texcoco lake

7/23/2019 texcoco lake

http://slidepdf.com/reader/full/texcoco-lake 1/22

PALEOLIMNOLOGY OF LAKE TEXCOCO, MEXICO.

EVIDENCE FROM DIATOMS

John P. Bradbury

Limnological Research Center, University of Minrlesota, Minneapolis 55455

ABSTRACT

A 46-m core from the lacsutrine sediments beneath Mexico City was analyzed to

establish a stratigraphic sequence of diatom assemblages for use in interpreting the climatic

and limnologic history of ancient Lake Texcoco.

Diatoms were found in nearly every 20-cm sample interval, and several major zones

were established. Planktonic and benthonic-epiphytic assemblages alternate throughout

the core, both of fresh- and brackish-water types.

The alternations reflect the fact that

the coring site is marginal to the main basin of the la’ke, and limnologic conditions change

as water levels rise and fall.

A freshwater planktonic assemblage dominated by Stephanodiscus niagarne in decpcr

parts of the core indicates that a large, cool, and possibly deep lake existed about 100,000

years ago, either because of pluvial or because of tectonic factors. This is replaced (depth

35 to 30 m) by a freshwater benthonic-epiphytic assemblage characterized by Denticula

elegans and other marsh diatoms.

The marshes were probably maintained by springs from

the shore when the lake was reduced to saline pools in the center of the basin.

As water levels rose again (core depths 30 to 5 m), brackish water flooded the marshes,

and brackish benthonic diatoms (such as Anomoeoneis cost&a, Campylodiscus clypeus,

and Nitzschia frustulum) replaced the earlier floras. These were periodically replaced

by brackish planktonic diatoms (as Cyclotebla striatn and CycZoteZZu quillensis) when the

lake was deeper, but the earlier frcshwatcr planktonic flora never recurred. The same

brackish planktonic and benthonic diatoms that prevailed for several tens of thousands of

years are found today confined to the brackish pools of Lake Texcoco that are remnants

of the former larger lake. The long interval of fluchlating brackish floras probably rcprc-

sents Wisconsin time.

The last 10,000 years of the lake’s history is marked by a return of the marsh flora,

sllggcsting a climate drier than that of Wisconsin time. A marked climatic change, how-

ever, is not necessary to explain this last change in the cliatom flora, and it seams likely

that the pluvial climate inferrecl for the southwcstcrn United States had less effect at the

latitude of Mexico City ( 19” 30’) than farther north.

INTRODUCTION

This paleolimnologic study of Lake Tex-

coca, Mexico, began in 1968 when I was

a postdoctoral fellow at Yale University

under the advisement of E. S. Decvey and

G. E. IIutchinson. It was partly supported

by a grant from the American Philosoph-

ical Society. The project was continued at

the Limnological Research Center of the

University of Minnesota where I was a

National Science Foundation postdoctoral

fellow under II. E. Wright. The cnthusi-

asm and support of thcsc individuals have

greatly facilitated this work. In addition,

I wish to acknowlcdgc the kind help of

P. 13. Scars, of L. Zccvaert, who provided

1 Contribution 92, Limnological Research Cen-

tcr, University of Minnesota.

material for study, and of Prof. J. L.

Lorcnzo and his colleagues at the Institut0

National dc Antropologia e Historia in

Mexico City, who have shared with mc

many of their insights about the cnvi-

ronmcntal history of the Basin of Mexico.

GI3OLOGIC SETTING

The general sequence of events that led

to the formation of the Basin of Mexico

h

as been traced (Mooser ct al. 1956;

Mooser 1963). The basin began as a gra-

bcn in the Tertiary trans-Mexico volcanic

belt, bounded on the cast and west by two

fault zones and their associated volcanoes,

Sierra de Las Cruces and the Sierra Nc-

vada, the latter containing the famous

volcanoes Popocatepctl and Iztaccihuatl.

The faults and volcanoes have been active

LIMNOLOGY AND OCEANOGRAPIIY

180

MARC11 1971, V. 16(2)



PALEOLIMNOLOGY OF LAKE TEXCOCO

181

throughout the Tertiary, progressively cov-

ering the Crctaccous basement; par ticu-

larly massive extrusions of acidic lava and

probable sinking of the graben floor oc-

curred in the Miocene and Plioccnc, de-

fining a valley whose integrated river

system drained to the south (Figucroa et

al. 1968). The northern limits of the val-

ley arc defined by lower mountains pro-

duced by faulting and volcanism in the

region of Pachuca. The valley was closed

in the late Pliocene when basaltic volcanic

activity from centers located in the south-

ern part of the basin (the massive Chichi-

nautzin lavas and Sierra Ajusco) dammed

the valley. The basin thus formed was

rapidly scdimented with elastic and pyro-

elastic material to a thickness of 800 m

(Figueroa et al. 1968), and the regional

drainage converged to the lowest part,

where a lake has persisted until modern

times.

LIMNOLOGIC SETTING

Today the Basin of Mexico is a plain

(elevation about 2,236 m) surrounded on

the east, south, and west by high moun-

tains (3,000-6,000 m). It was a closed

hydrographic system before being artifi-

cially drained in 1900, and precipitation in

the mountains and runoff from summer

rains drained into a chain of lakes that

nearly traversed it from north to south

( Fig. 1). After the rainy season ( May-

October) the lakes wcrc frequently joined

into a single sheet of water (clcvation 2,242

m), which during the dry winter months

was separated into a number of subbasins,

some artificially contained by dikes, The

principal ones and their elevation rclativc

to Lake Tcxcoco listed from north to south

arc [elevations from Zecvacrt (1952) and

Bonaparte et al. (ca. 1900) ] :

Zumpango

+6 m;

Xaltocan

+3 m;

San Cristobal

4-3 m;

Texcoco

0 m;

Mexico

+0.85 in;

Xochimilco

f3.5 m;

Chalco

d-3.5 m.

Lake Texcoco is the lowest in the scrics

and the most saline, both because of

evaporation and because thermal springs

flow into it (Mooser 1963). The Mexico

subbasin, artificially con taincd by Aztec

dikes to the west of Lake Texcoco, was

maintained by freshwater from Chalco,

Xochimilco, and numerous springs from

Chapultepcc,

southwest of Mexico City.

It drained into Lake Texcoco during times

of water surplus through an organized

system of canals and gates, In similar

fashion Lake Chalco was separated from

Lake Xochimilco. Because of the abun-

dance of frcshwatcr in the southern part

of the basin, Chalco, Xochimilco,

an d

Mexico were cxtcnsively used for chi-

nampa farming ( Dccvcy 1957).

The draining of the Basin of Mexico

and the growth of Mexico City have cre-

atcd some engineering problems.

The

early dikes, cspccially that of Nctzahua-

coyotl (ca. 1450 A.D. ), were built to pre-

vent seasonal floods of saline water from

cn tcring the highly productive chinampu

farms southwest of the capital. Flooding

continued in colonial times, and the need

for cffcctive sewage disposal for the city

that was rapidly grolwing onto the plain

of Lake Tcxcoco demanded that the lakes

bc systematically drained. The first cf-

forts began in the 17th century, and the

work was finally complctcd in 1945 with

the Tcquisquiac

tunnel, which has re-

duced the surface arca of Lake Texcoco

to a rainy-season arca of only 200 km2

( Mooscr 1963).

As the drainage of the lakes was cf-

fectcd, and more and more water was

pumped from aquifers of sands and silts

bcncath the lake plain, Mexico City began

to sink into the highly bcntonitic clays

that undcrlic it. This problem is cspccially

serious whcrc heavy buildings are con-

structcd in the metropolitan areas. Dr. L.

Zecvacrt has for many years conducted

stud& of the mechanical nature of the

lake sediments of the basin, principally

beneath Mexico City, to seek competent

strata for the placcmcnt of foundation

pilings. Hc has .taken numerous cores of



182

JOIIN I’. BRADBURY

99OlOO’

ChaDultevec <-

3% Reforma - Hovre core

- maximum level

0 5 40

I

I

I

km

w m

mean level

.~**~****~* minimum level

FIG.

1. Index map showing lakes in the Basin of Mexico.

Adapted in part from J. L. Lorenzo

(in Mooser et al. 1956).

the scdimcnts with a simple Shelby tube

tcr content arc studied. Through this work

sampler that is forced into the lake sedi- Zccvacrt (1952, 1953) has compiled a dc-

mcnts by a mechanized coring rig.

In

tailed stratigraphy of the Basin of Mexico

resistant material a jar hammer is used.

to clcpths of about 70 m. Hc has been

Cores about lo-cm diam and 2 m long

most intcrestcd and cooperative with rc-

arc taken in this fashion and their sedi-

spect to ancillary scientific studies on his

mentology, mechanical propcrtics, and wa-

cores and has provided both Sears and



Clisby ( 1955) and me ( Bradbury 1970b)

with samples.

LACUSTRINE STRATIGRAPIIY

The Plcistoccne stratigraphy in the Ba-

sin of Mexico was first studied in detail

by Bryan (1948). Hc worked with the

exposures of soils, tuffs, and alluvium on

the margins of the basin and divided them

into units of the following names, charac-

ters, and ages:

Nochc Buena-soils and alluvium, with

pottery shards;

Pre-Classic, Classic, and

Pus t-Classic.

Toltolzingo- dark brown soils, alluvium,

some eolian material.

Barrilaco -calichc-pedocal; Altithermal

4,500-7,500 BP.

Beccrra-alluvium-pedalfcr, Elephns,

Equus, Bison etc.; Cochranc-Mankato.

Morales-caliche-pcdocal; Wisconsin in-

tcrstadial.

Tacubaya-yellow-brown alluvium-pc-

dalfer; Tazcwell-Gary.

Tarango-acidic volcanic tuff, watcr-de-

posited; Plio/Pleistocenc?

Bryan (1948) assumed that these for-

mations or their cquivalcnts exist in the

lacustrine deposits in the center of the

basin, and Zeevaert (1952, 1953) corre-

lated Bryan’s strata with the alternating

layers of lacustrine clays, silts, and sands

undcrncath Mexico City. Lithology is used

as the basis for correlation of the coarser

units, whereas lake clays are thought to

bc contcmporanoous with soil formation

and roduccd alluviation. Deposits high in

calcium carbonate are considered cquiva-

lcnts of the calichcs on the basin margins,

Foreman (1955) more carefully analyzed

the sediments bcncath Mexico City, and,

although he did not USC: he formational

names ‘of Bryan and Zecvaert, he showed

their correlation to his findings. IIis stra-

tigraphy is generalized into scvcn zones,

but despite the complete lithologic dc-

scription, they do not have diagnostic

characteristics.

This seems to bc a result

oE high variability ,of the sediments and

the prodominancc of ash and wcathercd

as

h in them.

183

Bryan’s names applied to the stratigra-

phy beneath Mexico City are useful in

spcaking about the section.

Considering

the effects of erosion and the occurrence

of hiatuses in the marginal alluvium and

soils, as compared to the more oomplctc

dcpositional record in the central part of

this closed basin, temporal equivalents can

only be sporadic.

In addition, correlations

based on concepts such as pedalfers =

clays, pedocals = calichcs (which in scv-

cral cases

arc high concentrations of

ostracod carapaccs ) seem to be simplistic

representations of complicated and distinc-

tive cnvironmcnts. Mooser ct al. (1956)

pointed out that it is still adventurous to

identify the upper limit of the Tarango

formation bcncath Mexico City. For now,

despite the uscfulncss of Bryan’s forma-

tions, the lacustrine deposits of the Basin

of Mexico should bc characterized in their

own right and not equated with the mar-

ginal deposits.

FOSSIL STUDIES

Major advances in characterizing thcsc

deposits have come from fossil studies,

and foremost of these is the pollcn-strati-

graphic work of Sears (1952) and Sears

and Clisby ( 1955). Thcsc were preceded

by an exploratory study by Decvcy (1944).

Sears and Clisby studied two of Zcc-

vacrt’s cores to depths of more than 70 m.

They attempted palcoclimatic intcrpreta-

tion, but satisfactory pollen zonation is not

possible bccausc the percentage frcquen-

ties oE the major taxa are quite variable

throughout the core. In addition,

pollen

in the coarser scdimcnts is scarce and

poorly prcscrvcd.

Pollen abundance cor-

relates with clay zones and is used in

cnvironmcntal reconstruction.

Maxima in

the amount of oak, fir, and alder pollen

as opposed to pint pollen are considered

to represent warm-moist periods.

Foreman ( 1955)

noteld the prcscncc of

0s racods, sponge spiculcs, and diatoms;

he divides the latter into clongatc and

circular groups but dots not use any of

thcsc fossils for s ratigraphic zonation.





-l’ALEOLIMNOLOGY Ol? LAKE TEXCOCO

185

t

. . . . .

. . . .

i

ashy

. .I..

sand 1

-....<..::::.-------

n

-

-

-

....,

-

-

-

xi--

-

--

-

.- .

. -.

.-.

-.-.

.-.

1:

,_I

-=-

- 7-z.

.-.

, .*. .7.’

.* a,*.:.

*::::

and 8 si

I

---

1-T

._. . fresh and 1

FIG. 3. Correlation of core P 366-2 with the

stratigraphic divisions of Zecvacrt (1952) and

Foreman (1955). Difference in depths results

from the marginal position of P 366-2 relative to

their sections.

Earlier studies of diatoms from the Plcis-

tocenc sediments of Lake Texcoco include

the taxonomic work of Ehrenbcrg (1869)

and Lozano ( 1917) : P. Congcr used din-

toms of the Bccerra formation to intcrprct

the environment of deposition of human

and mammoth remains in the study of

Tcpcxpan Man by dc Terra et al. ( 1949),

DIATOM STUIHES

The prcs’cnt work represents the first

attempt to produce a diatom stratigraphy

in the Basin of Mexico. The coring site

was at the intersection of Paseo dc la

Reforma and Callc Havre, about 4 km S

70” W from the center lof Mexico City

(the Zocalo) and about 2 km N 60” E

of Chapultcpcc IIill ( Fig. 1). The core

(P 366-2)) taken by Zcevacrt in May 1967

in connection with the construction of a

large hotel at this site, was similar to those

studied by Foreman and Scars and Clisby

in 1955 and to the many dcscribcd by

Zccvaert ( 1952, 1953). It was initially 50

m long, but after Zecvaert’s mechanical

analysis of some sections only the upper

35 m were well rcpresentcd, although

there were a few samples from 44 to 46

m available for study.

The coring site is shown in Fig. 2, a

reproduction of the 1550 Alonzo dc Santa

Cruz map of the Basin of Mexico (from

Linnc 1948). The spot cannot bc precisely

indicated, but it is probably very near the

canoeist shown hunting water birds with

a spear at the top ( west) of the map.

Chapultepcc Hill is behind and Ito the left

of the hunter.

The Alonzo de Santa Cruz map shows

the lacustrine cnvironmcnt of the Basin of

Mexico before any attempts were made

to drain the lakes and therefore reprcscnts

a more or less normal state ‘of affairs, al-

though the activities of man (dikes, canals,

hunting, fishing, and so forth) had clearly

modified the environment. What is im-

portant hcrc is to notice the abundance

of aquatic vegetation, the apparent shal-

lowness of the lakes, the placement of

Nctzahuacoyotl’s

dike separating Lake

Mexico from saline Lake Tcxcoco to the

cast, and the presence of springs or other

sources of water at Chapultcpcc Hill.

The coring site is clearly marginal rela-

tivc to the main basin of Lake Texcoco,



186

JOHN I’. BRADBURY

and Zcevaert’s work shows that the forma-

tions beneath Mexico City dip basinward

and so are found at greater depths to the

northeast. The amount of clip is about 2

m/km.

SEDIMENTOLOGY AND CORRELATION

The strata found in the P 366-2 core can

be correlated with those of Zcevacrt and

Foreman only by lithology, bccausc the

depths arc not equivalent. The variability

of lithology makes this difficult, but the

scdimcntologic descriptions and analyses

of Zcevaert (unpublished but on file) in-

dicate that the upper units of his “Ta-

rango” formation arc reprcsentcd, as is the

“Tacubaya” formation. These sediments

are difficult to subdivide, and the likeli-

hood of lateral variation toward the ba-

sin margin further complicates correlation.

Noncthcless, they can be diffcrcntiatcd

from the overlying “Beccrra,” “Barrilaco,”

and “Toltolzingo” formations. The prob-

able correlation is shown in Fig. 3, but I

must stress that I do not feel thcsc names

are justly applied to the deposits beneath

Mexico City.

A detailed sedimcntologic study has not

been made of core P 366-2. The sediments

are generally similar to those of the cores

carefully described by Foreman ( 1955))

being predominantly fine sands, silts,

weathered ash and some clay, and occa-

sional unwcatl1ered ash. They are dia-

granlmatically characterized in Fig. 4 and

compared with Zcevaert’s (unpublished)

water-contcn t analyses. Fossils are abun-

dant; siliceous phytoliths, sponge spiculcs,

and diatoms are common, as are calcareous

ostracod carapaces.

Snails are less com-

mon but found in certain zones, especially

the coarser ones. Pollen is generally abun-

dant in the finer sediments, particularly

the clays and weathered ash, and occasion-

ally seeds and fish bones arc found. The

locations of high concentrations of ostracod

carapaccs and of fish fossils are indicated

in Fig. 4. The fish have been identified

by R. R. Miller and C. Barbour. In some

zones the sediments are penetrated by root

holes, suggesting emergent vegetation in

a shallow lake.

Diatom i?cology

Diatoms were examined at 20-~111 ntcr-

vals where possible throughout the core.

There wcrc 400 diatoms distributed in 12

to 60 taxa counted from each sample; 204

taxa were identified and their frequency

of occurrence calculated. Of those, 88 had

abundancies of at least 257, and their frc-

quencies are plotted in Fig. 4. The species

that had similar depth distributions wcrc

arranged in assemblage groups to facilitate

discussion of the limnological variation

against time;

these groups are identified

ecologically at the top of the figure. Each

spccics plotted is numbered consecutively

to help the reader locate its frequency sil-

houctte when it is mentioned in the text.

An alphabetical listing of all species found

is given in Table 1, with information about

their ecology and modern distribution.

Ideally, fossil diatoms that have similar

frequency distributions with depth consti-

tute fossil assemblages that reflect past

ecological associations. In practice the cf-

fects of reworking and transportation of

diatom frustules can obscure the internal

ecological coherence of fossil associations,

cspccially in shallow lakes, and it is not

always possible to interpret successfully

every elen1,cnt in a fossil assemblage. On

the whole, however, the species of the

diatom assemblage groups do reflect uni-

form ecology, within the rather broad lim-

its of diatom autecology. Some latitude in

ecological uniformity results from the jux-

taposition of similar but not identical dis-

tributions (in the interests of saving space).

Groups I-VI are dominantly freshwater

diatoms, the vast majority being bcnthonic

spccics preferring somewhat alkaline wa-

tcr and tolerant of small amounts of salt.

Group I has two species, Nitzschia tryhZi-

onella ( 1)

and N. hung&a (3), that

Cholnoky (1968) refers to as brackish-

water species, but IIustedt (1930) records

--- --

_- __~_____-

+

lb. 4.



D

A

O

S

R

G

O

L

A

T

M

C

D

F

M

C

A

y

:

J

P

B

a

1

/

F

A

B

C

A

E

P

C

S

E

/

L

m

o

a

O

V

1

N

2

M

1

F

A

B

S

-

W

A

S

E

-

_

/

/

P

C

A

/

/

p

B

C

/

P

l

O

C

-

/

S

L

S

E

C

D

(

8

P

H

O

o

d

w

I

A

Z

o

L

1

D

A

D

Z

(

m

S

G

’

4

N

f

-

-

q

l

k

2

I

I

I

‘

I

I

I

I

’

I

I

I

-

~

-

-

-

-

-

_

’

I

i

9

,

-

[

-

-

_

_

_

I

_

-

I

1

,

I

I

I

I

I

,

I

’

I

I

I

_

-

I

I

I

I

I I

_

,

I

C

s

-

(

2

C

4

-

3

D

-

e

-

3

I

-

-

-

-

-

-

I

L

-

-

-

I

I

;

-

I

l

i

’

I

I

’

-

_

_

-

I

I

I

_

_

’ ’ I

l

i

-

I

-



187

ALEOLIMNOLOGY OF LAKE l?EXCOCO

TABLE

1.

List of diatoms found in the Reforma-Ilavre (P 366-2) core

Ecological characteristics*

Occurrence in Mexico?

s

PII

IC

1 2 3 4 5 6

Achnanthes exigua Grun.

A. Jzauckiana Crun.

A. hungarica ( Grun. ) Grun,

A. lanceoluta ( Brelx ) Crun.

A. marginatula Grun.

A. minutissima Kutz.

Amphiprora alata Kutz.

Amphora acutiuscula Kutz.

A. coffaeiformis salina (W. Sm.) A. Cl.

A. macilenta Greg.

A. ovalis Kutz.

A. ovalis pedictdus Kutz.

A. veneta Kutz.

Anomoeoneis cost&a (Kutz.) IIust.

A. sphaerophora ( Ehr. ) Pfitx.

A. sphaerophora sculpta 0. Mull.

Caloneis hacillum ( Grun. ) Cl.

C. Zewisii influta (Schultzc) P&r.

C. Zimosa ( Kutz. ) Patr.

C. oregonica (Ehr. ) Patr.

C. pemqqxz (J. W. Bail.) Cl.

C. ventricosa subundulata ( Grun. ) Patr.

Cnmpglodiscus clyperu

Ehr.

C. noriczu Ehr.

Chaeloceras Ehr.

Cocconeis diminuta Pant.

C. placentzrlu Ehr.

C. thumensis A. Mayer

CycloteZZa comensis Grun.

C. kutxingiana Thwaites

C. meneghiniana Eaeuissima (v. Coor) IIust.

C. q uillensis Bailey

C. slriatu ( Kutz. ) Grun.

Cyclotella sp. cf. C. stylorum Brightwcll

Cymatopleura solea ( Breb. ) Wm . Sm.

CymbeZZa cistula (Hemp.) Grun.

C. cistula macuZata (Kutz. ) v. Heurck

C. helvetica Kutz.

C. mexicana (Ehr. ) Cl.

C. pusilla Grun.

C. ruttneri IIust.

C. triangulntum (I%. ) Cl.

C. turgida (Greg. ) Cl.

C. ventricosa Kutz.

Denticulu elegans Kutz.

F

8

OS

F 7.2-7.5 0,

F

8

02

B

F-B

F

F-B

B

F-B

F

F-B

B-M

8+

8.2

X X

8.0-8.2

8.0-8.5

x x x x

S-10 N&O:: x x

8.5

X

x x

X

A

X

B 8.6-8.7

B-M

A

X

F

8

F

8+

F 8

X X

F

8

F-B

>87

X X

B

B-M

(3.9-8.6 SS x

X X

F A

F G-9 OS

B A

B

7-8

F

8

X

X

F

8

X

* Ecological characteristics provided for the species plottecl in Fig. 4. S = salinity and F = freshwater, B = breck-

ish water, M = marine wntcr. pH = recorded pH and A = alkaline water, n = acidic.

IC = indicator choractertitics

and SS = stable salinity, AC = aerophil, W = warm water, E = cutrophic, hct = heterotroph, 0 = oligotrophic. This

information is largely from Cholnoky (1968), Patrick and Reimer (1966>, Hustedt (1930), and Bright (in prep.).

t 1 = Texcoco (10,000); 2 = Tlaxcal~ (3,500); 3 = Chalco (2,140); 4 = Tlaxcda (1,130); 5 = Xochimilco

(690); 6 = Zumpnngo (200). Numbers in pnrcnthescs represent conductivity in pmho/cm.



188

JOIIN I?. BRADBURY

TABLE

1. Continued

~-

Ecological characteristics*


s PI-1 IC 1 2 3 4 5 6’

Dintoma heimaZe (Roth) Heib.

Diploneis e2Ziptica ( Kutz. ) Cl.

D. obZonge2la (Naeg. ex Kutz.) Ross

D. palma Cl.

D. pseudovalis IIust.

F-B

s+

D. pueL?a ( Schum. ) Cl.

D. smithii (Breb. ex Wm. Sm.) Cl.

Epithemia hyndmanni Wm. Sm.

E. intermedia Fricke

E. sorex Kutz.

E. turgida (Ehr.) Kutz.

E. zebra (Ehr. ) Kutz.

Eunotia curvata ( Kutz. ) Lagcrcst.

E. diodon Ehr.

15. flexuosa Breb. ex Kutz.

E. glacialis Meist.

E. in&a Wm. Sm. ex Greg.

E. maior (Wm. Sm. ) Rabh.

E. pectin& (0. F. Mull) Rabh.

E. serra diadaema (Ehr.) Patr.

F 7.5-8

F 8.2

X

F-B 8.2-8.5

X x x

F -7.0

X

X

X

X X

Fragiluria brevktriata ( Grun. )

F. capuchina Desm.

F. construens venter (Ehr.) Grun.

F. leptostauron dubia (Grun. ) Hust.

F. pinnata Ehr.

F. vaucheriae ( Kutz. ) Peters

Frustulia rhomboides amphipleuroides (Grun.) Cl.

Gomphonema accuminatum coronata ( Ehr. )

Wm. Sm.

G. angustatum (Kutz. ) Rabh.

G. dubruvi-scense Pant.

G. gracile Ehr.

G. lunceolatum insignis (Greg. ) Cl.

G. longiceps subclavata Grun.

G. pawulum Kutz.

G. sphaerophorum Ehr.

G. tergestinum ( Grun. ) Fricke

G. ventricosum Greg.

Gyrosigma obtusatum (Sulliv. & Wormley) Boyer

G. spenceri (Quek.) Criff. & Henfr.

Ilantxschia amphioxys ( Ehr. ) Grun.

Mastoglwia smithii lacustris Crun.

Melosira granuluta (Ehr. ) Ralfs.

M. italica (Ehr.) Kutz.

M. variant Ag.

NavicuZa accomoda Host.

N. acceptata Hust.

N. agrestis IIust.

N. anglica subs&a (Gl-un.) Cl.

N. capitata hungarica ( Grun. ) Ross

F 7.5-7.8 02

F 7.7-7.8 On

F 7.6-7.7 02

F 7.5-7.7

F 7.2-7.4

X

F 8

x x

\/

F 7-9

01 lack

(;

X

x x

F-B 7.8-8.0 AC

F 7.9-8.2 W-E

F

8

F

8

X

X

X

X




189

TABLE

1. Continued

Ecological chnracteristics*


s PI-1 IC 1 2 3 4 5 6

N.

N.

N.

N.

N.

N.

N.

N.

N.

N.

N.

2

N:

N.

N.

N.

N,

N<

N,

N,

s*

N:

N.

N.

N.

N.

5

N:

N.

N.

N,

N.

2

cwi Ehr.

cinctu (Ehr.) Ralfs

circumtexta Meist. ex IIust.

cocconeiformzk Greg. ex Grcv.

consentanea Hust.

cryptocephala Kutz.

cuspid&a ( Kutz. ) Kutz.

cuspid&z ambigua (Ehr.) Cl.

cuspidata heribaudi Pcrgallo

exigua Greg. ex Grun.

festiva Krasske

fragilarioides Krnsske

graciloides A. Mayer

gregariu Donk.

grimmei Krasske

hn2ophiZa ( Grun. ) Cl.

huef2eri Grun.

huef2eri Zeptocephalu (Breb. ex Crml.) Patr.

laevissima Kutz.

lagerheimi Cl.

lanceoluta ( Ag. ) Kutz.

minima Grun.

minuscula Grun.

muralis Grun.

oblonga Kutz.

peregrinu ( Ehr. ) Kutz.

protracta Grun.

pseudoscutiformis IIust.

pupula rectangularis ( Ck?ff. ) Crun.

p ygmaea Kutz.

radiosa Kutz.

rhynchocephaZn Kutz.

rhynchocephala germ&G ( Wallace) Pntr.

sa2inarum Grun.

semen Ehr. emend. Donk.

seminuloides Hust.

subhamulata Grun.

N. submuralis I-lust.

N. texuna Patr.

N. tripunctata (0. F. Mull.) Bory

Neidium affine ( Ehr.) Pfitz,

N. iridis (Ehr. ) Cl.

Nitzschia acuta Hantz.

N. amph-ibia Crun.

N. amphibioides Hust.

N. angustuta (Wm. Sm.) Crun.

N. capitelluta Hust.

N. clausii Hantz.

N. communis R&h.

N. confinis Hust.

N. denticula Grun.

N. dissipata ( Kutz. ) Crun.

N. epithemioides Grim.

N. filiformis (Wm. Sm.) IIust.

N. fonticolu Grun.

N. frustulum Kutz.

F

F-B

F-B

F

F

8 X

8.3-8.6

X

X

8.3-8.6

758.0

13

A

1’

7.8

17

758.0 0, lack

F-B

A

F

B

F

8

A

X

7.3-7.6

F

F-B

A

6

8.5 O: lack

8

Ns het

8.2-8.5

9-10

7-10 02 lack

xxxxxx

x x

X

X

X

x x

X

X

x x

x x



190

JOHN P. BRADBURY

TABLE 1. Continued

Ecological chnracteristics*

Occurrence in Mexico+

s PEI IC 1 2 3 4 5 6

N. ganderscheimiensis Krasske

N. gracilis Hantz.

N. hantxschiunu Rabh.

N. hungaricu Grun.

N. kutxingiuna Hilse

N. linearis Wm. Sm .

N. microcephala Grun.

N. obtusa Wm. Sm.

N. palea (Kutz. ) Wm. Sm.

N. palea tenuirostris Crun.

N. puleaceu Grun.

Nitzschia sp. aff. N. punctatu (Wm . Sm.) Grun.

N. stugnorum Rabh.

N. sub& Kutz.

N. turds IIust.

N. tryblionella Hantz.

N. tryblionella victor& Grun.

N. vi&u Norman

N. vivax Wm . Sm.

Pinnularia acrosnhoeria Wm . Sm.

P.

P.

P.

P.

P.

P.

P.

P.

P.

uppendiculata ( Ag. ) Cl.

bogotensis ( Grun. ) Cl.

borealis Ehr.

bruunii ( Grun. ) Cl.

braunii amphicephalu (A. Mayer) IIust.

divergentissimn ( Grun. ) Cl.

globiceps Greg.

maior ( Kutz. ) Rabh.

microstauron ( Ehr. ) Cl.

Pleurosigma delicatulum Wm. Sm.

Rhoicosphenia curvnta ( Kutz. ) Grim.

Rhopalodia gibba (Ehr. ) 0. Mull.

R. gibber& murgnritifera Rabh.

R. gibber& protracta Grun.

Scoliopleura peisonis Grun.

Stnur0nei.s acuta Wm. Sm.

S. anceps Ehr.

S. kriegeri Patr.

S. Zapponica A. Cl.

S. legleri Hust.

S. phoenicentron ( Nitz. ) Ehr.

S. smithii Grun.

Stephunodisws niugurne Ehr.

SurireZla angustatu Kutz.

S. ovalis Brcb.

S. ovata pinnata Wm. Sm.

S. peisonis Pant.

S. strintulu Turpin

S. tenera Greg.

Synedra ecus Kutz.

S. rumpens familiaris (Kutz.) Hust.

S. rumpens Scotia Grun.

S. socia Wallace

S. dzn (Nitz.) Ehr.

F-B

F

F

F

F

F

F

F-B

F-B

F

F

F

F

F

F-B

F

B

F-B

H

B-M

F

7-9 0, lack

x

7.5-7.8

7.8

8.3-8.5 Na hct

G-9

O2 lack

x x x x x

7.8-8.2 E

a N2 het

Cl-

8 Ae

6

G-8 0, lack

A 02

7.8

A co,, sot

A

w-o

X

X

5.5-8

A

6.8

A

Na&O:,

A

7.6

X

X

X

x x



PALEOLIMNOLOGY

both as not uncommon in freshwater.

Group II dots not appear ecologically dis-

tinct from group I. Group III consists of

three species of Fra&ria that inhabit

shallow standing water with pH slightly

under 8. Group IV together with group

III form the dominants in several zones of

the core. Group IV contains all freshwa-

ter bcnthonic species, Cocconeis placentula

(21) being an epiphyte; thcsc species fre-

quently inhabit marshes with water of pH

8 ,or higher,

Group V is a large group

composed of minor elements of ‘the frcsh-

water bcnthonic flora; generally their pH

rcquiremcnts arc alkaline. Group VI, al-

though it has a distinctive distribution in

the core, is also composed of alkaline

freshwater bcnthonic species. This group,

characterized by Amphora ovnlis (Se), may

bc somewhat less tolerant of salinity varia-

tions (Hutchinson et al. 1956).

Group VII contains bcnthonic and plank-

tonic species of anomalous ecology; for

example, Anomoeoneis sphaerophora ( 60))

Melosira italica (61)) and Nitzschia com-

munk (63) arc all freshwater species, while

Stauroneis legleri (62) and Nitzschia epi-

themioides (64) are brackish-water spccics.

This may result from reworking and the

introduction of dead frustulcs from other

habitats, or possibly s,ome of the species

have a wider ecologic amplitude than pre-

viously suspcctcd.

Group VIII is composed largely of ben-

thonic brackish-water species. The pres-

ence of Amphora coffaeiformis salina (66),

Naviculu halophila ( 65)) and Cymbella

pusilla (68) suggest very saline conditions

(Hustcdt 1930).

Group IX is composed of a single spc-

ties, Nitzschia frustulum (70); it is a

brackish-water species ( Cholnoky 1968))

but it is found in a wide range of cnviron-

mcnts and appears to be tolerant of fluc-

tuating conditions. Some of its variability

may be the result of misidentification or

confusion with similar spccics, but reli-

able systcmatists have reported it in frcsh-

water environments (cg., Patrick ot al.

1967).

Group X contains brackish bcnthonic di-

OE’ LAKE TEXCOCO

191

atoms, two of which, Anomoeoneis costatn

(72)

and Surirellu peisonis (74), live in

water with high concentrations of sodium

carbonate ( Cholnoky 1968).

Group XI is composed of brackish-water

pkanktonic diatoms that do not tolerate

variations in salinity. It is ‘dominated by

Cyclotella striatn (79) and Chaetoceras sp.

(77) although locally CycZoteZZa quillensis

(80). and CycZoteZZa sp. cf. C. stylorum

( 81) are imbortant. Most authors rccog-

nize the close relationship bctwccn C. sty-

lorum, C. quillensis, and C. striuta (Boyer

1927; Hustcdt 1962) and their distribution

throughout the core suggests that they are

variations of the same thing, possibly eco-

types. The same possibly holds true for

Cyclotella meneghiniana laevissimn ( 78 )

(Hustcdt 1962).

Group XII contains alkaline frcshwatcr

diatoms, the most abundant being Melo-

sira gmnulata (83), a planktonic form.

The rest arc either cpiphytic or benthic

species.

A brief survey of the diatoms in modern

lacustrine cnvironmcnts in the Basin of

Mexico and nearby arcas was made to

serve as an ecological framework for in-

tcrprcting the diatoms in the core. Sam-

plcs were collected from the lakes and

ponds idcnitificd in Table 1. About 50% of

the common spccics from the core wcrc

also common elcmcnts of thcsc aquatic cn-

vironmcnts. Of the species in the frcshwa-

tcr environments (Tlaxcala 1,130, Chalco,

Xochimilco, and Zumpango) 82% bclongcd

to the freshwater groups I-VII from the

core, and 47% of the spccics in Lake Tcx-

coca and the saline Tlaxcala pond were

found in the brackish-water groups VIII-

XI. This was a small sample, and little

definitive ecologic

information can bc

gained from it,

but it indicates that a

subst‘antial fraction of the common diatom

flora of ancient Lake Tcxcoco can bc

found today in the Basin of Mexico, and

that the paleoccologist need not seek

vastly diffcrcnt environments from those

existing today to explain many of the flo-

ristic changes in the Plcistoccnc sediments

of the basin.



192

JOHN I?. BRADBURY

TABLE 2. The zones delineutecl in the core ancl their dominant diatoms (set also Fig. 4)

Z~;taha

Diatoms

Group

Ecology

1.

2-3 m

2.

3-4 in

3.

4-11 m

4.

11-14 111

5.

14--16 111

6.

16.-19 m

7.

19-20 m

8.

20-23 m

2::

25.10 m

10.

25.10-

27 m

DENTZCULA ELEGANS (22)

Cocconeis placentuln ( 21)

Nitzschia amphibiu ( 23 )

Rhopalodia gibberula

mnrgnritifern ( 24 )

Navicula cryptocephuln (20)

Fragilaria brevistrinta ( 19 )

Amphora veneta ( 8)

Amphora ovalis ( 56)

FAAGZLARZA BREVZSTRZATA (19)

Fragilnria construens venter ( 17)

Anomoeoneis costata ( 72 )

Campylodiscus clypeus (73)

Szcrirelln peisonis ( 74 )

Nitzschia frustulum ( 70 )

Anomoeoneis sphnerophorn ( 60)

Chaetocerns sp. ( 77 )

NITZSCHIA FRUSTULUM (70)

Cnmpylodiscus clypeus (73)

Anomoeonds costata ( 72 )

Rhopalodia gibber&z protracta ( 71)

Navicula hdophiln ( 65 )

Chaetocerns sp. ( 77 )

Cyclotello striata ( 79 )

CYCLOTELLA STRZATA (79)

Chaetoceras sp. (77)


Nitzschia communis ( 63 )


Cnmpylodiscus clypeus ( 73 )

Rhopnloclia gibber& protmctcl (71)

AMPHORA OVALZS (56)

Epithemin zebra ( 59 )

Fragikrin construens venter ( 17)


Campylodiscus clypeus ( 73 )


Rhopnlodin gibber& protracta ( 7 I)

Surirella peisonis ( 74 )

Cocconeis diminuta ( 75 )

Chaetoceras sp. (77)

Cyclotelln strinta (79 )

CYCLOTELLA STRIATA (79)

Chactocerns sp. ( 77 )

Anomoeoneis cost&a ( 72 )

Cnmpylodiscus clypeus ( 73 )


Navicula halophila ( 65 )

NAVICULA ZIALOPHILA (65)

Cyclotelln striata ( 79 )

Nitzschia frustulum (70’)

CYCLOTELLA QUZLLENSZS (80)

Cyclotella striintn ( 79 )

Cyclotella sp. cf. C. stylorum (81)

Anomoeoneis costata (72 )


Fragilnria construens venter ( 17)

FRAGILARIA BREVISTRZATA (19)

FTagilaria construens venter ( 17)

Fragilnriu pinnatn ( 18)

Rhopnlodia gibbn ( 15)

Nitzschia amphibia ( 23 )

Cocconeis placsntu lu ( 21)

Cyclotella striata (79)

NAVICULA HALOPHZLA ( 65 )

Nitzschia frustulum (70)

Anomoeoneis cost&n ( 72 )

Cyclotelln striata (79)

Cyclotella meneghiniana laevissima ( 78 )

Chaetoceras sp. ( 77 )

IV

IV

IV

IV

IV

III

II

VI

III

111

X

X

X

1X

VIII

XI

IX

X

X

~111

XI

XI

XI

XI

IX

VIII

X

X

X

VI

VI

III

IX

X

X

X

X

X

XJ

XI

2

X

X

XI

VIII

VT11

XI

IX

XI

XI

XI

X

X

TTI

IT1

III

TII

II

IV

IV

XI

VIII

IX

X

XI

XI

XI

Epiphytic and benthonic diatoms characteristic

of freshwater of high pH; most tolerate low

salinity.

Frngilaria spp. are commonly found in shallow

standing water and arc tolerant of a wide

range of salinity; the remaining species arc

found in brackish-water benthonic and plank-

tonic cnvironmcnts.

All forms are characteristic of brackish water;

C. strinta and Chaetoceras sp. are planktonic,

the others bcnthonic.

Flora similar to the preceding zone, but it shows

a dominance of planktonic species C. striatcl

and Chnetocclns ~1).

Flora is similar to the last two zones, but pres-

encc of A. utdis, F. construens venter, and

E. zebra suggests fresher water than the pre-

ceding floras.

Dominance of brackish planktouic (C. strinta

and Chaetoceras sp.), the remainder being

brackish bcnthonic.

Brackish-benthonic diatoms dominate.

Brackish planktonic dominate; presence of F.

construens venter suggests presence of fresh-

water ncarby.

Many diatoms from the freshwater groups plus

the Fmgilarin spp. indicate shallow, fresh,

alkaline water.

Bl;lckisll-l)cntllonic diatoms dominate.




193

TABLE

2. Continued

Zone and

depth

Diatoms

GrOUp

Ecology

11.

27-

27.75 m

12.

27.75-

29.50 m

13.

2x50-

33.50 111

14.

33.50-

35 m

15.

35 m-?

44-46 in

CYCLOTELLA

SP. CF.

C. STYLORUM

Campylodisczhs clypew ( 73 )


Rhopaloclia gibberula protracta (7 1)

CAMPYLODISCUS CLYPEUS (73)


Rhopalodia gibberala protracta (71)


Surirella peisonis ( 74 )

Cymbella mozicana (76)

Cyclotella striata (79)

DENTICULA ELEGANS (22)

Cocconeis placentula ( 21)

Rhopalodia gibberula margaritifera (24)

Nitzschia amphibia (23 )

Navicula cryptocephala ( 20 )


MELOSIRA GRANULATA ( 83 )

Cyclotella sp. (81)

FragiZaria brcvistriata ( 19 )

STEPIIANODISCUS NIAGARAE

(81)

$

X

X

X

X

X

X

x”

XI

IV

IV

IV

IV

IV

IX

XII

XI

III

Cyclotella

dominates,

and the remaining flora is

from brackish-bcnthonic habitats; possibly this

zone is a subzone of the following one.

Brackish-benthonic diatoms dominate.

l?reshwater, alkaline benthonic and epiphytic di-

atoms; many other spp. from groups I-V.

Planktonic and shallow, freshwater diatoms from

alkaline, warm, cutrophic lakes.

Cool planktonic freshwater diatom, characteristic

of deep, north-tcmperatc lakes.

[This zone has been determined from the core Bellas Artes 80, which was originally studied by Sears and

Clisby ( 1955) and Foreman ( 1955).

Only scattcrcd samples of this core exist, and the extent of this

zone is unknown.]

DENTICULA ELEGANS (22)

Cocconeis placentzila (21)

Nitzschia amphibia ( 23 )

R.

gibberula

margnritifera ( 24 )

IIantzschia amphiorys ( 25 )

Rhopalodia gibba ( 15)

IV

IV

IV

Freshwater, alkaline bcnthonic diatoms; II. am-

phioxys is an aerophilic spccics.

IV

V

11

(The limits of this zone are not fixed because it is separated from the other zones by several meters of scdi-

ments that may cont<ain more than one diatom asscmblagc.

For this reason it is not number&)

In addition to tho zones listed above, a diatom assemblage was noted in a sample from ~70 m in the Belhas Artes

80 core. Its equivalent probably exists beneath the depth reached by P 366-2. This zone, like the one at 35 m, is

dominated by Stephanodiscus niagarae.

Diatom Zonation

Distinctive assemblages of diatoms were

used to delineate 15 zones in the core

(Table 2). They have been numbered in

Fig. 4 and labeled with the name of the

diatom considered most charac tcris tic of

each zone.

A discussion of the paleolimnology of

Lake Texcoco must be prcccdcd by a con-

sideration of the mechanics of lacustrine

change in this shallow lake and how the

marginal diatom floras arc affected. As

those of many lakes in semiarid regions

with periodic rainfall, its level and salinity

fluctuate widely from season to season,

sometimes in rcsponsc to single storms.

In 1629 the waters of Lake Texcoco rose

8 m and submcrgcd a town for 5 consccu-

tive years (Bonaparte ct al., ca. 1900).

Such floods impose markedly differing

habitats to which the algae rapidly adjust.

During periods of mcagcr rainfall the

lakes evaporate and the water level rc-

treats from the short to lcavc mudflats

and salinc pools. In the center of the basin

a shallow saline lake can remain.

This simplified picture of lake-level fluc-

tuati.on is complicated by the fact that the

arca around Lake Tcxcoco, especially in

the area of the coring site and to the

southwest, has long been known for the

abundance of pcrcnnial freshwater seeps

and springs,

whose water supported the

prosperous chinampa agriculture in this

part of the basin (Cot 1964). Thus,

during the low-water stages it appears that



194

JOIIN I?. BRADBURY

the marginal, marshy, frcshwa ter environ-

ments supported by the springs extended

basinward and were rcpcatcdly invaded

by transitory floods of saline water from

the central basin of Lake Texcoco. Rc-

duccd spring flow, to be cxpcctcd during

long dry periods, can add another variable

to this picture.

The validity ,of this scheme is supported

by historic documentation of floods that

plagued the marginal marshland chinnmpa

agriculture of the area. For example, the

King of Texcoco, Netzahuacoyotl, built a

dike in 1450 to prevent the saline waters

of the main basin from entering the west-

ern subbasin of the lake ( Fig. 1).

The diatoms from the Pleistocene sedi-

mcnts of Lake Texcoco lbclong to four

broad ecological groups : brackish bcn-

thonic, brackish planktonic, freshwater

bcnthonic-cpiphytic, and freshwater plank-

tonic.

These groups and the lacustrine

environments that produced them arc

shown in Fig. 5, arranged in an idealized

climatic scqucncc.

Actually, the changes

between freshwater marsh, shallow saline

water, and deep saline water can occur

rapidly and can bc revcrscd. The fresh-

water planktonic diatoms represent more

stable lacustrine environments.

Paleolimnology

The earliest recorded diatom assemblage

comes from a depth of about 70 m in the

Bellas Artes 80 core studied by Foreman

( 1955) and Scars and Clisby ( 1955). It

is not shown in Fig. 4. It is dominated

by Stephanodiscus niagarae, a freshwater-

planktonic diatom, indicating that a large,

possibly deep, cool lake cxistcd at that

time. The salinity of the lake was low and

constant, and the lake probably had an

outlet. More than 20 m of sediment may

separate this zone from the next known

assemblage; it is not known what limno-

logic conditions they represent.

The lowest diatom assemblage from core

P 366-2 (4445

m, Fig. 4) is unnumbered

because it is separated from the overlying

sediments by 9 m.

It consists of marsh

diatoms,

principally Denticula elegans

FIG.

5. Schematic representation of the four

distinct lacustrine environments of Lake Texcoco

suggested by diatom analysis of core 1’ 366-2.

At the highest levels (bottom sketch), when the

lake may have overflowed, freshwater-planktonic

(fw-p) diatoms occurred, Decrease in the water

level and

increased salinities products brackish-

planktonic diatoms (br-p ). As water levels con-

tinue to drop brackish-benthonic diatoms ( br-b )

flourish in shallow, saline water. During the low-

est lake levels, spring-fed freshwater marshes cx-

tencl basinwarcl from the margins of the lake.

Associated with the marshes ca.re freshwatcr-bcn-

thonic ( fw-b ) cliatoms, and brackish-benthonic

forms arc present in the remnant saline pools in

the center of the basin. These assemblages can

be tentatively related to a climatic change from

moist to dry.

(22), and indicates that at that time the

lake was very shallow. It is not known

whether the disappearance of the earlier

large cool lake conditions resulted from

drainage or from desiccation.

This marshy environment ultimately was

followed by a return of the large lake, as

shown by zone 15 which is dominated by

S niagarae,

but these conditions wcrc

short-lived and it evolved into a shallower,

more eutrophic and possibly warmer lake

characterized by a M. granulata (83) as-

scmblagc (zone 14). Zone 13 shows an-

other return of marsh diatoms dominated

by D. elegans (22). Probably this time

the lake level fell by desiccation, and only

saline brines and salt flats occurred in the



PALEOLIMNOLOGY OF LAKE TE,XCOCO

195

central portions of the basin, where the

minimum lake levels have been rccordcd

in modern times ( Fig. 1). This is sug-

gested by the succcoding zones (12 to 3)

which have predominantly brackish-water

diatoms.

With the return of moister climate the

lake began to fill, and floods of salinc

water extended toward the margins of the

lake basin, Variable water depths are

suggested by the alternation of brackish-

benthonic and brackish-planktonic assem-

blages (zones 12, 11, and lo), followed

in zone 9 by a slight increase in FragiZari~~

spp. (group III) and other diatoms in

groups II, IV, and V, which indicate a

brief period of shallower but fresher wa-

tcr. The brackish lake continued to domi-

nate however. Zones 8, 6, and 4 reflect

deeper brackish water supporting brack-

ish-planktonic diatoms as C. quillensis (80)

and C. striate ( 79)) and zones 7, 5, and 3

reflect shallower water with brackish-bcn-

thonic diatoms.

The artificial zone boundaries (zones

11 to 3) obscure the complex variations

the lake undcrwcnt during the long period

reprcsen ted.

A somewhat better idea can

bc gained by following the fluctuations of

the frequency curves of the ,diatoms in

groups VIII-XI, Both the brackish-plank-

tonic diatoms of group XI and the very

saline-benthonic diatoms of group VIII in-

crcasc rapidly to high lcvcls and then fall

suddenly. Group IX, containing only the

diatom N. frustulum (70)) is important

throughout this interval, but cvcn with its

wide tolcrancc its abundance is highly

variable. The same is true to a lcsscr

extent for those forms of group X.

A note of caution may bc introduced

hcrc regarding the frcqucncy disltribution

of CycZotelZa sp. cf. C. stylorum (81). Its

highly erratic distribution at depths oE

23 and 27 m may bc the result of rcwork-

ing from lacustrine deposits left by the

shrinking lake after zone 14. Here it is

associated with M. gmnuZnta (83), and

the same association can bc found in a

fossil diatomitc surrounding the shores of

Lake Amatitlan in Guatemala. Thcsc as-

sociations,

incidentally, suggest that the

CycZoteZZu sp. in question is not ccologi-

tally rclatcd to C. stylorum, a marine spc-

tics (Hustcdt 1962), even though they

are similar morphologically, but shares the

alkaline, warm, cutrophic habitat of plank-

tonic M. granulutn. Its anomalous associ-

ations with group X diatoms at 27 m and

with C. striata (79) at 23 m, plus the fact

that at these levels it is mostly in a broken

condition, support the conclusion that its

appearance results from reworking. The

question cannot be scttlcd, howcvcr, until

adequate taxonomic work is done with

this species. This problem does not sig-

nificantly alter the paleolimnologic history

presented in the preceding paragraph.

Above zone 3, the sequence indicates

that lake levels fell, and frcshwatcr marsh

conditions again cxtcndcd basinward. In

zone 2 the FragiZaria spp. (17, 18, 19) in-

dicatc shallow stan,ding water of slightly

alkaline pH.

In zone 1, the occurrcncc

oE D. elegans (22) with Ilnntxschia amphi-

oxys (25), an acrophilic diatom, suggests

that the water level may have been cvcn

shallower.

In general, the coarser-textured parts of

the core, as indicated by low values in

Zccvacrt’s water-content curves ( Fig. 4))

contain diatom floras that belong either lto

the freshwater bcnthonic groups I-VII or

to the very salinc bcnthonic group VIII,

while the finer sediments correlate in most

instances with planktonic diatoms ( group

XI and XII) or with brackish-bcnthonic

diatoms of groups IX and X. Coarser sedi-

mcnts would be cxpcctcd in tither shal-

low-water environment, whether fresh or

salinc, while a dcepcr lake would favor

the accumulation of fine scdimcnt. The

coarser sediments also frequently have

root holes, which, however, are not con-

fined to these materials. There is a lim-

itcd ncgativc correspondence bc twccn the

number of species found in each sample

(Fig. 4)

and Zccvacrt’s water-content

curve.

Thcsc observations, although not all

quantified,

offer additional paleolimno-

logic information.

The freshwater marshes



196

JOIIN I?. BRADBURY

can be expected to contain more habitats

and hence more species, and the presence

of root holes helps to substantiate their

existence. Coarser sediment would tend

to accumulate in thcsc marshes if they

drained slowly basinward.

Correlations

‘Until additional cores arc studied, the

question of intrabasin correlation of the

diatom zones ccannot be definitively an-

swered. Zcevaert’s (1952, 1953) work

shows good correlation of his sedimento-

logic zones from core to core, and I suspect

that the diatom stratigraphy presented

will hold underneath all of Mexico City.

Probably toward the center of the basin,

where limnologic

conditions were fre-

quently very saline, the freshwater marsh

zones will not be present. The Stephano-

discus zone ( about 35+ m) has been

found more than 23 km to the northeast

in wells ,of the saline evaporation plant,

Sosa Texcoco, and there is some evidence

that the brackish-planktonic diatom zones

have similar distributions. The diatoms

found associated with the skeletal remains

of Tepexpan Man (P. Conger in de Terra

ct al. 1949)

might fit in any number of

zones beneath zone 2, and a longor pro-

file is required on ,the northeast side of

the basin before correlations can be made.

Zones 1 and 2 do seem to have local extent

around Mexico City and have tulncd up

in excavations for the new subway there.

To the southeast and into the basins of

Xochimilco and Chalco, zones 1 and 2 may

bc expected to increase in thickness, since

these chinampa areas are supplied by an

abundance of freshwater springs. A diat-

omite containing a dominance of Fragi-

Znria spp. occurs at a depth of 1.75 m at

Culhuacan in the Xochimilco basin (a

sample studied from Scars’ 1952 study),

and at Tlapacoya ( Fig. 1) the same spc-

tics and D. elegans arc found in sediments

that date from 24,000 years B.P. (Brad-

bury 1970n).

Fmgilarb and Denticulu

altcrnatc throughout the upper 6 m of the

Tlapacoya sections, but below this, in sedi-

ments dated at 24,060 to 35,000 years BP.

( Mooser 1967), diatoms of group X arc

common :

Campylodiscus clypeus, A. cos-

tata, and N. frustulum (Bradbury 1970a).

The radiocarbon dates on core P 366-2

(Fig. 4) place

the boundary between

zones 2 and 3 at a considerably later date

than this (assuming that the change found

at Tlapacoya is the same kind of tran-

sition), so we are faced here with the

likelihood that some ‘of thcsc zones have

a considcrablc time transgression. This is

what might be expected in Lake Chalco,

which is not only separate from and higher

than Lake Texcoco but also has an ample

supply of freshwater.

Another possibility exists that the gla-

cial substagcs that White (1962) cstab-

lishcd for the volcano Iztaccihuatl may bc

embraced by the ages of core P 366-2.

It is conceivable that the last three plank-

tonic diatom zones (zones 8, 6, and 4)

relate to glacial advances on the sides of

this volcano, but thcsc zones do not reflect

particularly dramatic climatic changes be-

cause they all consist of brackish-plank-

tonic diatoms. On the other hand, markod

changes in precipitation and tempcraturc

may not be needed to extend the glaciers

now existing on Popocatepetl and Iztacd-

huatl to lower elevations ( White 1954).

Probably the stratigraphy of Lake Chalco

is more sensitive to the glacial fluctuations

on these peaks, and correlation must wait

until they arc studiod in greater detail.

The age of the planktonic peaks in core

P 366-2 is 30,000 years or more according

to the radiocarbon dates, and therefore

they appear

rather early for White’s

(1962) tentative correlations of his gla-

cial advances with the Rocky Mountain

sequcncc.

Diatom, pollen, and chemical studies by

Hutchinson et al. (1956) on cores from

Lake Patzcuaro in Michoacan show a

ma&cd dry period beginning at depths

FIG.

6.

-+



r'Kl\'r\'/

-

--- ---

_--

-yAYAY-- 36 -

1 I

--d-m :

____ -

_---.

----

L.-z--:-40-

_---.

____

._--. -

_---

.---.

_---

--- 42

-I:&

__-_ -

._--.

__ __

'=---z-r- 4 4 -

.-_-.

_-__

_---. _

----

.---.

---_

---m---z-46-

.---.

_--_

_--_. _

.:.: ,'-5+

"Ai.: _

'. .:'

.

'::.::: -64-

-

. .

::::::: -66-

PALEOLIMNOLOGY Ol? LAKE TEXCOCO

-

: :

_---.

---_

-----* -

: : :

__--.

____

: : :

_---.

-___ -

: :

.---.

____

: : :

---

:------i- 40 -

----

.---_

____ -

-----

----

--~-~-~: - ,

2

____

---.

-_- -

---_

___

---.

\aaar-lrl-

____

____

____ -

_---_

____

--;-I-;<- ,6 _

-_ __.__-.

____ -

I

__-.

____

"""I-48-

:

__

____

_---.

____ -

.___.

__ __

; :

--------20-

k

+-III: -

._--.

~L~-~-~:

22

____

___.

___

_-_.

_ _

:~-~-~: - 24 -

----

.____

____ -

.---.

_-_-

----w-:-26-

.-_-.

____

.----

____ -

._--_

____

.__-.

>nr>nr-26-

197

POLLEN DIAGRAM

MADERO CORE

MEXICO CITY,

MEXICO

Percentages of total pollen

(Clisby & Sears i955)

-- .

Explanation

Weathered ash and clay

jzq

Fresh ash

m

Ostracod marl

Sand and silt

No pollen

+

Pollen types in counts

less than 90



198

JOHN I?. BRADBURY

around 6 m, and it is reasonable that this

dry period should be correlated with the

falling lake levels in the Basin of Mexico

indicated by zones 1 and 2. The Patzcuaro

material was not dated, and it was only

tentatively correlated with Scars’ ( 1952)

archaeologic scqucnccs, which implies that

the dry period was much later (say about

2,300 years B.P. ) than the transition in

Lake Texcoco ( about 6,000-10,000 years

1s.P.). Further studies in both lakes arc

nccdcd for conclusions about regional

correlation.

Of all possibilities for correlation, Sears

and Clisby’s (1955) pollen studies in simi-

lar cores from the sediments of Lake

Texcoco should prove the most rewarding.

Unfortunately, obvious correlations for

most stratigraphic levels have not been

found, although one would expect that the

pollen types labeled as moisture indicators

(Ahus, Quercus, and Abies) would corrc-

spond to peaks in planktonic diatoms. All

that can bc said for the time being is that

both parameters are highly variable and

that until pollen and diatoms are studied

from the same core the matter will remain

unsolved. All in all, the pollen from thcsc

sediments shows a somewhat monotonous

stratigraphic distribution when it is plot-

ted as pcrcentagcs of total pollen rather

than rclativc proportions of arborcal pollen

alone (Fig. 6). On this basis, the most

evident break in Sears and Clisby’s (1955)

counts is found in the upper 6 m, where

Gramincac, Amaranthaceae, Compositac,

and Zea bccomc very abundant as opposed

to arboreal pollen. This corresponds rather

well to the boundary between zones 3

and 2, suggesting a lowering of lake lev-

cls. This change in the pollen stratigraphy

seems clearly to bc the handiwork of agri-

cultural man in the Basin of Mexico

(Clisby and Scars 1955).

Most of Scars’ (1952) pollen work in

archaeologic horizons of the basin involves

sediments younger than those of core P

366-2, but some of his dccpcr profiles (Chi-

malhuacan and Chapingo ) may cover the

same time as the upper levels of this core.

His dry phase, zone D, may correspond

to some part of diatom zones 2 or 1 of

this study. Sears’ zone C, representing a

wet interval presumably corresponding to

the Zacatcnco high beach level ( 2,242 m )

of 3,500 years ago, is ltoo young to be

covered within the time span of core P

366-2. At any rate, it appears that the

elevation of this coring site was somc-

what higher than the 2,242-m level, so the

marshes that existed #there at that time

may not have becn flooded then.

Limnologic changes that appear in the

Basin of Mexico since about 6,000 years

RP., particularly in marginal areas of the

basin, arc not necessarily solely the result

of natural climatic or hydrologic changes,

because from this time on agricultural man

played an increasingly important role in

the area. It is not too presumptive to

speculate that early agriculturalists took

advantage of the favorable conditions on

the spring-fed shores of Lake Tcxcoco and

possibly initiated primitive systems for

farming those marshes ,that provided land

dry enough for planting and water shallow

enough for growth.

An attractive model

for such a system is the ridged fields

(Parsons and Denevan 1967) that pre-

Columbian agriculturalists used in many

parts of Mexico and Central and South

America. Thcsc consist of mounds or

ridges of earth built above the shallow-

water levels of flooded, low ground. The

ridges provide rich, aerated soil next to

abundant water that offers cxcellcnt con-

ditions for farming and at the same time

provides a certain amount of protein in

the form of fish, turtles, water-fowl, and

so forth. The construction of a ridged

field requires nothing more complicated

than a cligging stick to heap the marsh

sediments above water lcvcl, but even a

primitive system of ridges and canals and

possibly dikes would vastly alter a natural

limnologic environment.

The diatom evidence from Lake Tcx-

coca cannot prove these speculations indc-

pcndently. The abundance of Zea pollen

at equivalent levels in Scars and Clisby’s

( 1955) cores is supporting cviclcnce, and




199

the likelihood that the remarkable chi-

nampa agricultural system evolved from

the similar but more primitive ridged field

system seems logical. Archaeological cvi-

dencc to test these ideas might be obtained

near Mexico City or in the remaining chi-

nampa areas to the southwest. Possibly

some of it is covcrcd by the Rcccnt ba-

salt flows from Xitle and other volcanic

centers.

Although the diatoms from core P 366-

2 may be only suggestive with rcspcct to

the questions ,of early man and agriculture

in the Basin of Mexico, they offer much

more positive information on carlicr geo-

logic questions. Mooscr ct al. (1956)

discussed the difficulties in assigning max-

imum ages to the lake deposits and to the

formation of the basin, although thcsc

features are generally considered to be

Plio-Pleistoccnc.

This problem rests on

the identification of the Tarango formation

bcncath Mexico City. Where the Tarango

formation is developed on the margins of

the basin it predates the basaltic eruptions

that are #thought to have blocked the

south-draining v‘alley to form the Basin of

Mexico. Previous volcanic rocks were

nonbasaltic. Thus the Tarango formation

should bc an alluvial deposit free of ba-

saltic components. The prcscncc of deep

strata within this formation containing

abundant remains of S. niagame, a frcsh-

water-planktonic diatom characteristic of

large, cool tempcratc lakes in North Amcr-

ica, clearly suggests that thcsc deposits

are not entirely alluvial. Also, Foreman

(1955) found basalt fragments at similar

depths in the same formation.

Fish fo,ssils (Fig. 4) from core P 366-2

have been identified by R. R. Miller and

C. D. Barbour (pcrs’onal communication)

as Chirostoma humboltltianum, belonging

chiefly to the Lerma drainage basin, The

present fish fauna of the Basin of Mexico

is entirely of Lcrma affinities (Arellano

1953; Meek 1904). The carlicst fish fossil

from the core occurs at a depth of 35 m;

it indicates that at least since perhaps

100,000 years B.P. the Basin of Mexico

somctimcs contained a large lake, which

evidently drained northward to the Lerma

system.

The planktonic diatoms and fish fossils

from these deep levels support the conten-

tion of Mooscr ct al. (1956) that the for-

mation currently identified as Tarango

beneath Mexico City (Zccvaert 1952, 1953)

is not the same thing as Bryan’s (1948)

Tarango formation on the margins of the

basin. This change considerably lcsscns

the inferred age of these sediments as

Plioccnc ( Zccvaert 1953)) but it appears

that they represent more time than the

Wisconsin alone (Scars and Clisby 1955),

Perhaps the most important conclusion

that can bc drawn from the diatom stratig-

raphy of these deposits is that climatic

changes known to have occurred in more

northerly latitudes do not seem to have

been reflected in this region of Mexico.

Admittedly, lakes are sometimes not such

sensitive climatic indicators as we might

wish, but the remarkable pcrsistencc of

many diatom species throughout this long

time and their common distribution in the

basin today clearly suggest that truly plu-

vial climates did not reach this area. If

anything, the relatively small frequencies

of planktonic diatoms in the upper part

of the core suggest that the lake was shal-

low and that the climate was thcrcforc

arid during the late Plcistoccnc. This is

in agrecmcnt with the implications from

the late Pleistocene-early pos t-Pleistoccnc

vertebrate faunas of the Tchuacan Valley

200 km cast-southcast of Lake Tcxcoco.

Today the Tchuacan Valley has a hot,

semitropical thorn-scrub forest and a

small-mammal fauna charactcrizcd by the

cotton rat (Sigmodon) and the kangaroo

rat ( Dipodomys) . The fauna of the late

Pleistoccnc and early post-Pleistoccnc in

this area lacks these animals entirely but

has high percentages of small mammals

that charactcrizc the arid and seasonally

cooler interior plains of northern Mexico.

If this kind of environment prevailed in

the Tchuacan Valley arca (Flanncry 1967),

it seems reasonable to suppose that the

nearby Basin of Mexico was also drier at

that time than it is today.



200

JOHN I?.

REFERENCES

ARELLANO,

A. R. V. 1953,. Estratigrafia de la

Cuenca de Mexico.

Mem. Congr. Cient,

Mex. (Conmem.

400th Anniv. Univ. ) 3 :

172-186.

BONAPARTE, PHINCE ROLAND AND OTHERS, ~8.

1900.

In Le Mcxique au debut du XX”

siecle, p. 49-53.

Lib. Ch. Dclagrave, 15,

Paris.

BOYF.R, C S.

1927. Synopsis of North Ameri-

can Diatomaceac. Proc. Acad. Natur. Sc i.

Phila. 78( suppl. ) : 228 p.

BRADRURY, J. P.

1970a. Paleoecology of Tla-

pacoya, Mexico :

Evidence from diatoms.

(Abs.) Special Session: Excavaciones en Tla-

pacoya, E&ado de Mexico, Mexico. Proc.

Annu. Meeting Sot. Amer. Archaeol., 35th,

May, 1970.

-* 1970b. Diatoms from Plcistocenc sedi-

ments of Lake Texcoco, Mexico. Rev. Ge-

ogr. Phys. Geol. Dyn. 12( 2,) : 161-168..

BRYAN, K. 1948. Los suelos complejos y fosiles

de la altiplanicie de Mexico en relation a 10s

cambios climaticos. Bol. Sot. Geol. Mex. 13:

l-20.

CHOLNOKY, B. J.

1968. Die Ockdogie der Di-

atomecn in Binnengewassem. Cramer. 699 p.

CLISBY, K. H., AND P. B. SIRS. 1955. Paly-

nology in southern North America, Part 3.

Bull. Geol. Sot. Amer. 66 : 511-520.

COE, M. D. 1964. The chinampas of Mexico.

Sci. Amer. 277: 90-98.

DEEVEY, E. S. 1944. Pollen analysis and Mex-

ican ,a.rchaeology.

An attempt to apply the

method. Amer. Antiq. 10: 135-149.

-.

1957. Limnologic studies in Middle

America with a chapter on Aztec limnology.

Trans. Conn. Acad. Arts Sci. 39: 213-328.

DE

TERRA, II.,

J. ROMERO, AND T. D. STEWART.

1949. Tepexpan man. N.Y. Viking Fund

Publ. Anthro. 11, p. l-160.

EHRENBERG,C. G.

1869.

Ober machtige gc-

birgschichten verherrschend aus mikroskop-

ischon Bacillaries unter und bei der Stadt

Mexiko. Abh. Kgl, Acad. Wiss., p. 47, Pl.

1 and 2.

FIGUEROA,G. E., C. J. SCIILAEPFER,AND L. ZEE-

VAERT.

1968. An outline of the general

geology and problems of hydra- and cngi-

ncering geology of the Mexico City area.

Geol. Sot. Amer. Annu. Meeting Nov 1968,

Field Trip Guidebook, Mexico City. 22 p.

FLANNERY, K. V.

1967. Vertebrate fauna and

hunting patterns, Chap. 8. In D. S. Bycrs

[ed.], The prehistory of the Tehuacan Valley,

v. 1. Univ. Texas Press.

FOREMAN, F. 1955.

Palynology in southern

North America, Part 2. Bull. Geol.

SOC.

Amer. 66: 475509.

IIU~TEDT, F. 1930. Die Susswasscr-flora Mittcl-

europas,

Part 10. Jena. 466 p.

BRADBURY

-.

1962. Die Kieselalgen Deutschlands

Osterreichs und der Schweiz mit Beriich-

sichtigung der ubrigen Lander Europas so-

wie der aungrezenden Meercsgebiet. Zn L.

Rabenhorst [ea.], Kryptogammen-Flora von

Deutschland, Osterreich und der Schwciz, v.

7,

Sect. 1. 920 p.

HUTCIIINSON, G. E., R. PATRICK, AND E. S.

DEEVEY.

1956. Sediments OF Lake Patz-

cuaro, Michoacan, Mexico. Bull. Geol. Sot.

Amer. 67 : 1491-1504.

LINNE,

S. 1948. El Valle y la Ciudad de Me-

xico en 1550. Stockholm Ethnogr. Mus. 9:

220 p. + 12 maps.

L~ZANO, E. D. 1917. Diatomeas fosilcs mexi-

canas. Anal. Inst. Geol. Mex., 1, p. l-27.

MEEK,

E. S. 1904. The freshwater fishes of

Mexico. Collect. Field Mus. Zool. 5: 252 p.

MOOSER, F. 1963. La cuenca lacustre de1

Valle de Mexico. Ing. Hidro. Mex. 17( 2) :

47-56.

--. 1967. Tefracrondgia de la Cuenca de

Mexico para 10s ultimos treinta mil ahos.

Bol. Inst. Nat. Antropol. Hist., 30 , p. 12-15.

--,

S. E.

WHITE, AND

J. L.

LORENZO.

1956.

La Cuenca de Mexico: consideraciones gco-

logicas y arqueologicas. Inst. Nat. Antropol.

Hist., Dir. Prehist., Mexico, D.F., Publ. 2.

51 p.

PARSONS,J. J., AND W. M. DENEVAN.

1967.

Pre-Columbian ridged fields. Sci. Amer.

2117 : 93-100.

PATRICK, R., J. CAIRNS, AND S. S. ROBA~K.

1967.

An ecosystematic study of tho fauna and

flora of the Savanah River. Proc. Acad.

Natur. Sci. Phila. 118(5) : 109-407.

AND C. W. REIIMER. 1966. The dia-

tor& of the United States, v. 1.

Acad.

Natur. Sci. Phila., Monogr. 13. 688 p.

SEARS, P. B.

l952.

Palynology in southern

North America, Part 1. Bull. Gcol. Sot.

Amer. 63: 241-2Fj4.

AND K. I~. CLISnY. 1955. Palynology

in ’ southern North America, Part 4. Bull.

Gcol. Sot. Amer. 66: 521-530.

WIIITE, S. E. 1954. The firn fielcl on the vol-

cano Po,pocatepctl, Mexico. J. Glacial. 2

(16) : 389-392.

--.

1962. Late Pleistocene glacial sequence

for the west side of Iztaccihuatl, Mexico.

Bull. Geol. Sot. Amer. 73: 935-958.

ZEEVA~T, L. 1952, Estratigrafia y problemas

cle ingenieria en 10s depositos de arcilla la-

custre cle la ciudad de Mexico. Rev. Ing.

25( 1): 12-28.

-* 1953. Outline of the stratigraphical

and mechanical characteristics of the uncon-

solidated sedimentary deposits in the basin

of the Valley of Mexico. Proc. Congr. Int.

Ass. Quaternary Res. (INQUA), 4th, 30 Aug-

10 Sep 1953, Rome-Piss, Italy.

Documents

texcoco lake