenth and nineteenth centuries it is thought that many fossils were lost to the well established black market.

IPA Field trip to Central ! m a

Oct - Nov.2998

Evolution of A ustralopithecus and Honlo species (from Donaldson, 198

Between September 1891 and the end of 1892 some teeth, skuU caps and a femur (upper leg bone) were found. The femur was similar to the equivalent bone in modem man and indicated an upright posture and waking. The skull parts however were significantly different. The reconstructed hominid was called Pithecanthropus erectus (later Homo erectus) and was the first proposed "missing link between man and the apes. 'The name Pithecanthropus erectus was adapted from a hypothetical missing link suggested nearly thirty years before Dubois' actual find. In the 1860's Emst Haeckel, 's leading

[PA Field trip to Central lava

evolutionist, proposed a missing link that would have walked upright but had a smaller brain and he named it Pithecunthropus alulus, or ape man who could not speak. Recent studies have shown the age of the "Java Man" specimens to be within a range between 0.7 to 1.6 million years old. Dubois returned to the Netherlands in 1884 and published in 1885. Over the next few decades further work at the Trinil site yielded virtually no additional Homo erectus fossils. However in the 1930's work began in the Sangiran Dome some 50 kms to the west where several more H. erectus finds were made (von Koenigswald, 1940). The story does not end with publication. Naturally there was considerable controversy. One must that the publication of "Origin of the Species" in 1859 was not a shock because it proposed evolution; - many scientists had accepted a broad view of the phenomenon by then. What had pushed Charles Darwin and this theories into a major academic battletield was the natural mechanism he proposed and especially the direct implications that this covered human descent from an ape-like ancestor. Before setting out to Indonesia Dubois must have known that success in his search for a missing link would have placed hun under considerable pressure fiom opposing intellectual schools. Many references, text books and even previous versions of IPA Field Guides have described Dubois as becoming embittered by the controversy that fbllowed publication, eventually locking up his precious tbssils until a quarter of a century later he re-displayed them, - but then declared they were actually from a giant gibbon! Even ifone expects controversy it can still be unenjoyable or damaging, and Dubois' biographer, Bert Theunissen, concluded that Dubois was sensitive to much ofthe criticism. On the other hand Theunissen argues that at the time Dubois locked up his specimens (about 1899 or 1900) he was actually winning the debate. It was only in 1923 that he made his specimens available again, just before the next, and much more complete, Homo erectus fossils were found in China; the famous Peking Man. In the quarter century the specimens were locked away it is hard to portray Dubois as being a sulking recluse. He published extensively, and was a Professor at Amsterdam University and Curator of the Teylers Museum in Haarlem. It was during this period he worked on ideas that led to his publication in 1932 that the Trinil fossils were from a giant gibbon. This conclusion was far fi-om the act of "sour-grapes" interpreted in some text books. Dubois was in fact trying to keep his Java fossils at the honored position of "the missing link", but by way of his own theories of evolution. Throughout his teaching career Eugene Dubois argued strongly for mammalian evolution in major jumps, rather than Darwinian gradualism (not to be confused with the recent theories on punctuated equilibrium). Dubois' mechanism was based on the assumption that as neurons of the brains of mammals are formed in the embryo, then an additional phase of cell division in the embryo would double the numbers of neurons in the animal. On Dubois' scale where man occupied the primary position, the great apes were thought to have 114 the relative cranial capacity, carnivores and hoofed mammals 118, a group including rabbits at 1116, and another with mice at

IPA Field trip to Central lava

There are an increasing number of studies ingthe Out of Africamodel. The first was the now famous "Mitochondrial Eve" study on the DNA of cellular mitochondria that is only inherited through the maternal line. Another well publicised result is the more recent, complementary study on the male "Y" chromosome that is ed through the paternal line. The former study suggested a common ancestor about 250,000 years ago and the first results from latter study gave a value of 270,000 years. This appears to date the split of Homo supiens from H. erectus as being a long time after known Java Man fossils. For all H. erectus across the world to be potentially ancestral to H. supiens (the Candelabra model) this genetic divergence would have to be at least 1.8 MYBP. Early suggestionsthat the "Mitochondrial Eve" data pointed directly to a common Afiican origin have been shown to be invalid, but not necessarily incorrect. The computation required to absolutely determine the most efficient ranking ofbranching trees ofvariations is enormous, well beyond the best modem computers. Algorithms have been developed to try and achieve some meaninghl results using an incomplete number ofpossibilities, and it was in the design of this algorithm by the first workers in 1987 that some bias emerged. By 1992 it became apparent the alternative programs could find equally valid non-Afi-ican origin models. However Afica is still widely favored as the location of Homo supiens origin because of the wide genetic diversity in DNA in the African region, suggestingthis has always been the main gene pool, and the genetic home, for Homo supiens. The study of the male Y chromosome mentioned above used a technique known as determining the coalescence time ofalleles (alleles are variations within a gene, such as those seen expressed as variations in eye-co lour in humans, or wrinkled peas and smooth peas in the experiments of Gregor Mendel). Once an ancestral group starts to fbrm a new lineage then they will gradually begin to accumulate new varieties ofalleles. By attempting to back track variations in populations to a common ancestral point (the "coalescent") an idea ofthe time involved can be gained. This requires the study of many alleles in a population as some alleles can give falsely old coalescent times, not least because within the ancestralpopulation there will be some genetic diversity. The clustering of multiple coalescent times is the important data, not a single result. As of 1997, 14 diaerent coalescent times for various genetic loci in both rnitochondrialand nuclear DNA have been calculated. From the male Y chromosome, in addition to the 270,000 yr. result above a 188,000 yr. result has been obtained. These are part of the main cluster of results around 200,000 years ago.

Genetic analysis ofmodem humans continues to supply interesting data. One recent contribution has studied the unusually low level oftotal genetic diversity in mitochondria1 DNA in modem populations. Such low genetic diversity occurs when a population undergoes reduction to a small group but then expands again. 'This forms a genetic bottleneck, and Homo appear to have been through such a restriction quite recently. To quote one researcher, Henry Harpending, "Our ancestors survived un episode where they were us endungered us pygmy chimpanzees or mountain gorillas are today". This data implies that the multi-regional, Candelabra, evolutionary model cannot explain modern human origins. By a study of variation in mitochondria1DNA, and with some knowledge about the mutation rate ol'this material, Henry Harpending and Alan Rogers have suggested this bottleneck occurred quite recently; - about 70,000 years ago. They even suggested that a possible cause of such a near catastrophe (near catastrophe, except for those animals and plants now on the endangered species lists, thanks to lucky old H. supiens) may have been the gigantic eruption of Mount (now Lake) Toba in Sumatra, dated as 73,500 year ago. They have event tentatively identified a signal of this event in the rnitochondrial DNA variation of chimpanzees, although others have contested this.

IPA Field trip to Central b v a

Oct

-

Nov.1998

1/32,with "lowest" shrew-like animals at 1/64. The missing link was therefore predicted to have halfthe cranial capacity ofman. Naturally there is variability in brain size through life as well as across populations, so a measurement is standardised to body size to correct for this. This would have beena popular theory as it would place Homo sapiens at a distinct level above the apes, intermediaries and all "lesser" animals, and thus avoiding the discomtbrt of regarding oneself as a rather unspectacular type of vertebrate. While his Trinil Homo erectus were locked away, much of Dubois' work was on the measurement of the cranial capacity and body size of mammals, especially hominids. Unfbrtunatelyhi precious Homo erectus specimens did not fit the law of proportion he developed. The c r d capacity at around 900cm3is about 2/3Aoofurs while the femur bone suggested nearly identical body size. To keep H. erectus as the crucial missing link he turned around one of the earlier criticisms ofhis tinds; that the Trinil femur bone was fiom a gibbon. The body mass calculations fbr a lanky gibbon were a lot lower than the value for a human build with the same femur length. This apparent downgrading of Java Man to "Java Gibbon" resulted in a brain to body size ratio close to the half way value expected by his own evolutionary theory of cranial doubling, and this would have firmly placed the Trinil fossils as representatives of the true missing link. It is worth noting that this argument came at a time when the search for a ''missing link was at the forefiont of science and other claimants to the title, such as Raymond Dart's small Australopithecus had just been found (1 925; eventually dated as between I and 2 MYHP, and now regarded as a member of the genus that was ancestral to Homo erectus). Unfbrtunately for Dubois the Peking Man finds in the late 1920's showed Homo erectus to be much closer in form to man than gibbon. In addition the basis ofembryological cell doubling to achieve the brain size doubling in his theory of mammalian evolution was never widely accepted so a large part of Dubois' work was ed over by future generations.

Java Man and its ~eIations4ipto modern man While the field of human evolution is one of the most active and contentious fields of palaeontology there does appear to be a lull, or general consensus, at the time ofwriting to oEer an less controversial summary ofthe current knowledge on human evolution, and the position of Java Man. In the last decade much activity has centered on genetic data regarding the diversity ofthe human races and reconstructing age scales and rough lineages fiom statistical comparisons of DNA. In summary it now appears that Java Man was not a direct ancestor of modern humans, but African Homo erectus still probably was. Java man can be regarded as a distant cousin. Until very recently (and it would be wrong to say the argument is over) there were two schools ofthought; one group considered that Homo erectus emigrated fiom &ca and then, in several parts of the world sirnultaneously developed into Homo supiens. This is the multi-regionalist or candelabra model. The opposing view is the Out of Africa or Noah's Ark model which has the Homo erectus emigrants from Africa as evolutionary dead-ends, with H. supiens developing later on, in Afica, and then emigrating out to other continents.

IPA Field trip to Central \ma

RN RN1TNROPOWGI$7b VIEW O f J R M MRN

R Plio-PIeistocene Way of U& Norno

of Sangican Dome, Central Java 0. Frank Huffman Texas Archeological Research Laboratory University of Texas at Austin Austin, Texas, U. S. A. [email protected]

The remains ofhumans that lived over a million years ago have been discovered at Sangiran Dome, 12 km north of Surakarta (Solo). This article provides background infbrmation fbr a visit to the Dome, and specifically to the Sangiran Museum, where maps, displays, fossils, and other materials related to Homo erectus at Sangiran will be seen. At the museum, you can buy a booklet entitled "The Sangiran Prehistoric Site Museum as A Tourist Object in Solo," which contains both accurate and f ~ c i o uinfbmtion s on the site. A technically sound popular is available in Semah et al. (1 990). A detailed description ofthe geology and paleontology is found in Watanabe & Kadar (1985) which can be purchased in Bandung fiom the Geological Research and Development Centre. A coderence was held in Solo during September, 1998, and the Proceedings from the meeting, entitled "International Colloquium on 'Sangiran: Man, Culture and Environment in Pleistocene,' will contain contributions on both the scientific and public issues surrounding the Sanigran Dome anthropological site. In addition to the Sangirn Musuem, local museums with good displays oforiginal fossil material are open at Miri, about 25 km north of Sangiran, and at Trinil, the first early man site in Java, near Ngawi in East Java. As the introductory geologic infbrmation is presented here, broad questions about Javan Homo erectus are posed, such as "What was the way of life ofthese early people?'The questions and associated discussion help convey a sense ofthe broad anthropological significance of fossils such as those from Sangiran. The "Sangiran Early Man Site" is one of 550 locales worldwide that UNESCO has recognized for its special cultural and natural value. Sangiran is among the most important paleoanthropological sites anywhere for addressing questions about the ecology of early humans. The fossils come from a 3 x 6 km uplift of PlioPleistocene mudstone and sandstone. The remains of dozens of Homo erectus have been discovered over the last 60 years, making Sangiran the most prolific source of early human fossils in Java. Eastern Java is the only part of Southeast Asia where such remains have been recovered. The fossils and stratigraphic column at Sangiran indicate a variety of paleoenvironments in the landscape inhabited by early humans. The site therefore serves as an exemplar for complexity in Homo erectus' habitat. Sangiran is m h e r significant scientifically because of the long time span that human occupation is documented.

Homo erectus lived in the Solo area for several hundred thousand years in the Pleistocene. What part of the Pleistocene is a controversial matter. The traditional view, the one followed by most Indonesian and foreign specialists on the Javan record, places early humans at Sangiran fiom 1.3 million years ago to 0.7 or even .I25 ma (e.g., Watanabe & Kadar, 1985; Sernah et al., 1990). This viewpoint is based primarily upon fission-track

IPA Field trip to Central bva

dates and paleomagnetic stratigraphy. On the other hand, the age ofhabitation recently has been placed at 1.7 to 1.0 ma on the basis ofa new set of40Arl39Ar radioisotopic dates (Swisher, 1997). Sorting out the evidence will take some time. There are problems with the application of paleornagnetic studies in some parts ofthe section, some researchers suspect that key fission-track dates are unreliable, and while40Ar139Ar determinations are generally a dependable indication of age ofthe material dated, there are questions about how the human fossils relate to the stratigraphic sequence as dated. For example, there is a question about how deep in the section human fossils occur -- throughout the lagoonal and lacustrine section ( P u c q a n Formation) or only in its upper part.

Even ifthe earliest humans at Sangiran are mid- rather than basal-Early Pleistocene, Homo erectus lived in eastern Java as early as, or nearly as early as, anatomically similar humans inhabited East and South Africa, where the genus Homo is thought to have evolved. Plio-Pleistocene humans evidently inhabited both the Indonesian Archipelago and Afiica contemporaneously for long periods of time and were well suited ecologically to both areas. These records of ecological success are so long that they are hard to compare to more recent human history. It is therefore difficult to keep in mind the vast range ofpossible trajectories of early human history that such long time frames permit. Modern civilizations can be traced for only a few thousand years. Even the late prehistoric record, when humans living fi-omEurope to Australia left clear signals of cultural sophistication, is only 25,000 to 50,000 years or so. How many profound events in early human history took place in the course of Early and Middle Pleistocene? What behaviors enabled Homo erectus to "succeed" in Java, parts of Afi.ica, and possibly areas of Eurasia for so much ofthe Early Pleistocene? Current models of early human evolution do not oRer much help in answering this second question -- they portray Plio-Pleistocene humans as having little ecological (or cultural) versatility. Java alone is a place where versatility would seem to be a significant advantage. Moreover, Java and East Afiicaseem to differ a lot environmentally. And all the lands that lie between these distant points add to the spectrum of climates and terrains that Plio-Pleistocene humans apparently inhabited or traversed. It is not clear whether Homo erectus of Java and anatomically similar humans in Africa and Eurasia are the same species or several. The Javan horninid fossils dating from the Plio-Pleistocene (approximately 2.5 to 0.7 million years ago) are classified by dserent paleoanthropologists into one genus but multiple species, one species but multiple subspecies, and a single taxon -- that is, the same genus and species with no subspeciespresent. Some others include erectus in our own genus and species, Homo supiens. For the purposes here, I follow those who take the one-species approach but separate the Sangiran fossils from modem humans. 1 term all the older fossil horninids of Java Homo erectus -- that is, our genus but not our species. Eugene Dubois, its discoverer, originally put erectus in it's own genus, Pithecanthropus, aname that is still sometimes used. Also, some oversized hominid teeth and heavier-boned jaw fi-agmentsfrom Sangiran are given the genus name Meganthropus by some paleoanthropologists. Behind the issue oftaxonomical classificationof the hominid fossils is at least one important question of human history. Did Homo erectus of Java contribute to the genetic pool and the Culture that comprise Homo supiens today? Some anthropologists believe that the present day people in various regions ofthe Old World,

1PA Field trip to Central \ma

such as China or Southeast Asia, evolved from indigenous populations of Homo erectus. As such the ecological and culltural history of the Plio-Pleistocene horninids in each region is directly related to the more modern human history. This concept is called the multiregional hypothesis ofhurnanorigins. Other anthropologists conclude that modem humans derive from the human stock of Afi.ica alone, and that Homo supiens fiom a i c a "replaced" populations of more "primative" humans in Eurasia, including those in Java, in the last 200,000 years. This proposition, the "replacement hypothesis," finds in the genetic make up of modem human populations worldwide. Even $Homo supiens from Africa did overwhelm Eurasian populations biologically, however, the cultural composition of the new populations may have been less completely Af+ican. Local cultural traditions, especially those tied closely to the regional environment, flora, and fauna, might have been transcended the biological "replacement." In the early and middle parts ofthis century, much ofwhat we knew of our distant ancestors was based upon the early human fossils from Java and China (Zhoukoudian near Peking). Over the past thirty years, many important and frequentlypublicized fossil finds have been made in East Africa. Based largely upon the African record, Western science has turned Plio-Pleistocene humans into savanna (grassland and open woodland) creatures, so that we tend to visualize the earliest people living on African plains. Homo erectus of Java, if it is brought into the picture at all, is assumed to be consistent with the savanna model. 1believe, however, that the fossil finds at Sangiran and elsewhere in Central and East Java challenge this view. As the scientific community addressesthe issue, we all will better understand Homo erectus' longevity in Java and the way people lived over one million years ago. Perhaps, the record ofhumans in Asia might again assume a central place in theory-making about human evolutionand cultural development. The Sangiran record prompts us to ask a number of interesting questions about our past; here are some that occur to me: O What behaviors enabled humans to &bit eastern Java in addition to the interior of M c a at such an early date in the history ofthe genus Homo and for such a long time in the Plio-Pleistocene? Does Homo erectus of Java fit the Afiican savanna model? Were there anatomical differences between early hurnans in Afkica and Asia that suggest varying biological adaptations? Was Homo erectus limited to those parts of Java having an East Afiican-like environment'? Or did early humans occupy forest, mountain and coastal habitats in Java in addition to whatever grassy interior lowlands existed? Is Java an ecological microcosm that shows us better thananywhere else in the world the adaptive range that Plio-Pleistocene humans had in the tropics? Could Java be the basis for a more useful model of early human behavior than East Africa'? O Does the fossil record of Java indicate that Homo erectus inhabited all of Southeast Asia? Were early humans flexible enough ecologicaUy (culturally)to have lived in the rain (ever-wet) forests of Sumatra, Borneo, and Malaysia? Is the lack of fossils in the rain forests attributable to poor fossil preservation, lack ofoutcrop, the use ofperishable materials for tools, and low densities ofkequently moving human populations, rather than a total absence ofhumans living there in the Plio-Pleistocene? Andlor were early humans more attracted to Central and East Java because rain forests did not dominate the landscape and Homo erectus preferred drier clunates, less-dense forests, or some other special environmental condition that existed there? Is the situation in Southeast Asia an indication that early horninids flourished in the jungles of M i c a and along the f i c a n coasts?

O c t - Nov.2998

IPA Field trip to Central javn

1

Geological Index Map

/ 1

1

A

OFH/97

Volcanic Peak

Holocene Alluvium

Plio-Pleistocene Formations

Quaternary Volcanic Rocks

Pliocene and Older Formations

Figure I : Geologic map ofeastern Java, highlighting the formations that contain Plio-Pleistocene Homo erectus fossils, such as those at Sangiran Dome (Sg). Other labeled localities are Trinil (Tr) and Mojokerto(Mo). The Ngandong site lies northeast ofTrinil, the Sambungmacan locality is between Trinil and Sangiran, and Kedung Brubus lies between Trinil and Mojokerto.

O Did we evolve fiom hunters (or scavengers) in savannas, only to become a species with exceptionally diversified ways of life in the last several hundred thousand years? Or has our genus always been able to find the necessities of life in a variety of environments? And would this kind of versatility partially explain how Homo erectus was able to find adequate resources in Java and elsewhere throughout the Quaternary? O Was the human genus a small-brained, slow-witted, nontalking primate until the modern human species (Homosupiens) appeared in the last 200,000 years'? Or did the earliest humans, such as Homo erectus

[PA Field trip to Central lava

of Java, have intellect and culture similar to Late Pleistocene anatomically modem humans despite having a brain much smaller than they had? O What were the earliest humans like? Ifyou were to meet a Pleistocene f d y at Sangiran, would you see

in their eyes a people of our own kind, as we do when encounteringthe men, women and children from any culture ofthe present day world? Or would they seem remote to you, their visage being no more human than thoseofchimps, orangutans, and gorillas? Ifyou could watch the aectines do daily routines, would you immediately understand the intent oftheir actions or would they be inscrutable'? How does Sangiran help with such far-reaching inquiries? From my perspecitve, Sangiran and the other hominid-bearing outcrops in Java document environmental diversity and thereby indicate sources of food and ways of lifi:for Homo erectus that are not suggested in the savanna model. The Javan landscape contains raindrenched volcanic mountains, calm and stormy sea coasts, and a variety of intermediate habitats. All this comprises a decidedly diverent setting from the vast, dry East Afi-ican interior. Java would have provided early humans with mountain, sea coast, and lowland food resources that were a few days' walk from one another, a situationunlike East Africa. This description fits Java today. What evidence is there that such conditions as these existed in Java a million years ago or so? The geology ofthe horninid beds helps co& the similarity of ancient and present day environments, and thus ecological diversity in Homo erectus' homeland. Homo erectus fossils (80+ specimens) have been recovered from a 180 km long swath ofmedial eastern Java. The area extends fiom Sangiran Dome eastward to Mojokerto, near the Madura Strait (Figure 1). The fossils occur in and immediately south ofthe Kendeng Hills. The enclosing sediments largely are tilted or folded volcaniclastic beds. The deposits are made up of material derived from the axial volcanoes ofeastern Java. An isolated occurrence of mammalian fossils, including Homo erectus remains, is found on the southeast flank of Mt. Muriah volcano, which is north of the Kendeng Hills and east of Sernarang (Figure 1). In addition to the Plio-Pleistocene remains at Sangiran, Mojokerto, and Trinil (where the first Homo erectus was discovered in 1891), the Homo erectus has been found in flat-lying, Late Pleistocene terrace deposits at Ngandong. This site is situated about 20 m above the Solo River 65 krn northeast of Sangiran. On the whole, then, the Homo erectus fossils of Java may represent as much as 1.7- 1.8 million years of human history and certainly represent 1.2 my. However, this span of time appears to include an extended gap in the record of human fossils. By virtue of the geologic context and nature of the Ngandong terrace deposits, Late Pleistocene Homo erectus must have inhabited a terrain that looked much like modern Java. The long gap in the human record is represented by volcanic deposits (the Notopuro Formation, mainly) that occur south ofthe Kendeng Hills and indicate the activity at Lawu and Wilis volcanoes. Beds ofthis kind crop out around the perimeter of Sangiran Dome and underlie the Solo basin more generally (Figure 1). Whether or not early humans lived in Java during the fossil gap cannot be determined, but the absence of fossils is not an indication that they did not. Whatever the situation for the mid-Pleistocenemight have been, the formations that contain the older Homo erectus fbssils provide important evidence concerning earlier human ecology. In particular, the physiographic setting for earlier Homo erectus, including the early humans of Sangiran, can be read with confidence fiom the stratigraphic record ofthe horninid-bearing units (which are generally mapped as the Pucangan and Kabuh Formations). The analysis benefits fiom modeling the Plio-Pleistocene aflei interrelationships that exist today

Oct

IPA Field trip to Central Java

-

Nov.1998

-

I(

PHYSIOGRAPHIC MODEL HOMO ERECTUS HOMELAND

Volcanic Delta

UPLANDS

1

LOWLANDS

Embavment Ocean

WATER

1 7

I1 -

I

Non-Volcanic Vallevs LaaoonlLake

1 j~ n m a ~ ~ a d

Figure 2: Plio-Pleistocene physiographic model of the homeland of Homo erectus. Hominid sites are same as in Figure I, but the Late Pleistocene Ngandong locality is omitted.

between tectonic regime, depositional environments, and landtbrrns in the Indonesian Archipelago. Indeed, geologic studies demonstrate that the major features of the modem landscape in eastern Java were also present when Homo erectus first walked the land over a million years ago. Figure 2, which is the product of geological research of this kind', presents a portrait ofthe Plio-Pleistocene land in which Homo erectus lived. The landscape included: (1) The flanks and peaks of volcanoes at Wllis, Lawu, Ungaran, Muriah and elsewhere. (2) Calcareous uplands in the ancestral Kendeng Hik, Rembang

'

0. F. Huffinan, 1997, Physiographic Diversity in the Homeland of Homo erectus, Java, p. A3 19 in Abstracts with Program, v. 29, n. 6, Geological Society of America, 1997 Annual Meeting, Salt Lake City, UT; 0. F. Hufiinan, in press, PlioPleistocene Environmental Variety in Eastern Java and Early Homo etvctus Paleoecology -- A Geological Perspective," Proceedings ofthe International Colloquium on "Sangiran: Man, Cultureand Environment in Pleistocene" held in Solo on September 2 1-24, 1998.

1PA Field trip to Central Java

Hills / Madura Island, and Southern Mountains; each presumably was less highly elevated than the upper portions ofthe volcanoes, by analogy with modem Java. (3) Broad valleys and sandy river courses lying between the volcanic and nonvolcanic uplands, and presumably extensive low relief slopes at lower elevations on the volcanoes and in the areas of nonvolcanic bedrock. (4) Lowlands such as (a) a coarse-clastic delta plain near Mojokerto, and (b) the muddy fiinges ofa lagoon and lake that was >35 km long in the Solo basin, where (c) a fluvially dominated environment developed later. (5) Muddy and calcareous marine shorelines along (a) the Randulatung embayment and (b) the northern Madwa Strait. (6) A volcanic coastline along the southern Madura Strait. (7) The southem coast of Java, where rocky and sandy shorelines experiencing the fUry ofthe Indian Ocean were probably present. Volcanoes, nonvolcanic uplands, valleys, deltaic lowlands, the volcanic shorelines, and the open-ocean coast are still part ofthe landscape of eastern Java. This list of features is based upon generalized (i.e., time averaged) and patchy data. Undoubtedly, the actual physiographic diversity was more complex. Such a variety of physical conditions dictates, in turn, far greater diversity of biotic environments. The various ecological elements occurred in an area no more than 150 x 250 km -- volcanic peaks were not more than a few days'journey from the sea (<75 krn) for Homo erectus. Paleoecological study of Homo erectus has tended to be satisfied with environmentalgeneralizationssuch as "open woodland," "rain forest," and "forest-edge." Clearly, the land that Homo erectus occupied was far more diversified topographically and ecologically than these characterizationsimply. Any population ofearly humans that was large enough to be viable presumably would have frequented a variety of small-scale environments in Java to prosper. This activity would have provided Homo erectus with the opportunity to exploit a broad range of foods and other resources. In addition, because the tropics experienced climatic changes during Pleistocene glacial and interglacial episodes, the human population ofJava may have survived significant environmental variations. Paleobotanicalstudies and the modem ecology of Indonesia this portrait of environmental diversity and variability in Plio-Pleistocene Java. Indeed, eastern Java appears to have been a microcosm of ecological diversity and early human habitats in the tropics. What impact did this complex environment have on early human behavior? It is difficult to determine how Plio-Pleistocenehuman populationsadapted to different environments. Neither fossil evidence nor analogies with present day primates suffice to answer the question adequately at this point of time. People during the historic past have adjusted to a variety of environments by adopting various cultural practices, many ofwhich are expressed in their technologies. Stone tools are numerous in the African PlioPleistocene. Some, if not all, were used in food acquisition, and presumably were important in human ecology. However, Plio-Pleistocene stone implements, unlike the technology of more modern people, are similar in form over large areas and spans oftime. Technology therefore appearsto have played a different role in the behavioral repertoire of early humans than it did in more recent people. The culture of modem people gives them a great deal more flexibility in the environment than is the case for other primates. Monkeys have clear-cut biological differences fi-omplace to place and environment to environment, and do not use tools. Wild chimpanzees have some socially transmitted technologies, suggesting that their ecological behavior depends partially on technologically based cultural practices analogousto those in humans. But what about prehistoric fossil humans and hominid species intermediate between humans and great apes? To what extent did biological adaptations and cultural innovations contribute to their ability to live in various settings'? Anthropologists are cautious in attributing cultural behaviors and social-biologicaladaptations

IPA Field trip to Central lava

such as language to early prehistoric humans. The tendency has been to formulate theories about them that involve a limited range ofenvironments and ways of We. Plio-Pleistocenepeople are seen as "primitive" culturally in keeping with the limited diversity oftheir stone tools. The Javan record argues tbr an alternate perspective. Limiting Plio-Pleistocene humans to a narrow range of habitats and ways of life is difficult to do once the environmental complexity of Java is recognized. Rather than assuming that early humans relied on the resources fiom one part of Java's complex ecological microcosm, it is more reasonable to suppose that Homo erectus was capable oftaking advantage ofthe diversity and did so by developing cultural practices appropriate to a variety of food acquistion activities. This would help explain why early humans were able to live in eastern Java for so long, despite environmental (i.e., climatic or tectonic) fluctuations, and why early humans could inhabit parts of Africa, Java, and Eurasia during the Plio-Pleistocene,despite differences in climate, terrain, and biota. Actually, two possible modes of ecological behavior fit the situation in Java and only one ofthem requires multiple cultural adaptations;the alternatives are: (1) Homo erectus specialized in a limited number ofplant or animal foods that were present in a variety ofthe island's rnicroenvironrnents and survived the climatic cycles the region experienced; andlor (2) as early as the latest Pliocene or Early Pleistocene, Homo erectus was an ecological generalist, much like later humans, and flourished in Java because of an ability to exploit a range of resources and habitats, changing diet (thus, ecological and cultural behaviors) t7om time to time and place to place as environmentalconditions required. I favor the second hypothesis because it connects the ecological behavior of Homo erectus with that of Homo sapiens in a straighthrward way. ittedly, this proposal does not help in understanding why so little technological change is evident in Plio-Pleistocene stone tools and why Lithic artefacts are so rare in the hominid beds of Java. Homo erectus still could have lived largely on the meat it hunted or scavenged,just as is postulated for the early humans in At7ica. The animal resources eaten in Java did not necessarily inhabit savannalands, however. Some of the ungulates that have been found as fossils in the hominid-bearing formations of Java were long-lasting faunal elements, probably had broad ecological ranges, and therefore are consistent with both the scenarios listed above.

An example is the native cattle of Java, the bantengs. They are known in the fossil record fiom the latest Pliocene to the Holocene. They also survive in refbges on Java and Bali today. Bantengs had a broad ecological range in the historical past. Hints of this are found in the places they still occur. The same is true tbr many of Southeast Asia's larger mammals that might have constituted food for early humans, such as pigs and deer. Although they are generally grazing animals that therefore congregate in grasslands, cattle are noted (a) in the ever-wet parts of Java and in the more distinctly monsoonal ones, (b) in parts ofthe island where rain forest dominates and in areas where dry deciduous forests and savannas occur, and (c) along the coasts of southern Java and on mountain tops 2 to 3 krn above sea level. Native cattle of several taxonomic designations actually occur in many parts of Southeast Asia that are dominated by forests. For instance, bantengs occur in the rain forests of Borneo, where they occur with deer, mouse deer, muntjaks, pigs, rhinoceroses, elephants, macques, gibbons, orangutans, and other elementsthat are found as fossils withcattle in the horninid-bearing beds of Java. Other potential game for Homo erectus also had long-lasting and ubiquitous distributions. Presumably, there were plant foods with broad distributions also. For example, forest fruits, nuts, tubers, and possibly bamboo or palm shoots and sugar cane, come to mind. Some combination of these could have buf5ered Homo erectus against environmentalchange, as long as lowland forest were available.

lPA Field trip to Central Java

Still, it is unclear whether Homo erectus (1) was a dietary specialist or a generalist, (2) concentrated on plant foods or animals, (3) had a universal diet or included groups that lived on various combiitions of food, (4) used the same resources throughout the Quaternary or changed them from time to time. Answers to these questionswill impact theories ofhuman evolution greatly, and influence our understanding of our distant biological and cultural heritage. Without such answers, placing early humans in a single biotope, such as a savanna, and endowing themwith little culture presumes a "primitive" state that is not demonstrated. It may be a surprise to learn that such a range of alternative interpretations of early human behavior would still be possible after as much scientfic research as has been done on our distant ancestors. Early human fossils have been recognized and studied for over a century. And the record of Plio-Pleistocene humans started with the 1891 discovery by Eugene Dubois at Trinil only 60 kmeast of Sangiran. What explainsthe slow pace of scientific advance in understanding early human ecology and behavior? A partial explanation lies in the limited, often fiagmentary human fossils that one finds in Plio-Pleistocene age deposits. In Java, for instance, the vast majority ofthe human fossils are iiagments of skulls, jaw bones, or teeth. Long bones, such as the shafts of femurs, are discovered in small numbers. The original Homo erectus fossils from Trinil, for example, arejust a skullcap and partial thighbone. By no means do Plio-Pleistocene remains look like the skeletons that have been found in the human burials of the last 10,000 years or so. One East African site (Nariokotome) did yield an Early Pleistocene skeleton of a Homo erectus-like individual in the mid- 1980s. This is the most complete (40%) set of early horninid skeletal remains found anywhere. In order to recover the scattered fossils of this one individual, over 1,500 tons of earth had to be excavated and sitled by hand. The reconstructed Nariokotome skeleton nevertheless is dramatic substantiation of int'erences made as long ago as the 1 890s by Dubois that the main skeletal differences between humans ofthe Plio-Pleistocene and modern Homo supiens are the shape of the cranium (largely, a lower, smaller brain case) and the thickness of the bones. The Nariokotome fossils are the remains ofyoung boy, estimated to have been a 160 cm and 48 kg 10 to 12 year old when he died. His body, but not his skull, was quite modern in form and size. Stone tools have been found with early hurnan fossils in Mica, Europe, and China. The tools indicate a cultural side of Plio-Pleistocene humans that we otherwise might not suspect. Few implements have been discovered in situ in Java. Artifacts were found for many decades on the surface but none in the horninid formations. In the 1990s, however, chipped-stone cutting or chopping tools and stone hammerstones, including spheroids and polyhedric shapes, were recovered iiom the Pleistocene Kabuh beds at Sangiran Dome (Ngebung site). The rarity of stone implements in the horninid beds of Java led some archaeologiststo believe some years ago that Homo erectus was somehow less intelligent in Southeast Asia than elsewhere. Conclusionsof this kind are not warranted. Differences in the technology ofmodern hurnan groups, for example, do not indicate dBerences in their intelligence. Sophistication with tools is theretbre not necessarily a measure ofintellect in ancient hurnan species. Furthermore, stone tools survive in geological contexts where tools made ofperishable rnaterials, such as wood, do not. Homo erectus in Java, for example, might have used bamboo and rattan tools that quickly vanished fi-omthe geologic record, while early humans elsewhere made implements of lithic raw materials that survived burial and remain intact indefinitely. Even where lithic tools are abundant, the people who made them might have constructed many more objects out ofperishable m a t e d . In sum, finding no a r t k t s does not necessarily mean an absence of human presence or cultural sophistication.

IPA Field trip to Central Java

Recently, late Early Pleistocene stone tools have been reported fiom Flores Island, 600 km east of Java. The reports encourage the view that Homo erectus was even more of an explorer than previously thought. Ifthe limited amount of Plio-Pleistocene human fossils and stone tools hmders paleoecological and behavioral understanding of early humans, what light can the other fossil material in the hominid beds, or the nature of the deposits themselves, shed on these issues? The Homo erectus-bearing formations, such as the mudstones (Pucangan Formation) and sandstones (Kabuh Formation) at Sangiran Dome, not only give us information about the paleophysiography and surface conditions of Homo erectus' homeland but also are rich in large mammal skeletal remains and plant fossils, such as pollen and leaf imprints, with which to infer biotic conditions. The principal geologic evidence for each of the seven physiographic elements listed above is as follows (Figures 1 & 2): (1) Axial volcanoes are indicated by the volcaniclastic facies that comprises much of the horninid formations and their equivalents in eastern Java; the presence of a huge Plio-Pleistocene volcano at Mt. Wdis in the midst of the hominid area is indicated by the volcanic breccias (reportedly lahar deposits) that occur for a distance of 140 krn along the south side ofthe Kendeng Hills (Trinil to Mojokerto, approximately), including several hundred meters of breccia that crops out near Mt. Wilis 1 Pandan on the apparent flank ofthe volcano. (2) Nonvolcanic uplands --the sand and gravel that were derived fiom limestone and marl, the thinning of the Plio-Pleistocene sequence towards the central Kendeng Hills and the Rembang (Tuban) Hills, and the need to separate the volcaniclastic facies on the south side ofthe Kendeng Hills fiom a muddy marine facies on the north side. (3) River courses -- the fluvial deposits that contain vertebrate, nonrnarine mollusk-, and plant-fossils and typifL the Kabuh Formation and portions of the Pucangan Formation. (4) Lowlands -- the interfingering of fossiliferous volcaniclastic sediments and marine mudstone near Mojokerto (Mojokerto delta plain), and the fossil-rich Pucangan lagoonal and lacustrine mudstones exposed at Sangiran Dome and several other spots in the Solo basin (fringes of the lagoon and lake). (5) Muddv marine shorelines -- the mudstone equivalents of the hominid-bearing formations that occur north ofthe Kendeng Hills, on Madura Island, and under the modern Madura Strait. (6) Volcanic coast -- the scattered evidence of Early Pleistocene volcanic rocks and marine sediments along the southerncoast ofthe Madura Strait in easternmost Java. (7) Open-ocean coast along the Indian ocean -- the rapid deepening of the continental shelf and slope south of Java and the long time that the tectonic regime responsible for the bathymetry was active during the late Cenozoic. Palynologicalstudies ofthe hominid beds the presence ofvolcanic mountains and lowland features such as mangrovesand swamps. At Sangiran Dome, the mudstones (Pucangan) are lagoonal and lake deposits, and the sandstones (Kabuh) represent ancient river beds. Tuffs occur at many horizons, indicating active volcanism. Also a prominent laharic deposit is hund at the base ofthe Pleistocene at Sangiran; the unit is exposed at the Sangiran Museum and contains vertebrate remains as well as shelly fossils. The sands and associated gravels ofthe younger Kabuh Formation are mostly volcanic debris, reflecting the presence of volcanic edifices such as Mt. Lawu. The sediments also include nonvolcanic clasts from the limestone and marl terrains to the north (Kendeng Hills) and south (Southern Mountains) of the Solo basin. The lagoon and lake stretched for an east-west distance of 35 krn or more during the earlier part of the time span that the region was inhabited by Homo erectus. Thick soils and vegetation apparently blanketed the surrounding hills because little coarse sand and gravel made its way into the basin. This was not the case everywhere in eastern Java at the time; voluminous amounts of sand and gravel were laid down around Mt. Wilis (Figure 2). Most of the early human fossils at Sangiran Dome have been found in fluvially dominated Kabuh Formation, which overlies the lacustrine Pucangan. Apparently, new tectonic activity fostered bedrock exposure

IPA Field trip to Central lava

and rapid erosion in the volcanic and nonvolcanic uplands surrounding the basin during the deposition of the fluvial beds. The new conditions dramatically altered the landscape in Solo area on the whole. Homo erectus apparently did well in both sets of conditions. Other fossil occurrences the impression that Plio-Pleistocene Homo e ~ c t u lived s in a variety of settings, but none of the evidence is conclusive. The Trinil fossil was deposited in tluvial sediment also but the age of the deposit is poorly determined. The Mojokerto fossil apparently predates the human remains in the mudstones of Sangiran and lived at a time when marine conditions still prevailed at the Dome (Kalibeng Formation). Although it came to rest in a nearshore marine environment, the Mojokerto Homo erectus inhabited either the coarse clastic delta plain, the banks of the gravely river upstream of the delta, or the volcanically disturbed and rapidly eroding flanks of a large volcano. The uncertainty in these interpretations largely arises because the human fbssils, Like most of the other PlioPleistocene vertebrate fbssils in eastern Java, were transported before deposition. As a result, little direct evidence of human habitation has been found. One archaeological excavationthat was carried out during the last decade in the Sangiran area (Ngebung) reportedly has exposed a human habitation surface with stone tools. Homo erectus, therefore, can be placed in the basin when it was dominated by sandy and gravely river courses. The presence of early humans in other parts ofthe terrain must be int'erred. The presence of Homo erectus along the shores ofthe lagoon and lake that preceded the riverine phase is assured because of the numerous human fossils found in the (Pucangan) mudstones. Evidence documenting habitation outside lowlands is not available now and may never be. The compactness of Homo erectus' homeland (Figure 2 ) nevertheless pennits the assumption -- a preferred working hypothesis -- that early humans inhabited or frequented all ecologically attractive land areas in eastern Java. What potential foods did the land provide? What animals might early humans have hunted or scavenged'? The best general listing of the large-bodied animals that inhabited the early Homo erectus landscape comes fiom the "Jetis Fauna." This fossil assemblage appears to be older than (or as old as) any in the hominid-bearing formations. It dates to about 1.8 million years ago, based upon K-Ar and 40Arl39Ar radiometric results. The tbssils were collected at a number of localities near the Mojokerto horninid sites, at the east end ofthe hominid belt near Surabaya. Nearly all the same genera are found at Sangiran Dome. The primates recorded are hominid (Homo), orangutan, gibbon, macaque, and leaf monkey. Carnivores are tiger*, leopard*, large-tooth cat, hyena, wild dog*, sun bear*, otter*, and civet*. Elephantids include both elephant* and Stegodon. Ungulates are hippopotamus, rhinoceros*, tapir*, pig*, Asian buffalo*, wild cattle (banteng)*, an endemic ox, antelope, several species of deer*, muntjak*, and mouse deer*. Anteater*, turtle*, crocodile*, and fish* also occur. Porcupine*, rabbit*, rat*, and several bird* genera are known fiom other localities in the hominid-bearing formations. The fauna of Java lacked horses, camels, and giraffes -- elements common in Eurasia. The Javan assemblage is typically Southeast Asian, comprising mostly genera, although not species, that continued to inhabit Java throughout the Quaternary and in some cases still live in the region (indicated by *s). These genera occur widely in lowland ever-wet forests, but some are equally at home or more abundant in drier biotopes. One striking aspect of the terrestrial fauna of Java is its similarity fiom age to age over the last 1.8 million years or more. Cattle, deer, pigs, elephants, rhinoceros, tigers, monkeys, etc. inhabited Java with humans 1.7-

IPA Field trip to Central Java

1.8 million years ago and in the Holocene. Were these large mammals Homo erectus food'? The meat and bone of some, such as the wild cattle (bantengs) discussed above, are certainly candidates for early human dietary staples. Cattle fossils are abundant in Mesolithic (late preagricultural) archaeological sites in eastern Java, providing evidence that Homo supiens found cattle an important food in the Holocene forests of Java. Perhaps, Homo erectus did also.

What other sources of food might Homo erectus have found in medial eastern Java'? Smaller terrestrial mammals are greatly underrepresented in the reported collections of Plio-Pleistocene vertebrate fossils when compared to living faunas. Small mammals nevertheless would have populated the homeland forests of Homo erectus, and therefore must also be considered an abundant potential dietary resource. The remains of aquatic and semiaquatic animals, such as mollusks, fish and turtles, also are present in the fauna. They suggest other readily available animal foods. Perhaps, Homo erectus sometimes lived off river mollusks and turtles. Less is known of the plants that Homo erectus might have eaten. One study of surface wear on Homo erectus teeth suggests, however, that tough plant foods were important in the diet. Despite there being several studies of the Plio-Pleistocene flora of eastern Java, but not much about edible plants has been published. Temporal continuity is evident in the botanical record of eastern Java as it is in the mammalian record. Many of the elements of modern mangrove, swamp, lowland-rain, lowland-deciduous, montane, and disturbed (i.e., by volcanism) forests are found in palynological samples at Sangiran Dome and elsewhere in the Plio-Pleistocene sequence. Perhaps the ancient and modern records can be used together to determine what edible plants are likely to have been within the grasp of Homo erectus. During times when lowland rain forests were widespread in Java, so too would have been fruit- and nut-bearing trees, as well as forest tubers or other edible vegetables. Yams and palm pith are important staples for some indigenous forest people. Some rain forest fruits are commonly eaten throughout the region; Durio (durians), Nephelium (rambutan), Artocarpus (jackhit and breadh i t ) , Mungif1.r~(mangos), and Ficus (figs) are examples. If present during the Plio-Pleistocene in Java, plants of the forest could have offered Homo erectus abundant food resources. Long-term continuity is evident in landform (Figure 2), large mammals, and trees in the Homo erectus homeland. This suggests that a critical relationship existed betweenearly human ecology and the particular environmental setting of eastern Java. Hypotheses based on reasoning of this kind might emphasize the advantages offered by diverse sources of food, mentioned above, and high biotic output from the landscape. The soils that ed the high productivity might have been enriched by periodical explosive eruptions of andesitic volcanoes and the hot, wet, seasonal clunate of eastern Java. The latter elements of the modem ecology of Java are used to explain the high agricultural production and extraordinary population densities in the areas around the axial volcanoes of Central and East Java. One might suppose that Homo erectus was widely distributed in Southeast Asia but lived more densely around some ofthe volcanoes of Java than they did in places where no volcanoes existed or the climate was ever-wet. Much ofthe debate that has taken place about the early human environment in Java has focused on whether Homo erectus favored a habitat that was forested or not (in which case it was a drier biotope, such as open woodland or grassland). Savanna environments have been associated with our distant ancestors in various scientific theories since the time of Darwin. On the other hand, rain forests are the typical vegetation in Southeast Asia. The question of past vegetation patterns in Java remains unresolved. Forests of many kinds certainly existed during the PLio-Pleistocene. Grassy vegetation (possibly including bamboo and swamp grasses) is represented in palynological samples by nonspecific pollen. However, it remains undetermined what types of grasses

IPA Field trip to Central Java

were present and how much ofthe landscape, if any, was dominated by grasslands. Whatever the principal vegetation, the complexity of the paleophysiographic picture (Figure 2) suggests that forest and savanna resources were only part ofthose that early humans could have exploited. The sea coasts of Plio-Pleistocene Java, in particular, hint at the diversity of habitats and food resources that would have been available to Homo erectus. The coasts north of the axd volcanic belt (Figure 2) included low relief, muddy shorelinesthat doubtlessly were h g e d with mangrove forests. The pollen of mangrove trees not only is a consistent part ofthe pollen record of the hominid-bearing formations of eastern Java but is a floral component in deposits representing many millions of years in Southeast Asia. Modern mangroves are rich in crabs, snails, and bivalves -- animals available for gathering. Snakes, monitors, lizards, fi-ogs, monkeys, small otters, bats, and both nesting and migratory birds are potential hunting targets. A different assemblage of shoreline resources was likely to have been available along Java's south coast, assuming a Plio-Pleistocene geologic situation similar to the modern one. Extensive sandy beaches are found today along some stretchesofthe south coast. Sea turtle eggs and hatchlings, if not the adult animals, afford a periodic and abundant opportunity fbr human food. Beach forests occur locally where bantengs, deer, monkeys, bats, monitors, and turtles live much as they do at inland locales. Elsewhere, cliffs occupy the southern coast. Colonies of seabirds, such as boobies, terns, and tropic birds, indicate the possibility of easily gathered hod. Evidence of this kind suggests that virtually every corner of the landscape envisioned for Homo erectus (Figure 2) held food resources that early humans might have used. Distant ancestors of modern humans therefore might have exploited all the montane, upland, riparian, and coastal biotopes available in Java, less intensely than modern Indonesians do but with geographical coverage as complete. If Plio-Pleistocene Homo erectus possessed the ability to adjust dietary behaviors to meet this broad spectrum of circumstances, then fewer behavioral diff'erencesexist between early and more modem humans than is usually supposed. Perhaps nowhere else in the world is it easier to see this aspect of humanity's distant past than in eastern Java and particularly at Sangiran Dome. What foods early Homo erectus actually ate, whether the erectinestalked or only gestured, how they were organized socially, what intellectual life they had, and how they dealt with the environmentalchanges that took place over the decades and millennia in Java are questions that fhture discovery and study will be&' 7111 to answer. Whatever the course of paleoanthropological research around the world, the record of Plio-Pleistocene paleoenvironmental diversity in Java, exemplified by the geological sequence and human fossils at Sangiran Dome, should contribute substantially to understanding our past.

Selected References and Readings Bartstra, G.-J., Soegondho, S., and van der Wijk, A., 1988. Ngandong man: age and artifacts. Journal of Human Evolution 17: p. 325-337. Bellwood, P., 1997, Prehistory of the Indo-Malyasian Archipelago. University of Hawai'i Press, Honolulu, 384p.

[PA Field trip to Central Java

de Terra, H., 1943. Part V Pleistocene Geology and Early Man in Java, p. 437-464. in H. de Terra and H. Movius, Jr. (Editors), Research on Early Man in Burma. Transactions. The American Philosophical Society, Philadelphia. Kadar, A. P., 1992. Review ofthe Sangiran (Central Java) Plio-Pleistocene Environments fiom Marine and Non-Marine Floras and Faunas, p. 5 1-60, in Twenty-Ninth Annual Session. Committee for Co-Ordination of t Prospecting for Mineral Resources in Asian Offshore Areas (CCOP), Hanoi, Vietnam. Johanson, D., and Edgar, B., 1996. From Lucy to Language. New York, Simon & Schuster Editions, 272 p. (Photographsby David Brill). MacKinnon, K., Hatta, G., Halim, H., and Mangalik, A., 1996. The Ecology of Kalimantan. The Ecology of Indonesia Series. Volume Ill. Periplus Editions, Singapore, 802 p. Pope, G. G., 1995. The Influence of Climate and Geography on the Biocultural Evolution of the Far Eastern Hominids (Chapter 34) p. 493-506. in E. S. Vrba, G. H. Denton, T.C. Partridge and L.H. Burckle (Editors), Paleoclimate and Evolution, with Emphasis on Human Origins. Yale University Press, New Haven and London. Potts, R., 1996. Humanity's Descent The Consequences of Ecological Instability. William Morrow and Co., New York, 325 p. Rightmire, G.P., 1990. The Evolution of Homo erectus Comparative anatomical studies of an extinct human species. Cambridge University Press, Cambridge, 260 p. S e m i F., and Grimaud-Herve, D. (Editors), 1993. Le Pithecanthrope de Java A la Decouverte du Chainon Manquant, n. 184. Les Dossiers d'Archaeologie, Dijon, 77 p. Semah, F., Sernah, A.-M., and Djubiantono, T., 1990. I1y a plus d'un million d'annees...11s ont decouvert Java1 More than one million years ago...They discovered JavdLebih dari satu juta tahun yang lalu. ..Mereka menemukan pulau Jawa. Museum National d'Histoire Naturelle/Pusat PenelitianArkeologi Nasional, Paris/ Jakarta, 128 p. Sharp, I., and Compost, A., 1994. Green Indonesia. Tropical Forest Encounters. Oxford University Press, Oxford, 184p. Swisher, C., 111, Curtis, G.H., Jacob, T., and Getty, A.G., 1994. Age ofthe earliest known hominids in Java, Indonesia. Science, 263 (25 February 1994): p. 1118- 1 121. Swisher, C.C., 111et al., 1996. Latest Homo erectus of Java: Potential Comtemporaneity with Homo sapiens in Southeast Asia. Science, 274 (5294 13 December 1996): p. 1870- 1874. Tattershall, I., 1998. Becoming Human. Harcourt Brace, 264 p. Theunissen, B., 1989. Eugene Dubois and the Ape-Man from Java: The History ofthe First "Missing Link" and Its Discoverer. Dordrecht, Kluwer Academic Publishers. van Heekeren, H. R., 1972. The Stone Age ofIndonesia. Martinus Nijhoff,The Hague, 247 p. von Koenigswald, G.H., 1956. Meeting Prehistoric Man. Thames and Hudson, London, 21 6 p. Walker, A. and Shipman, P., 1 996. The Wisdom of the Bones In Search of Human Origins. Alfred A. Knopf, New York, 336 p.

/PA Field trip to Central Java

Watanabe, N., and Kadar, D. (Editors), 1985. Quaternary Geology ofthe Hominid Fossil Bearing Formations in Java; Report of the Indonesia--Japan t Research Project, CTA-41, 1976-1979. Geological Research and Development Centre, Geological Research and Development Centre, Bandung, Indonesia, No. 4. Whitten, A., Whitten, J., and Cubitt, G., 1992. Wild Indonesia. The wildlife and scenery of the Indonesian archipelago. New Holland Ltd., London, 208 p. Whitten, T., Soeriaatmadja, R.E. and Suraya, A.A., 1996. The Ecology of Java and Bali. The Ecology of Indonesia Series Volume 11. Periplus Editions, Singapore, 968 p.

Some Current WEB Sites about Homo erecfus

http:/lwww.geocities.comlTheTropics/3581/resource.html [I 992 Sangiran Guide] http://www.petra.ac.id/english/eastjava/cities/ngawi/tourobj/tnnil.tou[Trinil Museum] http:llanthro7.anthro.uiuc.edul-anth102/index.html [Course on human evolution] http://www.csus.edulanthlphysanth/ancestor.htm [Course on human evolution] http:llwww.handprint.comlLSIANCIHomo.html[Photos of fossil hominids]