m m

Ml NVIN0SH1HMS S3 I d VH a I1~L 1 B R AR ! E S ^SMITHSONIAN INSTITUTION NOIlfU

^ ^ ^ .^4# §

? t '• N^\ ^ vy

i 2 ^ £

00 2 00 A X CO) * 2

S SMITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1IIAIS S3!dVH8ll LIBRA

CO cz _ co — CO

CO

O o v _ ~-<ufusj>^ O

^1 1 ~*NVI NOSHXI lAJS S3iavyail~JLIBRARI ES^ SMITHSONIAN"* INSTITUTION 2 NOIlfU

•T v Z f” 2 ^

O ^<TstTt/7>^ ~~ O _ X5a VqX O X^vht

CO

rn v vss?" x^vasv^- rn

co — co n co

S SMITHSONIAN INSTITUTION NOIlfUllSNI NV1N0SH1IWS S3ldVU8ll LIBRA

2 CO 2 CO 2 >

2 ' v^' > 2 x^osv^x >' 2

^ CO -A' 2 co 2 CO

JI_NVIN0SH1MS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOIlfU

-p \ co _ ~ on z: co

^ x UJ

co

cm fiA- -^o\ _ /SJ&' r tyk'&i cm

< (f|r 3?) ^ (pit <

a: \Tfe, . A'o/ 5 kC Ji/ co

CD g m ^ A 2 m

X _J 2 j 2

S SMITHSONIAN INSTITUTION NOimillSNI NVIN0SH1IWS $3ldVdan LIBRA
2 r* _ 2

rn x _

co Z: co \ Jz co

N! NVIN0SH1IIAIS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOUflJ

co 2 CO z

< / V S , < .v^\ . - S <

2
o

co
X

r t

__ > ir 2 v > \g>^ 2 >

CO 2 CO *2 CO “ 2

[S SMITHSONIAN INSTITUTION NOIJ.fUIJ.SNI NVINOSHJ.IWS S 3 I d V 8 a IT _ L I B R A

CO _ CO — CO

X .-SiV, _ 1 1 I

CO

Q X^JUST^ O

NI^NVINOSHimS^Sa IdVyan^LIBRARIES2 SM 1THS0N I AN-1 INSTITUTION NOIIIU
r~ \» 2 r- 2 r~ 2

oo

O

CO

po/vp:

05T 'A w

m - xgmiysz m

W _ C /) \ tZ CO

winoshiiws S3iuvuan libraries Smithsonian institution Noiiniusr

< s

z if) z to

;mithsonian institution Nouniiism nvinoshiiws S3 1 Hvyan librarie

Z _ ^ _ —r ov ■ t/} _ -pr

O o

JVINOSHIIWS “S3 IdVHOIl LI B RAR I E SMITHSONIAN^ INSTITUTION Z NOlinillSI
> z _ r~ _ z _ z

33 w

~~ V:

m

jMITHSONIAN INSTITUTION NOIlfUllSNI ~ NVIN0SH1IWS S3IBV&ail LIBRARIE
Z \ 00 Z 00 . z

1o\ * 4 V I 1 /0$t\ =i 1

5 ' \^v > _

CO ' Z CO Z </>

JVINOSHJLMS S3IBVdai1 LIBRARIES SMITHSONIAN INSTITUTION NOlinillSI

tn - ^ CO “ CO

X ^ ^ CO y^L*oX u

\ H /§jtfWW°A “ wlT

O XfoosT^Z TO O

~l Z _J Z _J

SMITHSONIAN INSTITUTION NOIlDlllSNI NVIN0SH1IWS S3 I BVH8 11 LIBRARIE

m \ 11 m

co ti co \ co

MVIN0SHimS^S3 ! dVy 9 n^U B RAR I ES. ..SMITHSONIAN INSTITUTION NOIJ.rU.llSi
av^sovTx S

co A Z ^ \ §

SMITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1IWS S3IBVH8I1 LIBRARIE

CO — tn

CO

O "C _ X^lllS^X Q " _ Q

MVINOSHimS S3ldVdail"JLIBRARI E SMITHSONIAN-" INSTITUTION Z NOlinillSI

Z *- Z

v

THE TEXAS JOURNAL
OF SCIENCE

Volume XXXVII

1985

Published by

THE TEXAS ACADEMY OF SCIENCE

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Editor:

J. Knox Jones, Jr., Texas Tech University
Assistant to the Editor:

Marijane R. Davis, Texas Tech University
Associate Editor for Botany:

Randy Moore, Baylor University
Associate Editor for Chemistry:

Marvin W. Rowe, Texas A&M University
Associate Editor for Computer Science:

Ronald K. Chesser, Texas Tech University
Associate Editor for Mathematics and Statistics:

George R. Terrell, Rice University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Scholarly papers in any field of science, natural history, or
technology will be considered for publication in The Texas Journal of
Science. Instructions to authors are published one or more times each
year in the Journal on a space-available basis, and also are available
from the Editor (The Museum, Box 4499, Texas Tech University,
Lubbock, Texas 79409, 806/742-2487, Tex-an 862-2487).

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, No. 1 August 1985

CONTENTS

Instructions to Authors. . . . . . . 3

Suspended Sediment Variability in Surface Waters of the Lower

Rio Grande Fluvial System, South Texas. By Gerald L. Shideler . 5

Estimation of Dissolved Solids Accumulation in Reservoirs.

By George H. Ward, Jr. . . . . 25

Intrapopulational Variation in the Caudal Osteology of
Cnemidophorus tigris marmoratus Baird and Girard (Reptilia: Teiidae).

By Fred S. Hendricks . . . 33

Range Extension for Arcidens confragosus (Mollusca: Bivalvia:

Unionidae) in Southwestern Texas. By Charles M. Mather . 49

Nongeographic Variation, Reproduction, and Demography in the
Texas Kangaroo Rat, Dipodomys elator (Rodentia: Heteromyidae).

By Wm. David Webster and J. Knox Jones, Jr . . . 51

Variation in Vegetation Density and Foredune Complexity at

North Padre Island, Texas. By Michael Blum and J. Richard Jones . 63

Temperature Tolerance of Florida and Northern Largemouth

Bass: Effects of Subspecies, Fish Size, and Season. By W. Clell Guest . 75

Monodiffric Laplace Transforms and Fourier Transforms.

By Tahereh Daneshi and Charles R. Deeter . . . 85

SDB ABSTRACTS: Spring 1984 . . . . . 103

Announcements

120

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, Nos. 2 & 3

September 1985

CONTENTS

Joint Sets in the Precambrian-Paleozoic Rock Succession

of the Llano Uplift, Texas. By J. R. Alnes and S. E. Cebull . 123

Immune Response in the Bobwhite Quail, Colinus virginianus.

By Alfred C. Schram, Herbert F. Gonzalez, and Cynthia D. Meador . 133

Urnatella gracilis (Entoprocta) from Caddo Lake, Texas

and Louisiana. By Thomas M. Cusak and Jack D. McCullough . 141

Vesicle Contribution to Cyst Wall Formation of Posthodiplostomum

minimum Metacercariae. By Thomas G. Meade and Jose M. Garza . 143

On the Elements of Difference Equation Modeling in Social

Science. By Evans W. Curry and Deraid Walling . 147

Foods of Scaled Quail ( Callipepla squamata ) in Southeastern

New Mexico. By Troy L. Best and Richard A. Smartt . 155

Thyroid-Parathyroid Response to EDTA and CaCh Infusions

in White-tailed Deer By Chun Chin Chao and Robert D. Brown . 163

Effects of Pollution Effluents on Two Successive Tributaries and Village

Creek in Southeastern Texas. By C. Marc Barclay and Richard C. Harrel . 175

Spectrum Analysis of a Triple-Channel, Pulse-Slope-Modulated Wavetrain:

A Comparison of Two Methods. By Joseph H. Nonnast and Olan E. Kruse . 189

Studies of Vegetative Plant Tissue Compatability-Incompatability. VII.

Influences of Individual Organs on Graft Development. By Randy Moore . 201

Marine Fungi Associated with Spartina from Harbor Island, Texas.

By Robert D. Koehn . 213

Vertebrate Use of Nontidal Wetlands on Galveston Island, Texas.

By Allan J. Mueller . 215

Correlation Between Suspended Sediment and Other Water Quality
Parameters in Small Steams of Forested East Texas. By Alfredo

B. Granillo, Mingteh Chang, and Edward B. Rashin . 227

Atmospheric Carbon Monoxide in the El Paso-Cd. Juarez Area. By

Manuel Aguirre, Jr., and Howard G. Applegate . 235

Factors Influencing Cellulolytic Activity of the Soil Fungus,

Aspergillus candidus. By J. Ortega and E. J. Baca . 245

Comparative Behavior and External Color Patterns of Two Sympatric Centipedes
(Chilopoda iScolopendra) from Central Texas. By Raymond W. Neck . 253

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, No. 4 December 1985

CONTENTS

Recent developments in gas chromatographic trace analysis using new
concentration techniques. By Albert Zlatkis, Shary Weisner, and

Labib Ghaoui . 259

Strengths and weaknesses of damage assessment programs: the Ixtoc-I

and Burmah Agate oil spills, and the benthic macroinfauna of the Texas

continental shelf. By George S. Lewbel . 269

A record of Symbos (Artiodactyla:Bovidae) from Kaufman County,

Texas. By Jerry N. McDonald . 311

Reproduction data on Dionda episcopa from Fessenden Spring, Texas.

By Leslie M. Wayne and B. G. Whiteside . 321

Fecal and intestinal bacteria of swine maintained on atherogenic

diets. By D. H. Lewis . 329

)

Effects upon selected marine organisms of explosives used for sound
production in geophysical exploration. By Thomas L. Linton, Andre M.

Landry, Jr., James E. Buckner, Jr., and Robert L. Berry . 341

Records of the spotted skunk and long-tailed weasel from the Llano
Estacado of Texas. By J. Knox Jones, Jr., Robert R. Hollander, and

David A. McCullough . 355

Fine structure of the secondary walls of sclereids of Rauwolfia

serpentina. By A. J. Mia . 359

Studies on the use of boiled chicken egg yolk as a feed for rearing

penaeid shrimp larvae. By David M. Fuze, Joshua S. Wilkenfeld, and

Addison L. Lawrence . 371

Sea urchins from the Brazos Santiago Pass jetty, South Padre Island,

Texas. By Richard R. Fairchild and L. O. Sorensen . 383

Occurrence of mussels in the orbits of a blue crab. By George N.

Greene, David C. Me Aden, and William B. Baker, Jr. . 387

An instance of a largemouth bass, Micropterus salmoides, feeding on
a water snake, Nerodia erythrogaster transversa. By Dennis Parmley

and Charles Mulford. . . 389

Index (including list of authors and reviewers) . 391

Instructions to authors . 401

Dates of publication of volumes 34-36

402

August 1985

le XXXVII, Number 1

PUBLISHED QUARTERLY BY
THE TEXAS ACADEMY OF SCIENCE

SECTION I

MATHEMATICAL SCIENCES
Mathematics, Statistics,
Operations Research

SECTION X
AQUATIC SCIENCES

SECTION IX
COMPUTER SCIENCES

The
Texas
Academy
of

Science

SECTION VIII
SCIENCE EDUCATION

SECTION VII
CHEMISTRY

SECTION VI
ENVIRONMENTAL
SCIENCES

SECTION II
PHYSICS

SECTION III
EARTH SCIENCES
Geography
Geology

SECTION IV
BIOLOGICAL SCIENCES
Agriculture, Botany,
Medical Science,
Zoology

SECTION V
SOCIAL SCIENCES
Anthropology, Education,
Economics, History,
Psychology, Sociology

AFFILIATED ORGANIZATIONS
Texas Section, American Association of Physics Teachers
Texas Section, Mathematical Association of America
Texas Section, National Association of Geology Teachers
American Association for the Advancement of Science

GENERAL INFORMATION

MEMBERSHIP. Any person or group engaged in scientific work or interested in the pro¬
motion of science is eligible for membership in The Texas Academy of Science. Dues for
members are $20.00 annually; student members, $12.00 annually; sustaining members, at
least $30.00 in addition to annual dues; life members, at least $400.00 in one payment;
patrons, at least $500.00 in one payment; corporate members, $250.00 annually; corporate
life members, $2000.00 in one payment. Library subscription rate is $45.00 annually. Pay¬
ments should be sent to Dr. Fred S. Hendricks, TAS Secretary-Treasurer, Drawer H6,
College Station, TX 77844.

The Journal is a quarterly publication of The Texas Academy of Science and is sent
to all members and subscribers. Inquiries regarding back issues should be sent to the
Secretary-T reasurer.

The Texas Journal of Science is published quarterly at Lubbock, Texas U.S.A. Second
class postage paid at Post Office, Lubbock, TX 79401. Please send form 3579 and
returned copies to Texas Tech Press, Box 4240, Lubbock, TX 79409.

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, No. 1 August 1985

CONTENTS

Instructions to Authors . . . . .3

Suspended Sediment Variability in Surface Waters of the Lower

Rio Grande Fluvial System, South Texas. By Gerald L. Shideler . .................... .5

Estimation of Dissolved Solids Accumulation in Reservoirs.

By George H. Ward, Jr. . . . . . . . . . 25

Intrapopulational Variation in the Caudal Osteology of
Cnemidophorus tigris marmoratus Baird and Girard (Reptilia: Teiidae).

By Fred S. Hendricks. . . . . . 33

Range Extension for Arcidens confragosus (Mollusca: Rival via:

Unionidae) in Southwestern Texas. By Charles M. Mather . . .49

Nongeographic Variation, Reproduction, and Demography in the
Texas Kangaroo Rat, Dipodomys elator (Rodentia: Heteromyidae).

By Wm. David Webster and J. Knox Jones, Jr . . . . .51

Variation in Vegetation Density and Foredune Complexity at

North Padre Island, Texas. By Michael Blum and J. Richard Jones . 63

Temperature Tolerance of Florida and Northern Largemouth

Bass: Effects of Subspecies, Fish Size, and Season. By W. Clell Guest . 75

Monodiffric Laplace Transforms and Fourier Transforms.

By Tahereh Daneshi and Charles R. Deeter . . . 85

SDB ABSTRACTS: Spring 1984. . . . . . . 103

Announcements

120'

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Editor:

William H. Neill, Texas A&M University
Assistant to the Editor:

Fred S. Hendricks, Texas A&M University
Associate Editor for Chemistry:

Marvin W. Rowe, Texas A&M University
Associate Editor for Mathematics and Statistics:
George R. Terrell, Rice University

INSTRUCTIONS TO AUTHORS

Scholarly papers in any field of science, natural history, or technology
are eligible for publication in The Texas Journal of Science. Before a
paper can be accepted for publication, it must undergo critical review by
at least 2 appropriate referees and the editor.

A manuscript intended for publication in the Journal should be pre¬
pared in accordance with the following instructions and then submitted
to J. Knox Jones, Jr., The Museum, Box 4499, Texas Tech University,
Lubbock, Texas 79409.

The manuscript is not to have been published elsewhere. Triplicate
typewritten or machine-printed copies (or the original and 2 good xero¬
graphic reproductions) must be submitted. Text, table and figure cap¬
tions, and references should be double-spaced with 2-3 cm margins
(without right-justification) on 81/ X 11-inch paper. The title of the arti¬
cle should be followed by the name and business or institutional address
of the author(s). Be sure to include zip code with the address. If the paper
has been presented at a meeting, a footnote giving the name of the
society, date, and occasion should be included but should not be num¬
bered. Begin the body of the paper with a brief ABSTRACT to which is
appended a list of key words (abstracting services pick this up directly),
followed by the text subdivided into sections as appropriate. Use upper
/lower case letters, italics (indicated by single underscore), and indenta¬
tion in a way that mimics the pattern evident in the most recent issue of
the Journal.

In the text, cite all references by author and date in chronological
order, i.e., Jones (1971); Jones (1971, 1972); (Jones 1971); (Jones 1971,
1972); Jones and Smith (1971); (Jones and Smith 1971); (Jones 1971;
Smith 1972; Beacon 1973). If there are more than 2 authors, use: Jones et
al. (1971); (Jones et al. 1971).

References are then to be assembled, arranged alphabetically , and
placed at the end of the article under the heading LITERATURE
CITED. For a PERIODICAL ARTICLE use the form: Jones, A. P., and R. J.
Smith. 1971. Persistence of chlorinated hydrocarbons. J. Soil Chem.
37:116-123. For a paper published in the PROCEEDINGS OF A SYMPOSIUM,
etc., use the form: Jones, A. P. 1971. Persistence of chlorinated hydrocar¬
bons, p. 155-18sJ^rz A. P. Jones (ed.), Pesticides in soils. Soc. Soil Chem.,
New York, NY. For a REPORT use: Jones, A. P. 1971. Persistence of
chlorinated hydrocarbons. Texas Soils Institute (Austin, TX) Report No.
14, 46 p. A MASTERS OR Ph D. THESIS should appear as: Jones, A. P. 1971.
Persistence of chlorinated hydrocarbons in blackland soils. M.S. thesis,
Texas A&M Univ., College Station, TX. For a BOOK, NO EDITORS, use:
Jones, A. P. 1971. Environmental effects of chlorinated hydrocarbons.
Academic Books, New York, NY, 439 p. For a CHAPTER IN A BOOK WITH

4

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

EDITORS: Jones, A. P. 1971. Persistence of chlorinated hydrocarbons, p.
13-39. In A. P. Jones, B. R. Smith, Jr. and T. S. Gibbs (eds.), Environ¬
mental effects of chlorinated hydrocarbons. Academic Books, New York,
NY. For an IN-PRESS PERIODICAL ARTICLE use: Jones, A. P. In Press.
Persistence of chlorinated hydrocarbons. J. Soil Chem. For an IN-PRESS
BOOK use: Jones, A. P. In Press. Environmental effects of chlorinated
hydrocarbons. Academic Books, New York, NY.

References to unpublished data or personal communications should
not be listed in the LITERATURE CITED section. However, they
should be presented within the text as: (unpubl. data from C. J. Jones,
Dept. Zoology, Univ. Texas, Austin, TX) or (pers. comm, from R. C.
Smith, P.O. Box 133, Mexia, TX).

Any footnotes (except those referenced in the body of a table) should
appear on a separate sheet of paper, following the LITERATURE
CITED.

All tables are to be typed with a carbon ribbon, free of error, without
handwritten notations, and ready for photographic reproduction. Each
table, headed by its caption, should be placed on a separate sheet of
paper. Tables must have a text reference, i.e., Table 2 shows. . .or (Table
2).

Figures are to be original inked drawings or photographic prints no
larger than 4Zi X 6V2 inches and mounted on standard SV2 X 1 1-inch paper.
Each illustration should be marked on the back with the name of the
first author and the figure number. Captions for figures are to be pro¬
vided on a separate sheet of paper. Figures must have a text reference,
i.e., Figure 3 illustrates. . .or (Fig. 3).

Authors will receive galley proofs plus the edited typescript and
information concerning reprints and page charges. Proofs must be cor¬
rected (using ink) and returned to the editor within 5 days. Page charge
payment (check or purchase voucher) or a publication-grant request
must accompany return of the corrected proofs or a delay in printing the
manuscript could occur. Reprint orders should be returned directly to
Texas Tech Press, Box 4240, Lubbock, TX 79409.

THE EDITOR SHOULD BE NOTIFIED IMMEDIATELY OF ANY CHANGE IN
THE PRINCIPLE AUTHOR’S ADDRESS OR TELEPHONE NUMBER.

NOTE: Authors are encouraged to contribute $35.00 per published page
to defray printing costs, and authors of articles exceeding ten pages are
expected to make some contribution to the publication fund. However,
payment of printing costs is not a condition for publication in The
Texas Journal of Science, and NO AUTHOR, WHO WOULD OTH¬
ERWISE SUBMIT A MANUSCRIPT, SHOULD HESITATE TO DO
SO BECAUSE OF LACK OF FUNDS. Members without funds may
apply to the Texas Academy of Science for a grant to cover some or all
costs of publication.

SUSPENDED SEDIMENT VARIABILITY
IN SURFACE WATERS OF THE
LOWER RIO GRANDE FLUVIAL SYSTEM,

SOUTH TEXAS

by GERALD L. SHIDELER

U.S. Minerals Management Service

P.O. Box 6732

Corpus Christi, TX 78411

ABSTRACT

Suspended sediment variability in surface waters of the lower Rio Grande fluvial
system was studied along the 1400-km sector between El Paso and the Gulf of Mexico.
Sediment concentrations and textural characteristics were monitored during a sampling
period in each of three summers (1976, 1977, 1979).

Sediment quantity and quality varied both temporally and spatially. Sediment
concentrations ranged from 0.4 mg/liter to 3,648 mg/liter, with turbidity being
consistently lowest in Amistad Reservoir and highest in Big Bend National Park. Rio
Grande suspensate most commonly consisted of poorly sorted to very poorly sorted clay-
size particles, with coarest one-percentile fractions in the silt range. Suspended sediment
outflow from the Pecos River tributary was relatively meager (<20 mg/liter), and the
Pecos suspensate was more poorly sorted with coarser one-percentile fractions than the
Rio Grande suspensate.

The regional variability of Rio Grande sediment concentrations and textural
characteristics has both random and persistent components. Persistent trends result from
changes in stream discharge and competency, changes in bedrock and soil type, and
anthropogenic effects.

INTRODUCTION

The Rio Grande and its tributaries represent one of the major fluvial
systems of the western North American continent. The 1400-km Texas
sector of the lower Rio Grande, between El Paso and the Gulf of
Mexico, is the subject of this study (Fig. 1). This sector of the Rio
Grande valley contains a variety of bedrock materials in varying
physiographic settings; it is also a sector that is highly influenced by
man’s activities, in terms of multiple water usage and flow regulation.
Consequently, the lower Rio Grande provides an interesting and
challenging environment for regional studies of modern fluvial
sediments.

Previous studies of modern Rio Grande sediments have concentrated
on heavy-mineral variations in bedload deposits (e.g. Rittenhouse 1943,
1944; Shideler and Flores 1980). However, aside from routine water-

The Texas Journal of Science, Vol. XXXVII, No. 1 , August 1985

6

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

105° 100°

Figure 1. Map of lower Rio Grande study area showing locations of 13 sediment¬
sampling stations and 14 stream-gauging stations. Also shown on index map are
boundaries of physiographic provinces.

quality measurements, very little work has been done with the
suspended load of the Rio Grande. The purpose of the present study
was to evaluate the regional variability of suspended sediments in
surface waters of the Rio Grande, in terms of both sediment
concentrations and their textural characteristics, and to relate these
parameters to natural fluvial processes and anthropogenic activities.

STUDY AREA

The headwaters of the Rio Grande are located in the San Juan
Mountains of southern Colorado. The river flows southward through
central New Mexico, then southeasterly along the Texas/Mexico
border until it terminates in the western Gulf of Mexico. The two
major tributaries of the Rio Grande are the Pecos River and the Rio
Conchos (Fig. 1). The Pecos River originates in the Sangre de Cristo
Mountains of northern New Mexico, and joins the Rio Grande near
the city of Del Rio, Texas. The Rio Conchos originates in the Sierra
Madre Occidental of Chihuahua, Mexico, and joins the Rio Grande
near the city of Presidio, Texas. Two minor tributaries originating in

RIO GRANDE SUSPENSATE

7

Mexico, the Rio Salado and the Rio San Juan, join the Rio Grande
at Falcon Lake and east of the town of Roma, Texas, respectively.

Climatically, the Texas sector of the Rio Grande extends from the
Chihuahuan Desert in the north, to humid subtropical maritime
conditions at the Gulf of Mexico. Mean annual precipitation along the
lower Rio Grande increases downstream, ranging from about 20 cm (8
in) at El Paso to about 68 cm (27 in) at Brownsville; mean annual
temperatures range from 17.7 C (63.9 F) at El Paso to 23.0 C (73.5 F)
at Brownsville (Mueller 1975).

The studied sector of the Rio Grande traverses three major
physiographic provinces that are progressively lower in relief down¬
stream (U.S. Dept. Interior 1970). The river sector from El Paso to the
eastern margin of Big Bend National Park is within the Basin and
Range Province (Fig. 1), which is characterized by mountain ranges
and intermontane basins; elevations above sea level are generally
within the 2,000-5,000 ft (610-1,524 m) range. The river sector from Big
Bend National Park to approximately Del Rio is within the Great
Plains Province, which is characterized by a hilly terrain with
elevations within the 1,000-2,000 ft (305-610 m) range. From south of
Del Rio to the Gulf of Mexico, the Rio Grande traverses the relatively
flat Gulf Coastal Plain Province, which gradually slopes to sea level.

The regional geology of the lower Rio Grande valley is illustrated
by a generalized geologic map of the area (Fig. 2), which shows a wide
variety of bedrock, in terms of both age and lithology. The Basin and
Range Province (El Paso to eastern Big Bend area) consists of a
relatively heterogeneous assemblage of sedimentary, igneous, and
metamorphic rocks ranging in age from Quaternary to Precambrian.
Abundant Tertiary intrusive and extrusive igenous rocks are found
within this province. The Great Plains Province (eastern Big Bend to
Del Rio area) is more homogeneous, consisting mainly of alternating
carbonate and siliciclastic sedimentary rocks of Cretaceous age (Fig. 3).
Lower Cretaceous (Comanchean) deposits tend to be dominated by
carbonate strata, whereas the upper Cretaceous (Gulfian) section tends
to be dominated by siliciclastic rocks. Further downstream, the Gulf
Coastal Plain Province (south of Del Rio) consists mainly of
heterogeneous siliciclastic sedimentary deposits of Tertiary and
Quaternary age; these units gently dip and decrease in age in a
gulfward direction.

The water discharge characteristics of the lower Rio Grande fluvial
system are quite variable, and are influenced by both natural and
anthropogenic factors. The downstream increase in precipitation from
El Paso to Brownsville certainly influences stream flow conditions. In
addition, seasonal variations in precipitation are highly influential. As

8

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

105° 100*

Figure 2. Generalized geologic map of the lower Rio Grande valley. Geology adapted
and generalized from the Geological Highway Map of Texas (Amer. Assoc. Petroleum
Geologists 1973) and the Carta Geologica de la Republica Mexicana (Sanchez
Mejorada and Lopez-Ramos 1968).

noted by Mueller (1975), rainfall maxima along the Rio Grande from
El Paso to Del Rio occur during the summer in association with
thunderstorm activity; rainfall maxima near Brownsville occur both
during summer thunderstorms and during occasional fall tropical
storms and hurricanes. During rainfall maxima, the lower Rio Grande
is subject to flash flooding, with peak discharge possibly approaching
1 million cubic feet per second (28,320 m3/sec) during a very high
flood stage (Mueller 1975).

In addition to regional and seasonal variations in precipitation,
man’s activities greatly influence the discharge characteristics of the
Rio Grande. These activities include the construction of dams and
reservoirs for flood control and hydroelectric power, water diversions
for irrigation and municipal usage, and channel rectification and
canalization. The history of these activities along the Rio Grande has
been reviewed by Mueller (1975), who noted that the increased demand
for water by both the United States and Mexico since the turn of the
century has reduced the Rio Grande flow to only a fraction of its pre-
1900 discharge volume.

RIO GRANDE SUSPENSATE

9

Figure 3. Views of the fluvial system incised through sedimentary bedrock of Cretaceous
age within the Great Plains Province: A. View of the turbid Rio Grande
approximately 25 km upstream from the Pecos River junction, with Mexico in the
background; B. Upstream view of the nonturbid Pecos River near its junction with
the Rio Grande (Station 9 locality).

10

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

METHODOLOGY

For this study, a total of 13 monitoring stations was established for
sampling suspended sediments along the lower Rio Grande between
the Gulf of Mexico and El Paso (Fig. 1). Station sites were selected to
provide regional coverage and to permit evaluation of local effects of
major tributaries and reservoir construction. Accessibility was also an
influential factor in site selection, and all sites were located on the
United States side of the waterway; station 9 was located within the
Pecos River tributary just above its junction with the Rio Grande.

Field samples were obtained at the 13 monitoring stations during
each of three diffeent sampling periods, which extended over a 3-year
interval. The three suites of samples were obtained during the
following periods: 14-19 June 1976 (RGS-1); 21-24 June 1977 (RGS-2);
and 6-9 August 1979 (RGS-3). The series of sampling periods over a
relatively long observational time interval was used to distinguish
long-term, systematic variations in suspended sediment characteristics
from short-term variations. If the same local variation was observed
during all three sampling periods, then it was inferred to be systematic
rather than episodic or random in nature. Efforts were made to
minimize the duration of each sampling period, in order to obtain
quasi-synoptic sample suites.

At each station on each sampling occasion, stream flow conditions
were noted and representative surface-water samples were obtained a
few feet from the bank in quart-size, particle-free polypropylene
storage bottles. The bottles were opaque and contained a 5 percent
concentration of particle-free formalin to inhibit subsequent organic
growth.

In the laboratory, water samples were analyzed for suspended
sediment concentrations and texture. The mass of total particulate
matter (mg/liter) was determined by filtration through a pre-washed
0.45 jiim millipore filter. The texture of the particulate matter was
analyzed with a model TA Coulter counter, using a combination of
200-jum and 30-/xm tube apertures; this provided a frequency
distribution of particles within the 0.63 /xm to 81-/xm size range.
Statistical grain-size parameters of the particulate matter were derived
by computerized analysis. Derived grain-size parameters included
particle mean diameter and standard deviation (moment measures),
one-percentile diameter, and silt/clay ratio. The mean diameter,
standard deviation, and one-percentile diameter parameters were
calculated in terms of Krumbein’s (1934) phi equivalents (<f> — logzD,

where D = diameter in mm).

The sediment concentration and textural characteristics were plotted
on trend graphs to illustrate their regional variability. Mean values of

RIO GRANDE SUSPENSATE

11

Table 1. Ranges of Rio Grande water discharge rates and suspended sediment
characteristics during individual sampling periods.

Parameter

RGS-1
(June 1976)

RGS-2
(June 1977)

RGS-3

(August 1979)

Mean discharge
rate (m3/sec)

27-3,695

28-2,760

55-4,867

Sediment

concentration

(mg/liter)

0.4-2,356

0.4-3,112

0.4-3,648

Mean

diameter ( (f> *)

8.01-8.94

6.59-8.45

6.90-8.57

Silt/clay

ratio

0.29-1.08

0.48-2.55

0.40-2.03

One-percentile
diameter ( <f> )

4.44-5.89

3.56-4.96

3.58-4.95

Standard

deviation (</>)

1.10-1.66

1.30-2.11

1.19-2.53

*</> = — log2D, where D = diameter in mm

these parameters for the three sample suites were compared to
determine any significant differences, using a /-test at the 95%
confidence level. Sediment concentrations and textural parameters then
were related to water-discharge rates of the Rio Grande, using
discharge data (m3/sec) provided by the International Boundary and
Water Commission of the United States and Mexico from 14 stream
gauging stations (Fig. 1). At each station, mean daily water discharge
was averaged over the sampling period plus the 5 days prior to
sampling; these grand mean values then were used to characterize Rio
Grande discharge during each of the three sampling periods.
Suspended sediment characteristics also were related to mapped soil
changes (U.S. Dept. Interior 1970), and to changes in bedrock geology
as well as other natural and man-made features.

RESULTS AND DISCUSSION
Stream Discharge Characteristics

The ranges of mean discharge rates along the lower Rio Grande and
at the mouth of the Pecos River tributary during the three sampling
periods are presented in Tables 1 and 2, and the regional variability
of discharge rates along the Rio Grande is illustrated by trend graphs
(Fig. 4). The data indicate substantial temporal variability among the
three periods. Discharge along the Rio Grande, as well as outflow
from the Pecos River tributary, were generally highest during the third
sampling period (August 1979); this is attributed to greater rainfall
during that period. The temporal variablity among the three periods

12

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Table 2. Water discharge rates and suspended sediment characteristics of the Pecos River
(Station 9) during individual sampling periods.

Parameter

RGS-1
(June 1976)

RGS-2
(June 1977)

RGS-3

(August 1979)

Mean discharge
rate (m3/sec)

165

213

295

Sediment

concentration

(mg/liter)

20

6.4

1.6

Mean

diameter (</>*)

8.75

8.50

8.36

Silt/clay

ratio

0.37

0.40

0.83

One-percentile
diameter (0)

4.66

4.30

4.45

Standard
deviation (0)

1.65

1.42

1.78

*(f) — — log2D, where D = diameter in mm

was not consistent along the length of the lower Rio Grande, which
is attributed mainly to variations in the amount of anthropogenic
water usage and the degree of flow regulation at the times of sampling.

The regional variability of stream discharge characteristics along the
length of the lower Rio Grande shows some consistent trends for all
three sampling periods that appear to result both from natural
processes and anthropogenic effects. From north El Paso downstream
to north Presidio just above the junction of the Rio Conchos tributary
(stations 13,12), a consistent reduction in discharge occurred. This
reduction is attributed both to municipal water usage by the El Paso/
Juarez community, and to evaporation and irrigation usage. South of
Presidio just below the junction of the Rio Conchos tributary (Station
11), a consistent discharge increase occurred, which is attributed to the
added volume of the Rio Conchos. Volumetrically, the Rio Conchos
is the principal supplier of water to the lower Rio Grande fluvial
system (Mueller 1975). From south of Presidio to just north of the
Pecos River junction, the Rio Grande traverses the Big Bend region
where only small ephemeral creeks are confluent. This sector was
characterized by a fairly constant discharge during the first two
sampling periods, apparently reflecting a balance between evaporation
and creek inflow. In contrast, the third sampling period (August 1979)
showed a notable discharge increase downstream, apparently reflecting
an excess of creek inflow over evaporation; this probably resulted from
a late summer thunderstorm.

The discharge from the Pecos River tributary (Station 9) into the
Rio Grande was relatively low and fairly consistent (165-295 m3/sec

RIO GRANDE SUSPENSATE

13

Figure 4. Comparative trend graphs illustrating regional variability of mean water
discharge (m3/sec) along the Rio Grande during the three sampling periods.

range) for the three sampling periods. A short distance downstream
from the junction of the Pecos River, the Rio Grande enters Amistad
Reservoir (Station 8), after which the flow is regulated for hydroelectric
power generation, irrigation, and flood control. In the sector
extending from south of Del Rio (Station 7) to south of Eagle Pass
(Station 6), a significant amount of water diversion occurs for
hydroelectric power and irrigation (Maverick Irrigation District).
Further downstream, the Rio Grande enters Falcon Lake Reservoir
(Station 4), which also receives minor inflow from the Rio Salado
tributary of Mexico. The regulated Rio Grande outflow from Falcon
Lake Reservoir receives additional minor discharge from the Mexican
Rio San Juan tributary that joins the Rio Grande downstream from
Station 3 (Roma). Slightly upstream from Station 2 (Hidalgo), the Rio
Grande is diverted for flood control into a multi-channel floodway
system that flows into the Gulf of Mexico and adjacent lower Laguna
Madre. The last two gauging stations on the trend graphs are located
along the main channel of the Rio Grande, which has regulated flow;
the consistent downstream discharge reduction between the two gauge
stations probably reflects mainly municipal water usage by the
Brownsville/Matamoros community.

14

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Figure 5. Comparative trend graphs illustrating regional variability of suspended
sediment concentration (mg/liter) along the Rio Grande during the three sampling
periods.

In summary, the water discharge characteristics of the lower Rio
Grande reflect both natural processes and anthropogenic activities. The
effect of the latter is especially pronounced downstream from the Pecos
River junction, where the Rio Grande flow is highly regulated by dam
construction and reservoir storage, and by multiple-use water
diversions on both sides of this international waterway.

Sediment Concentration Variability

The ranges of suspended sediment concentrations, in terms of total
mass (mg/liter), along the lower Rio Grande and from the Pecos River
tributary during the three sampling periods are presented in Tables 1
and 2; the regional variability of sediment concentrations along the
Rio Grande is illustrated by trend graphs (Fig. 5). Some temporal
variations of sediment concentration occurred among the three
sampling periods at individual stations, but the concentration
differences were not regionally consistent along the length of the Rio
Grande. Furthermore, no statistically significant differences occurred
among mean concentration values during the three sampling periods
(Table 3). These relationships suggest that the time of sampling was
not an important factor influencing river turbidity, and that the

RIO GRANDE SUSPENSATE

15

Table 3. Results of /-tests for comparison of mean suspended sediment characteristics
during the three sampling periods.

/-Values

Parameter

RGS-1

vs

RGS-2

RGS-1

vs

RGS-3

RGS-2

vs

RGS-3

Sediment

concentration

(mg/liter)

0.382

0.319

0.023

Mean

diameter (0)

3.585*

2.484*

1.334

Silt/clay

ratio

2.924*

2.418*

1.042

One-percentile
diameter (0)

6.101*

4.225*

1.303

Standard

deviation (0)

3.904*

2.710*

0.098

^Significant difference at the 0.05 level of probability (degrees of freedom = 24, /o.os =
1.711).

observed concentration variations mainly reflect spatial rather then
temporal factors. During all three periods, the concentrations were
consistently lowest and of constant value (0.4 mg/liter) in Amistad
reservoir (Station 8), whereas they were consistently highest in the
Boquillas area of Big Bend National Park (Station 10) where they
reached a maximum value of 3,648 mg/liter.

The regional variablity of Rio Grande sediment concentrations had
persistent trends among the three sampling periods, indicating that the
causative factors were systematic rather than episodic or random in
nature. Beginning at El Paso (Station 13), an increase in turbidity
occurred downsteam to north Presidio just above the Rio Conchos
junction (Station 12). Physiographically, this entire stream sector is
within the Basin and Range Province with a land surface characterized
by high mountains and intermontane basins. The downstream
turbidity increase in this sector was inversely related to stream
discharge (Fig. 4), and is attributed largely to a reduction in water
volume. A downstream soil transition also may have been a
contributing factor. The northern third of the sector toward El Paso
is underlain mainly by a calciorthid soil (Calcisol), which is
characterized by large amounts of soluble calcium carbonate and
gypsum salts with no major horizons of clay accumulation. In
contrast, the southern two-thirds of the sector is underlain mainly by
a shallow torriorthent soil (Regosol) characterized by a loamy or clayey
composition that could serve as a more prolific source of detrital
sediment, possibly contributing to an increase in turbidity down-

16

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

stream. At south Presidio below the Rio Conchos junction (Station
11), a reduction in Rio Grande turbidity occurred, which was
associated with an increase in stream discharge. This turbidity
reduction is attributed to sediment dilution effects resulting from the
mixing of effluent from the Rio Conchos tributary.

Further downstream, the Rio Grande traverses the Big Bend
National Park where it consistently acquired its greatest turbidity
along the entire lower valley (Station 10). During two of the sampling
periods, the suspended sediment concentration at this locality
(Boquillas) was over an order of magnitude greater than in adjacent
areas. The Big Bend stream sector also occurs within the Basin and
Range Province, but has somewhat lower relief than further upstream;
the land surface is characterized by low mountains and intermontane
basins. The pronounced increase in suspended sediment concentrations
within the Big Bend sector of the Rio Grande was not associated with
any soil change, nor was it associated with any consistent change in
stream discharge. Consequently, it is interpreted as resulting mainly
from the presence of a highly prolific local source of fine-grained
detritus. Within Big Bend National Park, the Rio Grande traverses
Cretaceous-Tertiary bedrock of varying lithologies, which have been
described by Maxwell et al. (1967). The bedrock includes a variety of
intrusive and extrusive igneous rocks, carbonate units, and both clastic
and pyroclastic deposits. The clastic and pyroclastic deposits contain
substantial quantities of soft, fine-grained materials that include
bentonitic clays, tuffaceous clays and mudstones, ash beds, and chalky
shales. These relatively incompetent and easily erodable deposits
appear to be locally abundant, and are probably the source of most
suspended sediment for this highly turbid section of the Rio Grande.

Immediately downstream from the highly turbid Big Bend area, the
Rio Grande begins traversing the Great Plains Province, which is
characterized by high hills. The bedrock consists of Cretaceous strata
composed of both carbonate and siliciclastic lithologies. The Cretace¬
ous bedrock appears to contribute very little additional sediment to the
Rio Grande. The additional influx of suspended sediment from the
Pecos River tributary was consistently low (<20 mg/liter), apparently
reflecting a large proportion of lower Cretaceous (Comanchean)
carbonate units comprising the lower Pecos drainage basin. Most
suspended sediment remaining in the Rio Grande after its confluence
with the Pecos River is settled from suspension within Amistad
Reservoir (Station 8) where the water was nonturbid, having a constant
low value of 0.4 mg/liter during all sampling periods. This was the
least turbid water sampled along the lower Rio Grande. Slightly

RIO GRANDE SUSPENSATE

17

downstream from the reservoir at Del Rio (Station 7), the river showed
only a minor increase in turbidity (<5 mg/liter).

Downstream from Del Rio, the Rio Grande begins traversing the
upper Gulf Coastal Plain Province, which is characterized by an
irregular plain topography. An increase in turbidity occurred between
Del Rio and Eagle Pass (Station 6). This is attributed to a change in
bedrock and soil type. Along this sector, the Rio Grande traverses
Upper Cretaceous (Gulfian) strata, which are composed of a relatively
greater proportion of siliciclastic units, as compared with the more
calcareous lower Cretaceous section further upstream. The increase in
detritahsource rocks and changing physiography also have resulted in
a downstream change from a calcareous soil (Calciustoll) to a more
clay-rich soil (Vertisol) midway between Del Rio and Eagle Pass. The
soil and bedrock changes are interpreted as the cause of the
downstream increase in turbidity at Eagle Pass.

From Eagle Pass downstream to Falcon Lake Reservoir (Station 4),
the Rio Grande continues through the Gulf Coastal Plain Province.
The river traverses mainly siliciclastic strata of Late Cretaceous-
Tertiary age, with no major soil change. In this stretch, river turbidity
slightly increased. The regulated outflow from Falcon Lake Reservoir
(Station 3) had turbidity levels substantially below those upstream; this
is attributed to a depletion of sediment as a result of settling from
suspension within the reservoir. The pronounced effect that construc¬
tion of Falcon Lake Reservoir in 1955 has had on sharply reducing
both the discharge and the suspended sediment load of the Rio Grande
further downstream has been noted by Morton and Pieper (1975); they
suggested that this depletion of sediment may be a factor contributing
to accelerated shoreline erosion rates along the Gulf coast near the
mouth of the Rio Grande.

From the turbidity minimum near Roma, the sediment concentra¬
tions in the Rio Grande progressively increased to the mouth of the
river (Station 1). This increase in river turbidity is attributed both to
a reduction in stream discharge, as well as to a change in substrate
material and associated soil type. In this sector, the river flows through
the flat lower Gulf Coastal Plain Province composed mainly of
unconsolidated Quaternary clastic deposits, as opposed to the more
consolidated Tertiary strata further upstream. This transition occurs
slightly downstream from Roma, and it is associated with a soil
change from a clay-rich type (Vertisol) to a more friable organic-rich
type (Mollisol). This change in substrate composition enhances
erosion, resulting in the progressively higher concentrations of
suspended sediment from Roma to the mouth of the Rio Grande. A

18

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

downstream increase in agricultural cultivation is probably also a
contributing factor.

In summation, the spatial variability of lower Rio Grande turbidity
results from a combination of stream discharge variations, bedrock and
soil changes, and anthropogenic activities.

Sediment Textural Variability

The variability of suspended sediment textural characteristics along
the lower Rio Grande and from the Pecos River was determined in
terms of the following four parameters: 1) mean diameters (</>),
reflecting the average grain size of the entire suspensate; 2) silt/clay
ratios, reflecting the proportions of the two dominant detrital
components of the suspensate; 3) one-percentile diameters (</>),
reflecting the approximate size of the coarsest suspended particles — as
an index of turbulence and energy level; and, 4) standard deviations
((f>), reflecting the sorting of the entire suspensate. The ranges of these
four sediment parameters along the Rio Grande and from the Pecos
River tributary during the individual sampling periods are presented
in Tables 1 and 2; the temporal and regional variations of these
parameters along the Rio Grande are illustrated by trend graphs (Fig.
6).

The textural parameters along the Rio Grande indicate significant
temporal variability among the three sampling periods. In terms of
mean values, suspended sediments during the first sampling period
were significantly finer-grained, contained finer one-percentile frac¬
tions, and were better sorted than during the other two periods (Tables
1,3); no significant textural differences occurred between sediments of
the second and third sampling period. These relationships indicate
that some time-dependent factor had contributed to the observed
textural variability. Based on all three sampling periods, the overall
mean diameters of the suspensate at individual stations ranged from
fine silt (6.59 c f>) to clay (8.94 </>), but were predominantly within the
clay size range. The silt/clay ratio values ranged rom 0.29 to 2.55, but
were predominantly less than unity, thus illustrating the dominance of
clay-size particles within the suspensate. The coarsest one-percentile
diameters of the suspensate ranged from medium silt (5.89 </>) to very
fine sand (3.56 </>), but were predominantly within the silt size range.
The standard deviation values of suspended particles ranged from 1.10
</> to 2.53 <f>; using Folk’s (1965) classification, these values indicate that
the suspensate contained poorly sorted to very poorly sorted particulate
matter.

The regional variability of suspended sediment textural characteris¬
tics along the lower Rio Grande had some persistent trends among the

RIO GRANDE SUSPENSATE

19

RGS-I -------

RGS-2 - - -

RGS - 3 .

DOWNSTREAM — ►-

Figure 6. Comparative trend graphs illustrating regional variability of suspended-
sediment textural characteristics (mean diameter, silt/clay ratio, one-percentile
diameter, standard deviation) along the Rio Grande during the three sampling
periods.

three sampling periods, thus indicating some location-related syste¬
matic causes, in addition to temporal variations. From El Paso to
north of Presidio (stations 13, 12), the suspensate exhibited a consistent
downstream reduction in mean grain size, a reduction in the silt/clay
ratio, a reduction in the one-percentile size, and an improvement in

20

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

sorting. These attributes, which were associated with a downstream
reduction in stream discharge and an increase in turbidity, are
attributed mainly to an increased influx of clay-size particles consistent
with the downstream transition to a more clay-rich soil. The reduction
in the one-percentile size also suggests the concomitant downstream
depositional depletion of some of the coarsest particles (very fine sand
and coarse silt), resulting in a better sorted suspensate; this probably
reflected the observed reduction in stream discharge and velocity, and
an associated reduction in stream competency.

Below the junction of the Rio Conchos at south Presidio (Station 11),
the suspensate during the first two sample periods had an increase in
mean grain size, silt/clay ratios, and one-percentile diameters, and was
more poorly sorted. Very little variation in suspensate character between
Stations 1 1 and 12 occurred during the last period. These textural changes
were associated with an increase in stream discharge rate and a reduction
in turbidity, and may be attributed to the following causes: 1) mixing
of less turbid effluent from the Rio Conchos that contained relatively
coarser particulate matter; 2) increase in stream discharge and competency,
resulting in turbulent resuspension of coarser bedload sediment; or, 3)
a combination of both of the foregoing factors. The net result would
have been a somewhat coarser and more poorly sorted suspensate at the
south Presidio locality relative to further upstream. Further downstream
at the highly turbid Big Bend National Park locality (Station 10), the
suspensate showed a consistent reduction in mean grain size and one-
percentile diameters, and improved sorting; the silt/clay ratios were
reduced during two of the three sampling periods. In general, these
textural attributes suggest that the large local influx of detritus from
Big Bend source rocks was dominated by clay-size particles.

The suspended sediment discharged (at relatively low concentrations)
from the Pecos River (Station 9) did not show any consistent
differences in mean grain size or silt/clay ratios compared to the Rio
Grande suspensate; however, the Pecos sediments did consistently
contain coarser one-percentile fractions and were more poorly sorted
(Table 2). Within the nonturbid Amistad Reservoir (Station 8), most
of the suspended sediment had settled out of the water column. The
small amount of sediment remaining in suspension was predominantly
clay-size detritus, but did include some particles in the one-percentile
fraction as coarse as very fine sand during the last sampling period.
The composition of these relatively coarse particles is unknown, but
they probably represent phytoplankton; the presence of these size-
disparate particles resulted in relatively poor sorting of the reservoir
suspensate. Downstream from Amistad Reservoir at Del Rio (Station
7), the suspensate consistently was characterized by a coarser mean

RIO GRANDE SUSPENSATE

21

grain size, a higher silt/clay ratio, a finer one-percentile fraction, and
improved sorting. These characteristics were associated with regulated
stream discharge, which had a relatively high current velocity and only
a very slight increase in turbidity; bedload was sand and gravel. The
textural attributes of this dilute suspensate indicate relatively high
stream competency, a short transport distance, and a paucity of fine¬
grained siliciclastic source materials within the Del Rio area which is
underlain by carbonate-dominated lower Cretaceous strata. The size
reduction of the one-percentile fraction relative to Amistad Reservoir
is attributed to a depletion of the probable phytoplankton component
in this relatively high-energy stream sector.

Further downstream at Eagle Pass (Station 6), the increase in
turbidity resulting from bedrock and soil changes was not associated
with any systematic change in textural properties of the suspensate.
Lack of association could have been caused by the highly variable
stream discharge rates resulting from local water diversion activities;
this is suggested by the one-percentile fractions, which showed a
consistent direct relationship between grain size and stream discharge
within the Del Rio-Eagle Pass sector. From Eagle Pass to Laredo
(Station 5), the same relationship existed between the suspensate’s one-
percentile fraction and stream discharge rates, indicating that higher
discharge resulted in a coarser maximum particle size. In addition, the
Eagle Pass-Laredo sector had some consistent textural trends. The
suspensate increased in mean grain size and silt/clay ratios, and
became somewhat more poorly sorted downstream. These attributes
might reflect a downstream increase in the availability of coarser
siliciclastic units comprising the traversed upper Cretaceous-lower
Tertiary source rocks within this sector.

Downstream from Laredo, the suspensate within Falcon Lake
Reservoir (Station 4) showed a pronounced reduction in mean grain
size, silt/clay ratios, and one-percentile diameters, as well as slightly
improved sorting. These were the effects of the coarser detritus settling
from suspension within the reservoir. The suspensate of the regulated
outflow from the reservoir downstream near Roma (Station 3) was
characterized by a slight increase in mean grain size and silt/clay
ratios, and slightly poorer sorting; these characteristics apparently
reflect the renewed acquisiton of coarser suspended matter from clastic
Tertiary source rocks.

From Roma to the mouth of the Rio Grande (Station 1), the change
from a Tertiary substrate to an unconsolidated Quaternary substrate
(and associated soil change), which contributed to a progressive
downstream increase in turbidity, did not produce any systematic
textural changes in the suspensate. Within this terminal sector of the

22

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Rio Grande, where stream flow is highly regulated, the observed
textural variations appear to be largely of a random nature.

CONCLUSIONS

Stream discharge characteristics of the lower Rio Grande are the
result both of natural processes and anthropogenic activities; the latter
include flow regulation from channel modification, reservoir construc¬
tion, and multiple-use water diversions. The suspended load in surface
waters of the Rio Grande is variable in time and space, in terms both
of concentration and textural characteristics. During summer in 1976,
1977, and 1979, sediment concentrations ranged from 0.4 mg/liter to
3,648 mg/liter, with lowest turbidity occurring within Amistad
Reservoir and highest turbidity occurring within Big Bend National
Park. Sediment outflow from the Pecos River to the Rio Grande was
consistently low (<20 mg/liter).

Texturally, the mean grain size of the Rio Grande suspensate ranged
from fine silt to clay, with the latter being dominant. The coarsest
one-percentile fraction ranged from medium silt to very fine sand, but
was predominantly within the silt size range. The suspensate ranged
from poorly sorted to very poorly sorted. Suspended sediment outflow
from the Pecos River was consistently more poorly sorted and had
coarser one-percentile fractions than the Rio Grande suspensate, but
showed no consistent difference in mean grain size.

The regional variability of both sediment concentration and textural
parameters consists of random and episodic variations, as well as
persistent trends indicating systematic causes. The trends are related to
a combination of downstream changes in stream discharge and
competency, changes in bedrock and soil type, and anthropogenic
effects.

ACKNOWLEDGMENTS

The writer expresses his gratitude to his wife, Marie Jeannette, for
her cheerful field assistance during the three long, hot, and dusty
sampling trips along the Rio Grande. Appreciation is also expressed
to D. Owen for assistance in data tabulation, and to R. Flores for
reviewing the manuscript.

LITERATURE CITED

American Association of Petroleum Geologists. 1973. Geological highway map of Texas,
map no. 7, American Assoc. Petrol. Geologists, Tulsa, OK.

Folk, R. L. 1965. Petrology of sedimentary rocks. Hemphill’s, Austin, TX, 159 p.
Krumbein, W. C. 1934. Size frequency distributions of sediments. J. Sed. Petrol. 4:65-77.

RIO GRANDE SUSPENSATE

23

Maxwell, R. A., J. T. Lonsdale, R. T. Hazzard, and J. A. Wilson. 1967. Geology of Big
Bend National Park, Brewster County, Texas. Publ. 6711 of Univ. Texas Bureau of
Econ. Geology, Austin, TX, 320 p.

Morton, R. A., and M. J. Pieper. 1975. Shoreline changes on Brazos Island and South
Padre Island (Mansfield Channel to mouth of the Rio Grande): an analysis of
historical changes of the Texas Gulf shoreline. Geol. Circular 75-2, Univ. Texas
Bureau of Econ. Geology, Austin, TX, 39 p.

Mueller, J. W. 1975. Restless river. Texas Western Press, Univ. Texas at El Paso, El Paso,
TX, 151 p.

Rittenhouse, Gordon. 1943. Transportation and deposition of heavy minerals. Geol. Soc.
America Bull. 54: 1725-1780.

Rittenhouse, G. 1944. Sources of modern sands in the middle Rio Grande Valley, New
Mexico Jour. Geol. 52:145-183.

Sanchez Mejorada, S. H., and E. Lopez-Ramos. 1968. Carta geologica de la Republica
Mexicana: Comite de la Carta Geologica de Mexico, 1:2,000,000.

Shideler, G. L., and R. M. Flores. 1980. Heavy-mineral variability in fluvial sediments
of the lower Rio Grande, southwestern Texas. Texas J. Sci. 32:73-91.

U.S. Department of Interior. 1970. Physiographic divisions of the United States: The
National Atlas of the United States, Washington, D.C., 417 p.

, l

ESTIMATION OF

DISSOLVED SOLIDS ACCUMULATION
IN RESERVOIRS

by GEORGE H. WARD, JR.

Espey, Huston & Associates, Inc.

P.O. Box 519

Austin, TX 78767

ABSTRACT

An analytic model for estimating long-term accumulation of dissolved solids in
reservoirs is presented. The model yields solutions amenable to hand calculation;
therefore, it is useful for preliminary mass-budgeting analysis and for testing accuracy
of more complex numerical models. The model also offers valuable insight into the
physics of the dissolved solids budget of a reservoir. Application of the model to a
power-plant cooling reservoir in central Texas and to Lake Meredith in the Texas
Panhandle demonstrates its validity in forecasting long-term dissolved solids trends.

INTRODUCTION

A common problem in reservoir design and management is the
accumulation of dissolved solids due to evaporation, which is a
significant portion of the typical reservoir’s volume budget (e.g., Gray
1970). Evaporative accumulation of solids can become particularly
acute if throughflow is small or if the reservoir is subject to a heat
load; small drainage-area cooling ponds are therefore especially
susceptible. This problem is now routinely treated (e.g., Rich 1973;
Viessman et al. 1977) by numerically modeling the dissolved solids and
volume budget, which involves simulating reservoir volume (or stage)
and solids concentration through time with long period-of-record
inputs of inflow volume and solids concentration, evaporation,
outflow, etc. Indeed, recourse to the computer has become so ingrained
that a reminder seems worthwhile that the problem is amenable to
analytic solution. The analytic solution, in fact, has some intrinsic
value: apart from offering an economical calculation for those of us
still too impecunious to have ready access to a computer, the analytic
model also provides insight into the physical processes involved,
which can be indispensible in data analysis or reservoir design.

THEORY

For a reservoir of volume V and surface area A, the rate of increase
in dissolved-solids concentration c due to net evaporation E (gross

The Texas Journal of Science, Vol. XXXVII, No. 1, August 1985

26

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

evaporation less precipitation, in dimensions of water depth per unit
time) is EAc/V. The change in concentration due to inflow Q at
concentration a is (q— c)Q/V. Therefore, the total change in
concentration is given by

V^£=EAc + (Q-c)Qi (1)

dt

which for constant parameter values has the solution

c = °Qi( 1 - e~Rt) + c0e"Rt (2)

RV

where R = (Q— EA)/V. Inflow includes runoff from the drainage area
of the reservoir and any makeup pumpage, with a in (2) being the
flow-weighted mean concentration of these sources of inflow. Outflow
Qo (which includes spills, releases and seepage) does not appear in (2)
because, being at concentration c, it does not affect concentration in
the reservoir. It is, however, implicit in the volume budget in that in
order to maintain reservoir volume

Qo — Qi — EA. (3)

The rate constant R is the inverse of the throughflow “residence time,”
the time required for the outflow Q0 to deplete the volume V, and
must be positive in (2); R measures the responsiveness of the reservoir
to fluctuations in the controlling parameters. In order to validly apply
(2) to the long-term mean dissolved solids concentration with long¬
term mean values of E, Q, and q, three conditions are required: (i)
the reservoir volume (and, hence, surface area) must be substantially
constant, (ii) q and Q must be uncorrelated in time (satisfied, e.g., if
q is constant), and (iii) c must be uncorrelated in time with Q and
with E. When these requirements are met, the differential equation (1)
can be averaged in time to yield an equation of the same form with
average values of the parameters E, Q and q, and therefore has the
solution (2). Actually, (ii) can be eliminated as a requirement if the
average q is the flow-weighted time mean. This requires a data record
for both flow and dissolved solids for each inflow to the reservoir. The
last condition (iii) is implied for R sufficiently small, i.e., when the
reservoir responds slowly to fluctuations in E, a or Q.

One immediate conclusion is that, if inflow exceeds net evaporation,
the dissolved solids concentration asymptotically approaches an
equilibrium value

DISSOLVED SOLIDS ACCUMULATION

CiQi

Qi-EA

27

Coo = lim c(t) =
t— oo

(4)

whereupon (2) can be rewritten in the more transparent form

(5)

The values of the two parameters Coo and R determine the form of c(t).
The effect of the initial concentration cG declines exponentially in
time. The greater the magnitude of R, the faster this asymptote is
approached and the faster the influence of initial concentration c0
declines.

An important special case is when R-^0, i.e., Qi balances EA. (In
practical application makeup might be required to maintain reservoir
volume, for a cooling pond in southerly latitudes, say, whereupon Qi
is forced to equal EA.) Then the solution of (1) becomes simply

C(t) = fiQ1 t + Co

V

(6)

and c increases linearly with time, without any asymptotic limit. Here
c(t) is of course given by the dilution of the dissolved solids load qQ
in the reservoir volume V.

The above analysis applies to any species of dissolved solids or to
the total dissolved solids. It can be viewed as a special case of the
solution for a well-mixed reactor (e.g., Metcalf and Eddy, Inc. 1972)
with first-order coefficient EA/V, and is readily generalizable to a
nonconservative constituent subject to a first-order decay. Similarly,
the analysis can be extended to a series of reservoirs following the
method of O’Connor and Mueller (1979), if the outflow from each
reservoir is properly decremented by the evaporation.

Even when a numerical dissolved-solids model is available, the
analytical solution (5) or, if appropriate, (6) is useful in two respects:
as a standard to test the validity and accuracy of the computer code,
and as an expedient means for testing sensitivity to the various input
variables and identifying thereby potential deficiencies in the data base.

APPLICATIONS

A cooling reservoir in central Texas and a water supply reservoir in
the Texas High Plains, both troubled with dissolved solids accumula¬
tion, provide two convenient cases for application of the analytical
model.

28

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Table 1. Average physical and hydrological parameters for example applications.

A. Cooling Reservoir, central Texas, 1976-79 Average

Surface area

330 hectares

Volume

1.63 X 107 cubic meters

Inflow

7.40 X 106 cubic meters per year

Inflow TDS

391 parts per million

Evaporation

5.21 X 106 cubic meters per year

B. Lake Meredith, Texas, Water Years* 1973-80 Average

Surface area

4,047 hectares

Volume

4.72 X 108 cubic meters

Inflow

1.06 X 108 cubic meters per year

Inflow chlorides

220 ppm

Evaporation

5.09 X 107 cubic meters per year

*A water year is the 12-month period beginning 1 October of the preceding calendar
year.

The first case is a 370 ha reservoir near Rockdale, Texas, that
provides cooling water for a 360 MW power plant rejecting an average
of 464 MW (thermal). Total dissolved solids (TDS) have increased
nearly monotonically in this reservoir since its creation in 1953. For
the four-year period 1976-79, average physical and hydrological
parameters are summarized in Table 1. Releases are rarely made from
this reservoir; rather, make-up is usually required from two sources, a
nearby stream and ground water, which for 1976-1979 averaged 6.39 X
106 m3/yr and 1.01 X 106 m3/yr, respectively, with corresponding
average TDS concentrations of 350 ppm and 650 ppm. Net natural
evaporation was estimated from the Lake Somerville data, the nearest
National Weather Service (NWS) class-A pan, using a pan-to-lake
coefficient of 0.7. This was augmented by a forced evaporation of 3.18
X 106 m3/yr computed by the method of Ward (1980) using the above
power-plant heat rejection and meteorological data from the Austin
NWS station. In addition, the plant consumes 1.13 X 106 m3/yr, but
returns the TDS in its waste stream, so that this acts like additional
evaporation. Thus total evaporation (Table 1) is the sum of natural
net, forced, and plant consumption.

The values of the determining parameters of (5) become R = 0.134
yr-1 and Coo = 1325 ppm. In Figure 1 are shown TDS measurements
from monthly grab samples for the period 1976-1979, with the curve
given by (5) superposed. Although (5) is apparently a good predictor
of the long-term trend, one must consider its sensitivity to some of the
parameters. For example, peripheral runoff into the lake is ignored in
the data of Table 1. This probably is justified, considering the small
drainage area and physical alterations; however, the curves for
hypothetical ungauged runoff values of 6.0 X 105 and 1.2 X 106 m3/
yr at 200 ppm, shown as broken curves in Figure 1, indicate that this

DISSOLVED SOLIDS ACCUMULATION

29

Figure 1. Observed and computed total dissolved solids (parts per million) accumulation
in a central Texas cooling reservoir. Curves (descending) computed for peripheral
runoff equal to 0 (solid curve), 6 (broken, short dashes), and 12 (broken, long dashes)
X 105 m3/yr.

could alter the predicted TDS buildup significantly. (The value of 1.2
X 106 m3/yr comes from applying a drainage area ratio to the 1976-
9 mean flow at USGS Gauge 08109700. However, this ratio is 0.02754,
disproportionately small to justify such a data transfer.) Similarly, the
concentration of the inflowing water can produce as great an alteration
by 8% and 13% decreases in magnitude, respectively. Given the infrequency
and variability of the measurements, these are well within the error of
estimation of q.

Lake Meredith, near Amarillo, Texas, is a 6,680 ha reservoir that
supplies water for industrial and municipal use on the High Plains.
It has been plagued in recent years by an increasing concentration of
chlorides. The primary source of inflow is the Canadian River, which
transports the chlorides into the lake from a leaky brine aquifer in
New Mexico (Bureau of Reclamation 1979). However, most of the
discharge into the lake is produced by severe thunderstorms, so that
chlorides are inversely correlated with flow. Therefore, q was taken to
be the flow- weigh ted concentration, which is routinely computed by
the U.S. Geological Survey (USGS) as a part of its surface-water
measurement program. The values of Table 1 were based upon
averages for water years 1973-1980 from reports of the USGS (1973-
1980). Net evaporation rates came from data of a Class- A pan near

30

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Figure 2. Observed and computed chlorides (parts per million) accumulation in Lake
Meredith, Texas. Curves (descending) calculated for pan data X 1.1 (broken, long
dashes), 1.0 (solid), and 0.9 (broken, short dashes).

Amarillo, obtained thrugh the Texas Natural Resources Information
System (Ferguson 1981) with a pan-to-lake coefficient of 0.7, and
precipitation data from the NWS Amarillo Station.

Chlorides data for the period 1972-1981 are shown in Figure 2,
together with predictions of (5). The pan evaporation data proved to
be sparse, only 47 monthly values over the twelve-year period 1969-80,
so the estimate of EA is rough. Sensitivity of the chlorides trend to
evaporation is indicated by variation of the pan coefficient by ±10%.
From Figure 2, it is apparent that (5) provides a reasonable
representation of the chlorides trend in Lake Meredith, and that a
better measurement of evaporation is needed to refine the long-term
equilibration concentration of chlorides.

LITERATURE CITED

Bureau of Reclamation. 1979. Lake Meredith salinity study. U.S. Department of the
Interior, Amarillo, TX.

Ferguson, D. L. 1981. Access to water-related data in Texas. J. Water Res. Planning and
Mgmt. Div., ASCE, 107 (WR2): 375-383.

Gray, D. M. (ed.). 1970. Handbook on the principles of hydrology. Water Information
Center, Inc., Huntington, NY.

Metcalf and Eddy, Inc. 1972. Wastewater engineering. McGraw-Hill Book Company,
New York, NY.

DISSOLVED SOLIDS ACCUMULATION

31

O’Connor, D. J., and J. A. Mueller. 1970. A water quality model of chlorides in Great
Lakes. J. San. Eng. Div., ASCE, 96 (SA4): 955-975.

Rich, L. G. 1973. Environmental systems engineering. McGraw-Hill Book Company,
New York, NY.

U.S. Geological Survey. 1973-80. Water resources data for Texas (published annually).

Department of the Interior, Government Printing Office, Washington, D.C.

Viessman, W., J. Knapp, G. Lewis, and T. Harbaugh. 1977. Introduction to hydrology.

IEP— A Dun-Donnelley Publisher, New York, NY.

Ward, G. H. 1980. Accuracy of Harbeck diagram for forced evaporation. J. Energy Div.,
ASCE, 106 (EY1): 23-31.

'

INTRAPOPULATIONAL VARIATION
IN THE CAUDAL OSTEOLOGY OF
CNEMIDOPHOR US TIGRIS MARMORATUS
BAIRD AND GIRARD (REPTILIA: TEIIDAE)

by FRED S. HENDRICKS

Department of Wildlife and Fisheries Sciences

Texas A&M University

College Station, TX 77843

ABSTRACT

Examination of caudal vertebrae from 94 specimens of Cnemidophorus tigris
marmoratus from Presidio, Presidio County, Texas, yielded the following information
on caudal osteology and its non-geographic variation: The first two vertebrae in the
caudal sequence typically lack a chevron bone. The intervertebrally placed chevron bone
usually first appears after the third caudal vertebra and is repeated posteriorly for 30 to
40 vertebrae, forming a complete hemal arch. Usually the anterior-most 14 or 15
vertebrae of the caudal sequence bear a pair of primary transverse processes. A secondary
transverse process is present anterior to the primary transverse process, beginning usually
with the ninth through twelfth vertebrae. The autotomic septum is usually present in
the anterior-most vertebrae with a secondary transverse process, continuing posteriorly
until it is absent again in the last 10 to 20 vertebrae. The total number of post-sacral
vertebrae may vary from 62 to 72 in complete tails. As the tail increases in length, the
vertebrae in the middle of the caudal sequence exhibit the greatest rate of increase in
length. Tails with regenerated tips contained 8 to 41 vertebrae.

Neither sex nor ontogenetic stage accounted for most variation in the caudal sequence.
However, there is a tendency for smaller specimens to have fewer vertebrae with
secondary transverse processes.

INTRODUCTION

Interpretation of osteological data has been an important tool for
classifying lizards and evaluating evolutionary lineages. Etheridge
(1967) examined representatives of all families of lizards and found that
even though lizard tails exhibit remarkable diversity in form and
function, the basic osteological differences generally are consistent
within established genera and higher categories. The taxonomic value
of caudal vertebrae within the family Teiidae has been discussed briefly
by Cope (1892), Etheridge (1967), Hoffstetter and Gasc (1969), and
Presch (1974). Specific information regarding caudal osteology for the
genus Cnemidophorus is limited to Cope’s (1892, 1900) brief collective
description of three specimens of “C. tessellatus” and one specimen of
“C. sexlineatus" , Etheridge’s (1967) examination of 12 specimens of
Cnemidophorus which he discussed collectively with other Teiidae,

The Texas Journal of Science, Vol. XXXVII, No. 1 , August 1985

34

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

SNOUT-VENT LENGTH

Figure 1. Size-frequency distribution, by sex, of the 94 cleared and stained specimens
of Cnemidophorus tigris marmoratus examined from Presidio, Presidio Co., Texas.

and Presch’s (1974) observation that Cnemidophorus has more than 60
caudal vertebrae. Cope’s (1900) assertion that “the discrimination of
the North American species of the genus Cnemidophorus is the most
difficult problem in North American herpetology” may still be valid.
Osteological studies have the potential to help alleviate this problem.

In light of the taxonomic importance of caudal osteology to delimit
certain taxonomic categories, it seems apropos to assess the intra-
populational variation of caudal vertebrae as a basis for determining
the reliability of these features as taxonomic characters. The purpose
of this paper is to describe the caudal sequence, and to evaluate the
variability of the caudal vertebrae in a large series of Cnemidophorus
tigris marmoratus from a single locality.

METHODS AND MATERIALS

A series of 94 specimens of Cnemidophorus tigris marmoratus was
collected in the summer of 1968 from the vicinity of Presidio, Presidio
County, Texas (Texas Cooperative Wildlife Collection, Cleared 8c
Stained 198-291). The sample consists of 57 males and 37 females
representing the range of snout-vent lengths depicted in Figure 1.

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

35

Pianka (1970), who examined the more western subspecies of C. tigris ,
reported that the smallest female in reproductive condition was 71 mm
snout-vent length. Scudday (1971) examined specimens from western
Texas, and found a 68 mm female reproductive. For purposes of this
study, males are considered sexually mature adults when they grow to
a size comparable to the size class for sexually mature females. The
sample, therefore, consists of nearly equal proportions of adults and
subadults. The series contains 15 specimens with regenerated tails, 40
with complete tails and 39 with at least some portion of the tail
missing. Preliminary scale counts, measurements of snout-vent and
total length, and sex were recorded prior to skeletal preparation. The
specimens were cleared and stained with Alizarin following the Taylor
(1967) enzyme method, incorporating the modifications of Zug and
Crombie (1970).

All specimens were examined with the aid of a dissecting
microscope. The following characteristics of the caudal sequence were
recorded: lengths of vertebrae, number of pygals, number of vertebrae
with a chevron bone, number of vertebrae with the primary transverse
process, first and last vertebrae with a secondary transverse process,
first and last vertebrae with an autotomic septum, total number of
post-sacral vertebrae and number and condition of vertebrae in
regenerated tails.

RESULTS

The tail of C. tigris marmoratus is circular in cross-section with the
greatest diameter in the sacral region, tapering gradually to a fine
point distally. The complete tail comprises approximately 72% of the
lizard’s total length.

The basic caudal sequence of C. tigris marmoratus is shown in
Figure 2. Immediately posterior to the two sacral vertebrae begins a
series of vertebrae with one pair of posterolaterally directed primary
transverse processes. This series continues posteriorly for 12 to 16
vertebrae, where the primary transverse processes abruptly disappear. A
shorter series of laterally directed secondary transverse processes is
present anterior to the primary transverse process at the eighth to tenth
vertebra. The series with secondary transverse processes continues
posteriorly for two to eight vertebrae. The origin of the latter series
typically marks the beginning of the vertebrae with an autotomic
septum. The secondary transverse processes, anterior to the septum,
usually are not present in vertebrae as far caudal in the series as are
the primary transverse processes, which are posterior to the septum.
The remaining caudal vertebrae lack transverse processes and are
similar to the sixteenth vertebra shown in Figure 2. However, not all

36

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

Figure 2. Dorsal view of sacral and anterior caudal vertebrae of Cnemidophorus tigris
marmoratus. More distal vertebrae are similar to the sixteenth.

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

37

Figure 3. Anterior view of the posterior half of the tenth caudal vertebra (at the
autotomic septum) with primary transverse processes (PTP) and partly occluded
anterior view of the eleventh caudal vertebra with secondary transverse processes
(STP) of Cnemidophorus tigris marmoratus. The chevron bone (CB) is intervertebral.

of the remaining vertebrae contain autotomic septa. The septum
disappears between the 46th and 57 th vertebrae, although an
incomplete septum occasionally is present several vertebrae posterior to
the last one with a complete septum. Not all vertebrae are similar in
length. The vertebrae in the mid-caudal region are the longest and the
most posterior vertebrae are the shortest.

The ventrally located Y-shaped chevron bones (Fig. 3) are typically
intervertebral in position and articulate equally with adjacent centra.
In some other lizard species, the chevron bones undergo an anterior
migration (Hoffstetter and Gasc 1969). The chevron bone is treated
here as being associated with the vertebra anterior to it.

Vertebra Length

Cnemidophorus tigris marmoratus tends to maintain a constant
ratio of tail length to total length throughout ontogeny. The first post-
sacral vertebra is the shortest vertebra in the anterior half of the tail
(Fig. 4). The lengths of successive vertebrae increase steadily until
approximately the twentieth vertebra, vertebrae are similar in length
between the twentieth and fortieth, and then the vertebrae rapidly
decrease in length. The vertebrae posterior to about the fifty-fifth
vertebra are nearly equal in length regardless of specimen size. No
differences in the rate of increase in vertebra length could be found
between males and females.

38

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 1, 1985

CAUDAL VERTEBRA NUMBER

Figure 4. Vertebra length related to caudal vertebra number of five specimens of
Cnemidophorus tigris marmoratus (A = 85mm SVL; B = 72mm SVL; C = 59mm SVL;
D = 46mm SVL; E = 40mm SVL).

Pygal Vertebrae

The pygal vertebrae are those immediately posterior to the sacrum
that lack a chevron bone (Hoffstetter and Gasc 1969). At least one
pygal is always present, with two pygals the most frequent case (85 of
94), and no specimen had more than two. Four of the nine specimens
having chevrons associated with the second vertebra are females, each
a subadult with poorly developed chevrons. Romer (1956) documented

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

39

VERTEBRAE WITH CHEVRONS

Figure 5. Frequency distribution of the number of vertebrae with complete chevron
bones in 46 specimens of Cnemidophorus tigris marmoratus with undamaged tails.

sexual variation in the position of the first chevron in crocodilians and
turtles in which females lacked proximal chevrons due to the position
of the cloaca and associated structures. The number of pygal vertebrae
in C. tigris marmoratus is not significantly different between the sexes
nor ontogenetically variable.

Caudal Vertebrae with Chevron Bones

The morphology, number, and region of occurrence of chevron
bones in C. tigris marmoratus is similar to that described by Reese
(1923) for Tupinambis nigropunctatus (= T. teguixin). The chevron
bones in C. tigris marmoratus project ventrally and distally, beginning
generally at the third vertebra. The chevron bones form a complete
hemal arch for the next 30 to 40 vertebrae. About a third of the most
anterior chevron bones are large and do not vary greatly in size. The
remaining two-thirds of the complete chevron bones are sequentially
smaller until the ventral fusion of the lateral elements is incomplete.
Figure 5 includes only chevron bones with complete hemal arches. The
paired lateral processes are progressively smaller in size until they are
minute and then absent.

The mean number of caudal vertebrae with chevron bones is 34.8
(SD = 2.4) in 46 specimens in which a definite count can be made.
Figure 5 indicates a rather normal distribution of the number of
caudal vertebrae associated with chevron bones. No ontogenetic or
sexual variation could be found.

Primary Transverse Process Series

Etheridge (1967) calls the series of post-sacral vertebrae exhibiting
two pairs of transverse processes the “diverging transverse process
series.” However, I have discarded this teminology in favor of the more
consistent terms, “primary transverse processes,” referring to those

40

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

processes occurring most cephalad in the caudal sequence, and
“secondary transverse processes,” referring to those first appearing
more caudad in the sequence. Etheridge’s diverging process series
terminated when either pair of transverse processes disappeared. This
resulted in the series being terminated sometimes by the disappearance
of the anterior pair and sometimes by the disappearance of the
posterior pair of processes. The terminology suggested here removes
this ambiguity, and may also be applied to caudal sequences of other
species regardless of the orientation of transverse processes.

The primary transverse processes are always found on the first post-
sacral vertebra and posteriorly for 12 to 16 vertebrae (Fig. 6), with
counts of 14 or 15 most common in 83 specimens (mean 14.4,
SD = 0.8). The processes are largest anteriorly and gradually decrease
in size posteriorly until they abruptly disappear. Neither sexual nor
ontogenetic variation was observed. Reese (1923) presented the only
comparative data concerning the number of transverse processes in the
Teiidae. He reported that in Tupinambis nigropunctatus the transverse
processes are not present distally from about the twentieth vertebra.

Secondary Transverse Process Series

The first caudal vertebra with a secondary transverse process is the
ninth in 72 specimens, the eighth in 18 specimens, and the tenth in
only 3 of 93 specimens. No ontogenetic or sexual differences were
found in the inception of this series. The number of vertebrae in the
secondary transverse process series ranges from two to eight (Fig. 7).
Smaller specimens tend to have fewer vertebrae with secondary
transverse processes, and specimens having the greatest number of
processes are large males. The last vertebra with a secondary transverse
process is usally the tenth to twelfth, but the series continues as far
as the sixteenth in some specimens. This process diminishes in size
gradually, occasionally with only one member of the pair present on
the distal vertebra of the series. The series is considered terminated
when the basal width of the process exceeds its length, on both sides.
The secondary transverse process series terminates one to five vertebrae
anterior to the most distal vertebra of the primary transverse process
series in 80 specimens, one to two vertebrae posterior in five
specimens, and at the same vertebra in three specimens. Since the
beginning of the secondary transverse process series is nearly constant
in the caudal sequence, the end point of this series is correlated with
the number of vertebrae in the sequence. Consequently, ontogenetic
and sexual differences in the end point parallel those found in the
length of the series.

CAUDAL OSTEOLOGY OF CNEM1DOPHORUS 41

12 13 14 15 16

VERTEBRAE WITH PRIMARY
TRANSVERSE PROCESSES

Figure 6. Frequency distribution of the number of caudal vertebrae with primary
transverse processes of Cnemidophorus tigris marmoratus in 83 specimens clearly
showing termination of the series.

42

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

I

£

LU

<

cl

m

LU
I —
CL
LU

>

Li.

o

CL

LU

CQ

Z>

z

CD

LU

lO

U1

LU

SNOUT-VENT LENGTH

Figure 7. Number of vertebrae with secondary transverse processes related to snout-vent
length and sex in Cnemidophorus tigris marmoratus. Solid horizontal bars represent
males; open bars represent females. Vertical bars indicate means of each category. N
= sample size in each category.

Autotomic Septum

In most teiids the beginning of the vertebral series bearing secondary
transverse processes also marks the first vertebra with an autotomic
septum (Etheridge 1967). The first vertebra with an autotomic septum
was never more than one vertebra before or after the beginning of the
secondary transverse process series. The autotomic septum first
appeared in the caudal sequence in the ninth vertebra in 51 specimens,
the eighth in 38, and in the tenth in only four of 93 specimens. The
fracture plane was present in each vertebra from the anteriormost
through the 46th to 57th (Fig. 8). Occasionally an incomplete septum
occurred in a vertebra posterior to the regular termination point.
Etheridge (1967) stated that in some species fusion of the intravertebral
fracture planes begins distally and progresses anteriorly as the
individual becomes older. In the specimens examined here, no
significant ontogenetic or sexual differences for the first or last vertebra
with an autotomic septum, nor septal fusion occurring in the posterior
vertebrae of large specimens, was found.

Total Caudal Vertebrae

The number of post-sacral vertebrae in the series of 40 specimens
with complete tails ranges from 62 to 72 (Fig. 9) and has a mean value

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

43

46 48 50 52 54 56 58

LAST VERTEBRA WITH AUTOTOM 1C SEPTUM

Figure 8. Frequency distribution of the last vertebra in the caudal sequence with an
autotomic septum in Cnemidophorus tigris marmoratus in the 39 specimens clearly
showing the termination of the series.

of 66.0 (SD = 2.5). The most distal vertebrae rapidly decrease in size,
do not have autotomic septa, and in smaller specimens, lack complete
ossification, often rendering accurate counts of the last two or three
vertebrae difficult. Neither sexual nor ontogenetic trends explain the
variation in total number of vertebrae.

Regenerated Tails

In the specimens with entire tails, 18% of the females and 32% of the
males have regenerated tails. Cagle (1946) observed a similar ratio in

15—1

62 64 66 68 70 72

TOTAL CAUDAL VERTEBRAE

Figure 9. Frequency distribution of the total number of caudal vertebrae in 40 specimens
of Cnemidophorus tigris marmoratus with complete tails.

44

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Hemidactylus garnotti (29% of the females and 33% of the males had
fully or partially regenerated tails). One could speculate that females
present themselves for predation less frequently than do males. Pianka
(1970) observed inactivity and wariness of C. tigris females carrying
large oviducal eggs. This could account for fewer broken tails in
females, although he, using a much larger sample, found no overall
significant differences between the sexes in percent with broken tails.
The 15 specimens examined here with regenerated tails have 8 to 41
caudal vertebrae present. The specimen with regeneration after the
eighth vertebra has a deformed sacrum, and asymmetrically deformed
caudal vertebrae. This specimen is also the only one of the 15 in which
the regenerated portion of the tail does not begin at an autotomic
septum, but begins intervertebrally. Moffat and Bellairs (1964) reported
similar regeneration following experimental cutting of the tail in
Lacerta vivipara. The regenerated tails in the other fourteen specimens
begin at split centra posterior to the secondary transverse process series
and never anterior to the twelfth caudal vertebra.

DISCUSSION

The entire caudal sequence of these specimens of Cnemidophorus
tigris marmoratus is summarized in Figure 10. Certain characteristics—
such as the number of pygal vertebrae, the number of vertebrae with
primary transverse processes, the first vertebra in the secondary
transverse process series, and the first vertebra with an autotomic
septum — are less variable than the number of caudal vertebrae with
chevron bones, the number of vertebrae involved in the secondary
transverse process series, the last vertebra with an autotomic septum,
and the total number of vertebrae. It is apparent that some variation
does exist and is manifested in some characters more than others.

Although the autotomic septum has a relatively constant position of
first occurrence, the number of vertebrae having it shows an interesting
correlation with the length of post-sacral vertebrae. Differences in
vertebra length among specimens of different size (Fig. 4) are greatest
in the region between post-sacral vertebrae 10 and 50; the first and last
vertebrae with an autotomic septum are numbers 8 to 10 and 46 to 57
respectively. It is obvious that as the lizard grows in overall length,
some caudal vertebrae undergo disproportionate growth relative to
others. All of these longer vertebrae have a transverse fracture plane.
It seems plausible that the presence of an autotomic septum may
contribute to the increase in length of a post-sacral vertebra.

Moffat and Bellairs (1964), working with Lacerta vivipara, stated
that the split in the centrum seems to be the result of bone and
notochordal cartilage being invaded by vascular connective tissue. The

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

45

VERTEBRA TYPE

Figure 10. Summary of the range and variability of the constituents of the caudal
vertebral sequence of Cnemidophorus tigris marmoratus. Solid vertical bars represent
the total range of each character. Open boxes to the right and left of each bar show
the range of the proximal and distal termination, respectively, of each sequence.
Horizontal bars indicate the mean values within the respective ranges. Caudal
vertebrae are numbered sequentially, beginning with the first postsacral. (Py = pygals;
CWCh = caudal vertebrae with chevron bones; PTP = primary transverse processes;
STP = secondary transverse processes; AS = autotomic septum; Tot = total caudal
vertebrae).

separately arising split in the neural arch joins the split in the centrum
soon after birth to complete the autotomic septum. Such
vascularization within the body of the vertebra, if it persists in post¬
natal development, would seem to be conducive to osteoblast
proliferation, resulting in bone being laid down proximal to the
septum and lengthening of the vertebra. Moreover, Pratt (1946)
observed that in L. vivipara the autotomic septum is bordered by
distinct lips which he postulated are due to growth in the vertebra
after formation of the septum. A thickened region proximal to the
septum was observed in specimens examined here which might

46

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

indicate a growth center in the septum. Osteoblasts in non-autotomic
vertebrae would probably not be as richly vascularized and, as a result,
not proliferate as rapidly as those in autotomic vertebrae. Also, the
presence of four epiphyses per autotomic vertebra as opposed to two
for a non-autotomic vertebra might be conducive to disproportionate
lengthening.

Hoffstetter and Gasc (1969) indicated that the first vertebra with an
autotomic septum was the fifth or sixth caudal in the family Teiidae.
Clearly this is not the case in Cnemidophorus tigris marmoratus,
where the septum almost always first appears in the eighth or ninth
caudal, nor is it the case in Tupinambis nigropunctatus, in which the
autotomic septum is reported to begin at about the fourteenth caudal
(Reese 1923). It is apparent that generalizations concerning the first
vertebra with an autotomic septum may be misleading when applied
to the Teiidae.

Etheridge (1967) indicated that the presence of the secondary
transverse process was a result of the splitting of the primary transverse
process by the autotomic septum. Observations in this study support
this interpretation. As did Etheridge, vertebrae were found in this
study with partially split transverse processes (Fig. 2, caudal vertebra
number 8). The first vertebra with a secondary transverse process was
usually the first vertebra with an autotomic septum. Also the large
male specimens were likely to have more secondary transverse process
vertebrae than were small specimens. The reason for this ontogenetic
variation is not immediately clear. Pratt (1946) documented muscle
attachment to the transverse processes in L. vivipara. That same
muscle attachment was observed in this study in incompletely cleared
specimens of C. tigris marmoratus. The secondary transverse process
series may afford additional points of attachment for tail muscles,
resulting in more force being required to cause autotomy in this area.
Only one specimen with a regenerated tail had a break prior to the
termination of the secondary transverse process series. That specimen
had regeneration between vertebrae rather than at the autotomic
septum. It is possible that there is a selective advantage in larger
lizards to have a shorter effective autotomic series. Predators may be
kept adequately occupied by a smaller percentage of a large lizard’s
tail, allowing the larger male lizards to protect their hemipenal
structures, and larger lizards of both sexes to retain more of the tail
for fat storage, balance, and other functions.

Romer (1956) stated that the tail was elongate in primitive reptiles,
probably containing 50 to 70 vertebrae, and that primitive lizards had
50 to 75 vertebrae, a count found in a number of teiids. Without
elaboration, Presch (1974) considered a decrease in caudal vertebrae, to
fewer than 60, to represent a more advanced condition. He indicated

CAUDAL OSTEOLOGY OF CNEMIDOPHORUS

47

that within the macroteiids, the genera Ameiva, Cnemidophorus ,
Kentropyx, Dicrodon, Callopistes and Tupinamhis had more than 60
caudal vertebrae, and that the genera Teius, Crocodilurus and Dracena
had fewer. The counts presented here (62 to 72) conform with the
arrangement presented by Presch.

ACKNOWLEDGMENTS

I thank Torsten Duren, Richard Frysinger, Todd Gorden, Robert
Howard, Tommy Jenkins, Regina Psencik, Yale Swatzell and Henry
Wood for their assistance in clearing and staining specimens. I am
indebted to James R. Dixon for his useful suggestions throughout the
study, and to Douglas Albaugh, James R. Dixon, Robert A. Thomas,
John A. Weist, and Henry Wood for their critical reviews of the
manuscript.

Chester O. Martin kindly provided the illustration for Figure 3.

This study was supported by research funds from the Texas
Agricultural Experiment Station, Project 1678, and by the Department
of Wildlife and Fisheries Sciences, Texas A&M University. This paper
is contribution number TA 11242 of the Texas Agricultural
Experiment Station.

LITERATURE CITED

Cagle, F. 1946. Tail loss and regeneration in a Pacific island gecko. Copeia 1946:45.

Cope, E. 1892. The osteology of the Lacertilia. Proc. Amer. Phil. Soc. 30( 1 38): 1 83-22 1 .
Cope, E. 1900. The crocodilians, lizards, and snakes of North America. Ann. Rept. U.
S. Nat. Mus. 1898:153-1270.

Etheridge, R. 1967. Lizard caudal vertebrae. Copeia 1967:699-721.

Hoffstetter, R., and J. Gasc. 1969. Vertebrae and ribs of modern reptiles, p. 201-310. In
C. Gans (ed.), Biology of the Reptilia, Vol. I. Academic Press, New York, NY.

Moffat, L. , and A. Bellairs. 1964. The regenerative capacity of the tail in embryonic and
post-natal lizards (Lacerta vivipara Jacquin). J. Embryol. Exp. Morph. 12:769-786.
Pianka, E. 1970. Comparative autecology of the lizard Cnemidophorus tigris in different
parts of its geographic range. Ecology 51:703-720.

Pratt, C. 1946. The plane of fracture of the caudal vertebrae of certain lacertilians. J.
Anat. 80:184-188.

Presch, W. 1974. Evolutionary relationships and biogeography of the macroteiid lizards
(Family Teiidae, Subfamily Teiinae). Bull. S. California Acad. Sci. 73:23-32.

Reese, A. 1923. The osteology of the tegu, Tupinamhis nigropunctatus. J. Morphology
38:1-17.

Romer, A. 1956. Osteology of the reptiles. Univ. Chicago Press, Chicago, IL.

Scudday, J. 1971. The biogeography and some ecological aspects of the teiid lizards
( Cnemidophorus ) of Trans-Pecos Texas. Ph.D. thesis, Texas A8cM University, College
Station, TX.

Taylor, W. 1967. An enzyme method of clearing and staining small vertebrates. Proc. U.
S. Nat. Mus. 1 22(3596): 1 - 1 7.

Zug, G., and R. Crombie. 1970. Modifications of the Taylor enzyme method of clearing
and staining for amphibians and reptiles. Herpetol. Rev. 2:49-50.

RANGE EXTENSION FOR ARCIDENS CONFRAGOSUS
(MOLLUSCA: BIVALVIA: UNIONIDAE)

IN SOUTHWESTERN TEXAS

by CHARLES M. MATHER

Department of Science and Mathematics

University of Science and Arts of Oklahoma

Chickasha , OK 73018

ABSTRACT

The known range of the freshwater mussel Arcidens confragosus (Say) is extended
southwestward to the San Marcos River drainage in the Guadalupe River system.

Arcidens confragosus (Say) is a moderate sized freshwater mussel that
has been reported from rivers draining into the Gulf of Mexico from
Alabama to Texas. It is generally “found in a sand or mud bottom in
sluggish water a few feet deep” (Johnson 1980). Shira (1913) reported
A. confragosus from Caddo Lake on the Louisiana /Texas border.
Strecker (1931) recorded Texas specimens from Skull Creek, Trinity
River, Elm Fork of Trinity River, Sabine River, Neches River,
Kickapoo Creek, Buffalo Bayou, West Yegua Creek, San Jacinto River,
Poe Lake, Angelina River, Chambers Creek, Mussel Shoal Creek and
Colorado River. Johnson (1980) listed it from Texas’ Sabine River,
Neches River, Trinity River, San Jacinto River, Buffalo Bayou, Brazos
River and Colorado River. Lake LBJ on the Colorado River was
reported as a new west-most record for the species by Murray (1972).
Finally, the latest report of localities in Texas (Clarke 1981) includes
the Sabine River, Neches River, Angelina River, Trinity River, San
Jacinto River, Buffalo Bayou, Brazos River, Navasota River and
Colorado River. In all of these reports, the Colorado River system is
regarded as the western limit of distribution in Texas.

Collections made during the drought of 1980, when many Texas
rivers and lakes were at very low levels, produced a number of A.
confragosus from various sites around the state. Among these
specimens were some found stranded on the mud of an ox-bow lake
at Palmetto State Park, Gonzales County. This site is on the San
Marcos River, which is part of the Guadalupe River drainage.

Of 11 specimens found, ten were freshly dead paired valves and one
was living. In all, the periostracum was dark greenish /brown to black.
The nacre was bluish white with some irridescence on the posterior
half. The teeth were typical of Arcidens, the pseudocardinals

The Texas Journal of Science, Vol. XXXVII, No. 1 , August 1985

50

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Table 1. Measurements of ten A. confragosus specimens collected from the ox-bow lake
at Palmetto State Park, Gonzales County, Texas.

Range (mm)

Average (mm)

Length

64.6 - 92.5

80.9

Height

45.0 - 63.0

55.3

Width

29.7 - 42.7

37.1

compressed with the posterior tooth of the left valve being flared and
arched. The lateral teeth were vestigial. External sculpture of the shell
consisted of nodulous umbos, vertical zig-zags on the disc and fine
radiating ridges extending toward the posterior wing. All sculpture
was rather subdued and became less distinct toward the margins, a
feature which characterizes the subspecies A. c. jacintoensis Strecker,
1931. Table 1 summarizes the measurements of ten of the specimens.

The Palmetto State Park record extends the range of A. confragosus
westward by an additional major river system, with only the small
Lavaca system discharging into the Gulf of Mexico between the
mouths of the Colorado and Guadalupe rivers.

Specimens are deposited in the author’s collection (CMM 52-2794),
the University of Science and Arts of Oklahoma Mollusk Collection
(USAO 53-427) and the Ohio State University Museum of Zoology
(OSUM 1980:687).

Other bivalves collected with Arcidens confragosus in the ox-bow
lake were Anodonta grandis, A. imbecillis, Amblema plicata,
Carunculina parva, Lampsilis teres and Corbicula manilensis. One
fragment seems to be Cyrtonaias tampicoensis.

I thank Dr. David H. Stansbery for verification of identification of
the Arcidens specimens.

LITERATURE CITED

Clarke, A. H., Jr. 1981. The Tribe Alasmidontini (Unionidae: Anodontinae), Part I:

Pegias, Alasmidonta, and Arcidens. Smith. Contr. Zool. 326:1-101.

Johnson, R. I. 1980. Zoogeography of North American Unionacea (Mollusca: Bivalvia)
north of the maximum Pleistocene Glaciation. Bull. Mus. Comp. Zool. 1 49(2): 1-1 89.
Murray, H. D. 1972. Fresh-water mussels of Lake LBJ, Texas. Amer. Malacol. Union
Bull. 1971:36-37.

Shira, A. F. 1913. The mussel fisheries of Caddo Lake and the Cypress and Sulphur
rivers of Texas and Louisiana. U. S. Bur. Fish., Econ. Circ. 6:1-10.

Strecker, J. K., Jr. 1931. The distribution of the naiades or pearly fresh-water mussels
of Texas. Baylor Univ. Mus., Spec. Bull. 2:1-71.

NONGEOGRAPHIC VARIATION, REPRODUCTION,

AND DEMOGRAPHY IN THE TEXAS KANGAROO RAT,
DIPODOMYS ELATOR (RODENTIA: HETEROMYIDAE)

by WM. DAVID WEBSTER and J. KNOX JONES, JR.

The Museum and Department of Biological Sciences

Texas Tech University

Lubbock, TX 79409

ABSTRACT

Nongeographic variation, reproduction, and demography were examined in specimens
of the Texas kangaroo rat, Dipodomys elator, from Hardeman County, Texas. Males
were significantly larger than females in 9 of 13 external and cranial measurements. Molt
occurs annually. Reproductive activity is not restricted temporally, although young-of-
the-year are more abundant from May through October. Postnatal growth is rapid, and
sexual maturity is attained at an early age.

INTRODUCTION

The Texas kangaroo rat, Dipodomys elator, is known only from
nine counties in north-central Texas (Martin and Matocha 1972;
Cokendolpher et al. 1979), except that one specimen was taken in
adjacent Oklahoma (Comanche County) near the turn of the century
(Bailey 1905). D. elator apparently is restricted to habitats with clay-
containing sandy-loam soils that support buffalo grass and mesquite
(Dalquest and Collier 1964; Roberts and Packard 1973). Because of the
limited geographic distribution of this rat and its protected status by
the State of Texas, relatively few specimens have been collected over
the years. Consequently, virtually nothing is known about the limits
of morphological variation, reproduction, or demography in D. elator,
which were the topics of this investigation.

MATERIALS AND METHODS

One-hundred and sixty-four specimens of D. elator were collected
from the ranch of Claude Holcomb, approximately 5 km NE Quanah,
Hardeman Co., Texas, between July 1969 and July 1970. Specimens
examined from that place were listed by Martin and Matocha (1972);
all are deposited in The Museum, Texas Tech University (TTU).

We analyzed the following 13 dimensions for each individual: total
length, length of tail, length of hind foot, length of ear, weight,
greatest length of skull, least interorbital breadth, breadth across
zygomatic plates, breadth across auditory bullae, nasal length, least

The Texas Journal of Science, Vol. XXXVII, No. 1, August 1985

52

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

depth of rostrum, maxillary alveolar length, and mandibular alveolar
length. External measurements were recorded in millimeters (mm)
directly from specimen labels, as was weight (in grams) if available.
Cranial and mandibular measurements (mm) were taken with dial
calipers, in accordance with the methods described by Setezer (1949),
Likicker (1960), and Nader (1978); length of nasals was measured with
the aid of a stereoscopic dissecting microscope. The interparietal was
divided in some individuals, so the morphology of the interparietal
region was mapped for all specimens. Reproductive information was
recorded directly from specimen labels. Patterns of molt were mapped
for all specimens in the process of pelage replacement.

Statistical analyses were performed on the IBM 360/50 computer at
Texas Tech University using SAS programs (Barr et al. 1976).
Univariate analyses yielded standard statistics (mean, range, standard
deviation, standard error of the mean, variance, and coefficient of
variation), and a single-classification analysis of variance (ANOVA)
was used to compare two or more samples for significant differences
(E-test, significance level 0.05) among or between sample means. If
means were found to be significantly different, the Sums of Squares
Simultaneous Test Procedure (SS-STP) was used to determine
maximally nonsignificant subsets.

Each specimen was placed in one of four age classes (modified after
Lidicker 1960; Genoways and Jones 1971; Best and Schnell 1974; Nader
1978) as follows (representatives of which are shown in Fig. 1):
Juvenile — deciduous premolar still present on one or both sides above
or below. Subadult — permanent dentition complete; p4 unworn, lophs
still clearly evident; ml with distinct reentrant notch on labial margin.
Adult — all cheekteeth slightly worn but p4 with distinct lingual
reentrant notch separating protoloph from metaloph and usually small
labial notch as well. Old adult — all cheekteeth with distinct wear; p4
subquadrangular to subcircular in occlusal view, lacking labial and
lingual notches.

NONGEOGRAPHIC VARIATION
Secondary Sexual Variation

Analysis of variance was used to test for significant differences
between the sexes for each measurement; only individuals classified as
old adults were used in this analysis. Males were significantly larger
than females in 9 measurements, and averaged larger than females in
2 others (Table 1). Females averaged larger than males in 2
measurements (length of tail and length of ear), but neither differed
significantly between the sexes. The interparietal was divided in 57%
of the females and 42% of the males examined, but the difference

BIOLOGY OF THE TEXAS KANGAROO RAT

53

Figure 1. Representative left mandibular rami of Dipodomys elator in each age class,
showing varying amounts of wear on teeth. From left to right: juvenile (TTU 13529),
subadult (TTU 24754), adult (TTU 24759), and old adult (TTU 1 1443).

between the sexes was not significant (x2-value 3.42, P=0.07, 1 df).
However, because of the discrepancy in size, the sexes were treated
separately in subsequent analyses.

Variation with Age

ANOVA and SS-STP were used to test for significant differences
among the age classes in 13 external and cranial mesurements for
males and females. The results were similar for both sexes (Table 2).
Adults and old adults were significantly different in only 4 of 13
measurements for males and only 1 of 13 for females; subadults and
adults were significantly different in 2 of 13 for females; and between
juveniles and subadults, 11 of 13 for both males and females.
Subadults and old adults were significantly different in 7 of 13
measurements for males and 8 of 13 for females. Therefore, we regard
juveniles and subadults as separate age classes, but would be inclined
to treat adults and old adults together in future systematic analyses.

We could not determine the sequence of tooth replacement in D,
elator. All juvenile specimens available to us had permanent incisors,
and 9 of 1 1 juveniles possessed the full complement of deciduous
premolars. One rat (TTU 11431) collected in September had shed most
of the upper right dp (a few small spicules remained), and both lower

54

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

Table 1. Secondary sexual variation of Dipodomys elator. Statistics include mean ± 2
standard errors, (sample size), (range), and CV. Only old adults (age class 4) are
included. Measurements in mm except for weight (g).

Measurement

Males

Females

F

P

Total length

313.74+4.69(43)

312.06+3.83(31)

0.27

(274-345)4.90

(286-333)3.42

0.6022

Length of tail

184.76+4.08(42)

184.84+2.84(31)

0.00

(135-205)7.15

(162-195)4.28

0.9771

Length of hind foot

46.35+0.39(43)

45.55+0.36(31)

8.53*#

(43-49)2.74

(44-47)2.18

0.0047

Length of ear

13.67+0.30(43)

13.77+0.26(31)

0.23

(10-16)7.26

(13-15)5.21

0.6351

Weight

85.86+3.18(40)

77.43+2.96(29)

14.00***

(50.0-102.2)11.71

(54.2-93.5)10.28

0.0004

Greatest length of skull

40.53+0.26(42)

39.73+0.36(27)

13.95***

(38.7-42.6)2.08

(38.4-41.2)2.33

0.0004

Least interorbital breadth

13.20+0.16(41)

12.91+0.18(29)

5.90*

(12.2-14.4)3.87

(11.8-13.9)3.75

0.0178

Breadth across zygomatic plates

23.85+0.22(40)

23.26+0.22(29)

14.25***

(22.1-25.0)2.87

(22.3-24.7)2.52

0.0003

Breadth across auditory bullae

25.07+0.18(43)

24.56+0.21(31)

14.03***

(23.6-26.2)2.32

(23.6-25.9)2.37

0.0004

Nasal length

15.80+0.17(42)

15.54+0.23(27)

3.36

(14.5-17.2)3.44

(14.3-16.4)3.77

0.0712

Least depth of rostrum

7.62+0.07(42)

7.49+0.10(30)

5.47*

(7.1-8.1)2.87

(6.9-8.0)3.72

0.0222

Maxillary aveolar length

5.39+0.06(43)

5.29+0.07(31)

4.69*

(5.0-5.8)3.43

(4.7-5.6)3.90

0.0336

Mandibular aveolar length

5.29+0.05(41)

5.19+0.07(31)

6.20*

(5.0-5.6)2.91

(4.7-5.5)3.97

0.0152

milk premolars were fractured by the erupting permanent teeth.
Another individual (TTU 13538) taken in November had shed both
lower deciduous premolars, and the upper left dp as well. In subadult
D. elator, the lower permanent premolars usually evidenced greater
wear than upper premolars, so the lower premolars may reach the
occlusal plane earlier in ontogeny than those in the upper toothrow.

Juvenile pelage differed from that of older individuals. In some
specimens, juvenile pelage appeared slightly darker than adult pelage
mid-dorsally and on the flanks, because the tips of individual hairs
were more heavily pigmented. In others, however, no difference was
evident dorsally in color of juvenile and adult pelages. The ventral fur
of juveniles and adults were indistinguishable in color. However,
juvenile pelage was finer than adult pelage and more sparsely
distributed over the body, particularly on the belly. Similar
observations have been made for Dipodomys phillipsii by Genoways
and Jones (1971).

BIOLOGY OF THE TEXAS KANGAROO RAT

55

Individual Variation

Coefficients of variation (CV) for cranial measurements ranged from
2% to 4% in both sexes for old adults, whereas CV’s for external
measurements were somewhat larger (Table 1). Males were more
variable than females in external measurements, but females tended to
be more variable in most cranial dimensions. The CV’s are well within
the range for rodents of similar size (Long 1968, 1969).

Morphology of the interparietal was extremely variable in our
sample of D. elator. Although a single undivided interparietal is
typical in most species of Dipodomys, we examined individuals of D.
elator with as many as four bones occupying the area between the
posteromedial border of the parietals and the anteromedial border of
the occipital, and in two Texas kangaroo rats (TTU 11780 and 24775)
the interparietal was absent. The most common anomaly was a
completely or partially divided interparietal, the bones on either side
of the midline being approximatley equal in size. Several specimens
had small- to medium-sized bregmatic bones located along the sutures
between the parietals, occipital, and interparietal.

Dental anomalies were found in three kangaroo rats. The labial
reentrant notch of the lower right premolar was absent in one
specimen (TTU 11442), but a notch of about equal size was present
on the posterior border of that tooth. Two specimens had supernumary
teeth. One (TTU 24727) had an extra lower molar posterior to m3 on
both dentaries, and another (TTU 24769) had an additional upper
right molar posterior to M3; all supernumary teeth were molariform
in shape but reduced in size.

Seasonal Variation

Molt apparently occurs once a year in D. elator. Juvenile and
subadult specimens in the process of molt were taken in the months
from August through November, whereas molting adults and old
adults were captured in all months from April through September, and
in November and January. Our examination of pelage patterns
suggested that molt in all age classes begins on the nose and between
the shoulders, and proceeds anteriorly and posteriorly as well as
ventrally from these places. The pelage on the head, back, cheeks,
rump, and sides is replaced next, and the fur on the venter and flanks
is replaced last.

The pattern of pelage replacement in D. elator is similar to that
described for D. spectabilis (Nader 1978). The occurrence of an annual
molt in D. elator, as opposed to the semiannual molt in D. phillipsii
(Genoways and Jones 1971), is like that of D. merriami (Lidicker
1960), D. microps (Hall and Dale 1939), and D. spectabilis and D.
deserti (Nader 1978).

56

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Table 2. Age-associated variation in external and cranial dimensions in males and
females of Dipodomys elator. Lines to the right of sample means connect maximally
nonsignificant subsets (by SS-STP) at the 0.05 level of probability for juvenile (1),
subadult (2), adult (3), and old adult (4) age classes. Measurements in mm except for
weight (g).

Females

Males

Age

Sample

size Mean

SS-STP

Age

Sample

size Mean

SS-STP

TOTAL LENGTH

4

31

312.06

4

43

313.74

3

16

300.88

2

7

310.71

2

12

294.67

3

19

302.83

1

5

257.80

1 i

5

264.80

LENGTH OF TAIL

4

31

184.84

4

42

184.76

3

16

178.50

2

7

184.29

2

12

176.75

3

19

177.74

1

5

152.40

1 1

5

143.00

LENGTH OF HIND FOOT

4

31

45.55

4

43

46.35

3

19

45.16

2

7

46.29

2

12

45.08

3

20

45.75

1

6

43.17

1 1

LENGTH OF EAR

5

44.40

4

31

13.77

2

7

13.71

2

12

13.58

4

43

13.67

3

19

13.32

3

20

13.60

1

6

12.00

1 1

WEIGHT

5

11.00

4

29

77.43

4

40

85.86

3

18

74.30

3

20

76.92

2

9

67.94

2

7

71.93

1

5

43.40

1 1

4

51.88

GREATEST LENGTH OF SKULL

4

27

39.73

4

42

40.53

3

18

39.33

3

19

39.57

2

11

38.74

2

7

38.99

1

5

35.30

; i

4

36.43

REPRODUCTION AND DEMOGRAPHY

Little has been published on the natural history of D. elator, save
the comments by Chapman (1972), Dalquest and Collier (1964),
Packard and Roberts (1973), and Roberts and Packard (1973) that
address certain aspects of behavior and ecology. We discuss here two
other topics — reproduction and demography — that are poorly known
in this species.

BIOLOGY OF THE TEXAS KANGAROO RAT

57

Table 2. (Continued)

Females

Males

Sample

Sample

Age

Size

Mean

SS-STP Age

size

Mean

SS-STP

LEAST INTERORBITAL BREADTH

4

29

12.91

4

41

13.21

1

3

19

12.66

3

18

12.57

2

10

12.36

2

7

12.41

1

6

11.37

1 1

5

11.80

29
18
12

6

31

19

12

6

28

18

11

5

30
18
11

5

31
12

6
19

12

6

BREADTH ACROSS ZYGOMATIC PLATES
23.26 | 4 40

22.51 | 3 20

21.50 | 2 7

19.33 | 1 3

BREADTH ACROSS AUDITORY BULLAE

24.56

24.13

23.61

21.95

15.54

15.16

14.82

12.88

4

3
2

I 1

NASAL LENGTH

4
3
2

I 1

43

20

7

4

42

19

7

5

LEAST DEPTH OF ROSTRUM
7.49 4 42

7.37 3 19

7.23 2 7

6.66 | 1 5

MAXILLARY ALVEOLAR LENGTH

5.29

5.28

5.23

5.20

43

5

7

20

MANDIBULAR ALVEOLAR LENGTH

5.48

5.37

23.85
22.59
21.63
19.77

25.07

24.18

23.89

22.55

15.80

15.27

14.86
13.04

7.62

7.46

7.27

6.78

5.39
5.36
5.36
5.22

5.50

5.40

4

31

5.19

4

41

5.29

3

19

5.15

3

20

5.20

Pregnant females, carrying an average of three embryos, were
collected in the months of February, June, July, and September. Males
with enlarged testes were taken in May, June, July, and October.
Evidently, D. elator matures rapidly because subadult females collected
in July and September were pregnant, and a juvenile male taken in
October had enlarged testes. Early attainment of sexual maturity also

58

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

occurs in D. microps (Hall and Dale 1939), D. heermanni (Fritch
1948), D. merriami (Reynolds 1960), D. deserti (Nader 1978), and D.
ordii (Blair 1943).

Because of the protected status of D. elator and the limited number
of specimens available to us, it was not possible to examine
demographic parameters directly. However, assuming the relative
number of individuals in each age class and month to have been more
or less independent of each other, certain aspects of the demography
of D. elator can be deduced from the specimens available. The
numbers of juvenile, subadult, adult, and old adult kangaroo rats were
recorded for all months, except March, as follows: 0, 0, 3, 7 (January);
0, 0, 1, 12 (February); 0, 0, 1, 7 (April); 2, 1, 2, 9 (May); 0, 0, 1, 11
(June); 5, 1, 9, 23 (July); 1, 6, 4, 5 (August); 1, 6, 6, 4 (September);
1, 3, 8, 3 (October); 1, 1, 2, 8 (November); and 0, 1, 2, 6 (December).
We tested to determine if age and month were independent variables
with a modified R X C contigency table, and found that they were not
independent (G-value 63.63, P < 0.005, 30 df).

We hesitate to infer too much from these data because trapping was
not standardized (procedures explained in Martin and Matocha 1972),
capability may not be equal for each age class and sex, and the effects
of immigration and emigration on population dynamics are not
known in this species. However, if we assume that males and females
of each age class were equally capable and that migration to and from
the population was negligible (or similar in magnitude), then the
number of individuals in each age class in each month can be
standardized by calculating an index of relative availability (IRA),
which we devised, as follows: IRA = (n^/naHnm/N), where nam is the
number of individuals in a given age class in a given month, na is the
total number of individuals in that age class, nm is the total number
of individuals in that month, and N is the total number of individuals
for all age classes and months.

The IRA for these data (Fig. 2) indicates a peak period of
recruitment in May and July when juveniles were relatively abundant.
Subadults were most common in August and September, and adults
were most abundant in October. Old adults have indices that were
relatively stable from November through July, but were lowest in
August, September, and October when rats of other age classes were
relatively more abundant. Martin and Matocha (1972) noted an
increase in the total number of D. elator in May (they did not define
age classes), an increase that we attribute to juvenile recruitment.

The Texas kangaroo rat is apparently reproductively active
throughout the year, but the index of relative abundance indicates that
maximum recruitment is limited to the late spring and early summer.
This period of recruitment follows perforce the time when females are

BIOLOGY OF THE TEXAS KANGAROO RAT 59

MONTH

Figure 2. Relative proportion of individuals of Dipodomys elator in each age class by

month (except March) as determined using an index of relative availability (see text).

pregnant, as well as the time spent by neonates in the nest. From these
data, it appears that adult and old adult females, pregnant in January
and February, give birth early in the year and suckle their young until
spring when juveniles first enter the trapable population. Additionally,
a second peak of reproductive activity occurs in the summer and
autumn when the rapidly-maturing subadults join the established
breeding population.

Dipodomys merriami also has an extended breeding period, and is
seasonal with two or more litters each year (Lidicker 1960; Bradley and
Mauer 1971), as does D. ordii (Blair 1943) and D. heermanni (Fitch
1948). In D. merriami, the ability to produce more than one litter each
year is related to the increased ingestion of green plants in the spring
and early summer (Reynolds 1960; Chew and Butterworth 1964;
Bradley and Mauer 1971; Van De Graaff and Baida 1973; Reichman
and Van De Graaff 1975), hence a noticeable weight gain in both sexes
during this period. D. elator consumes more grasses and annual forbs
than perennials throughout the year (Chapman 1972), but seasonal
data on food habits unfortunately are lacking.

The absence of highly synchronized molting also implies that
reproductive activity is prolonged in D. elator. Kangaroo rats born
early in the year molt in late summer and early autumn. (There
probably is an earlier maturational molt of which we are not aware.)
However, the annual molt of adults and old adults is more evenly

60

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

distributed throughout the year, indicating that a single reproductive
cycle is highly unlikely. A critical evaluation of the population
dynamics of D. elator over several years is necessary to provide more
insight into these aspects of its biology.

LITERATURE CITED

Bailey, V. 1905. Biological survey of Texas. N. Amer. Fauna 25:1-222.

Barr, A. J., J. H. Goodnight, J. P. Sail, and J. T. Helwig. 1976. A user’s guide to
SAS76. SAS Inst., Raleigh, NC, 329 p.

Best, T. L., and G. D. Schnell. 1974. Bacular variation in kangaroo rats (genus
Dipodomys). Amer. Midi. Nat. 91:257-270.

Blair, W. F. 1943. Populations of the deer-mouse and associated small mammals in
the mesquite association of southern New Mexico. Contrib. Lab. Vert. Biol., Univ.
Michigan 21:1-40.

Bradley, W. G., and R. A. Mauer. 1971. Reproduction and food habits of Merriam’s
kangaroo rat, Dipodomys merriami. J. Mammal. 52:497-507.

Chapman, B. R. 1972. Food habits of Loring’s kangaroo rat, Dipodomys elator. J.
Mammal. 53:877-880.

Chew, R. M., and B. B. Butterworth. 1964. Ecology of rodents in Indian Cove
(Mojave Desert), Joshua Tree National Monument, California. J. Mammal. 45:203-
225.

Cokendolpher, J. C. , D. L. Holub, and D. C. Parmley. 1979. Additional records of
mammals from north-central Texas. Southwestern Nat. 24:376-377.

Dalquest, W. W. , and G. Collier. 1964. Notes on Dipodomys elator, a rare kangaroo
rat. Southwestern Nat. 9:146-150.

Fitch, H. S. 1948. Habits and economic relationships of the Tulare kangaroo rat. J.
Mammal. 29:5-35.

Genoways, H. H., and J. K. Jones, Jr. 1971. Systematics of the southern banner-tailed
kangaroo rats of the Dipodomys phillipsii group. J. Mammal. 52:265-287.

Hall, E. R., and F. H. Dale. 1939. Geographic races of the kangaroo rat, Dipodomys
microps. Occas. Papers Mus. Zook, Louisiana State Univ. 4:47-63.

Lidicker, W. Z. , Jr. 1960. An analysis of intraspecific variation in the kangaroo rat
Dipodomys merriami. Univ. California Publ. Zool. 67:125-218.

Long, C. A. 1968. An analysis of patterns of variation in some representative
Mammalia. Part I. A review of estimates of variability in selected measurements.
Trans. Kansas Acad. Sci. 71:201-227.

Long, C. A. 1969. An analysis of patterns of variation in some representative
Mammalia. Part II. Studies on the nature and correlation of measures of variation,
p. 289-302. In Contributions in mammalogy (J. K. Jones, Jr., ed.), Misc. Publ. Mus.
Nat. Hist., Univ. Kansas 51 : 1-428.

Martin, R. E., and K. G. Matocha. 1972. Distributional status of the kangaroo rat,
Dipodomys elator. J. Mammal. 53:873-877.

Nader, I. A. 1978. Kangaroo rats: intraspecific variation in Dipodomys spectabilis
Merriam and Dipodomys deserti Stephens. Illinois Biol. Monogr. 49:1-113.

Packard, R. L., and J. D. Roberts. 1973. Observations on the behavior of the Texas
kangaroo rat, Dipodomys elator Merriam. Mammalia 37:680-682.

Reichman, O. J., and K. M. Van De Graaff. 1975. Association between ingestion of
green vegetation and desert rodent reproduction. J. Mammal. 56:503-506.

Reynolds, H. G. 1960. Life history notes on Merriam’s kangaroo rat in southern
Arizona. J. Mammal. 41:48-58.

BIOLOGY OF THE TEXAS KANGAROO RAT

61

Roberts, J. D. , and R. L. Packard. 1973. Comments on movements, home range and
ecology of the Texas kangaroo rat, Dipodomys elator Merriam. J. Mammal. 54:957-
962.

Setzer, H. W. 1949. Subspeciation in the kangaroo rat, Dipodomys ordii. Univ. Kansas
PubL, Mus. Nat. Hist. 1:473-573.

Van De Graaff, K. M., and R. P. Baida. 1973. Importance of green vegetation for
reproduction in the kangaroo rat, Dipodomys merriami merriami. J. Mammal.
54:509-512.

VARIATION IN VEGETATION DENSITY
AND FOREDUNE COMPLEXITY AT
NORTH PADRE ISLAND, TEXAS

by MICHAEL BLUM and J. RICHARD JONES

Department of Geography

The University of Texas at Austin

Austin , TX 78712

ABSTRACT

Variation in vegetation density and foredune complexity were evaluated for five sites
along North Padre Island, Texas. Three-dimensional plots of foredune topography with
superimposed vegetation-density isolines graphically illustrate differences, which were
statistically significant, among the five test sites. The study suggests that this variation
is the result of differences in recreational use, in particular vehicular traffic, along this
section of the Texas barrier-island system.

INTRODUCTION

Vegetation plays an integral part in the establishment and
stabilization of a barrier-island foredune system. A stable foredune
protects interior points on the island from heavy storm surges and
high winds, and constantly adjusts itself to changing physical
processes and climatic conditions (McGowen et al. 1972). Critical to
achieving this dynamic equilibrium, as Clark (1974) noted, is the
presence or absence of an associated plant community. The plant
community in effect acts as a barrier which traps eolian sediments and
holds them in place with sufficient strength to resist all but the
strongest natural forces.

The dependence of a foredune system for stabilization is well
documented. Davis (1957) and Woodhouse and Hanes (1967) were able
to stabilize foredune systems with artificial plantings along the Outer
Banks of North Carolina. Gage (1970) found vegetation to be the most
effective means for stabilizing artificially established experimental
dunes on North Padre Island, Texas. Similar results were reported by
Dodd and Webb (1975), who used vegetation to stabilize foredunes and
other shoreline features at Galveston Bay, Texas. Dahl and Goen
(1977) artificially established pioneering beach grasses such as Uniola
paniculata (sea oats) and Panicum amarum (bitter panicum) along the
Texas coast and found sand accumulation to be significantly greater
in these experimental areas than in naturally established control areas
where vegetation was less dense.

The Texas Journal of Science, Vol. XXXVII, No. 1 , August 1985

64

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Barrier islands along the Texas Gulf Coast, in particular Mustang
and North Padre islands, have in the past forty years experienced a
dramatic increase in recreational use and residential development. The
1970s alone saw an increase of 43% in the amount of traffic to and
from North Padre Island (Davenport 1980). Texas regulations have
permitted essentially unlimited vehicular access to much of the beach
area. Although by law this access is limited to the backbeach, in
practice the heaviest concentration of traffic is near “the break in slope
between the backbeach and the foredunes” (White and Morton 1978).
Unfortunately, intensive recreational use has imposed tremendous
pressures on the natural foredune system and its associated plant
community.

Godfrey and Godfrey (1973) investigated the effects of intensive use
on the natural foredune system along North Carolina’s shoreline; they
found natural vegetation succession was disrupted, and the
topographic complexity of the foredune decreased, on intensively used
areas. The effects of recreational use on dune vegetation in Britain also
have been examined (Liddle and Greig-Smith 1975; Hylgaard and
Liddle 1980). Hylgaard and Liddle (1980) monitored a newly opened
recreational area, and found a 50% decrease in vegetation density over
a 4-month time interval.

As Gage (1970) notes, historical records substantiate the opinions of
coastal geomorphologists that many island foredune systems were
much higher and wider prior to intensive recreational use. White and
Morton (1978) also surmised that recreational use controls the
establishment and zonation of the pioneering plant community. This
in turn affects the development and stabilization of the natural
foredune system.

In general, however, the exact influence of intensive recreational use
on foredune systems and their plant communities is poorly understood
along the Texas coast. The objectives of this study were to evaluate
differences in foredune height and complexity and the associated plant
community among five selected sites along Padre Island, Texas, and
to suggest a possible cause for these differences.

All five sites are located on the northern end of Padre Island (Fig.
1), which is a Holocene sand body overlying Pleistocene-age mud and
sands. Hunter et. al (1972) suggest that the northern part of the island
is generally stable with no long-term accretion or erosion. The climate
of North Padre Island is sub-tropical to semi-arid; precipitation is 66
cm (26 inches) annually. The prevailing wind is onshore from the
southeast, which provides ample energy for eolian sediment transport.

VEGETATION VS. TOPOGRAPHY OF FOREDUNES

65

Figure 1. Location of study area. Numbers on “North Padre Is.” correspond to sample
sites.

METHODS

Five sites on a 20-km section of North Padre Island (Fig. 1) were
selected at which vegetation density and foredune height were to be
measured. Two of the sites are located within the Malaquite Beach
area. These sites served as controls, as this section of the beach has
been closed to vehicular traffic since 1968. The remaining three sites
are subject to varying amounts of recreational use, including vehicular
traffic. The natural factors involved in dune formation such as wind
energy, sand supply and climatic conditions are essentially similar at
all five sites; thus, the plant community at each site should be similar
under natural conditions. By comparing the vegetation densities and
foredune heights of the five sites, we hoped to evaluate the influence
of vehicular traffic on the foredune and its plant community.

Each of the five sites comprised a 100-m-long area from a
continuous, well developed foredune system. On each site vegetation
density and foredune height were measured along five transect lines.
The transects were spaced 20 m apart and extended 5 m on either side
of the dune field. All data were collected in November 1982, following
the summer tourist season.

Vegetation density (percent area covered) was sampled systematically,
using the quadrat method described in Grieg-Smith (1957) and
Kershaw (1964). Along each transect, five one-meter-square quadrats,
equally spaced, were measured for percent area covered. The relative
contributions of each species found were also recorded. Thus, for each
site a total of 25 quadrats provided representation not only of
vegetation density but also of the distribution of various members of
the local plant community.

66

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 1, 1985

Figure 2. Three-dimensional view of dune topography at site 1. Superimposed isolines

represent vegetation density (percent cover).

Dune-height measurements were taken using methods similar to
those described in Birkmeier (1981). Height and distance from
beginning of transect were recorded at each break in slope within the
dune field to obtain an accurate cross-sectional profile for each dune
transect. Measurement tools included a fiberglass stadia rod 8 m in
height, a clinometer, and a 100-m tape measure.

Data were analyzed using the SPSS (Nie et al. 1975) and SYMAP/
SYMVU (Dougenik and Sheehan 1975) computer packages. The
elevation of the five sites are presented as three-dimensional
illustrations with super-imposed isolines representing vegetation
density, allowing a visual comparison of the differences among sites
in terms of vegetation density and dune topography. Analysis of
variance (Iverson and Norpoth 1976) was used to test if the observed
differences in vegetation densities and foredune heights were
statistically significant.

RESULTS

Site one, located 800 m south of the Bob Hall pier, was characterized
by a tightly packed, gently sloping forebeach and lacked incipient
bedforms and vegetation. The foredune system was continuous but
topographically simple (Fig. 2). Vegetation density increased leeward
across the foredune, with the community dominated by Uniola
paniculata (sea oats), Panicum amarum (bitter panicum) and Croton
punctatus (beach croton).

VEGETATION VS. TOPOGRAPHY OF FOREDUNES

67

Figure 3. Three-dimensional view of dune topography at site 2. Superimposed isolines

represent vegetation density (percent cover).

Site two was 3500 m north of the Malaquite Beach boundary. A
loosely packed, gently sloping forebeach gave way to a series of
smaller, discontinuous hummocky mounds and coppice dunes. These
bedforms were sparsely vegetated, mostly with Ipomoea pre-caprae
(railroad vine) and Sesuvium portulacastrum (sea purslane). The
foredune system was continuous and topographically complex (Fig. 3),
extending the entire length of the site and covered mostly with Uniola
paniculata, Panicum amarum and Siltelianthus argophyllas (silverleaf
sunflower).

Site three was one of the two control areas (no vehicular traffic
permitted) and was located 1 km south of Malaquite Beach’s northern
boundary. A well-developed berm, approximately 8 m from mean sea
level, was present at this site. A series of hummock mounds and
coppice dunes as well as incipient barchan dunes were developed
behind the berm. These bedforms were densely vegetated with Ipomoea
pre-caprae and Sesuvium portulacastrum, with occasional patches of
Croton punctatus (beach croton). The foredune system was continuous
and topographically complex, with some transects having more than
one peak (Fig. 4). The dunes were densely covered with a variety of
vegetation, including Uniola paniculata , Croton punctatus , Oenothera
drummondii (beach evening primrose), and Siltelianthis argophyllas .

Site four was the other control area, located 100 m north of
Malaquite Beach’s southern boundary. A well-developed berm and
bedforms similar to those at site three were present. Vegetation, typical
of the beach, consisted of Ipomoea pre-caprae , Sesuvium

68

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

Figure 4. Three-dimensional view of dune topography at site 3. Superimposed isolines

represent vegetation density (percent cover).

portulacastrum and Croton punctatus. Also similar to site three, the
foredune ridge was continuous and topographically complex (Fig. 5).
The largest dunes encountered in any of the sample areas were found
at site four with some elevations exceeding 13 m above mean sea level.
Vegetation on the dunes was very dense and highly diverse, including
Uniola paniculata, Cassia fasciculata (partidge pea), Siltelianthis
argophyllas, Spartina patens (saltmeadow cordgrass), and assorted
yellow composites.

Site 5 was located 8 km south of Malaquite Beach and 800 m north
of the “four wheel drive only” area. As at sites one and two, a berm
was absent, and the broad sandy forebeach was devoid of bedforms and
vegetation. Sand material on the forebeach was moderately well
packed. The foredune ridge was continuous and very peaked, with
steep slopes both on the windward and leeward sides (Fig. 6). Most of
the foredune system was sparsely vegetated with Uniola paniculata,
Panicum amarum and Sesuvium portulacastrum.

Table 1 presents a summary of mean vegetation density, dune
height, and species of vegetation dominant at each site. Dissimilarities
in plant density and dune height among sites were apparent. Density
values ranged from 22.4% at site one, to 52.8% at site three. Mean
foredune height (average height from the toe of the foredune to the
point where the foredune merged with the barrier flat) ranged from
4.68 m at site one to 7.68 m for site four. Table 1 clearly illustrates

VEGETATION VS. TOPOGRAPHY OF FOREDUNES

69

Figure 5. Three-dimensional view of dune topography at site 4. Superimposed isolines
represent vegetation density (percent cover).

that sites three and four, the control sites, were more densely covered
by vegetation and had higher mean dune heights than sites one, two
and five. For both vegetation density and mean foredune height, the
two control sites ranked first and second followed by the remaining

Figure 6. Three-dimensional view of dune topography at site 5. Superimposed isolines
represent vegetation density (percent cover).

70

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

Table 1. Mean vegetation density, mean dune height and dominant plants at the five
study sites. Values in parentheses following dominant plants are individual density
(percent cover) at the respective sites.

Mean Vegetation
Density (% cover)

Mean Dune Height
(meters)

Dominant Plants

Site 1

22.4

4.68

Panicum ( 7.6%)

Croton ( 3.8%)

Sea Oats ( 3.0%)

Railroad Vine ( 2.2%)

Site 2

29.0

5.10

Sea Oats ( 9.0%)
Sunflower ( 7.6%)
Composites ( 4.4%)
Panicum ( 2.6%)

Site 3

52.8

5.49

Sea Oats (14.2%)
Sunflower (11.2%)
Composites ( 3.2%)

Croton ( 5.6%)

Site 4

48.6

7.68

Sunflower (22.4%)
Spartina ( 6.4%)

Sea Oats ( 5.6%)

Railroad Vine ( 4.2%)

Site 5

27.4

5.04

Panicum (13.6%)

Sea Oats ( 5.2%)

Primrose ( 4.2%)

Croton ( 2.0%)

three sites. These relationships are more evident by comparison of the
three-dimensional plots which illustrate foredune height, topographic
complexity, and vegetation density among the five sites (Figs. 2-6). Up
to 40% of the surface area was covered by vegetation on the windward
slopes of sites three and four. Such high density values did not appear
at sites one, two and five until one crossed the peaks of those foredune
systems; and even then only rarely did vegetation density exceed 40%.
Density values up to 70% were found on the leeward sides of sites three
and four.

The question of whether the measured differences in mean foredune
height and vegetation density were statistically significant remains. As
shown in Table 2, analyses of variance yielded highly significant (P
< 0.01) F-ratios both for mean foredune height and vegetational
densities among sites. We infer that vehicular traffic was the causal
factor for the significant differences among sites for both variables.

VEGETATION VS. TOPOGRAPHY OF FOREDUNES

71

Table 2. Analysis of variance in vegetation density and dune height by site.

Source

D.F.

Sum of Squares

Mean Square

F Ratio

DENSITY BY

SITE

Between Groups

4

18722.80

4680.70

14.66**

Within Groups

120

38292.00

319.10

Total

124

57014.80

DUNE HEIGHT

BY SITE

Between Groups

4

173.06

43.26

10.19**

Within Groups

120

509.47

4.24

Total

124

682.54

**P<0.01

DISCUSSION

As Godfrey and Godfrey (1973) reported, natural plant succession as
well as the composition of the plant community in foredune areas are
altered by heavy use. Our results are consistent with this view. In
general, the two control sites, three and four, had a more diverse and
dense vegetational assemblage, with as many as 15 species present on
each. Mean vegetation density values on sites one, two and five were
also at least 20% less than at the two control sites (Table 1).

On sites one, two and five, the vegetational density isolines run
roughly parallel to the shoreline (Figs. 2,3,6). On the control sites,
three and four, no such zonation patterns are apparent (Figs. 4,5).

As demonstrated by Table 1, those sites with the greater vegetation
density also had higher mean foredune heights. This is consistent with
results reported in Dahl and Goen (1977). They found that rates of
sand accumulation increased with increase in vegetation density.

It is appropriate to mention some possible consequences of decreased
vegetation densities and mean foredune height as a result of vehicular
traffic. Increased usage of North Padre Island has occurred not only as
a result of recreational use, but also from residential and commercial
activities (Davenport 1980). Thus, an interesting paradox has
developed. In order to protect the various homes and commercial
establishments from heavy storm surges and high winds, it is necessary
to preserve the natural foredune system. However, the pressures exerted
on the natural foredune system by heavy recreational, residential and
commercial use appear to be adversely affecting the foredunes’ ability
to protect those areas further inland. Hurricane Carla in 1961 produced
storm surges ranging from 4 to 7 m above mean sea level (Wiese and
White 1980). Such a storm probably would destroy foredune systems at
sites one, two and five while doing less (but still significant) damage
to the two control sites, three and four.

72

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

ACKNOWLEDGMENTS

An anonymous reviewer is thanked for comments which improved
the manuscript. Ms. Beverly Beaty-Benadom typed and proofed the
final copy. The University of Texas at Austin is acknowledged for
support of this research.

LITERATURE CITED

Birkmeier, W. A. 1981. Fast accurate two person beach surveys. CERC Technical Aid No.
81-11. U.S. Army Corps of Engineers, Fort Belvoir, VA, 182 p.

Clark, J. R. 1974. Coastal ecosystems. The Conservation Foundation, Washington, DC,
178 p.

Dahl, B. E., and J. P. Goen. 1977. Monitoring of foredunes on Padre Island, Texas.
CERC Miscellaneous Report No. 77-8. U.S. Army Corps of Engineers, Fort Belvoir,
VA, 69 p.

Davenport, S. 1980. Texas 1980 — Year of the coast. General Land Office, State of Texas,
Austin, TX, 108 p.

Davis, J. H. 1957. Dune formation and stabilization by vegetation and plantings.
Technical Memorandum No. 101. U.S. Army Corps of Engineers, Fort Belvoir, VA,
47 p.

Dodd, J. D., and J. W. Webb. 1975. Establishment of vegetation for shoreline
stabilization in Galveston Bay. Miscellaneous Paper No. 6-75. U.S. Army Corps of
Engineers, Fort Belvoir, VA, 67 p.

Dougenik, J. A., and D. E. Sheehan. 1975. SYMAP users reference manual. Laboratory
for Computer Graphics and Spatial Analysis, Harvard University, Cambridge, MA.

Gage, B. O. 1970. Experimental dunes of the Texas coast. CERC Miscellaneous Paper
No. 1-70., U.S. Army Corps of Engineers, Fort Belvoir, VA, 30 p.

Godfrey, P. J., and M. M. Godfrey. 1973. A comparison of ecosystems and geomorphic
interaction between altered and unaltered barrier island systems in North Carolina,
p. 239-258. In D. F. Coates (ed.), Coastal geomorphology. Publications in
Geomorphology, State University of New York, Binghamton, NY.

Greig-Smith, P. 1957. Quantitative plant ecology. Academic Press, New York, NY, 198
P-

Hunter, R. E., R. L. Watson, and G. W. Hill. 1972. Padre Island National Seashore field
guide. Corpus Christi Geological Society, Corpus Christi, TX, 61 p.

Hylgaard, T., and M. J. Liddle. 1980. The effects of human trampling on a sand dune
ecosystem dominated by Empetrum nigrum. Journal of Applied Ecology 18:559-569.

Iversen, G. R., and H. Norpoth. 1975. Analysis of variance. Sage Publications, Beverly
Hills, CA.

Kershaw, K. A. 1964. Quantitative and dynamic ecology. Edward Arnold Publishers,
London, England, 183 p.

Liddle, M. J., and P. Greig-Smith. 1975. A survey of tracks and paths in a sand dune
ecosystem. Journal of Applied Ecology 12:909-930.

McGowen, J. H., L. E. Garner, and B. H. Wilkinson. 1972. The Gulf shoreline of
Texas — processes, characteristics, and factors in use. Bureau of Economic Geology,
University of Texas at Austin, Austin, TX, 27 p.

Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. H. Bent. 1975. SPSS,
statistical package for the social sciences. McGraw-Hill, New York, NY, 675 p.

White, W. A., and R. A. Morton. 1978. Land and water resources, historical changes and
dune criticality — Mustang and North Padre Islands, Texas. Bureau of Economic
Geology, University of Texas at Austin, Austin, TX.

VEGETATION VS. TOPOGRAPHY OF FOREDUNES

73

Wiese, B. R., and W. A. White. 1980. Padre Island National Seashore: a guide to the
geology, natural environments, and history of a Texas barrier island. Bureau of
Economic Geology, University of Texas at Austin, Austin, TX.

Woodhouse, W. W., and R. E. Hanes. 1967. Dune stabilization with vegetation on the
outer banks of North Carolina. Technical Memorandum No. 22, U.S. Army Corps
of Engineers, Fort Bel voir, VA.

TEMPERATURE TOLERANCE OF

FLORIDA AND NORTHERN LARGEMOUTH BASS:

EFFECTS OF SUBSPECIES, FISH SIZE, AND SEASON

by W. CLELL GUEST

Texas Parks and Wildlife Department
Fort Worth Research Unit
6200 Hatchery Road
Fort Worth, TX 76114

ABSTRACT

Heat and cold tolerance of several size groups of Florida and northern largemouth bass
[Micropterus salmoides jloridanus (FLMB) and M. s. salmoides (NLMB)] was examined
in the laboratory. Seasonal effects on lower temperature tolerance also were investigated.
All size groups of NLMB survived to both higher and lower temperatures than
corresponding size groups of FLMB. However, effects of fish size on lethal temperatures
were inconsistent between subspecies and, in the case of lower lethal temperatures, may
have been confounded with seasonal effects. Large fish of both subspecies survived to
much lower temperatures when tested in the winter than when tested in the summer.

INTRODUCTION

Florida largemouth bass, Micropterus salmoides floridanus (FLMB),
now have been introduced in many areas outside their native range
(Sasaki 1961; Addison and Spencer 1971; Buchanan 1973; Graham
1973; Stevenson 1973; Reiger and Summerfelt 1976; Hall 1977; Latta
1977), sometimes unsuccessfully (Graham 1973; Stevenson 1973; Reiger
and Summerfelt 1976; Hall 1977; Latta 1977). Recent studies suggest
that some of these stocking failures involved the FLMB’s relative
intolerance to cold (Johnson 1975; Nieman 1978; Cichra et al. 1980;
Guest in press).

Temperature tolerance of both FLMB and northern largemouth bass,
M. 5. salmoides (NLMB), might vary with fish size, but little research
has been conducted to evaluate this premise. Field observations have
generated several hypotheses. Guest (1980) thought adult FLMB were
more sensitive to cold than yearlings. Venables (1976) observed that
small NLMB were more eurythermal than larger NLMB, and others
have agreed (Stevenson 1973; Coutant 1975; Siler and Clugston 1975).
Hall (1977) felt that small FLMB were more heat sensitive than small
NLMB.

The present study examined the effects of fish size on the upper and
lower lethal temperature tolerances of FLMB and NLMB. Due to the

The Texas Journal of Science, Vol. XXXVII, No. 1, August 1985

76

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

protracted time required for lower temperature tolerance tests (May
1978 through March 1980), seasonal effects on temperature tolerance
also were investigated.

METHODS

Experiments evaluating effects of fish size (Tables 1 and 2) and
season (Table 3) on the temperature tolerances of FLMB and NLMB
were conducted in the laboratory at Heart of the Hills Research Station
(HOHRS), Kerr County, Texas. Fish <100 mm total length (TL) were
provided by the San Marcos State Fish Hatchery, Hays County, Texas,
and maintained in 750-liter holding tanks receiving a continuous flow
of aerated spring water prior to testing. Larger fish were collected by
electrofishing and/or angling from lakes in which only one subspecies
was known to occur; afterwards, these fish were maintained in ponds
at HOHRS until testing.

Fish < 200 mm TL were tested in either 20- or 75-liter aquaria
suspended in 700-liter tanks; larger fish were tested in the 700-liter
tanks. Due to the antagonistic behavior of test fish > 100 mm TL,
aquaria and tanks were partitioned to separate fish individually. This
greatly reduced the number of fish that could be used in each test.
Each aquarium usually held four fish (same subspecies) and the 700-
liter tanks each held eight fish (four fish of each subspecies). Whenever
possible, eight fish of each size group from each subspecies were used
in each test. Therefore, replicate aquaria or tanks were used, and
replicates were combined in data analyses. Four fish of each subspecies
and size group were maintained at acclimation temperature as controls
during testing.

Water temperature in tanks was thermostatically controlled with
heaters and chillers. Temperature-change rates were controlled by
electrically timed, gear-driven thermoregulators (modified from Abell
et al. 1977). Laboratory lighting was regulated with timers to provide
a 12h:12h photoperiod. Fish were fed live fathead minnows,
Pimephales promelas , or goldfish, Carassius auratus, during
acclimation and testing. Test containers were cleaned daily.

During the acclimation process, water temperatures were adjusted
from ambient to acclimation values at a rate of 1 C/day. Fish were
held at acclimation temperature at least 2 weeks before testing. Fish of
all size groups were acclimated at 10 or 20 C for lower and 30 C for
upper lethal temperature tests. In upper lethal tests, one group of
small fish from each subspecies also was acclimated at 36 C (Table 2).
After acclimation was assumed to be complete, test fish were exposed
to a temperature change of 1 C/day until death. The number
surviving was recorded at each 0.5-C interval.

Table 1. Results of lower lethal temperature tolerance tests with four size

TEMPERATURE TOLERANCE OF LARGEMOUTH BASS

-< 0-1 ©
o £ . sp

C _ <U
£ © £
be ©

.§ V •§

s> t g>

c/3

gj —

2 e c

6 g 2

c sS s

s- re

^ -Q

be

cSc

oj

-O © >

c c
re be

W

So^

2 c
J w -d

U<

©

4J

c

re ©

^ U £
c o 8
E <” «u
T3 S3

° g *
1/3 _
a o 5
3 ^ re
o w -a
re

-a £
<u c
re tl

£ S3
© ©
u w

u . „

re co

be

S. I

ll

>. </3

re ^

\ 3

- 1?

«E2

<u re
2 <D

_ 're 2"
V> -£ £
§ 3 a

<u

¥-s J

” X! V

C/3 C

co re T3

S o c
© <© re

8
S 3

<U

- 3

re re

s-.

<u

a

£

qj

<u ^

3 s £

re J§ rS'

a 113 £ 2

C S-.

| ■“ 1

w >-
c ©
<u
u

re

r<u e
H o

oo re"

© ©

1 +

1 ^

ere co

© <30

■ ©

1 ©

ere ere

<30

© ©

©

+1+1

1 +1

1 1

1 1

© t>»

©

— I ere

<30

© ©
™ ere

© ©

©

©

© ©

© ©

1 s

1 ©

re" ©

— ’ ©

ere

© ©

r- ©

© ©

© ©

© ©

ere ©

re"

©

©

©

©

c

be

u

3

3

re

qj

<

s

c

re

s

"3

©

<u

Uh

0

0

© ©
ere ere

© ©

© ©

© so

© ©

© ©

© ©

© t'-

06 G

© ©

© ©

00 SO

ere ©

+1+1

+1+1

+1+1

+1+1

re" re"

© t'*

© ©

© ©

— < SO

re" ©

r-» c—

ere so

ere <—i

© so

© ©

+1 +1

+1+1

+1+1

+1+1

ere ere

ere

— ©

© C"

<30 <30

00 ©

t" !>•

© ©

00 00

«W

Eg

Ss

+ +

d +

d H

Z Uh

d +

Z Uh

Z u*

Z Ui

77

78

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

-b -c

o g
£ ^
be -b

« §

rx a

J u
Z .£

^ as

pa \
S U
+ -

«

OS O
TJ w
'£ T3
O 2

E £

O?

(A 1/5

Cl xj

5 c

O o3

w -O

qj <L>

2 £

-5 £

■5 1 ~

£ W CO

* u £
2 © 3
5 50 2

w be £

<U C 0,

■ > £

u

| 'S 5

-a S «

o — X!
<u § ,2
s a -o

2 ° §

<D £ </5

p ^ £

I o M

— ^ <D

« &■ *

ja u «2

, -r c

<u d <u

a © c

3 - g

<*- CS c

3 1 5

1 r*

* b >

CM ^ ’be

w 2 W
J JS c/5

03 " — '

<

h

£ rt

8 -S

s <u

as

<D

S-h — '

a $

os "7

u

'57

— CM CM CM

— GO

— CM

3

d d d d

d ©

d d

+1 +1 +1 +1

+ +I

+1+1

Sh

<U

00^ GO-

as cm

— oo

a

as as as as

od od

as oo

£

GO GO GO GO

GO GO

GO OO

u

<U wo 3

a.fcj 5

£ - 9

r<u C

H o

CO _ C" © !>■ OO m

as as as oo co id oo oo

GO OO GO GO SO CO SOSO

oooo OO mm

© p © ©

as as as oo

SO SO GO SO

in m mo

be

3

<

o — o — <
oooo
+1+1 +1+1
m m ■mh
oooo

+1 +1 +1 +1

OOOO 00 00 t"> 00

c-~ ©
d ©
+1+1
© — <
m as

SO GO

+1+1

© ©

d m

+1 +

© p

oo d
— o

CM CM

© ©

+1+1

P3 ro PQ ?o CQ ^

S§S§ s|

zSzE zE

pa og

si

u

TEMPERATURE TOLERANCE OF LARGEMOUTH BASS

79

-fi J2

3 -JS

o M)

s '£>

U

£ £

©

m © © ^ so; | —

bo „

5m u.

h

— < od © ■ + ed ed ' ©

- -5

nJ

— Sb

PQ C

5 c

u

2 2

+r
. ^

SO 00 eo t>» ©

c g

'5b
* w
« &

5 .3

1 «3

H K3

<u

u

«s <u

rs n3

= 1
O |

a s

3

2 ►*

bo c«

<U \

■s U

D —

2 «s

6 o

■5 xs
.«

£ u
<u

& 5

<u xs
u C
C «$
cS s-,

fc B

o .s
: *

es s-.

Sm 43

<& c
a c

S |

<u 3

--3 <u
.3 6

<v

CJ &

> 2

? o b

0-5

— w 03

-g « fc

«2 <y g

— <U

2 S "

^ 135 c«

* a -s

• TO

CO — '
M3 __,
1*3 M3 X
J fl c
a £ n
<
h

g £ «
§ jy fc
" a

U

43

3 % h

03 "T —

^ M 2

a.s E
S - 2

£ s

s •-
8 5

o

g

+1 +1 | +1 +1+1 | I

SO — © © ^

— I ed ■+ ed go

O ©

—3 04
© ©
—3 id

o

© ©
—3 ed

A ©

-3 id

© ©
id id

bo

3

<

bO

3

<

© ©
© ©

+I-H

^ GO

© ©

r~*

© ©

+1 Tl

GO ©
+ ©

©© oooo m© oo m

+1 +1

© ©
id ©
04 r-
t-~ Tf

© ©

© oi

GO C-»
+1+1
© ©
04 +
00 04
© ©

+1 +1 +1 +1 +i +1 +1 +1

03 03 t- © ©04 00 04

0404 oo© r- 04 ©in

—i GO GO CO so

P Pffl PQ on PQ eQ M efl

£§ S§ si si

^ J

80

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

All upper lethal tests were conducted during the summer; however,
lower temperature tests were more protracted, owing to limitations of
equipment (chiller) and fish availability. Therefore, as a test of
possible seasonal effects on lower lethal temperature tolerances, a
group of large fish was subjected to the temperature decline both in
the summer and in the winter (Table 3).

Data were analyzed using analysis of variance (ANOVA) techniques.
Because all fish did not die in some lower temperature tolerance tests
(Table 1), the ANOVA for these tests analyzed percentage mortality of
each subspecies at 1.5 C, the lowest test temperature common to all
size groups. For each test, the temperature at which 50% of the fish
died was calculated by regression analyses (after arcsin \f%
transformation of the data) for determination of lower and upper
lethal temperatures (LT50) as defined by Otto and Rice (1977).
Covariance analyses were used to examine the effects of subspecies and
fish size on LT50 values. All statistical evaluations were made at the
0.05 significance level.

RESULTS AND DISCUSSION

No FLMB or NLMB mortalities occurred in controls during lower
lethal temperature tests. Among cold-treated fish, mortality usually
began when water temperatures declined below 5 C, but results
probably depended on season. ANOVA showed no differences in
mortality between acclimation temperatures (F=0.01; df=l,3); so, these
data were combined (Table 1). There were significant differences
between subspecies (F=28. 18; df=l,3): in all size groups, NLMB
exhibited greater tolerance to low water temperatures than FLMB.

Generally, as fish size increased, resistance to low water temperatures
increased for both subspecies (Table 1). However, smaller fish were
tested during the summer, while larger fish were tested during the
winter. Seasonal conditions probably influenced these results more
than fish size: adults of both subspecies tested in the summer died at
less extreme low temperatures than similar-sized fish tested in the
winter (Table 3).

Only three size groups of each subspecies were tested for lower
temperature tolerance during the same season (summer); all size groups
were acclimated at 10 C. Analyses indicated inconsistent size effects
(Table 3). The two smallest size groups of NLMB appeared more
resistant than adult NLMB and more resistant than all size groups of
FLMB. The middle size group of FLMB was more sensitive to low
water temperatures than other FLMB size groups, but this difference
was not significant (F=2.09; df=2,21).

TEMPERATURE TOLERANCE OF LARGEMOUTH BASS

81

Calculation and comparison of LT50 values for lower lethal
temperature tests were complicated also by incomplete mortalities or
no mortalities in some tests and seasonal variations during testing
(Table 1). There were no significant differences (F=0.29; df=4,20) in
slopes of LT50 regressions for tests in which total mortality occurred;
the pooled slope (b= 19.64) was used to calculate LT50 values in tests
where total mortality did not occur.

Previous studies comparing the lower temperature tolerance of
FLMB and NLMB have used short-term tests comparing effects of
single-step decreases in temperature, usually on one size group of fish
(Nieman 1978; Cichra et al. 1980). Nieman (1978) and Cichra et al.
(1980) found FLMB (150-200 mm TL) to be more susceptible to cold
shock than similar-sized NLMB, while Hart (1952) found FLMB and
NLMB (25-90 g) acclimated to 20 C to have lower incipient lethal
temperatures of 5.2 and 5.5 C, respectively. Because the above studies
were short-term, seasonal variation did not affect the comparisons
being made. However, these studies usually were conducted when fish
were available rather than purposefully testing during a particular
season. The authors most probably assumed (as I had done at the
outset of the present study) that the acclimation process would make
tests comparable. However, as shown in Table 3, lethal temperature
values (mean and LT50) varied with season even though fish size and
acclimation conditions were similar. The 2+ week period of
acclimation to temperature and the 12:12 photoperiod apparently were
insufficient to negate seasonal effects. Others (Hart 1947; Black 1953;
Fry 1967; Kowalski et al. 1978) also reported seasonal variations in the
lethal temperatures of several fish species. Kowalski et al. (1978) found
variation in lethal temperatures within each of three fish species that
had been acclimated to constant temperatures for 4 weeks. He
suggested seasonal variation should be considered in laboratory studies
on the temperature tolerance of fish. The present study is a good
example of the difficulties in interpreting results from tests conducted
during different seasons.

Upper lethal temperature tests were not complicated by seasonal
conditions because all fish were tested in the summer (Table 2). No
mortality occurred in controls during testing. Once fish began dying
in treatment tanks, mortality was rapid, with total mortality usually
occurring within a range of 1 or 2 Celsius degrees. There were
significant differences in lethal temperatures between subspecies
(F=8.82; df=l ,45) and among size groups (F-18.31; df=2,45). Mean
upper lethal temperatures were higher for NLMB than for FLMB in
all size groups. LTso’s also were higher for NLMB than for FLMB in
all size groups, but these differences were not significant (subspecies
F= 1 .5; df=l ,3; size group F=0.5; df=3,3). Computed upper lethal

82

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

temperatures (mean and LT50) differed by less than one Celsius degree
between subspecies for each size group tested. Nevertheless, the
difference between subspecies was consistent and supports Hart’s (1952)
conclusion that FLMB are less heat tolerant than NLMB.

Samples from the smallest size group of test fish were acclimated at
30 and 36 C for over 30 days. The latter acclimation temperature
approaches the “ultimate” upper lethal temperature of 36.5-36.8 C
reported for NLMB (Hart 1952; Coutant 1975) and surpasses the upper
incipient lethal temperature of 33.7 C for 30 C-acclimated FLMB
determined by Hart (1952). McCormick and Wegner (1981) noticed
biologically significant differences in mortality rates between northern
and southern stocks of NLMB juveniles (0.6-1.05 g) exposed to 36 C.
Almost all of the northern stock (from Wisconsin and Minnesota) were
dead after 2 weeks, while mortality rates decreased in the southern
stock (from Tennessee) after the second and third weeks of exposure.
The authors did not mention the season in which their test was
conducted, which could have affected the comparison, but heat
tolerances of both their stocks seem to have been less that that of the
“Texas” stocks of FLMB and NLMB in the present study.

Results of the present study support findings of others. There are
subspecific differences in the temperature tolerances of FLMB and
NLMB. However, the effects of fish size on lethal temperatures were
inconsistent and inconclusive. These discrepancies could have been due
to the small sample sizes used in each test as well as the variation that
can be expected in temperature tolerance trials replicated over time.
Seasonal variation was shown to be very important in determining
lethal temperature limits and should be a major consideration when
designing experiments to determine temperature criteria for fish.

ACKNOWLEDGMENTS

Thanks are extended to Texas Parks and Wildlife Department
personnel who provided fish used in the study, statistical data
analyses, and critical review of the manuscript. Special appreciation is
expressed to W. H. Neill and R. L. Noble, Texas A8cM University,
College Station, who offered valuable suggestions and assistance. This
study was funded by the Federal Aid in Fish Restoration Project,
Texas, F-31-R.

LITERATURE CITED

Abell, P. R., L. B. Richardson, and D. T. Burton. 1977. Electronic controller for
producing cyclic temperatures in aquatic studies. Progressive Fish-Culturist 39: HO¬
ME

Addison, J. H., and S. L. Spencer. 1971. Preliminary evaluation of three strains of
largemouth bass, Micropterus salmoides (Lacepede), stocked in ponds in south

TEMPERATURE TOLERANCE OF LARGEMOUTH BASS

83

Alabama. Proceedings of the Annual Conference of the Southeastern Association of
Game and Fish Commissioners 25:366-374.

Black, E. C. 1953. Upper lethal temperatures of some British Columbia freshwater fishes.
Journal of the Fisheries Research Board of Canada 10:196-210.

Buchanan, J. P. 1973. Separation of the subspecies of largemouth bass, Micropterus
salmoides salmoides, and M. s. floridanus and integrades by use of meristic
characters. Tennessee Valley Authority, Muscle Shoals, AL, 12 p.

Cichra, C. E., W. H. Neill, and R. L. Noble. 1980. Differential resistance of northern
and Florida largemouth bass to cold shock. Proceedings of the Annual Conference
of the Southeastern Association of Fish and Wildlife Agencies 34:19-24.

Coutant, C. C. 1975. Responses of bass to natural and artificial temperature regimes, p.
272-285. In FI. Clepper (ed.), Black bass biology and management. Sport Fishing
Institute, Washington, D.C.

Fry, F. E. J. 1967. Responses of vertebrate poikilotherms to temperature, p. 375-409. In
A. Rose (ed.), Thermobiology. Academic Press, New York, NY.

Graham, L. K. 1973. The introduction of Florida largemouth bass in Missouri ponds.
Project F-l-R-22, Study No. 1-14, Job No. 2, Missouri Department of Conservation,
Jefferson City, MO.

Guest, W. C. 1980. Florida largemouth bass temperature tolerance study. F-31-R-6, Study
No. 36, Texas Parks and Wildlife Department, Austin, TX.

Guest, W. C. In Press. Survival of adult Florida and northern largemouth bass subjected
to cold temperature regimes. Proceedings of the Annual Conference of the
Southeastern Association of Fish and Wildlife Agencies, 1982.

Hall, C. R. 1977. An evaluation of Florida largemouth bass ( Micropterus salmoides
floridanus) introductions into a power plant cooling reservoir. M.S. thesis, Texas
A&M University, College Station, TX.

Hart, J. S. 1947. Lethal temperature relations of certain fish of the Toronto region.
Transactions Royal Society of Canada, Series 3, 41: 57-71.

Hart, J. S. 1952. Geographic variations of some physiological and morphological
characters in certain freshwater fish. Biological Series 60, Ontario Fishery Research
Publication 72, Toronto, Canada.

Johnson, D. L. 1975. A comparison of Florida and northern largemouth bass in
Missouri. Ph.D. dissertation, University of Missouri, Columbia, MO.

Kowalski, K. L. , J. P. Schubauer, C. L. Scott, and J. R. Spotila. 1978. Interspecific and
seasonal differences in the temperature tolerance of stream fish. Journal of Thermal
Biology 3:105-108.

Latta, W. C. 1977. Survival and growth of Florida largemouth bass in ponds and lakes
in Michigan. F-35-R-3, Study No. 27, Michigan Department of Natural Resources,
Lansing, MI.

McCormick, J. H., and J. A. Wegner. 1981. Responses of largemouth bass from different
latitudes to elevated water temperatures. Transactions of the American Fisheries
Society 110:417-429.

Nieman, D. A. 1978. Some aspects of the ecology of Florida and northern largemouth
bass in a thermally enriched reservoir. M.S. thesis, Oklahoma State University,
Stillwater, OK.

Otto, R. G., and J. O. Rice. 1977. Responses of a freshwater sculpin ( Cottus cognatus
gracilis) to temperature. Transactions of the American Fisheries Society 106:89-94.

Reiger, P. W. , and R. C. Summerfelt. 1976. An evaluation of the introduction of Florida
largemouth bass into an Oklahoma reservoir receiving a heated effluent. Proceedings
of the Annual Conference of the Southeastern Association of Fish and Wildlife
Agencies 30:48-57.

84

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Sasaki, S. 1961. Introduction of Florida largemouth bass into San Diego County. Inland
Fisheries Administrative Report No. 61-11, California Department of Fish and Game,
Sacramento, CA.

Siler, J. R., and J. R Clugston. 1975. Largemouth bass under conditions of extreme
thermal stress, p. 333-341. In H. Clepper (ed.), Black bass biology and management.
Sport Fishing Institute, Washington, D.C.

Stevenson, F. 1973. A preliminary evaluation of a Florida strain of largemouth bass,
Micropterus salmoides floridanus, in Ohio ponds. Project F-29-R-13, Job No. VI-C,
Ohio Department of Natural Resources, Columbus, OH.

Venables, B. J. 1976. Aspects of the thermal ecology of largemouth bass ( Micropterus
salmoides) in north central Texas. Ph.D. dissertation, North Texas State University,
Denton, TX.

MONODXFFRIC LAPLACE TRANSFORMS
AND FOURIER TRANSFORMS1

by TAHEREH DANESHI

Department of Mathematical Sciences
Cameron University
Lawton, OK 73505

and CHARLES R. DEETER

Department of Mathematics
Texas Christian University
Fort Worth, TX 76129

ABSTRACT

Monodiffric functions are discrete analogs of analytic functions of a single complex
variable. This paper defines a two-sided monodiffric Laplace transform, a monodiffric
Laplace transform, Z-transform, and various monodiffric Fourier transforms. The two-
sided Laplace transform is used to provide a monodiffric extension of a function defined
on the x-axis to the whole discrete plane. The basic operational properties of the
monodiffric Laplace transform are presented, and a brief table of corresponding
functions and transforms is given to facilitate the determination of both direct and
inverse transforms. With these, the solution of initial-value problems involving linear
monodiffric differential and difference equations with constant coefficients can be easily
obtained.

INTRODUCTION

The set of all Gaussian integers (the points of the complex plane
with integer coordinates) is called the discrete plane and is denoted by
G. Throughout this paper all functions will be complex-valued and
will have domains which are either the discrete plane or a subset of
the discrete plane.

Isaacs (1941) called a function f monodiffric on G if, for every z e
G,

f(?+l ) — f(7.) = f(Z+i) ~ f(z)

The class of all functions monodiffric on G is denoted by M(G).

■This paper is part of a doctoral dissertation presented by the senior author in partial
fulfillment of the requirements for the Doctor of Philosophy degree in Mathematics,
Texas Christian University, Fort Worth, TX 76129, August, 1981.

The Texas Journal of Science, Vol. XXXVII, No. 1, August 1985

86

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Given a function f monodiffric on G, a function f', the monodiffric
derivative of f, is defined by the relation

f'(z) = f(z+l) - f(z) = f(z+i) - [ (z) .

i

A finite sequence C = < zi, Z2, . . zn> in the discrete plane G is
called a discrete curve from zi to zn if | Zj+i — zj| = 1, for each j = 1,
2, . . . , n — 1.

Let f be a complex valued function defined on the discrete plane G,

and let C == <a = zo,. . . , zn = b> be a discrete curve in G from a to

b. The discrete line integral of f from a to b along C is

fW 2 f(Zk) (Zk Zk-l) , (1)

J a k= 1

where zf is the lower or left point of Zk-i and Zk. It is known (Isaacs
1941) that this integral is independent of the discrete curve in G
connecting a to b if, and only if, f e M(G).

It is easily verified that the pointwise product of monodiffric
functions is not necessarily monodiffric. Berzsenyi (1970) defined a
“double dot” integral in the discrete plane G by

f\l) ■ g({)«£ = 2 f(*t) [g(*0 - g(zk-i)] , (2)

where C = <a =z0, . . zn = b> is a discrete curve in G with Z* the
lower or left point of Zk-i and zk. If f, g e M(G), then fi f(£) : g(Qd(
is independent of the discrete curve in G connecting a to b. Also, many
basic properties of ordinary complex line integrals are valid for both
the discrete line integral (1) and the :-integral (2) (Berzsenyi 1970).

Utilizing the .-integral, Berzsenyi (1972) defines the following
convolution-type product operations:

(f*g) (z) = J f(z-£) : g(()S(,
cz

and

(f°g) (z) = (f*g) (z) + g(0)f(z),

TRANSFORMS OF MONODIFFRIC FUNCTIONS

87

where Cz is a discrete curve starting at the origin and ending at the
point z.

Isaacs (1941) defined the n-th monodiffric pseudo-power of z,
denoted here by p„(z), as follows. Let G be the discrete plane. For all
z e G,

po(z) = 1 , and pn(z)

nX

P»-l(£)S£,

1,2,

where the integral is the discrete line integral (1). Some of the basic
properties of these functions are (i) pn e M(G), for all n = 0, 1,2, . .
(ii) pn(0) = 0, for all n = 1, 2, . . (iii) pn — npn-i, for all n = 1, 2,
. . . , where pn is the monodiffric derivative of pn; and, (iv) pj(x) =
x(x— 1). . . (x— j+1), x an integer.

In general, °-powers of a monodiffric function can be defined in the
whole discrete plane. Indeed, Berzsenyi (1972) proved that

f»(g»h) - (f*g)*h = g(0) (f*h).
A simple calculation shows that

f°(g°h) - (fog)oh = f(z) (g*h) (0) = 0.

Hence, the °-product of monodiffric functions is associative in the
whole discrete plane. Therefore, the n-th °-power of a monodiffric
function f is defined to be

Isaacs (1952) defined the “exponential function,” e(z,t), by the
relation

e(z,t) = (l+t)x(l+it)y,

where z — x+iy and t is any complex number. He proved that, for a
fixed t, e(z,t) e M(G), and that e'(z,t) = t e(z,t). Also, discrete-cosine
and discrete-sine functions are defined by the relations

Sin az = e(z,ia)-e(z,-ia)

2i

Cos az = e(z,ia) + e(z,-ia)

2

88

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

TWO-SIDED LAPLACE TRANSFORM

Definition 1 : Let f(x) be a function defined for all integer values of
x. The monodiffric two-sided Laplace transform of f(x) is defined by

L2(f(x)) = F(s) =J e(— x,s) : f(x)Sx,

where e(z,t) = (l+t)x(l+it)y. Note that, as a function of the ordinary
complex variable t, e(z,t) has — 1, +i and/or 00 as zeros or singularities,
depending on z.

Duffin and Duris (1964) defined Laplace transforms for discrete
analytic functions and proved some useful properties. Those properties
are also valid for monodiffric functions with Definition 1.

Let L = _ L_. Then e(— x,s) = ( _ ! _ )x = Lx, and

s+1 s+1

L2( f(x)) = F(s) = F (_L- 1 ) = F( L) = f “ L* : f(x)Sx,

where

Lx:f(x)6x= 2 Ln [f(n) — f(n — 1 )]. (3)

n=— oo

As in discrete analytic function theory, this integral converges
absolutely when

lim sup [|f(— n)— f(— n— 1)| ]1/n<|L|
n — 00

< - ^ (4)

lim sup [|f(n)~ f(n— l)|]/n

n ^ oo

Due to the similarity of two-sided discrete Laplace transforms in
monodiffric function theory and discrete analytic function theory, the
proofs of the following two theorems are omitted.

Theorem 1\ Let f(x) and g(x) be two functions satisfying inequality
(4). Then, in the annulus of the L-plane where L2(f(x)) and L2(g(x))
exist, we have

U (f(x)) Li (g(x)) = L2 T” f(x-t) : g(t)6t^

TRANSFORMS OF MONODIFFRIC FUNCTIONS

89

Theorem 2: Let f(x) and g(x) be functions defined on the set of all
integers; then,

L2 (f(x)) = L2 (g(x)) if, and only if, f(x) = g(x) + k,

where k is an arbitrary constant.

Next, the problem of determining a function when its transform is
given will be considered. The following theorem deals with the inverse
of the monodiffric two-sided Laplace transform.

Theorem 3: Consider F(L) as a function of the complex variable L,
analytic on an annulus R centered at the origin. Then, there exists a
function f(x), — 00 < x < °°, such that L2 (f(x)) = F(L). This function
is given by

f(x) = k + _L

2 th

r _ _ dL,

J Lx+1(l— L)

r

where k is an arbitrary constant and T is any positively oriented simple
closed coutour inside R and around the origin. Proof : The Laurent
series

oo

F( L) = 2 anLn ,

with

r

converges inside R. From equation (3), it must be true that

f(n) - f(n-l) = an.

This condition is satisfied if one defines

f(x) = k +

2 7ri

r _ TC _ dL.

J Lx+I(l-L)

r

90

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Substitution of L = _ ! _ yields

s+1

f(x) = k - _L f e(x,s) F<s> ds, (5)

2niJ s

c

where C is the image of T under the transformation S = — 1.

Equation (5) is called the inversion integral for the monodiffric two-
sided Laplace transform. The value of the constant k will change
depending on whether or not the contour C encloses the point s = 0.

Corollary 1 : Let f(x) be a monodiffric function which satisfies
inequality (4). Let Z.2(f(x)) = F(s). Then f(x) is given by equation (5)
for some choice of the constant k.

Berzsenyi (1970) showed that when a function f is given on the x-
axis and the negative y-axis, it can be continued uniquely to a
monodiffric function on the discrete plane. In the following theorem,
with the aid of the monodiffric two-sided Laplace transform F(s), a
monodiffric extension for functions f(x) defined on the integers is
obtained. A similar theorem is given by Duffin and Duris (1964) for
discrete analytic functions.

Theorem 4: Let F(x) be a function defined on the integers x, — 00 <
x < o°, such that inequality (4) is satisfied. Let F(s) be the monodiffric
two-sided Laplace transform of f(x). Then f(x) can be extended to a
function monodiffric on the discrete plane by the formula

f(z) = k-J_fe(z,s)JM. ds,

2ttiJ s

c

where k is a constant and C is a positively oriented closed contour in
the region of absolute convergence of the two-sided Laplace transform
such that C does not pass through S = i. Proof : Clearly, f(z) is a
monodiffric function, and, by Theorem 2 and equation (5), f(z)|z=x =
f(x). Note that this continuation is not unique and depends on
whether or not the point i is enclosed by the contour C.

THE MONODIFFRIC LAPLACE TRANSFORM

In continuous theory, the Laplace transform of a function f(x) is
defined, for positive values of x, by Jo e_sxf(x) dx (Churchhill 1958).

Definition 2: The monodiffric Laplace transform for f(x), 0 < x <
°°, x an integer, is

TRANSFORMS OF MONODIFFRIC FUNCTIONS

91

e(— x,s) : f(x)Sx + f(0)

f(x)<5x + f(0),

where L = _

s+1

The monodiffric Laplace transform can be obtained from the two-
sided Laplace transform by defining

so that

for x = 0, 1,2,
for x = — 1 , —2, . . . ,

L(f(x)) = L2(g(x)).

The region of absolute convergence of the Laplace transform is

|L|< - r — - tt/ — ' (6)

lim sup [|f(n) — f(n — 1 )| ] /n

n — oo

Theorem 5: Let f(x) and g(x) be defined for all positive integers x.
Then, in the common region of absolute convergence of both L(f(x))
and L(g(x)),

L(f(x)) L(g(x)) = L (Jo t(x~l) ■ g(t)5t)+ g(0)L(f(x)).

OO

Proof: L( f(x)) = S Lx[f(x) — f(x— 1)], and

x=0

L(g(x))= S Lx[g(x) — g(x— 1)],
x=0

where f(— 1) and g(— 1) = 0. Then,

OO OO

L(f(x))L(g(x)) = 2 L*[f(x) — f(x 1 )] 2 Lx[g(x) — g(x— 1)].

x=0 x=0

92

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

This product can be computed as a Cauchy product of two series, since
both series converge absolutely in the common region of absolute
convergence of the Laplace transforms. Therefore, a simple calculation
shows that

£-(f(x)) L( g(x)) = L (Jo f(x-t) : g(t)St)+ g(0)L(l(x)),
which proves the theorem.

Next the inversion integral for the monodiffric Laplace transform is
considered, as well as its extension to a quadrant or half of the discrete
plane.

Theorem 6: Let F(L) be a complex valued function, analytic on the
disk |L| < k, k a positive constant. Then, there exists a function f(x),
0 < x < o°, such that L(f(x)) = F( L). This function is given by

f(x) = k+_Lr _ LT _ dL,

2th J Lx+1(1— L)

where k is a constant and T is a positively oriented simple closed
contour in the region of convergence. Proof : By Definition 2 and
equation (6), F has a Taylor series expansion:

F(L) = X a„Ln ,
n=0

where an = f(n) — f(n — 1).

The proof is completed in a manner entirely similar to the proof of
Theorem 3.

When L = _ I _ is substituted, one obtains

sT 1

f(x) = k — _L f e(x,s) ds.

2mJ s

c

Theorem 6 makes it possible to state a theorem for the monodiffric
Laplace transform similar to Theorem 4 for the two-sided monodiffric
Laplace transform.

Theorem 7: Let f(x) be a function defined for all positive integers,
such that inequality (6) is satisfied. Let F(s) be the monodiffric Laplace
transform of f(x). Then f(x) can be extended to a function which is

TRANSFORMS OF MONODIFFRIC FUNCTIONS

93

monodiffric in the first quadrant (or the half-plane x > 0) by the
formula

e(z,s) ds,

s

c

where k is a constant and C is a positively oriented closed contour in
the region of absolute convergence of the monodiffric Laplace
transform such that C avoids s = i.

PROPERTIES OF THE MONODIFFRIC LAPLACE TRANSFORM

Among the most useful properties of monodiffric transforms are the
following:

(i) L(af(x) + bg(x)) = aL(f(x)) + bL(g(x));

(ii) L(f<n’(x)) = s"L(£(x)) = [s"f(0)+s”~1f'(0)+. . .+sf(n_1,(0)];

(iii) L((f°g)(x)) = L((f*g)(x)) + g(0)L(f(x));

(iv) L((f°g)(x)) = L(f(x))L(g(x));

In these equations, a and b are complex constants and n in an integer.
Also, it is assumed that all transforms have a common disk of
convergence, |L| < k.

Property (i) is the linearity property of monodiffric Laplace
transforms. Its proof follows from the linearity properties of the
integral.

Property (ii) is one of the most important properties of monodiffric
Laplace transforms. Its proof is as follows:

L(F(x)) = f e( — x,s) : f'(x)6x + f'(0)

o

= 2 (_J_)x[f'(x)-f'(x- l)] + f'(0)

X=1 s+1

But, f'(x) = f(x+l) — f(x). Therefore,

94

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

L(f'(x)) = _L_ 2 ( — 1 — )x [£(x+l) — £(x)]
s+1 x=0 s+1

oo

= s 2 (

' 1 )x+l [f(x+l)-f(x)l

x=0

s+1

OO

' 1 )X [f(x — f(x — 1)]

= s 2 (

X=]

s+1

= s (L(f(x)) - f(0)).

That is,

L(f'(x)) = s (L(f(x)) - f(0)).

Application of the foregoing relation to f'(x) yields

L(f"(x)) = s (L(f'(x)) - f'(0))

= s (sL(f(x)) - sf(0) - f'(0))

= s2L(f(x)) - s2f(0) - sf'(O).

Using mathematical induction will lead to property (ii).

Property (iii) is a consequence of property (i) and the fact that
(f°g)(x) = (f#g)(x) + f(x)g(0). Finally, property (iv) is simply a
combination of Theorem 5 and property (iii).

Table 1 displays the monodiffric Laplace transforms of several well-
known functions, along with their regions of convergence. The
transforms are obtained by applying Definition 2 and the properties
noted above to the functions listed.

SOLUTION OF ORDINARY MONODIFFRIC DIFFERENTIAL AND
DIFFERENCE EQUATIONS BY THE USE OF THE MONODIFFRIC
LAPLACE TRANSFORM

The solution of ordinary differential equations by Laplace transform
methods can be analogized to provide monodiffric Laplace transform
methods to be applied in the solution of ordinary monodiffric
differential equations.

Consider the general monodiffric differential equation with constant
coefficients of the n-th degree,

f,n, + a„-,f,n_1,+...+ aof = g,

(7)

TRANSFORMS OF MONODIFFRIC FUNCTIONS

95

Table 1. Monodiffric Laplace transforms.

Monodiffric

Region of

Function

Laplace Transform

Convergence

_!_pk(x),x>k

1

l-L-l

<

1

k!

sk

s + 1

(e(x,a))°k

sk

l_l_l

<

1 1 1

(s - a)k

s + 1

1 + a

_J_(e(x,a) — l)°k

1

l_if

<

1 1 1

k

a

(s - a)k

s + 1

1 +a

J_(Sin «x)°k

sk

|_!_|

<

1 1 1

k v

a

(s2 + a2)k

s + 1

1 + ia

(Cos ax)°k

s2k

1 JL_|

<

1 1 1

(s2 + a2)k

s + 1

1 + ia

_J_( 1 — Cos ax)°k

1

1 1 1

<

1 1 1

a

(s + a )

s + 1

1 + ia

where oto, au, . .., an-i are complex constants, g e M(D), and primes
and parenthesized superscripts denote monodiffric derivatives. The
initial conditions are given by

f(0) = Co, f'(0) = ci

= c„-,.

By property (ii) of the monodiffric Laplace transform,

L(f,n)) = s"L(f) - s"f(0) - sn_1f'(0) sCAO),

or

L(f(n)) = s"L(f) — SnCo — Sn_1Cl — . . . — SC(n-l).

If the monodiffric Laplace transform is applied to both sides of
equation (7), the result can be written as

(s"L(f) — SnCo — Sn_1Ci — . . . — SCn-l)

+ (*n-l(sn-1f — Sn_1Co — Sn_2Ci — ... — SCn-2) + . . .

+ oo(sf — sco) + aof — L(g),

96

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

or

snL(f) ~b ftn-isn *f + ... + oasf + ftof

— /3nsn + /Jn-is" 1 +. . .+ j8is + L(g),

(8)

where

fik — Cn-k + ofn-iCn-k-i + • . . + ftkCo, for k = 1 , . . . , n.

From equation (8), it follows that

— fins" + /3n-lSn 1 +. . .+ PiS + L(g)

S" + Q!n-lSn 1 +. . . + ft lS + fto

or

L( f(x)) = ^nSn +. . .+ Pis + _ ^(g(x)) _

Sn + Q!n-lSn 1 + . . .+ fto Sn + ftn-lSn 1 + . . . + fto

The first of the functions in the immediately preceding relation can
be decomposed into powers of simple partial fractions in the form of
the transforms given in Table 1. The second fraction may be modified
similarly if the function g is known. Hence, L(f) can be expressed as
a linear combination of the transforms. The corresponding inverse
functions of the transforms can be found in Table 1, and f(x), the
solution of equation (7), can be written as a linear combination of
those inverse functions.

As an example, consider

f" + X2f = 0 ,

with initial conditions

We get

f(0) = tt , f'(0) = j8

L(f)=asT6s =F(S)
s2+X2

which has the solution

f(x) = q; Cos Ax + p/\ Sin Ax

TRANSFORMS OF MONODIFFRIC FUNCTIONS

97

We extend this solution to the first quadrant by Theorem 7. Therefore,

f(z) = k — _L_ f e(z,s) _^s±g_ ds ,

27ri Jc s2+X2

where the region of absolute convergence of F(s) is |s+l|> 1. Let C be
a closed path in that region, such that C encircles s = ±iX once and
avoids s = i. Then

res ( e(z,s) as+^ , iX ) = (<*iX+/3)e(z,iX)

V s2+X2 7 2iX

res ( e(z,s) y — iA ) = ( uiX+ff)e(z, iX)

V s2+X2 7 — 2iX

and n(C,iX) = n(C,— iX) = —1 where n(C,f) denotes the index of C
with respect to f. Therefore,

_L f e(z,s) ds = — ^_(e(z,iX)+e(z,— iX))

2tti Jc s2+X2 2

-JL (e(z,iX)— e(z,iX)) ,

2iX

J_ f e(z,s) ds = —a Cos Xz - j8/X Sin Xz

2rri s2+X2

Now,

f(0) = k - J_ f as+^ ds ,

27ri-K s2+X2

or

k =J_f as+^ ds + f(0) = — a+a = 0

2jriJc s2+X2

Finally, then

f(z) = a Cos Xz + /3/k Sin Xz

98

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

Next, we consider the linear homogeneous monodiffric difference
equation with constant coefficients,

f(x+n) + an-if(x+n — 1) . .+ aif(x+l) + aof(x) = 0.

Applying the definition of the monodiffric Laplace transform to
f(x+n), a simple calculation shows that

L (f(x+n)) = e(n,s)L(f(x)) — _5 _ |"e(n,s)f(0) +

s+1

e(n - l,s)f(l) +. . .+ e(2,s)f(n - 2) +

e(1’s> f(n - 1)]
s

where f(0), f(l),. . f(n— 1) are given initial values of f. Assuming that
all Laplace transforms exist, we can solve the resulting simple
equation for L(f(x)) and find f(x) (x > 0) as an inverse Laplace
transform.

Z-TRANSFORM DEFINITION AND RELATIONSHIP
BETWEEN LAPLACE AND Z-TRANSFORMS

Definition 3\ Let f(x) be a monodiffric function defined for x > 0.
The Z-transform of f(x) is the function

oo

Z(f) = 2 z~xf(x)

x=0

where z = _L = s + 1 . We use Z to denote the z-transform of f.

L

We can obtain the z-transform directly from the Laplace transform
as follows:

oo oo

L(f)= X Lxf(x) — L X Lxf(x)
x=0 x=0

= (1— L) X Lxf(x)
x=0

TRANSFORMS OF MONODIFFRIC FUNCTIONS

99

-i- 2 Lxf(x).
s+ 1 x=0

Therefore,

Z(f) = i±L L({)

S

or

L(l) = i_L Z(f).

Z

Note that Z is a linear operator. This operator can be used to solve
linear differential and difference equations analogous to the method of
Laplace transform.

FOURIER TRANSFORMS

Definition 4 : If f(x) is a function defined for all integers, and if
/-»|f(x)|<5x exists, the monodiffric Fourier transform of f is defined by

F(

u) =J e( — x,iu) : f(x)Sx, — 00 < u < °°,

or

F(

(1+iu) x : f(x)6x.

Let

_ ! _ = P. Then, u = ^ _ LL, and

1+iu P

F(u) = f(T£zL) = F( P) = f " P1 : f(x)6x,

or

o©

F( P)= 2 Px[f(x) - f(x-l)].

x=— O©

The integral is absolutely convergent if

lim sup [| f(— n)— £(— n— 1)| ]1/n < |P| < _ l _

n— 00 lim sup [| f(n) — f ( n — 1)| ]1/n

n— -00

(9)

100

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Equation (9) permits the determination of the function f(x) in terms
of an integral involving the transform F, as was shown in Theorem
3:

£(x) = k + J_ f _ _ dP,

2ni J r Px+1(l— P)

or

f(x) = k — _i_ f Jl iHL e(x,iu)du.

27ri^c u

The equation

e(x, — iu) = Cos ux — i Sin ux

makes possible the definition of the monodiffric Fourier Sine and
Cosine transforms.

Definition 5: Let f(x) be a function such that fo |f(x)|dx exists. Then
the monodiffric Fourier Sine transform of f is

S(f(x)) =J Sin ux : f(x)6x = Fs(u), u > 0.

Let P = _ ! _ Then, u = _ 11 and

1 Tiu P

F (u) = Fs(i(p~1.)) = F.(P) = f °° Sin L(P~ll x : f (x)dx.

P Jo p

oo

By assumption, X |f(k) — f(k — 1 )| converges and, since
k=l

| (Sin uk)[f(k) - f(k-l)]| <|f(k) ~ f(k - 1)|,

the Weierstrass M-test ensures that / Sin ux : f(x)<5x converges.

Jo

Definition 6: Let f be a function which satisfies the same conditions
as the function in Definition 5. The monodiffric Fourier Cosine
transform of f is

C(f(x)) =J Cos ux : f(x)<5x = Fc(u), u > 0.

TRANSFORMS OF MONODIFFRIC FUNCTIONS

101

Reasoning similar to that above shows that the integral converges
absolutely.

Note that, for x > 0,

F(u) = Fc(u) - i Fs(u),

which is analogous to continuous theory.

The theory and properties of the monodiffric Fourier transforms are
being studied and developed in order to use them to convert
differential forms into algebraic forms involving boundary values.
That, in turn, will make possible the solution of boundary-value
problems in ordinary monodiffric differential equations by Fourier
transform methods.

LITERATURE CITED

Berzsenyi, G. 1970. Line integrals for monodiffric functions. J. Math Anal. Appl. 30:99-

112.

Berzsenyi, G. 1972. Convolution products of monodiffric functions. J. Math. Anal. Appl.
37:271-287.

Churchhill, R. V. 1958. Operational mathematics. McGraw-Hill Book Company, Inc.,
New York, NY.

Duffin, R. J., and C. R. Duris. 1964. A convolution product for discrete analytic
function theory. Duke Math. J. 31:199-220.

Isaacs, R. P. 1941. A finite difference function theory. Univ. Nac. Tucuman Rev. 2:177-

201.

Isaacs, R. P. 1952. Monodiffric functions. Nat. Bur. Standards Appl. Math. Ser. 18:257-
266.

ABSTRACTS

OF THE SECOND ANNUAL
SOUTHWESTERN DEVELOPMENTAL BIOLOGY
CONFERENCE

Department of Biology
and

Institute of Developmental Biology
Texas A8cM University
College Station, TX 77843
March 22-24, 1984

HOST COMMITTE:

G. Bhaskaran, W.L. Rickoll, H.W. Sauer (chairman), and T.L. Thomas
Department of Biology
Texas A&M University
College Station, TX 77843

SPONSORS:

Department of Biology, Texas A&M University
College of Science, Texas A&M University
Office of University Research Services, Texas A&M University
National Science Foundation
American Society of Zoologists
Society for Developmental Biology

104

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

FUNCTIONAL AND PHYSICAL CORRELATIONS AT THE ANTENNAPEDIA
LOCUS OF DROSOPHILA MELANOGASTER. Thomas C. Kaufman and Michael K.
Abbott, Department of Biology, Indiana University, Bloomington, IN 47405.

Function of the Antennapedia ( Antp ) locus is required to specify proper segmental
identity in the thorax of the embryo and adult of Drosophila melanogaster. This
function is similar to that found for the Ultrabithorax ( Ubx ) locus of the bithorax
complex (I. Duncan, Genetics 100:20, 1982; G. Struhl, Proc. Natl. Acad. Sci. USA
79:7380-7384, 1982; T.C. Kaufman and M.K. Abbott, In Molecular aspects of early
development, Plenum Press, in press.) However, unlike Ubx mutations, the structures
into which thoracic elements are homeotically transformed in Antp animals are different
in the embryo as compared to the adult. Embryos lacking a functional Antp locus have
three thoracic segments, all of which express a phenotype more characteristic of only
the first thoracic segment. In adult flies, Antp cells develop normally in the head and
abdomen but cause transformation of legs into antennae. Additional defects are produced
in the anterior portions of the dorsal mesothorax. We now have identified Antp
mutations that apparently affect these two transformations separately. Thus some alleles
cause only the leg-to-antenna transformation and permit otherwise normal embryonic
development. Other alleles produce the embryonic transformation previously described
but have no effect upon leg development. Both types of lesions fail to complement
presumed null alleles and deletions of Antp but complement one another to produce
normal adult flies.

The recent molecular characterization of the DNA from the Antp locus has allowed
us to begin an analysis of these lesions at the macromolecular level. Mutations of the
Antp locus are distributed over somewhat more than 100 kb of DNA as are those regions
of the locus that are transcribed. The lesions associated solely with the leg-to-antenna
transformation, however, map to a discrete 1 kb region within the locus, while those
lesions associated with the dominant antenna-to-leg transformations serve to dissociate
this segment from the normal 5' end of the Antp transcription unit. The correlation of
specific functional defects with certain subregions of the DNA of the locus may indicate
the positions of differing functional domains within the gene. One interpretation of the
complementation of the various Antp mutations is that the function of each of these
units is, to some extent, autonomous.

This research was supported by PHS-GM 24299 and PHS-GM 29709.

REGULATION OF CORTICOSTERONE PRODUCTION IN THE YOUNG RAT.
Glen Genovese and Susan J. Henning, Department of Biology, University of Houston,
Houston, TX 77004.

Circulating concentrations of corticosterone in the rat increase dramatically during the
second and third weeks of life, initiating ontogenic changes in other organ systems such
as the gut, pancreas and liver. The biochemical control of the corticosterone increase has
not been investigated previously. Examined in this study were cholesterol ester hydrolase
(CEH), the enzyme responsible for supplying cholesterol as substrate to the steroidogenic
pathway, and mitochondrial and microsomal cytochrome (cyt) P-450 which are
important cofactors in the regulation of steroidogenesis. Cytosolic CEH activity was
measured by following hydrolysis of 14C-cholesterol oleate. Cyt P-450 was measured in
microsomal and mitochondrial suspensions by difference spectroscopy. CEH specific
activity, total activity, and activity/g body weight decreased between postnatal days 10
and 16, and then increased between days 16 and 24. The amount of mitochondrial
protein/g adrenal tissue increased significantly between days 10 and 15 while the amount
of mitochondrial cyt P-450/mg mitochondrial protein remained the same. Between days
15 and 21 the amounts of mitochondrial protein/g adrenal tissue decreased while
mitochondrial cyt P-450/mg mitochondrial protein increased greatly. The amount of

SDB ABSTRACTS, SPRING 1984

105

microsomal cyt P-450/mg microsomal protein increased between days 10 and 21, with
most of the increase occurring between days 10 and 15. This suggests that the increase
in corticosterone production is due, first, to an increase in the number of adrenal
mitochondria and, second, to an increase in the concentration of cyt P-450 within the
mitochondrial . To a lesser extent, the increase of microsomal cyt P-450 may contribute
to increased steroid production. Changes in CEH activity are not important during the
initial increase of corticosterone production but may play a role during the later phase
of the corticosterone rise.

ABSORPTION OF LEAD BY SUCKLING AND WEANLING RATS. Susan J.
Henning and Lucy L. Leeper, Department of Biology , University of Houston-U niversity
Park, Houston, TX 77004.

In many species (including humans), the young absorb much greater proportions of
a given dose of Pb than do adults.. The aim of this study was to determine the site of
absorption of Pb by suckling rats. When 203PbCl2 (carrier-free) plus Pb-acetate (50 /zg/
g BW) was intragastrically administered to rat pups aged 10 and 14 days, mucosal
uptake of Pb, measured 2 h after intubation, was many-fold greater in the duodenum
than in other regions of the small intestine. By 24 days of age, this duodenal uptake
was no longer apparent. In suckling pups the duodenal content of Pb became
undetectable by 24 h post-intubation; in contrast, ileal uptake was minimal at 2 h but
increased progressively through 24 h. The ileal uptake component was also age-
dependent, having disappeared by 24 days of age. When mucosal uptake of Pb was
measured in suckling pups which had received cortisone acetate 5 days earlier, the
hormone treatment suppressed ileal uptake and had no significant effect on duodenal
uptake. Furthermore, when blood Pb was determined, there was no significant difference
between hormone- treated pups and their vehicle-treated littermates. These findings,
together with the temporal details, suggest that it is the duodenum where Pb absorption
occurs, whereas ileal uptake represents intestinal retention. To investigate the hypothesis
that duodenal absorption of Pb is enhanced in the presence of milk, we examined the
effect of fasting versus suckling following gastric intubation of Pb. Duodenal content
and blood Pb were significantly higher in the fasted animals, suggesting that milk
actually reduced the absorption of Pb. (This work was supported by NIH grant #HD
14094.)

GILL CALCIUM CONCRETIONS ARE MOBILIZED DURING GLOCHIDIA
FORMATION IN THE FRESHWATER MUSSEL. Harold Silverman, Thomas H.
Dietz, and W.L. Steffens, Department of Zoology and Physiology , Louisiana State
University , Baton Rouge, LA 70803.

Unionid clams have a gill marsupium in which they brood their embryos up to a
shelled glochidial stage. The large mussel Anodonta grandis broods enough larvae for
the weight of the gill to increase 15 times during brooding. The amount of calcium
necessary for embryonic shell formation is too great to come directly from the freshwater
environment. Apparently the source of calcium is the large deposit of extracellular
concretions found in the gill. Between 25-50% of the dry weight of gills in pre-
reproductive unionids is composed of extracellular calcium/phosphate concretions.
These concretions are organized in masses around the nerve tracts supplying individual
gill filaments, as demonstrated either by whole-mount photography, whole-mount x-
rays, or en face frozen sections. During egg laying and subsequent brooding of embryos,
the amount of concreted material steadily declines, as indicated by a drop in isolatable
concretions, loss of x-ray dense material, and loss of visible concretions in whole
mounts.

106

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

BINDING OF LECTIN-COATED MICROSPHERES TO EMBRYONIC MOUSE
PALATES. Jackie Duke, L. Janer, and M. Campbell, University of Texas Dental
Branch, Dental Science Institute , Houston, TX 77225.

Fusion of the shelves of the mammalian secondary palate involves the appearance of
glycoproteins along the medial epithelial edge (MEE) of the palate. SEM studies have
disclosed the presence of a filamentous material in this region, presumably comprised
of glycoproteins. The present study explored the Con-A binding capabilities of the MEE
prior to fusion, in order to determine if glycosylation was present. The binding was
examined by a simple SEM technique involving lectin-coated microspheres.

Maxillary blocks were dissected from mouse embryos of gestational ages 12.5 to 14.5
days. After incubation with Con-A-coated microspheres, or Con-A-coated spheres plus a-
methyl mannoside, tissue blocks were rinsed and prepared for scanning electron
microscopy. Very few spheres adhered to the vertical shelves of 12.5 day palates, and
those seen were scattered randomly over the palate surface. No areas of MEE breakdown
were seen. By 13.5 days, elevation had begun, and a small amount of MEE breakdown
had occurred, with heavy concentrations of spheres (>1 00/cell) bound to a few cells in
these areas. By 14.5 days, shelves were horizontal and in contact medially. MEE
breakdown was observed along the entire MEE surface between rugae 1 and 5 and most
of the cells in these areas, including the tips of the rugae, were coated with a
filamentous material and bound heavy concentrations of spheres.

This study demonstrated that the glycosylated nature of cell surfaces and filamentous
material in breakdown regions of the mouse palatal MEE could be determined via SEM
with a simple sphere binding technique.

This research was supported by NIDR Small Grant 1R03 DE06486-01.

EMBRYONIC AXIS AND GROWTH-REGULATOR-INDUCED MODULATION OF
STORAGE-PROTEIN DIGESTION IN SUNFLOWER COTYLEDONS. Randy D.
Allen, Craig L. Nessler, and Howard J. Arnott, Department of Biology, Texas AirM
University , College Station, TX 77843.

Cotyledons of sunflower seedlings expand and their protein content first rises, then
begins to decrease, during the first three days of growth. Storage-protein structures,
which are visible with scanning electron microscopy, undergo modification which leads
to storage-protein disappearance by day 4 post-imbibition. Expansion of cotyledons
detached from seeds prior to imbibition is greatly reduced, total protein levels remain
high, and storage-protein structures remain visible. Incubation of cotyledons in 1 fxM
benzyladenine or kinetin increases the rate of cotyledon expansion and results in storage-
protein loss to levels below those in intact seedling cotyledons. Incubation in
indoleacetic acid appears to inhibit cotyledon expansion and protein mobilization. More
rapid hydrolysis of storage proteins in cotyledons of intact seedlings or those treated with
cytokinin is further indicated in day 2 specimens by SDS-polyacrylamide gel
electrophoresis. These results suggest a possible mechanism for regulation of cotyledon
development by a balance of the promotive effects of cytokinin and inhibitory effects of
auxin.

PATTERN DISCONTINUITIES IN AN APPARENT MUTANT, SPONTANEOUS
DOUBLE ABDOMEN, OF CHIRONOMUS SP. Jean Percy1, Kristen Wacker, and Klaus
Kalthoff, Center -for Developmental Biology, Department of Zoology, University of
Texas at Austin, Austin, TX 78712.

We have isolated what appears to be a mutant strain, spontaneous double abdomen
(sda), from our laboratory culture of Chironomus sp. Most embryos showing the sda

'Study conducted while on leave from the Forest Pest Management Institute, Box 490,
Sault Ste. Marie, Ontario, Canada, P6A 5M7.

SDB ABSTRACTS, SPRING 1984

107

phenotype have anterior segments replaced by posterior segments with reversed polarity.
Similar phenotypes can be generated by UV irradiation of the anterior pole of normal
Chironomus eggs.

Most double abdomen phenotypes are symmetrical, showing a mirror-image
duplication of the posterior six abdominal segments joined by an intermediate complex.
In addition, we have found various asymmetrical phenotypes. These include embryos
with an apparently normal series of segments from the first thoracic (Ti) through the
last abdominal (Aio) segment, and a reversed Aio joined anteriorly to Ti. Other embryos
show both head and Aio structures anterior to Ti. Still other embryos have normal head
and Aio structures but fewer segments than normal, or are truncated, either anterior to
a thoracic segment or posterior to an abdominal segment.

The sda phenotypes of Chironomus are a connecting link between the symmetrical
double abdomens induced by UV in other chironomids and the phenotypes of the
hicaudal mutant of Drosophila melanogaster. The similarities add to the evidence for
a common mechanism underlying the determination of anteroposterior polarity in
dipteran eggs. Of particular interest are the phenotypes juxtaposing structures that are
widely separated in normal embryos. We plan to analyze whether the discontinuities
arise in situ or develop secondarily. The former case would be difficult to reconcile with
models of pattern formation that rely on gradients of diffusable signal substances.

RNA CHAIN INITIATION AND ELONGATION BY RNA POLYMERASE IN SEA
URCHIN EMBRYOS. J. Akif Uzman1 and Fred H. Wilt2, 1 Center for Developmental
Biology, Department of Zoology, University of Texas, Austin, TX 78712, and
2 Department of Zoology , University of California, Berkely, CA 94720.

The involvement of RNA chain initiation and elongation by RNA polymerase II in
regulating the rate of total RNA synthesis and particularly that of the early histone
mRNAs was investigated. Nuclei were isolated from developing sea urchin embryos
( Strongylocentrotus purpuratus) at the 4 to 600 (gastrula) cell stages. The transcription
of template-engaged RNA polymerase complexes was studied in a run-off transcription
assay. This assay allows the transient elongation of already initiated RNA chains by
template-bound RNA polymerases; we optimized our assay by using high levels of
nucleoside triphosphates.

The rate of RNA synthesis for total RNA and histone mRNA increases between the
16 and 100 cell stages to a maximal rate between the 100-200 cell stages; thereafter, it
declines steadily. In our run-off transcription assay conditions, the relative level of run¬
off RNA synthesis from stage to stage closely paralleled the known rates of sy thesis in
vivo. If sarkosyl, which allows the transient elongation of already engaged but stalled
RNA polymerases, is included in the assay, the nuclei exhibit a greater and greater
stimulation of the run-off assay as development proceeds. However, sarkosyl did not
reveal any run-off transcripts from the early histone genes at later stages of development.
Hence, there are many template-engaged RNA polymerase II complexes that do not
elongate initiated RNA chains rapidly at later stages; but, at one set of loci, the early
histone genes are inactive at later stages because they do not possess any RNA chains
initiated by template-bound RNA polymerases.

REGULATION OF PROTEIN SYTHESIS IN THE SEA URCHIN EGG. Matthew
Winkler1, Ellen Nelson1, and John Hershey2, 1 Center for Developmental Biology,
University of Texas, Austin, TX 78712, and 2Department of Biological Chemistry,
University of California, Davis, CA 95616.

A 15-30 fold increase in the rate of protein synthesis follows fertilization of the sea
urchin egg. The major portion of this increase is mediated by the mobilization of stored
maternal mRNA into polysomes. The molecular mechanisms responsible for regulating

108

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

the entry of stored mRNA into polysomes are poorly understood. We have used cell-free
translation systems derived both from sea urchin eggs and rabbit reticulocytes to
determine which steps in the process of protein synthesis are rate limiting in the sea
urchin egg. Addition of exogenous mRNAs to the sea urchin system was used to
determine if mRNA availability limits protein synthesis. As in the Xenopus oocyte,
protein synthesis does not appear to be limited by the availability of mRNA, but rather
is regulated at the level of the translational machinery. Mixtures of the sea urchin cell-
free system and various fractions of rabbit reticulocyte lysates were used to identify rate-
limiting steps in the translational machinery. Purified protein-synthesis-initiation factor
eIF-2 stimulated protein synthesis, suggesting that protein synthesis may be regulated by
the level or activity of this factor. We have found also that crude preparations of sea
urchin mRNA are not active in initiation assays performed in the reticulocyte lysates.
This suggests that in the egg, mRNA is not available to the protein synthetic machinery
(masked message). Thus, protein synthesis appears to be regulated at both the level of
mRNA availability and the activity of components of the translational machinery.

ENDODERM HETEROGRAFTS BETWEEN ANURAN SPECIES AND PRIMOR¬
DIAL GERM-CELL MIGRATION. S. Subtelny, Department of Biology, Rice
University , Houston, TX 77251.

Xenoplastic grafts of primordial germ cell (pcg)-containing endoderm regions from
Bufo valliceps into UV-irradiated, sterile stg 17 Rana pipiens tailbud hosts adversely
affect embryogenesis of the latter and consistently yield no pgc migration. When Bufo
grafts are extirpated from the ventral abdominal halves of UV-irradiated Rana hosts at
stg 20 or 21, after dorsal intraendodermal pgc migration is typically initiated, normal
host development is restored, but the tadpoles still possess sterile gonads. This suggests
that the grafted pgcs do not undergo intraendodermal migration. Bufo grafts into
anterior or posterior abdominal regions of nonirradiated stg 17 Rana embryos elicit the
adverse graft reaction but do not prevent migration of host pgcs. This indicates that the
intolerance reaction is not directly involved in the absence of xenograft pgc migration.
The results suggest that the grafted pgcs either fail to interact with the foreign host
tissues or fail to respond to a putative functional host attractant, or both.

A HIGHLY REPETITIOUS DNA SEQUENCE SPECIFICALLY REPLICATED IN
THE GERMLINE, POLYPLOID NURSE CELLS OF CALLIPHORA ERYTHROCE-
PHALA. M.E. Nazimiec and K. Beckingham, Department of Biochemistry, Rice
University, Houston, TX 77251.

Transcription of highly repetitious DNAs has been detected in oocytes of several
species, although no function for these transcripts has yet been established. In the
Diptera, the highly polyploid germline nurse-cell nuclei perform most of the
transcriptional functions associated with the oocyte nucleus in other phyla. In somatic
polyploid and polytene cells, non-functionality for many families of repetitious DNA
sequences is implied by the severe under-replication of these sequences. In order to
identify repetitious DNA families which might have functional significance in polyploid
nurse cells, we therefore have looked for repetitious sequences which are preferentially
replicated in these cells. To date we have identified one cloned, highly repetitious DNA
sequence which is highly represented in nurse-cell nuclear DNA of Calliphora
erythrocephala. This clone, 3B55, shows more than a one-hundred fold greater
representation in nurse-cell DNA as compared to somatic DNA of equivalent ploidy
(salivary gland DNA). This representation is at least equivalent to that seen in diploid
DNA from early embryos.

The 3B55 clone consists of ~25 copies of a 250 bp repeating unit with sites for
restriction enzymes Rsa I and Hae III. In genomic DNA fractionated on CsCl gradients,

SDB ABSTRACTS, SPRING 1984

109

these sequences are preferentially localized in a satellite band of higher density than bulk
DNA. In situ hybridization of 3B55 DNA to the trichogen (bristle-forming) cell polytene
chromosomes has localized this DNA to the pericentric regions of chromosomes 2, 3, 4
and 5 with no hybridization to chromosomes 1 and 6. Initial colony hybridization
experiments with 32P-end labelled nurse-cell nuclear RNA as a probe have indicated that
3B55 is transcribed in these nuclei. This possibility is being actively investigated.

TEMPORAL AND SPATIAL PATTERNS OF MUSCLE-TYPE ACTIN mRNA
ACCUMULATION DURING DEVELOPMENT OF THE ASCIDIAN STYELA
PLICATA. Craig R. Tomlinson, Center for Developmental Biology, Department of
Zoology, University of Texas, Austin, TX 78712.

The egg of Styela plicata has three cytoplasmic regions — ectoplasm, endoplasm, and
myoplasm — that delimit the presumptive fates of different embryonic cell lineages. After
fertilization, each of these differently pigmented regions participates in ooplasmic
segregation and is distributed to specific blastomeres during embryonic development.
Different larval tissues ultimately are formed from these regions, which can be followed
up to the tadpole stage. A previous study (W.R. Jeffery et al., Develop. Biol. 99:408-417,
1983) showed by in situ hybridization that actin mRNA accumulates in the myoplasm,
the pigmented region that eventually is distributed to the tail muscle and mesenchyme
cells of the tadpole. A goal of the present investigation is to determine if muscle actin
mRNA precociously accumulates in the myoplasm. To accomplish this, actin clones
were selected from a cDNA library of adult S. plicata muscle RNA, and two muscle actin
clones were identified by DNA sequencing using the Sanger method. The 3'-end of the
non-codogenic region of the muscle actin cDNA was subcloned into the M13 mp8 vector
to produce a specific, complementary, single-stranded probe to muscle actin mRNA. The
probe now is being used in Northern blotting analyses to determine the timing of
muscle actin mRNA accumulation and for in situ hybridization to thin sections of the
egg and later developmental stages to determine the location of muscle actin mRNA.
Also, probes from the codogenic and non-codogenic regions of muscle actin mRNA are
being used to determine the number and identity of the muscle actin genes.

ISOLATION AND MOLECULAR CHARACTERIZATION OF THE YELLOW
CRESCENT, A MUSCLE-FORMING CYTOPLASM IN ASCIDIAN EMBRYOS.
William R. Jeffery, Center for Developmental Biology, Department of Zoology,
University of Texas, Austin, TX 78712.

The yellow crescent (YC) is a cytoplasmic region of specific morphogenetic fate
localized in the posterior-vegetal region of fertilized ascidian ( Styela ) eggs. The YC is
formed from the cortical region of the unfertilized egg by a spectacular episode of
cytoplasmic rearrangement between fertilization and the first cleavage. During
embryogenesis, the YC cytoplasm is evenly distributed between the first two blastomeres,
but in subsequent cleavages it becomes restricted to the larval tail muscle and
mesenchyme lineage cells. Cytoplasmic transfer experiments suggest that the YC may
contain factors that can promote the development of muscle-specific markers in non¬
muscle cell lineages. To search for factors specifically located in the YC, we have
developed a procedure for the mass isolation of this cytoplasmic region and have begun
to characterize its protein and nucleic acid components. The isolation procedure exploits
the previous observations that the YC contains large numbers of cytoskeletal filaments
and that the latter are known to be insoluble in high levels of Na+ and Ca+2. To isolate
YCs, fertilized eggs were permeabilized with 0.1% Triton X-100 in 50 mM Tris-HCl, 500
mM NaCl, 10 mM MgCh, and 5 mM CaCh (pH 7.2), gently homogenized, and the YCs
(which remain intact after homogenization) were fractionated by differential
centrifugation and purified by centrifugation through sucrose step gradients.

110

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

Transmission and scanning electron microscopy showed that isolated YCs, which are
obtained in more than 80% yield, contain all the YC organelles found in vivo; these
organelles include pigment granules, mitochondria, rough ER, cytoskeletal filaments,
and plasma membrane. The isolated YCs contain about 20% of the total protein and
about 8% of the total RNA of the egg. Two-dimensional gel electrophoresis indicated
that the YC includes actin, other cytoskeletal proteins, and a specific subset of egg
proteins. It is depleted in yolk proteins. In vitro translation and product analysis of
RNA extracted from YCs indicated that most of the prevalent mRNAs of this region,
with a few notable exceptions, are also found in other parts of the egg. It is concluded
that YC contains a specific subset of the macromolecular complement of the egg. This
procedure for YC isolation may be useful in the eventual identification of ascidian
muscle-cell determinants.

ISOLATION AND CHARACTERIZATION OF A DEVELOPMENTALLY REGU¬
LATED GENE FAMILY IN A NEMATODE. M R. Klass, S. Kinsley, and L. Lopez,
Biology Department, University of Houston, Houston, TX 77004.

The major sperm protein (MSP) of the nematode Caenorhabditis elegans is a low-
molecular- weight, basic protein (15K) implicated in the pseudopodial movement of
mature spermatozoa. Its synthesis occurs in a specific region of the gonad and is
regulated at the level of transcription (M. Klass and D. Hirsh, Devel. Biol. 84:299-312,
1981; S. Ward and M. Klass, Devel. Biol. 92:203-208, 1982; M. Klass, B. Dow, and M.
Herndon, Devel. Biol. 93:152-164, 1982). A developmentally regulated gene family that
codes for this major sperm protein has been identified. Whole genomic blots, as well
as analysis of genomic clone banks, indicate that there are between 15 and 25 copies of
the MSP gene in the nematode genome. Three distinct members of this MSP gene family
have been cloned and their nucleotide sequences determined. Differential screening of a
cDNA clone bank made from PolyA mRNA from adult males yielded 45 male-specific
clones, 32 of which were clones of MSP genes. One of these cDNA clones was found
to contain the entire nucleotide sequence for the major sperm protein, including part
of the 5' leader and all of the 3' trailing sequence. Genomic clones bearing copies of
the MSP genes have been isolated. At least one of the members of this gene family is
a pseudogene. Another member of the MSP gene family that has been cloned from
genomic DNA contains the entire uninterrupted structural sequence for the major sperm
protein in addition to a 5' flanking sequence containing a promoter-like region with
the classic “TATA” box, a sequence resembling the “CAAT” box and a putative
ribosome binding sequence.

Recombinant plasmids have been constructed containing the MSP promoter region
attached to an E. coli /8-glucuronidase gene fused to an MSP terminator sequence (pMSP
2. l:/3glu). This plasmid has been used to transform the nematode via injection into the
gonad of /Sglu-strains to analyze the tissue specificity of the MSP promoter.

COMPUTER-ASSISTED ANALYSIS OF ENZYME ACTIVITY PROFILES DURING
DEVELOPMENT. William S. Bradshaw, Sheila A. McGrath, and Craig Harris,
Department of Zoology, Brigham Young University , Provo, UT 84602.

Developmental profiles from late gestation through the neonatal period until
adulthood have been generated for a number of enzymes in various tissues of the rat.
A computer-assisted procedure (MINITAB) has been adapted to fit the raw data with a
series of polynomial functions, thus generating a standardized mathematical description
of each profile. The relative contribution to the whole pattern of each polynomial, and
its sign, are among the parameters used to create a code system with which to catalogue
the unique profiles observed. A quantitative comparison between two developmental
profiles can be performed using a second program (RUMMAGE), which utilizes a

SDB ABSTRACTS, SPRING 1984

111

general linear model. The analysis of variance shows where two profiles diverge, to what
extent the specific activities differ, and the statistical significance of the differences.
Specific applications of the procedure include a) the effect on normal ontogeny of
prenatal exposure to fetotoxic chemicals (induction of UPD-glucuronyltransferase —
UDPGT — in rat liver by PCBs); b) the observation of tissue-specific patterns for a single
activity (/3-glucuronidase in brain, intestine, kidney and liver); c) differences between
related organisms in the profile for a single activity (UDPGT in rat, rabbit, and guinea
pig); d) a comparison of differences among enzymes employing the same or similar
catalytic mechanisms (rat liver dehydrogenases); e) a comparison of enzymes
participating in a specific biochemical pathway (the glycolytic enzymes).

DEVELOPMENT OF IMMUNOREACTIVE INSULIN IN THE FETAL BOVINE
PANCREAS. J.B. D’Agostino, J.B. Field, and M.L. Frazier, Diabetes Research
Laboratory , St. Luke’s Episcopal Hospital, Baylor College of Medicine, Houston, TX
77225.

Although maturation of the fetal endocrine pancreas has been studied in a number
of species, the bovine pancreas has not been thoroughly characterized. The aim of this
study was to characterize the development of immunoreactive insulin (IRI) in the fetal
bovine pancreas. Pancreatic IRI was acid-extracted, and both pancreatic and serum IRI
quantified by radioimmunoassay. Standardization of IRI for wet tissue weight revealed
that first trimester concentrations were similar to those seen in the adult (8.2+0. 7 and
5. 9+1. 7 U/g pancreas, respectively). IRI increased progressively during gestation,
attaining 39.2 ±6.5 U/g pancreas in the third trimester which was approximately seven
fold higher than that seen in the adult. Normalization of IRI concentrations for acid-
soluble protein resulted in a significant decrease in IRI between the mid-second and
third trimesters, presumably due to dilution of the endocrine pancreas by the rapidly
growing exocrine pancreas. IRI was also detectable in fetal sera from all three trimesters.
In contrast to the profile for pancreatic concentrations of IRI during fetal development,
serum concentrations remained constant throughout gestation at approximately 20 /zU/
ml. We conclude that 1) the endocrine pancreas is functional with respect to insulin
biosynthesis during all three trimesters of development in the bovine fetus and 2)
pancreatic, but not serum, concentrations of IRI increase progressively as development
proceeds.

ALPHAFETOPROTEIN AND c-Ha-ras GENE EXPRESSION IN THE NEOPLASTIC
MOUSE LIVER. Randy Mifflin and John Papaconstantinou, Department of Human
Biological Chemistry and Genetics, Division of Cell Biology, University of Texas
Medical Branch, Galveston, TX 77550.

Synthesis of alphafetoprotein (AFP), the major serum protein of the mammalian fetus,
is normally repressed in the adult liver but can be induced by certain hepatotoxins
during the early stages of chemically induced hepatocarcinogenesis and in hepatic
tumors. The cellular Harvey ras oncogene (c-Ha-ras) is also expressed at elevated levels
in certain fetal and cancerous tissues including the liver. This study was initiated to
determine if these genes are regulated by similar factors in mice. The initial experiments
involved monitoring AFP and c-Ha-ras mRNA levels in the liver during development,
carbon-tetrachloride-induced regeneration of the liver, and azo-dye-induced
hepatocarcinogenesis. The carcinogenic regimens consisted of feeding mice a diet
containing either 0.06% 3'-methyl-4-dimethylaminoazobenzene (3'-MeDAB — a weak
carcinogen in mice) or 0.06% o-aminoazotoluene (oAT — a more potent mouse
carcinogen). The carcinogens were absorbed into laboratory chow using olive oil as
carrier. AFP and c-Ha-ras mRNA levels were estimated by dot-blot hybridization in
which poly(A) enriched RNA is immobilized on nitrocellulose and hybridized to probes

112

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

specific for each gene. Transcripts of each gene were also analyzed by the Northern
blotting technique. A comparable degree of c-Ha-ras mRNA induction was found in the
fetal, regenerating and carcinogen-treated livers. Surprisingly, a similar level of c-Ha-ras
induction was achieved on a control diet containing 10% olive oil. Of the carinogen-fed
mice, only those fed oAT had significantly elevated AFP mRNA levels. We conclude that
although both genes are expressed under similar conditions, their regulation is not
strictly coupled; otherwise, c-Ha-ras induction could not have occurred in the absence
of AFP induction.

INVASION OF MURINE MESENTERIES IN VITRO BY B16 MELANOMA CELLS.
Paul B. Bell, Jr., Tien-Ling Lee, and Ming-Ling Yang, Department of Zoology, The
University of Oklahoma, Norman, OK 73019.

Pieces of mesentery from C57 B16 mice, maintained on coverslips in liquid medium,
became covered by a continuous squamous epithelium of mesothelial cells. Beneath the
epithelial cells was a loose connective tissue containing fibroblasts, macrophages, and
mast cells immeshed in an extracellular matrix (ECM) of collagen and elastic fibers. B16
F10 murine melanoma cells, which metastasize spontaneously in syngeneic C57 B16
mice, rapidly adhere to and invade intact mesenteries in culture. Scanning electron
microscopic (SEM) observations showed that freshly seeded B16 cells initially remain
rounded on top of the mesothelium, but they adhere to the mesentery by extending small
cell processes through holes in the mesothelial cell layer and making contact with the
under-lying ECM. Such holes were not present in control mesenteries, and, therefore,
appeared to form in response to the B16 cells. By 2 h after the addition of B16 cells,
many endothelial cells had retracted from each other, exposing large patches of ECM
onto which the B16 cells spread, creating a mosaic of mesothelial cells and tumor cells
on the mesentery surface. B16 cells also extended processes under adjacent mesothelial
cells and down into the loose connective tissue. By 12 h the mesothelial cells had
respread to form an almost continuous epithelium. A few gaps persisted and a few
tumor cells remained on the mesentery surface, but most of the B16 cells were then in
the interior of the mesentery where they could be seen in SEM by virtue of their
backscattered electron signal. Time-lapse cinemicrography showed that the B16 cells
continued to move about slowly in the interior of the mesentery, deforming the collagen
fibers of the ECM as they moved past. B16 cells continued to divide inside the mesentery
and eventually formed small tumor masses.

TROPONIN-C-RELATED PROTEINS ENCODED BY A FAMILY OF GENES
EXPRESSED IN THE EMBRYONIC ECTODERM OF THE SEA URCHIN. William
H. Klein, Department of Biology , Indiana University , Bloomington, IN 47405.

We have isolated and characterized several cDNA clones representing a family of
mRNAs expressed in the embryonic ectoderm of Strongylocentrotus purpuratus. These
mRNAs (termed Spec for S. purpuratus ectoderm) accumulate in the presumptive dorsal
ectoderm of postcleavage stage embryos and code for a group of 10 to 12 low-molecular-
weight acidic proteins. Studies using antibodies prepared against the major Spec
proteins indicate that these polypeptides are localized in the cytoplasm of dorsal
ectoderm cells. Hybridization analysis and DNA sequencing show that the mRNAs
coding for these proteins, although related, can be divided into two subfamilies.
Comparison of the translational reading frames of the Spec mRNAs with known protein
sequences shows a significant homology with troponin-C-related proteins, especially in
the calcium binding domains. We speculate that the Spec proteins are previously
uncharacterized members of the troponin-C superfamily. Analysis of short- and long-
range genomic topography of these genes is now underway to detect possible regulatory
elements involved in their ontogenic expression.

SDB ABSTRACTS, SPRING 1984

113

GENETIC AND CORTICAL PHENOTYPES ASSOCIATED WITH EARLY CLONAL
DEATH IN PARAMECIUM TETRAURELIA. Karl J. Aufderheide, Department of
Biology and Institute of Developmental Biology, Texas A&M University , College
Station, TX 77843.

The clonal cycle of P. tetraurelia starts with a sexual event. Vegetadvely grown clones
then proceed through a series of stages, culminating in senescence and eventual death.
This aging process is analogous to that of tissue-culture cells, and is a good model for
cellular aging. The aging characteristics of various mutant cell types were compared
with wild-type (stock 51s). Cell lines unable to excrete trichocysts showed two general
patterns of aging. Some trichocyst mutants ( ndB , nd6, nd7 , nd9) had clonal life-spans
that were normal or sub-normal. However, another group (ndA, ftA, ptA, tam6) had
clonal life-spans that were 1/3 to 1/4 that of wild-type. The details of trichocyst
phenotypes within each group are varied, but all the members of the latter group show
a pleiotropic effect of misdivision of the macronucleus at cell division (the “amac”
effect). Another mutant type (am), which has normal trichocysts but shows the amac
phenotype, also has an abnormally short life-span. Additionally, homopolar doublet
cells, which are cytotactically propagated and not genetically different from wild-type,
show the amac phenotype. Doublets also have an unusually short clonal life-span. Thus,
an inability to properly divide the macronucleus is correlated with an abnormally short
clonal life-span. J.D. Berger (J. Protozool. 26:18-27, 1979) has shown that cells with the
amac phenotype frequently end up with a tiny fragment of a macronucleus. In a process
of recovery of normal DNA content, these cells prolong macronuclear S and do more
DNA synthesis than normal. The hypothesis being tested is that clonal aging is a
function of the number of rounds of DNA synthesis throughout the life of the clone,
and that amac lines, by doing more synthesis per cell cycle than normal, age
prematurely. (Supported by NIH-NIA grant AG02657.)

GENETIC AND PHYSIOLOGICAL REGULATION OF THE AMOEBAL-
PLASMODIAL TRANSITION IN THE MYXOMYCETE PHYSARUM POLYCEPHA-
LUM. Werner Nader1, Gregory L. Shipley2, Aloys Huettermann1, and Helmut W. Sauer2,
institute for Forest Botany, University of Goettingen, Buesgenweg , D-3400 Goettingen,
F.R.G., and 2Department of Biology, Texas A&M University , College Station, TX 77843
USA.

During the life cycle of the acellular slime molds, uninucleate amoebae differentiate
to multinucleate plasmodia. Genetic analysis of Physarum polycephalum reveals that
this process is under control of the multiallelic mating type gene matA (for review see
R.W. Anderson and C.E. Holt, Develop. Genet. 2:253-267, 1981). Amoebae differentiate,
if they either carry two different alleles of matA, as is the case in the zygote after mating,
or if they carry certain mutations called gad (greater asexual differentiation). Cell fusion,
on the other hand, is regulated by the multiallelic gene matB. Combined genetic (P.J.
Youngman et al., Genetics 97:513-530, 1981) and cinematographic (C.E. Holt et ah, Film
B1337, Insdtut fuer den wissenschaftlichen Film, Goettingen) analysis indicates that
amoebae fuse rapidly only when they carry different matB alleles. In the fused cell,
nuclei with different matA alleles fuse in interphase and the zygote develops into a
plasmodium. If nuclei carry like matA alleles, fusion occurs in the prophase of the first
nuclear division after the nuclear membrane disappears. The zygote then divides into
two diploid amoebae. In order to fuse or to differentiate apogamically to plasmodia,
amoebae have to be induced to competence by an extracellular diffusable factor, which
has been recently isolated and partially characterized. Amoebae and plasmodia exhibit
numerous differences in cellular organization, thus indicating drastic changes in gene
expression during the transition. Based on the current knowledge of cell-differentiation,
we are characterizing some of those changes on the molecular level and analyzing their
cause in order to comprehend their role in development.

114

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

METABOLITE CONTROL OF DEVELOPMENT OF PHYSARUM. Hiltrud U. White
and Henry R. Henney, Jr., Department of Biology, University of Houston, Houston,
TX 77004.

Previous work has established that certain extracellular metabolites are effective
regulators of developmental processes in Physarum flavicomum (H.R. Henney and P.
Chu, Exper. Mycol. 1:41-51, 1977; H.R. Henney and A. Tavana, Exper. Mycol. 6:153-
160, 1982). Recently, we have extended those studies by determining the effect of various
purines, pyrimidines and nucleosides on the encystment process of P. flavicomum (H.
White, M.S. thesis, University of Houston, Dec. 1982). Of the compounds tested
guanine, guanosine, cytidine, cytosine, 5-methylcytosine and uracil had no effect on
encystment. Adenosine, thymine, uridine and 3-methyladenine only slightly delayed
encystment and protein degradation. Adenine and, to a lesser extent, hypoxanthine
produced a significant inhibition of encystment and greatly increased rates of protein
and RNA degradation, which eventually led to 75% cell death in the adenine-exposed
cells. The treatment with adenine resulted initially in elevated intracellular levels of S-
adenosylmethionine (up to 3.5 times the level of untreated control cells), which may be
involved in controlling the encystment process.

CHRONOLOGY OF GENOME REPLICATION AND ATTEMPTS AT CLONING
THE ACTIN GENE FAMILY OF PHYSARUM. Gerard Pierron, Werner Nader, and
Helmut W. Sauer, Department of Biology, Texas A&M University , College Station, TX
77843.

We have focused on genome replication and transcription in the naturally
synchronous mitotic cycle of Physarum and determined the order of replication of the
four unlinked actin loci — ardA,B,C,D — employing two independent strategies: (1)
preparative separation of in vivo bromo-substituted, newly replicated (HL) DNA from
late replicated (LL) DNA on CsCl gradients, electrophoresis of endonuclease Hind III
restriction fragments on 0.6% agarose gels, transfer onto nitrocellulose membrane
(Southern blots) and hybridization with a nick-translated specific heterologous cloned
actin probe from sea urchin; or, (2) determination of the “gene dosage” from
hybridization intensities of actin restriction fragments by densitometry. Both sets of data
establish the invariant chronological order of replication of loci C and D before 8 min,
locus B at 8-10 min and locus A at 80-90 min of the 3-h S-phase, i.e. at times when
5, 5-6 and 73-75% of the genome has replicated.

Determination of whether the expression of the actin genes of Physarum — to be
monitored by Northern-blot analysis — is developmental^ regulated requires homologous
cloned probes of the four actin genes and gene-specific probes of their untranslated
flanking sequences. As a first step we have succeeded in cloning genomic DNA of 15-
20 kb, obtained by partial digestion with endonuclease Mbol and preparative sucrose
gradients (10-30%) in the lambda replacement vector EMBL 3. This genomic library of
about 75,000 clones, the equivalent of 5 Physarum genomes, was screened. Twelve
positive clones have been identified by hybridization with the sea-urchin probe, and four
clones have been isolated and shown by restriction analysis to contain distinct inserts.
We are interested in the relative position of these genes with respect to the origin of
replication of their respective replicons. Furthermore, we shall investigate whether the
early actin genes become activated upon their replication, as has been shown for a
number of transcription units in EM chromatin spreads and taken as additional evidence
for replication-transcription coupling in the cell cycle of Physarum.

SDB ABSTRACTS, SPRING 1984

115

FLOW-AUTORADIOGRAPHY: AN ADAPTATION OF THE AUTORADIOGRA¬
PHIC METHOD FOR USE WITH FLOW-SYSTEMS. David P Bloch, Barbara
McArthur, and Jean Smith, Botany Department, University of Texas, Austin, TX 78713.

A fluorescent developer prepared by conjugating fluorescein isothiocyantate to
paraphenylenediamine was used to convert a silver autoradiographic image to a
fluorescent image. The intensity of fluorescence was linearly related to the number of
silver grains, as indicated by quantitative image analysis of autoradiographs made with
stripping film.

Dispersed cells and nuclei can be microencapsulated in photographic emulsion using
gelatin-gum arabic coacervates as a vehicle. The encapsulated cells are dispersable and
the emulsion appears to respond to internal radioisotope in a typical manner.

The relationship between fluorescence and radioactivity is still to be quantified.
Simulated results of dual parameter analysis of DNA amounts and labelling were used
to describe the potential value of this approach to cell-cycle analysis.

APPLICATION OF MOLECULAR CLONING TECHNIQUES TO DEVELOPING
SYSTEMS. David S. Durica1, Jeffrey L. Nordstrom2, and Terry Thomas3, Departments
of 1 Medical Biochemistry, 2Biochemistry and Biophysics, and 3 Biology , Texas A&M
University , College Station, TX 77843.

With the application of appropriate technologies it is possible, in principle, to isolate
and identify any gene and its corresponding gene product(s). One approach is to clone
cDNA libraries representing mRNA populations in vectors capable of expressing the
cloned genetic information as a polypeptide. Recombinants then are identified by
screening these libraries with antibodies directed against the gene products of interest.
This approach is illustrated by a description of the bacteriophage Agtll expression
vector system. Use of cloned DNA sequences is considered in terms of the analysis of
mRNA and protein expression in developing systems. This includes use of DNA-
sequence information to generate gene-specific antibodies directed against synthetic
peptides specific for individual gene-family members. Analysis of RNA expression by
nucleic-acid-hybridization techniques, including in situ hybridization, is discussed.
Particular emphasis is given to utilizing these approaches to gain information on the
spatial and temporal patterns of embryonic gene expression.

GROWTH AND DIFFERENTIATION OF PHYSARUM AMOEBAE IN LIQUID
CULTURE. G.L. Shipley and J. Marrs, Department of Biology, Texas AirM University ,
College Station, TX 77843.

Growth of wild-type Physarum amoebae in liquid culture was achieved by culturing
the cells in a basal salt solution: citric acid, 15.6mM; K2HPO4, 18.2mM; NaCl, 42.8mM;
MgS04, 8.3mM and CaCl2*2H20, 3.4mM; titrated to pH5.0 with 30% KOH. Amoebae
were provided with live concentrated Escherichia coli (5X109 cells/ml final
concentration) as a food source. Amoebae grew in stationary culture or at 60 RPM with
an average doubling time of 12 h (26 C). We found that only the divalent cations in
the medium were essential for growth but that either one at a proportionally higher
concentration could substitute for the other. Mating between genetically compatible
heterothallic strains (CH508 x LU648) as well as differentiation of mutant selfing
(apogamic) strains (Colonia and CHI) in liquid culture were compared with results
previously reported for differentiation on solid media: (1) The amoebal densities at
which two selfing strains differentiated were almost identical, unlike the results seen on
agar plates. The only difference between these two strains is a mutation at the IND
locus, believed to be the gene for the inducer (pheromone). The clear differences seen
between the two strains on agar media may be due to different diffusion rates rather than
an alteration in functional capability. (2) The rate of differentiation and number of

116

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

plasmodia formed by the selfing strains in stationary culture were similar to that on agar
plates. However, the same cells gyrated at 60 RPM formed fewer plasmodia (4% vs 90-
100%) and at a slower rate. These results suggest that in addition to being induced to
differentiate by an extracellular pheromone, intraclonal cell-cell contacts may also be
required for the amoebae to become “committed” to form plasmodia. (3) Mating, which
requires amoebal fusion, did not occur in either stationary or gyrated cultures. The
failure of the gyrated culture to mate may also point to a need for cell-cell contact.
However, the failure of the stationary cultures to mate cannot be explained by the
culture conditions alone, as evidenced by the successful differentiation of the selfing
strains. Experiments at both the genetical and physiological levels aimed at explaining
the failure of amoebae to mate in liquid culture are in progress.

XENOPUS EGG POLARITY EXPRESSED BY UNEVEN DISTRIBUTION OF
SPECIFIC mRNA. Elisabeth Buchanan, Center for Developmental Biology, Department
of Zoology , University of Texas, Austin, TX 78712.

The Xenopus egg is a highly polarized cell exhibiting a distinct animal-vegetal axis.
To determine whether polarity extends to the molecular level, the distribution of specific
mRNAs during oogenesis and maturation were examined by in situ hybridization with
cloned DNA probes. Previous studies using poly(U) probes showed poly(A)+ RNA was
uniformly distributed in the oocyte cytoplasm until Dumont stage III, when a ring of
poly(A)+ RNA appeared at various distances between the germinal vescicle (GV) and the
cortex (D. Capco and W.R. Jeffery, Develop. Biol. 89:1-12, 1982). By stage IV, the ring
reached the cortex; it became attenuated into the vegetal hemisphere in stages V and VI,
and it disappeared during maturation. A second localization of poly(A)+ RNA was seen
below the GV at stage V. Poly(U) hybridizations are difficult to interpret since
fluctuations in signal could arise from changes in the distribution and steady-state levels
of poly(A)+ RNA or by the addition and removal of adenylate residues. The earlier
studies were extended by examining the distribution of actin and histone mRNA.
Similar distributions were observed for poly(A)+ RNA, actin mRNA and histone mRNA
in previtellogenic oocytes. During vitellogenesis, however, actin mRNA exhibited a
unique localization, while histone mRNA continued to be distributed like poly(A)+
RNA. At stage III, actin mRNA was present in the perinuclear ring and also at the
boundary between the clear and yolk cytoplasms. The ring did not reach the cortex at
stage IV; instead, the mRNA was maintained in a vegetal-subcortical localization
through stage VI. Extensions of this localization to the vegetal side of the GV and into
the animal hemisphere appeared at stages V and VI. At stage V, histone mRNA was
present in the subcortical regions of both hemispheres and below the GV. The
subcortical localization was maintained in stage VI oocytes and another localization
appeared as a ring around the GV. Maturation altered these localizations. Actin mRNA
first appeared in an inverted cup-like structure in the area formerly occupied by the GV,
spread out toward the animal cortex as two concentric disks, and eventually became
localized in the cortex. Histone mRNA also was localized in inverted cups in maturing
oocytes and some ootids. The results indicate that egg polarity includes a non-uniform
distribution of actin and histone mRNA. However, these distributions are sometimes
distinct from those of poly(A)+ RNA and are subject to rearrangement at maturation.
The rings and cup-shaped localizations of mRNA may represent cycles of synthesis,
transport and localization of maternal mRNA during oogenesis and egg maturation.

SDB ABSTRACTS, SPRING 1984

117

DEVELOPMENTAL MODULATION OF THE SINGLE DROSOPHILA CALMOD¬
ULIN GENE. S.L. Tobin1, M. Yamanaka2, J.A. Saugstad1, and B.J. McCarthy2.

1 Department of Biochemistry and Molecular Biology, University of Oklahoma Health
Sciences Center, Oklahoma City, OK 73190, and 2 Department of Molecular Biology and
Biochemistry , University of California, Irvine, CA 92717.

The Drosophila calmodulin gene was isolated from the Maniatis library of
recombinant phage by virtue of its homology with mRNA from electric eel electroplax.
DNA sequence information to date indicates that of amino acid residues 59-139, only
one differs from the amino acid sequence of bovine brain calmodulin. Hybridization of
calmodulin probes to genome DNA blotted onto nitrocellulose indicates a single band
of hybridization; therefore, it appears that there is only a single calmodulin gene in the
Drosophila genome, in contrast to the multigene families for actin and tubulin. This
gene has been localized to region 49A of the salivary gland chromosomes. We have
examined the proportion of transcript homologous to probes containing calmodulin
protein-coding nucleotides and differing amounts of flanking sequences by hybridization
of these probes to a developmental series of unfractioned RNAs “dotted” onto
nitrocellulose. Results suggest that the proportions of calmodulin-homologous message
vary widely during development. In addition, there is apparently a very actively
transcribed, developmentally modulated sequence near, or possibly within, the 5' end of
the calmodulin gene. We are continuing to investigate the characteristics of this gene
as they relate to its developmental role(s).

PROGRAMMED EXPRESSION OF CELL-SURFACE DETERMINANTS DURING
ENDODERM FORMATION. Raymond J. Ivatt, Department of Tumor Biology, M.D.
Anderson Hospital, Texas Medical Center, Houston, TX 77030.

Embryonal carcinomas and early embryonic cells express an unusual class of
carbohydrates on their cell surfaces. These carbohydrates are lost in a programmed way
during early embryogenesis and have a very restricted distribution in the adult. Their
disappearance during development coincides with the major period of histogenesis. In
the adult, they are associated with cells of the reticuloendothelial system. Therefore in
both the embryo and the adult, this unusual class of carbohydrate is associated with
cellular interactions which are transient in nature. There are several lines of evidence
which implicate these carbohydrates in cellular recognition in the early embryo. We have
examined the regulation of these carbohydrates during the formation of endoderm using
the embryonal carcinoma system as a model. We have developed a sequential lectin-
affinity chromatography procedure which fractionates these complex embryonic
carbohydrates into discrete subclasses that share common structural features. This
approach has allowed the first real opportunity to analyze these complex glycans. We
have characterized the major structural changes that accompany endoderm formation
and have used two approaches to address the mechanisms that regulate these changes.
The first approach involves comparative biochemical studies on established cell lines
that have the biochemical phenotype of either stem or endodermal cells. The second
approach involves isolation of stem cells that are deficient in selected, developmentally
regulated carbohydrate determinants and comparative biochemical studies on these
variants and the parental cells. We are currently exploring the role that cellular
interactions may play in regulating the expression of these carbohydrates.

MELANOCYTE DIFFERENTIATION. M. Lynn Lamoreux, Department of Biology,
Texas Air M University , College Station, TX 77843.

Tyrosinase (EC 1.14.18.1) catalyzes three steps in the biochemical pathway leading to
production of eumelanic (nonyellow) pigment. These are the hydroxylation of tyrosine

118

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1, 1985

to produce dopa (3,4-dihydroxyphenylalanine), the oxidation of dopa to dopaquinone,
which spontaneously converts to dopachrome, and the conversion of 5,6-dihydroxyindole
to melanochrome. Skin extracts from 6-day-old nonyellow C57BL/6J-a/a mice also
catalyze these steps (Murray et al., Develop. Biol. 100:120-126, 1983). Skin extracts from
6-day-old C57BL/6J-/4y/a or C57BL/6J-e/e mice, which are producing pheomelanic
(yellow) pigment, catalyze the conversion of dopa to dopachrome at a much reduced
rate. The three isozymes of tyrosinase present in eumelanic pigment cells are the result
of post-translational processing of the nascent protein (Hearing et al., J. Biochem.,
13:99-103, 1981). Pheomelanic mice contain only one major brand of active tyrosinase
as determined by gel electrophoresis and dopa staining of hair bulb extracts (Holstein
et al., Proc. Soc. Exp. Biol. Med. 126:415-418, 1967). Thus the difference in activity of
the tyrosinase is correlated with the agouti- or extension-locus genotype of the mouse
and with a difference in the isozymes as assessed by their ability to convert dopa to
melanin. This correlation suggests the hypothesis that the post-translational processing,
which results in the presence of specific isozymes of tyrosinase, also influences the
activity of the enzyme, and that the agouti and extension loci influence the post-
translational processing of the tyrosinase.

ECTOPIC TESTES IN THE NORWAY RAT Laurence G. Gumbreck1, Allan J.
Stanley1, John E. Allison1, and Edward Peeples2, 1 Department of Anatomy and
Physiology , University of Oklahoma Health Sciences Center, Oklahoma City, OK 73190
and 2 Department of Biological Sciences, University of Northern Colorado, Greeley, CO
80639.

Animals with ectopic testes, some unilateral (on either the right or left side) and some
bilateral, have appeared in a colony of King-Holtzman rats. This heritable defect has
been given the genetic symbol “ect.” Females carrying the gene are unaffected by it.
When such females are bred to apparently normal males they produce no ect offspring.
However, 25% of their F2 generation carry the gene; therefore, it is assumed ect is
autosomal recessive. Studies involving selective breeding for males with ectopic testes
indicate that some males have genomes resistant to the expression of the gene and that
the apparent thresholds essential to left side, right side and bilateral expression appear
to be a secondary control of the multifactorial type. (This research was supported in part
by National Institutes of Health grants HD-0192T03, HD-01075-01 and HD-0309-01.)

MONOCLONAL ANTIBODY FORMATION TO PHYSARUM AMOEBAL ANTI¬
GENS. M.E. Schelling and G.L. Shipley, Biology Department, Texas A&M University ,
College Station, TX 77843.

Twelve stable murine hybridoma clones producing monoclonal antibodies against
Physarum polycephalum LU648 amoebal antigens have been established. Physarum
amoebae were harvested from agar plates into buffered saline solution and assayed for
fusion competence in mating assays. About 107 fusion-competent amoebae were injected
subcutaneously into Balb/C mice. The injection schedule consisted of a primary
injection followed in three weeks by a second injection. The third injection, occurring
1 week later, was followed in three days by fusion. SP2/0 AG 14 myeloma cells were
fused with spleen cells from immunized mice during centrifugation in the presence of
polyethylene glycol. Cells were plated at a density of 106 cells per well of 96 well plates
and hybridomas were grown at 37° C, 7% CO2. Selection was by HAT medium. When
the hybridomas were 2/3 confluent, hybridoma supernatants were assayed for the
production of monoclonal antibodies by a horseradish peroxidase ELISA. Cloning was
done by limiting dilution. The antibodies were characterized according to class, subclass
and type of light chain. Reactivity to strains CH508 and axenic strains MAI 85 and CLD

SDB ABSTRACTS, SPRING 1984

119

AXE was determined. State-specificity of the monoclonals was determined. Further
characterization of the antibodies is in progress.

INFLUENCE OF B-HAPLOTYPE DISPARITY ON ADOPTIVE IMMUNITY BY
BLOOD LEUCOCYTES TRANSFERRED INTO CHICK EMBRYO HOSTS. Frank
Seto, Zoology Department, University of Oklahoma, Norman, OK 73019.

Immunologically immature embryos and newly hatched chicks were used for in vivo
culture of transfused cells. Spleen cells or blood leucocytes from mouse-erythro-
cyte(MRBC)-primed donor chickens, when transferred with MRBC, will produce
antibodies adoptively in young hosts as measured with hemagglutinin and plaque¬
forming cell (PFC) assays. The influence of B blood group, the major histocompatibility
complex in chickens, on the adoptive immunity was investigated. Leucocytes from three
B-haplotype donor types — Bn/Bn, B19/B19, and B/13/B19 — were cultured in recipients of
the three B haplotypes, in nine different donor-host combinations. With 14-day embryo
hosts, substantial adoptive immunity was observed in all nine host groups. Splenic PFC
numbers were significantly higher in hosts that received cells from B-haplotype-
mismatched donors of the B13 or B19 type as compared with those that received B-
haplotype-matched or B13/B19 donor cells. Enhanced adoptive immunity occurred in
those combinations that resulted in a graft-versus-host reaction and resembled the
allogeneic effect. In comparable experiments with baby chick hosts, PFC formation was
as high as that observed with embryo hosts in those donor-host combinations where the
recipients were of the B13/B19 type or of the same B haplotype as the donors. In contrast,
adoptive immune responses were considerably reduced in B 1 3 or B19 homozygous chicks
that received cells from B-mismatched donors. Depressed PFC formation in the older
allogeneic hosts seems more consistent with the interpretation of a host allograft
reactivity existing in neonatal chicks rather than the consequence of neonatal
suppressor-cell activity.

COMPUTER ANALYSIS OF HAMSTER SPERM MOTILITY. L.E. Franklin, Biology
Department, University of Houston at University Park, Houston, TX 77004.

When mounted appropriately on a glass slide, sperm of the golden hamster typically
exhibit a two-dimensional, circular swimming pattern in balanced culture media such
as RPMI. The flagellar beat is rapid but its amplitude is small. Consequently, deviation
of the head to either side of the path is small. During the first few minutes after sperm
inoculation, the frequency of the flagellar beat decreases with a concomitant increase in
amplitude. Deviation of the head to either side of the path increases, and the sperm path
becomes more linear.

The circular swimming track lends itself to measurement, and thus perhaps to
meaningful comparisons of sperm motility in different samples. Our pilot studies have
employed gamma irradiation to induce measurable changes in the motility of sperm
populations. Of the several parameters measured, only sperm velocity and yaw (side-to-
side movement of the head) were significantly altered. The purpose of the investigation
is to identify some sperm motility characteristic which can be quantified and used to
evaluate male fertility.

120

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 1 , 1985

ASZ AND FIVE OTHER SOCIETIES TO MEET
IN BALTIMORE, MD, DECEMBER 27-30, 1985

The 1985 Annual Meeting of the American Society of Zoologists with the American
Microscopical Society, Animal Behavior Society, The Crustacean Society, International
Association of Astacology, and the Society of Systematic Zoology will be held at the
Baltimore Convention Center during the traditional post-Christmas period.

The Call for Contributed Papers has been issued by ASZ, and an August 12 deadline
for receipt of abstracts of papers (oral or poster presentations) has been set.

Symposia presently scheduled are (1) Behavior as a Factor in the Population of
Cricetid and Muroid Rodents, (2) Processing of Environmental Information in
Vertebrates, (3) Comparative Immunopathological Aspects of Natural Killer and
Antibody-Dependent Cytotoxic Function, (4) Muscle Fiber Typing as a Bioassay of
Nerve-Muscle Interaction, (5) Mechanisms of Physiological Compensation in Intertidal
Animals, (6) The Significance of Protein Gylcosylation in Molecular and Cellular
Recognition, (7) Pattern Formation and Recognition in Complex Biological Systems, (8)
Functional Morphology of Feeding and Grooming in Selected Crustacea, (9) Speciation
Patterns in the Southern Appalachian and Ozark Regions of Eastern North America,
(10) Developmental and Evolutionary Aspects of the Neural Crest, (11) Questions,
Explanations, Models and Tests in Morphology: The Interaction Between Hypotheses
and Empirical Observations, (12) The Role of Johns Hopkins University in the
Development of Experimental and Quantitative Biology in America, and (13) Science as
a Way of Knowing — Genetics.

In addition, workshops will be offered on (1) Computer-Assisted Analysis of 14C-2-
deoxy-D-Glucose Autoradiographs, (2) Intercellular Communication — Signal Molecules,
Cell Surface Receptors, and Cell Responses, (3) Biological Experiments in the
Microgravity of Space and (4) Application of Histochemical Techniques to Microscopy.
I. Preparation of monoclonal antibodies and discussion of selected applications to
cytochemistry.

Meeting plans include several socials, special programs, commercial exhibits and a
Job Placement Service. Hotel rates are $44 for single/double rooms at the Omni
International Hotel and $42 for single rooms and $50 for double rooms at the Hyatt
Regency Baltimore Hotel. Housing and Registration deadline is November 20. Forms
will be available mid-September. This meeting is hosted by Towson State University
with Donald C. Forester and Philip D. Creighton chairing the Local Arrangements
Committee. For more information contact: Mary Adams — Wiley, Executive Officer,
American Society of Zoologists, Box 2739 California Lutheran College, Thousand Oaks,
CA 91360 (telephone 805 492-3585).

FUTURE MEETINGS OF THE TEXAS
ACADEMY OF SCIENCE

1986 — Texas A8cl University, Kingsville, 6-8 March

1987 — Sam Houston State University, Huntsville, 5-7 March

THE TEXAS ACADEMY OF SCIENCE, 1984-85
OFFICERS

President:

President-Elect:

Vice-President:

Immediate Past President:
Secretary-Treasurer:

Editor:

AAAS Council Representative:

Michael J. Carlo, Angelo State University
William J. Clark, Texas A&M University
Billy J. Franklin, Lamar University
Bernard T. Young, Angelo State University
Fred S. Hendricks, Texas A&M University
William H. Neill, Texas A&M University
Arthur E. Hughes, Sam Houston State University

DIRECTORS

1982 Ethel W. McLemore, Dallas
Donald H. Lokke, Richland College

1983 D. Lane Hartsock, Austin
Katherine Mays, Bay City

1984 E. D. McCune, Stephen F. Austin State University
Jim Neal, U.S. Fish and Wildlife Service

SECTIONAL CHAIRPERSONS

I — Mathematical Sciences : Patrick L. Odell, University of Texas, Dallas

II — Physical Sciences : Herbert D. Schwetman, Waco

III— Earth Sciences: Austin Sartin, Stephen F. Austin State University
I Biological Sciences: Robert I. Lonard, Pan American University

V — Social Sciences: Don Matlock, Southwest Texas State University

VI— Environmental Sciences: Dean V. Ferguson, Southwest Texas State University

VII — Chemistry: Richard Langley, Stephen F. Austin State University

VIII — Science Education: Barbara Lee ten Brink, Texas Education Agency

IX— Computer Sciences: H. P. Haiduk, Amarillo College

X — Aquatic Sciences: David Buzan, Texas Department of Water Resources

COUNSELORS

Collegiate Academy: Shirley Handler, East Texas Baptist College

Helen Oujesky, University of Texas, San Antonio
Junior Academy: Ruth Spear, San Marcos

Peggy Carnahan, San Antonio

COVER PHOTO

Map of Corpus Christi Bay, Texas
from Blum and Jones, pp. 63-73

2nd CLASS POSTAGE
PAID AT LUBBOCK
TEXAS 79401

3024 I 85

Library Acquisitions
Smithsonian Institute
Washington DC 20560

ISSN 0040-4403

e XXXVII, Numbers 2 & 3

September 1985

PUBLISHED QUARTERLY BY
THE TEXAS ACADEMY OF SCIENCE

1122

(KFI)

Rd 327

2

3km

SECTION I

MATHEMATICAL SCIENCES
Mathematics, Statistics,
Operations Research

SCIENCES

AFFILIATED ORGANIZATIONS
Texas Section, American Association of Physics Teachers
Texas Section, Mathematical Association of America
Texas Section, National Association of Geology Teachers
American Association for the Advancement of Science

GENERAL INFORMATION

MEMBERSHIP. Any person or group engaged in scientific work or interested in the pro¬
motion of science is eligible for membership in The Texas Academy of Science. Dues for
members are $20.00 annually; student members, $12.00 annually; sustaining members, at
least $30.00 in addition to annual dues; life members, at least $400.00 in one payment;
patrons, at least $500.00 in one payment; corporate members, $250.00 annually; corporate
life members, $2000.00 in one payment. Library subscription rate is $45.00 annually. Pay¬
ments should be sent to Dr. Fred S. Hendricks, TAS Secretary-Treasurer, Drawer H6,
College Station, TX 77844.

The Journal is a quarterly publication of The Texas Academy of Science and is sent
to all members and subscribers. Inquiries regarding back issues should be sent to the
Secretary-T reasurer .

The Texas Journal of Science is published quarterly at Lubbock, Texas U.S.A. Second
class postage paid at Post Office, Lubbock, TX 79401. Please send form 3579 and
returned copies to Texas Tech Press, Box 4240, Lubbock, TX 79409.

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, Nos. 2 & 3

September 1985

CONTENTS

Joint Sets in the Precambrian-Paleozoic Rock Succession

of the Llano Uplift, Texas. By J. R. Alnes and S. E. Cebull . 123

Immune Response in the Bobwhite Quail, Colinus virginianus.

By Alfred C. Schram, Herbert F. Gonzalez, and Cynthia D. Meador . 133

Urnatella gracilis (Entoprocta) from Caddo Lake, Texas

and Louisiana. By Thomas M. Cusak and Jack D. McCullough . 141

Vesicle Contribution to Cyst Wall Formation of Posthodiplostomum

minimum Metacercariae. By Thomas G. Meade and Jose M. Garza . 143

On the Elements of Difference Equation Modeling in Social

Science. By Evans W. Curry and Deraid Walling . 147

Foods of Scaled Quail ( Callipepla squamata ) in Southeastern

New Mexico. By Troy L. Best and Richard A. Smartt . 155

Thyroid-Parathyroid Response to EDTA and CaCh Infusions

in White-tailed Deer By Chun Chin Chao and Robert D. Brown . 163

Effects of Pollution Effluents on Two Successive Tributaries and Village

Creek in Southeastern Texas. By C. Marc Barclay and Richard C. Harr el . 175

Spectrum Analysis of a Triple-Channel, Pulse-Slope-Modulated Wavetrain:

A Comparison of Two Methods. By Joseph H. Nonnast and Olan E. Kruse . 189

Studies of Vegetative Plant Tissue Compatability-Incompatability. VII.

Influences of Individual Organs on Graft Development. By Randy Moore . 201

Marine Fungi Associated with Spartina from Harbor Island, Texas.

By Robert D. Koehn . 213

Vertebrate Use of Nontidal Wetlands on Galveston Island, Texas.

By Allan J. Mueller . 215

Correlation Between Suspended Sediment and Other Water Quality
Parameters in Small Steams of Forested East Texas. By Alfredo

B. Granillo, Mingteh Chang, and Edward B. Rashin . 227

Atmospheric Carbon Monoxide in the El Paso-Cd. Juarez Area. By

Manuel Aguirre, Jr., and Howard G. Applegate . 235

Factors Influencing Cellulolytic Activity of the Soil Fungus,

Aspergillus candidus. By J. Ortega and E. J. Baca . 245

Comparative Behavior and External Color Patterns of Two Sympatric Centipedes
(Chilopoda \Scolopendra) from Central Texas. By Raymond W. Neck . 253

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Editor:

J. Knox Jones, Jr., Texas Tech University
Assistant to the Editor:

Marijane R. Davis, Texas Tech University
Associate Editor for Botany:

Randy Moore, Baylor University
Associate Editor for Chemistry:

Marvin W. Rowe, Texas A&M University
Associate Editor for Computer Science:

Ronald K. Chesser, Texas Tech University
Associate Editor for Mathematics and Statistics:

George R. Terrell, Rice University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Scholarly papers in any field of science, natural history, or
technology will be considered for publication in The Texas Journal of
Science. Instructions to authors are published one or more times each
year in the Journal on a space-available basis, and also are available
from the Editor (The Museum, Box 4499, Texas Tech University,
Lubbock, Texas 79409).

JOINT SETS IN THE PRECAMBRI AN-PALEOZOIC
ROCK SUCCESSION OF THE LLANO UPLIFT, TEXAS

by J. R. ALNES and S. E. CEBULL

Department of Geosciences
Texas Tech University
Lubbock, Texas 79409

ABSTRACT

The strike of 3881 joints was measured at 40 widely scattered localities across the
Llano Uplift of central Texas. Exposed at these 40 localities are 15 rock units that range
in age from Precambrian to Cretaceous. At least three generations of joints are present.
These are of probable Precambrian, Pennsylvanian-Cretaceous, and post-Cretaceous age.
The second is the best defined; it is characterized mainly by three regional joint sets
(groups of joints that comprise more than five percent of the total number and that
within two standard deviations of the group), the general azimuths of which are N88°W,
N46°W, and N46°E. The latter is the most prominent set. It parallels the normal faults
of the Llano Fault System of Atokan age, and, like the faults, displays no geometric
relation to the domal shape of the Uplift, which developed during Mississippian-
Pennsylvanian time. It is possible that the joint set and Fault System are related; if so,
second generation joint sets are of probable Pennsylvanian age. The joint sets appear
to be superimposed on the domal structure. Key Words : joint sets, Llano Uplift,
structural geology.

INTRODUCTION

The Llano Uplift of central Texas is an enigmatic domal structure
that lies only a few kilometers north and west of the buried Ouachita
tectonic belt, an orogen of Pennsylvanian age. The Uplift’s possible
origin and tectonic significance remain unexplained. Although much
detailed structural data in the Uplift area is available, no
comprehensive study of joint patterns has been undertaken, perhaps
because such studies too commonly prove unproductive. Nonetheless,
this study focuses on this aspect of the structure of the region.

The interior of the Llano Uplift consists of exposed cratonal,
igneous and metamorphic rocks of Precambrian age. These rocks are
principally granite, schist, and gneiss that display radiometric ages on
the order of a billion years. The exposures of crystalline basement rock
are rimmed by a succession of middle Cambrian to Pennsylvanian and
Cretaceous strata, which also occur as erosional outliers in the interior
of the Uplift. These strata are primarily of limestone, dolostone, and
sandstone.

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

124

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Although many of the Precambrian units were subjected to intense
deformation about 1.0 to 1.15 billion years ago (“Llano Orogeny’’ —
King 1969), post-Precambrian deformation is unspectacular, resulting
chiefly in doming, normal faulting, and jointing. Doming produced
generally low, quaquaversal dips of the sedimentary units. Such dips
are mainly on the order of a few degrees, with greater dips occurring
only locally. The pattern is disrupted by faults, however.

The structurally high position of the region is the product of two
to three intervals of domal uplift during Carboniferous time,
beginning in middle Mississippian and possibly ending in or before
late Pennsylvanian time (first noted by Sellards and Baker 1934). An
angular unconformity in Atokan strata coupled with clastic
sedimentation northward off the Uplift during Atokan and
Desmoinesian time suggests that the latest and principal doming
occurred in Atokan time.

The Uplift is transected by a system of normal faults (“Llano
system’’ of Sellards and Baker 1934) that strike chiefly northeast, a
pattern that clearly fails to conform to the domal shape. Most of the
faults display modest displacement, although some offsets are in excess
of 600 meters (Clabaugh and McGehee 1972). These faults developed
after deposition of the Bend Group and before that of the Strawn
Formation (Sellards and Baker 1934). The relationships are evident on
the Llano Sheet of the Geologic Atlas of Texas (Barnes 1981). Faults
cut all units up to and including those of lower Atokan age but not
those of upper Atokan age. Hence, most faults are of approximately
middle Atokan age. They extend well beyond the limits of the Uplift.

As suggested above, published studies of the joint pattern of the
Llano region are not extensive. Hutchinson (1956) described northeast
and northwest joint sets in the Enchanted Rock batholith, and Boyer
et al. (1961) reported the presence of four joint sets in the seven-square-
mile (18.1 km2) area of the Red Mountain Gneiss. The latter sets strike
in east, northeast, northwest, and north directions; dips range from
75°-90°. Northeast and northwest sets are the most prominent. Boyer
et al. (1961) also reported that the northwest set is the oldest, followed
by the east, northeast, and north sets, in order of diminishing age.
They noted that the older joints are mineralized and that the
mineralization is of Precambrian age and suggested that reactivation of
joints, especially the northeast set, occurred later as a result of
Paleozoic uplift.

This study expands on those of these earlier works. Its primary
purpose is to delineate the pattern of joint sets over the whole of the
Uplift by sampling techniques and procedures similar to those utilized
by Holst and Foote (1981) in their recent study of joints in Devonian
rocks in Michigan.

JOINT SETS OF LLANO UPLIFT

125

PROCEDURE

The strike of 3881 joints was measured at 40 localities at which 15
different rock units are exposed. Among the 40 localities, widely
scattered over the Uplift, seven Precambrian units (19 localities) are
represented. These include the Valley Spring Geniss, Packsaddle
Schist, Big Branch Gneiss, Red Mountain Gneiss, Town Mountain
Granite, Oatman Creek Granite, and the Sixmile Granite. Also
included in the 40 localities are the Wilberns and Riley formations of
Cambrian age (eight localities), the Tanyard, Gorman, and Honeycutt
formations of Ordovician age, and the stribling Formation of
Devonian age (five localities in Ordovician-Devonian rocks combined),
the Marble Falls Limestone of Pennsylvanian age (one locality), and
the Glen Rose Formation of Cretaceous age (four localities). The
Precambrian units include both plutonic igneous and metamorphic
rocks as well as rocks with both isotropic and anisotropic (well-
foliated) fabrics. The Phanerozoic rocks are exclusively of sedimentary
(chiefly carbonate) origin. Thus, to the degree possible, units studied
include rocks not only from diverse locations but also of different
lithologies and age.

At each locality the strike of the first hundred joints encountered was
measured, except where less than a hundred fractures were exposed;
then all were measured. Holst and Foote (1981) found this procedure
to be statistically valid. At many places the dips of joints also were
determined, but these showed little variation, mostly dipping 80
degrees or more.

For each locality, groups of joint strikes were delineated and the
median strike and approximate confidence level for each group were
determined. Strikes that fell outside an approximate 95 percent
confidence level were discarded, and new median and confidence levels
were calculated. If more than five percent of strikes at a single location
remained within the group, it was designated a joint set, albeit a local
one.

RESULTS— JOINT SETS

Strikes of joints from all localities are grouped into different
categories by age of unit and, partly, by lithology and displayed on
rose diagrams in Figure 1. Strikes in Precambrian rocks are shown on
four diagrams (Fig. 1A-D) each representing different gross litologies:
gneiss (Fig. 1A; Valley Spring, Red Mountain, and Big Branch
gneisses), schist (Fig. IB; Packsaddle Schist), granite with aligned
feldspar (Fig. 1C; Town Mountain Granite), and granite with isotropic
fabric (Fig. ID; Oatman Creek and Sixmile granites). Strikes in
Phanerozoic rocks are shown on diagrams E-H, which represent

126

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Figure 1. Rose diagrams of strikes of joints (A-H), faults (I) and regional joint sets (J).
A-I are plots of “raw” field data; J is based on data summarized in Figure 2. A-D
represent Precambrian units of gneiss, schist, granite with aligned feldspar, and
granite with an isotropic fabric, respectively. E-H represent Cambrian, Ordovician-
Devonian, Pennsylvanian, and Cretaceous units, respectively. I illustrates strike of
faults. J illustrates strikes of regional joint sets (described in text). For A-H, first
number in parentheses is number of joints measured, and the second is the number
of locations; for I, number in parentheses is number of faults.

JOINT SETS OF LLANO UPLIFT

127

Cambrian, Ordovician-Devonian, Pennsylvanian, and Cretaceous
units, respectively.

When these “raw” data were analyzed by individual locality, it was
found that each locality is characterized by one to five local joint sets.
Percentage of joints that fall into local joint sets (as defined above) at
all localities range from 29 to 86.

Data from these individual localities were then combined by rock
group and analyzed by the same technique (Holtz and Foote 1981) to
obtain the regional pattern of (local) joint sets. Table 1 illustrates the
results of this analysis. It shows regional joint sets for each group of
rocks as well as the standard deviation within each set. Overall, the
data show that Precambrian through Devonian units are characterized
primarily by three regional joint sets, the average median strike of each
being N88°W, N46°W, and N46°E (see Fig. 1 J). Principal exceptions to
this pattern were found in Precambrian granitic rocks, which exhibit
an additional joint set with an average strike of N15°W, and in
Ordovician-Devonian units, which contain the N46°W set only locally
(but display two NE sets — Table 1). A Pennsylvanian unit was studied
only at one locality, so data are meager. However, three joint sets were
delineated at that locality (Fig. 1G). At the Precambrian-Pennsylvanian
localities, more than 68 percent of joints fall into well-defined,
regional joint sets (although Ordovician-Devonian units have only 39
percent within the sets). Cretaceous rocks display a different pattern.
They possess a single joint set comprised of only 14 percent of joints
in rocks of this age. However, this set has an average strike of
approximately N45°E and, therefore, is similar to the northeast-
striking set of the older rocks. The azimuth of this set also is similar
to the strike of many of the faults of the Uplift, 70 percent of which
are in the range of N20°-60°E (Fig. II).

Interestingly, joint sets in rocks with well-defined anisotropic
fabrics, such as schist and gneiss, give no clear indication that their
orientations were affected by that of the foliation. Not only are joint
sets similar in rocks with both isotropic and anisotropic fabrics
(compare gneiss and schist with isotropic granite in Table 1 and
Figure 1A and IB with ID), but joints were observed in the field to
cut indiscriminantly across foliation surfaces. However, the cherty
nature of Ordovician-Devonian rocks may be responsible for the lack
in these units of the regional NW-striking joint set, the occurrence of
an additional NE-striking set, and a reduction in the percentage of
joints within joint sets. This possibility is suggested because units with
lower percentages of chert, but otherwise similar lithologies, were
observed to contain fewer but better defined joint sets, thereby
indicating that chert-rich units are characterized by a more diverse
pattern of joints.

128

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Finally, the similarity of joint-set number and orientations in the
Precambrian through Devonian, and probably Pennsylvanian, rock
succession raises the question as to whether joints of the Cambrian and
younger rock units mimic reactivated older joints, suggesting the
presence of at least two joint generations, or if only a single generation
of joints is represented. As noted by Boyer et al. (1961), some of the
joints in Precambrian rocks are mineralized with quartz or granitic
material and, hence, are of Preambrian age. Although data on joints
of clearly Precambrian age are sparse, constituting only about 2.5
percent of joints in Precambrian rocks, such joints commonly strike
northeast or northwest and are everywhere cut by unmineralized joints.
Also, field observations show that strata in close proximity, or even in
contact, may contain different patterns and numbers of local joint sets.
Therefore, although mimicking due to reactivation of older joints may
be a factor in development of the joint-set pattern, it probably is not
profound.

RELATIONSHIP OF REGIONAL JOINT SETS TO UPLIFT AND FAULTS

The foregoing evidence suggests that the joints of the Llano Uplift
are of at least three generations. The first generation is represented, in
part, by the mineralized joints; the second by the three regional joint
sets; the third by joints in Cretaceous units, the pattern of which is
unlike that of older rocks. The age of the first generation is clearly
Precambrian. The second, which is expressed regionally in rocks
probably as young as Pennsylvanian (Marble Falls Limestone), is
between Pennsylvanian and Cretaceous in age. The third is post
Cretaceous (that is, post-Glen Rose). Each may be influenced
somewhat by its predecessor(s), accounting especially for the
commonality of the northeast sets in rocks of all ages, but probably
each is mostly independent in terms not only of age, but also of
pattern and origin. Of these, the second generation is the most
interesting, partly because it constitutes the best defined of the joint
sets, but also because its age is similar to that of other significant
tectonic features of the region.

As outlined earlier, the dominantly northeast-striking normal fault
system of the Uplift (Fig. II) is of approximately middle Atokan age.
Doming developed during Mississippian through Pennsylvanian time,
with perhaps the most significant doming in Atokan time. These ages
approximately coincide with the earliest possible time for second-
generation jointing. Approximate relative timing of jointing, faulting,
doming, and Ouachita orogenic activity is summarized in Figure 2.
The close temporal relation suggests that jointing might be associated
with doming or faulting, or both. As regards the latter, a close

Table 1.— Median strike of regional joint sets and standard deviation for each group of units. Numbers in parentheses are percent of joints
in set. "Average” includes only those sets that consistently overlap within two standard deviations. Resulting three main joints sets are

JOINT SETS OF LLANO UPLIFT

129

fao v

Cfa 13

TJ 60

13 73
c/3 ^

o B

RJ £
<*_, 03

0 c/3

E .-g

£ c/$

13 C/3

XI xi
w u

C 2
u

~a -

~ 'S

£ fac

CU o

O if) N tO
oi ifi o cfi

W W Ui w

03 tf)

^ m 80

2 2 2 2

CO N If3 - O

6 d d ^ d

+1 +1 +1 +1 +1 +1

£ £ £ £ £ £

— CO O — I — I

— I CM »T3

z z z z z z

—

gt> — cd

+1 +1 +1 +1

£ £ £ £
S £ « §
z z * z

c/5

kO I>-
^6 in

+i +i +i +i +i +i +i

w w

cm m

GO *03

2 2

+1 +1

c S c c
E o .E

u £ £ ^

l-s S&l

u

03

c

03

C C «*

.2 03 3

•C c > o
’> > *T o
o i; c 5

"g Q e <u

O (X CJ

#

03 CO

£ £

co m

_c

2 2

130

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Mississippi/^ Pennsylvanian Permian, , .Cretaceous

3>

-i

O

>

? 7 _ 7 _ 7 _ 7 . . 7

JOINTING

FAULTING

— ? - ►

UPLIFT

OUACHITA OROGENY

- - - -►

Ouachita Marathon

Region Region

Figure 2. Summary of age relationships between regional, second-generation joint sets
in Precambrian-Pennsylvanian rocks, faults, and domal uplift. Also shown is the
range of probable ages of the Ouachita Orogeny, which is older in the Ouachita
region to the northeast and younger in the Marathon region to the west.

geometric relation also exists, the strike of the fault system being close
to that of the northeast set of second-generation joints (Fig. 1). Joints
of this generation constitute the most prominent set, and it is also this
set that may have been present in Precambrian time and at least partly
“resurrected” in post-Cretaceous time. Thus, jointing and faulting
may well be associated. However, neither jointing nor faulting display
patterns that conform in any understandable way with the domal
geometry of the Uplift. Accordingly, they must be regarded as being
independent of the doming process.

SPECULATION

It is tempting to speculate as to the tectonic origin of the second-
generation regional joint sets in the Precambrian-Paleozoic rock
succession, especially inasmuch as their interpreted age interval
extends through the period of Ouachita Orogeny (chiefly
Pennsylvanian, Fig. 2) and that of the continental breakup (Triassic)
that initiated the opening of the present Gulf of Mexico. If this
jointing is a product of the former, it probably is mostly of
Pennsylvanian age; if of the latter, it is likely to be Triassic. There is
no adequate way to solve this question at present. Despite the fact that
normal faults in general suggest an extensional regime, those discussed
here appear to be temporally associated with the compressional
orogenic belt that is thought by many to be a product of continent-
continent collision. On the basis of temporal considerations, we

JOINT SETS OF LLANO UPLIFT

131

suspect, but cannot demonstrate, that the joint sets also are affiliated
with the Ouachita event.

CONCLUSIONS

The Llano Uplift displays at least three generations of joints. The
most prominent one is the second. It affects rocks of Precambrian
through Pennsylvanian age and is expressed chiefly in the form of
three regional joint sets. These sets may reflect the orientations of
joints formed in Precambrian time, and post-Cretaceous joints may
mimic earlier joint patterns, but the general effects of mimicking are
limited. The prominent regional joint sets appear unrelated to the
doming process; it appears they are superimposed on the dome.
Probably they are related to faulting and, therefore, are of
Pennsylvanian age, but this conclusion is uncertain.

LITERATURE CITED

Barnes, V. E. 1981. Geologic atlas of Texas, Llano sheet. Univ. Texas Bureau Eco.

GeoL, Austin, one sheet, scale 1:250,000, 15 pp.

Boyer, R. E., S. E. Clabaugh, C. H. Gates, and J. R. Moffett. 1961. Comparison of
two joint study methods applied to the Red Mountain Gneiss, Llano County, Texas.
Texas J. Sci. 13:478-486.

Clabaugh, S. E., and R. V. McGehee. 1972. Precambrian rocks of Llano region. Pp.
9-23, in Geology of the Llano region and Austin area, Univ. Texas Bureau Eco.
Geol. Guidebook 13.

Holst, T. B., and G. R. Foote. 1981. Joint orientations in Devonian rocks in the
northern portion of the Lower Peninsula of Michigan. Bull. Geol. Soc. Amer. 92:85-
93.

Hutchinson, R. M. 1956. Structure and petrology of Enchanted Rock Batholith,
Llano and Gillespie counties, Texas. Bull. Geol. Soc. Amer. 67:763-806.

King, P. B. 1969. The Tectonics of North America — a discussion to accompany the
tectonic map of North America. Scale 1:5,000.000, U.S. Geol. Surv. Prof. Paper 628.
Sellards, E. H., and C. L. Baker. 1934. The geology of Texas. Vol. II, Structural and
economic geology. Univ. Texas Bull. no. 3401.

Present address of Alnes: Union Oil Company of California, P.O. Box 671, Midland,
Texas 79701.

W\

IMMUNE RESPONSE IN THE BOBWHITE QUAIL,
COLIN US VIRGINIA NUS

by ALFRED C. SCHRAM, HERBERT F. GONZALEZ,
and CYNTHIA D. MEADOR

Departments of Chemistry and Biology
and the Killgore Research Center
West Texas State University
Canyon , Texas 79016

ABSTRACT

Immunizations of bobwhite quail, Colinus virginianus, with a mixture of bovine
serum albumin (BSA) and formalinized rabbit erythrocytes (FRE) elicited two types of
immune response. FRE-agglutinating antibodies were found for a short duration,
whereas a BSA-binding antibody response lasted longer and was significantly enhanced
by a secondary immunization. Immunoelectrophoresis indicated a closer phylogenic
relationship between the chicken, Gallus domesticus, and the bobwhite quail than
between the bobwhite quail and the scaled quail, Callipepla squamata.

INTRODUCTION

The immune response of the bobwhite quail, Colinus virginianus ,
was investigated as part of a study of induced immunity in the local
fauna. The bobwhite quail is a small bird found from Alaska to
Mexico. Although the number of bobwhite quail in the vicinity of
Canyon has been reduced in recent times, coveys are still observed
along roadsides and creek banks. A survey of the recent literature (1968
through 1980) failed to reveal mention of bobwhite quail immunology.

MATERIALS AND METHODS

The experimental group of quail consisted of a breeding pair
obtained from a wild covey, and three of their offspring, which had
been hatched in an incubator and were 10 months old at the beginning
of the experiment. During the experiment, the birds were kept in a
large cage, located near an east window in an air-conditioned
building. They were supplied with water and chicken feed ad libitum ;
their diet was occasionally supplemented with seed of wild grasses and
with small insects. The birds were immunized with a total of 200 pg
of bovine serum albumin (BSA) and 50 /xl of packed formalinized
rabbit erythrocytes (FRE) (Csizmas 1960) in a 50 percent emulsion in
Freund’s complete adjuvant, distributed between two intramuscular
injections, one in each thigh. Nine weeks later, a booster injection of

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

134

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

100 ng of BSA and 25 /jl\ of FRE in Freund’s incomplete adjuvant was
given in the breast muscle.

Small amounts of blood (0.5 to 1.0 ml) were obtained by heart
puncture at 0, 2, 5, 6, 8, and 11 weeks following the initial
immunization. The serum was kept frozen until fractionation by gel
filtration on Sephadex G-200; the conditions for the fractionation have
been described earlier (Schram 1970a; Schram and Christenson 1971),
and are summarized in the caption of Figure 1. The protein content
of each fraction was determined both by the formula (Layne 1963): mg
protein/ml = (1.55 X Absorbance at 280 nm — 0.76 X Absorbance at
260 nm), and by the Lowry method (Lowry et al. 1953). The
carbohydrate content was measured by the anthrone method (Morris
1948). Agglutination of FRE was carried out in 5 X 50 mm microtubes
containing 50 /jl 1 of two-fold serial dilutions of each fraction in normal
saline buffered with phosphate buffer at pH 7.2 (PBS) and 50 yul of a
0.2% (v/v) suspension of FRE in PBS. BSA-binding antibodies were
detected by a radioimmunoassay (Farr 1958) using 200 yul of each
fraction and 1 /xg of iodine- 125 labeled BSA (prepared following the
method of Day et al. 1967, to a specific activity of 0.025 mCi/mg).
Finally, cholesterol levels were determined (Leffler 1959) on one-half
of the ethyl ether extract of the portion of each fraction left after the
other assays had been performed. Cholesterol determinations were
carried out to provide some reference on the level and form of
cholesterol in the circulatory system of the bobwhite quail. For the
immunoelectrophoresis, rabbit antisera to chicken, to pigeon serum,
and to scaled quail serum were obtained a week after the tenth
monthly injection of 200 /xl of the appropriate serum mixed with
incomplete Freund’s adjuvant (Schram and Lowe 1973). The chicken
serum for the immunization was pooled from samples obtained from
six White Leghorn roosters; the pigeon serum was pooled from 12
pigeons; the scaled quail serum was pooled from only two quail.
There was not enough bobwhite quail serum available to raise a rabbit
antiserum to bobwhite quail.

RESULTS

Through gel filtration on Sephadex G-200 (Fig. 1), bobwhite quail
serum proteins separated, by eluting order, into macroglobulins, a
gammaglobulin-like fraction, and an albumin-like fraction. Porcine
gammaglobulin and BSA eluted with a maximum at fractions 25 and
35 respectively, under identical fractionating conditions. The total
serum protein content, obtained by integration of the area under the
corresponding portions of the curve and corrected for the volume of
serum used in the fractionation averaged 41 mg/ml (Layne method),

IMMUNE RESPONSE IN QUAIL

135

Figure 1. Sephadex G-200 fractionation of 0.25 ml of bobwhite quail serum, obtained
11 weeks after the first immunization (two weeks after the booster immunization).
The proteins and the carbohydrate contents of each fraction have been expressed as
a percent of the total amounts eluted. Agglutination of FRE and jug of 125I-BSA
bound have been converted to values corresponding to 1 ml of serum. Conditions for
fractionation: 1.5 X90 cm column; 0.1 5M NaCl containing 0.02M Tris buffer pH 8.6
and 0.02 percent NaN3; rate of collection: 0.1 ml per minute; fraction size: 2.0 ml.

corresponding to 24 percent macroglobulins and 76 percent albumin
and proteins of intermediate molecular size; or 38 mg/ml (Lowry
method), corresponding to 18 percent macroglobulins and the balance
in albumin and proteins of intermediate molecular size. Similarly, the
total carbohydrate content averaged 1.97 mg/ml, with 0.54 mg/ml
contributed by free glucose. The total cholesterol level averaged 2.8
mg/ml of serum, of which 0.75 mg/ml was eluted with the
macroglobulin fraction and the rest, as free cholesterol. Agglutination
of FRE was maximal in the macroglobulin fraction (fractions 15 and
16), whereas BSA-binding antibodies were eluted in the next group of
proteins (maximum at fractions 23 and 24). Agglutination titers and
BSA-binding levels, measured on whole serum dilutions rather than on

136

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

E

3

30 ®

CD

Cl

T3

c

Z3

O

XI

<

CO

OQ

Figure 2. Antibody response in a bobwhite quail. The first immunization was given at
time 0 and the second one at nine weeks. Agglutination titers are expressed as the
maximum serum dilution in PBS showing a clear agglutination of FRE (supplied
as an equal volume of a 0.2 percent suspension in PBS). Antigen binding titers are
expressed as /zg of l25I-BSA specifically bound by the proteins of 1.0 ml of serum.
An aliquot of the eleventh week serum was used for the fractionation recorded on
Figure 1.

the fractions, were maximal six weeks after the primary injection,
decreased rapidly by the eighth week, and were stimulated by the
booster injection given at the ninth week (Fig. 2). But the anamnestic
response in hemaglutination titer was weaker than the response to the
primary injection. On the other hand, following the second injection,
the antigen-binding titer rose considerably above the original
maximum level.

The serum proteins of the bobwhite quail were compared to those
of other birds by immunoelectrophoresis. Not enough quail serum was
available for a prolonged course of immunization in rabbits, to obtain
antiserum to the bobwhite quail serum. But rabbit antiserum to
chicken serum revealed a high degree of similarity between chicken
and bobwhite quail serum (Fig. 3a and 3b), as would be expected
between two galliform birds. Rabbit antiserum to scaled quail serum
however, reacted with fewer bobwhite quail serum components. The
differences are most obvious in the albumin region of the
electropherograms. Bobwhite quail albumin on the other hand must
be immunologically and electrophoretically similar to pigeon albumin
(Fig. 3g and 3h). Chicken albumin, although immunologically

IMMUNE RESPONSE IN QUAIL

137

Figure 3. Tracing of immunoelectrophoretic patterns. Wells: a, c, e — chicken serum (10
yul); b, d, h — bobwhite quail serum (10 /jl\ ); f, g — pigeon serum (10 /d). Troughs: 1 —
rabbit antiserum to chicken serum; 2 — rabbit antiserum to scaled quail serum; 3 and
4 — rabbit antiserum to pigeon serum.

Electrophoresis was carried out for 150 minutes at 10 ma and 200 VDC in 1 percent
agar and 1 percent sodium barbital, pH 8.6. The antisera were allowed to diffuse for
five days at 4° C; the agar slides then were dialyzed overnight versus normal saline;
they were stained for two hours with 0.02 percent Aniline Blue in 10 percent acetic
acid, then cleared for 2 hours in several changes of distilled water, and finally
immersed for 30 minutes in 1 percent glycerol containing 0.5 percent acetic acid,
before air-drying.

reacting with antiserum to pigeon serum, has a higher electrophoretic
mobility than both pigeon and bobwhite quail serum albumin.

DISCUSSION

The pattern of elution of bobwhite quail serum proteins on
Sephadex G-200 was similar to that of the Japanese quail (Schram and
Christenson 1971) and of other galliform birds (Schram 1970b; Schram
and Lowe 1973; Schram and McNabb 1975). Although bobwhite and
Japanese quail are phenotypically similar, there seem to be two major
differences in their serum proteins. Bobwhite quail serum has a higher
proportion of proteins of molecular size in the neighborhood of
175,000 daltons and the ratio of tyrosine to the other aromatic amino
acids in this portion of the bobwhite quail serum is higher than in
the corresponding portion of the Japanese quail serum (based on the
comparison of the protein determinations by the Layne and the Lowry
methods).

The immune response to a particulate antigen (FRE) as measured by
active hemagglutination, was relatively weak, of short duration, and

138

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

was not brought back to its original maximum level by a second
injection (Fig. 2). Th agglutinating antibodies were found mostly in
the macroglobulin fraction and thus must be of the IgM type. The
response was detected over a six-week period; the second immunization
did not markedly raise the agglutination titer, but induced the
formation of agglutinating antibodies of lower molecular size (found
in fractions 23 and 24, Fig. 1). On the other hand, BSA-binding
antibodies of lower molecular size (found in fractions 23 and 24, Fig.
1) must be of the IgG molecular size and would be expected to increase
after a booster injection. There were individual variations in serum
antibody titers and for that reason, the responses recorded in Figure 2
were measured on sequential serum samples of the same bird. On the
whole, the immune response in the bobwhite quail was similar to that
observed in other birds and in mammals. The relatively low response
to FRE and the higher response to BSA are, of course, dependent on
the artificial method of immunogenic stimulation (intramuscular
injections in Freund’s adjuvant emulsions). Unfortunately, there were
no other birds of the same flock available for a study of their immune
response to bacterial antigens, following a less artificial exposure to
the immunogens.

The carbohydrates and cholesterol levels measured were only
approximate, because the amounts available for analysis were small
and close to the limit of detection by the anthrone and Leffler methods
respectively. Nevertheless, it appeared that two-thirds of the total
carbohydrate and one-third of the total serum cholesterol were
associated with large molecular aggregates, eluting with the void
volume of the Sephadex column. Some carbohydrate was also
associated with intermediate size proteins.

Through immunoelectrophoresis, bobwhite quail and chicken serum
proteins appear to share a large number of determinants (Fig. 3a and
3b); rabbit antiserum to scaled quail serum also reveals similarities in
the gamma globulins of chicken, bobwhite quail, and scaled quail
serum (Fig. 3c and 3d). The rabbit antiserum to pigeon precipitates
several chicken serum proteins in the albumin through the beta-
globulin region of electrophoretic separation, but not chicken gamma
globulins (Fig. 3e). The same antiserum reacts in a similar fashion
with bobwhite quail serum (Fig. 3g).

Altogether, the cross-reactivities reveal more similarities in the
antigenic determinants of the serum proteins of chicken and of
bobwhite quail than in the determinants of serum proteins of scaled
quail and bobwhite quail. These similarities suggest a closer
phylogenic relationship between chicken and bobwhite quail than
between scaled quail and bobwhite quail. Chicken serum albumin has
a higher electrophoretic mobility than bobwhite quail albumin,

IMMUNE RESPONSE IN QUAIL

139

although both appear to have the same molecular size, because they
elute in the same fraction (maximum at fraction 35, Fig. 1). The
strong reaction with rabbit antiserum to chicken serum suggests also
that both albumins share a large number of common antigenic
determinants, but differ in the number of acidic or basic amino acids,
or in their ratio. Because it appears as a single electrophoretic
component, it must be more homogeneous than the other serum
proteins, which segregate in at least eight minor and one major
component (Fig. 3b).

ACKNOWLEDGMENTS

This investigation, which was carried out in the Killgore Research
Center, was supported in part by a grant from the Committee on
Organized Research, West Texas State University. We also wish to
thank the Department of Surgery at Texas Tech University Health
Sciences Center in Lubbock for use of equipment.

LITERATURE CITED

Csizmas, L. 1960. Preparation of formalinized erythrocytes. Proc. Soc. Exp. Biol. Med.
103:157-160.

Day, E. D., S. Lassiter, and R. B. Fritz. 1967. Radioiodinadon of antibodies adsorbed
to insoluble antigens. J. Immun. 98:67-71.

Farr, R. S. 1958. A quantitative immunochemical measure of the primary interaction
between 131I-BSA and its antibody. J. Infect. Dis. 103:239-263.

Layne, E. 1963. Methods of enzymology (S. Colowick and M. O. Kaplan, eds).,
Academic Press New York, 6:451-452.

Leffler, H. H. 1959. Estimation of cholesterol in serum. Amer. J. Clin. Path. 31:310-
313.

Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1953. Protein
measurements with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

Morris, D. L. 1948. Quantitative determination of carbohydrate with Dreywood’s
anthrone reagent. Science 107:254-255.

Schram, A. C. 1970.a. Serum proteins of the domestic goose, Anser anser. Comp.
Biochem. Physiol. 31:81-87.

- . 1970b. Serum proteins of the silver pheasant, Gennaeus nychteremus. Comp.

Biochem. Physiol. 36:481-492.

Schram, A. C., and C. P. Christenson. 1971. Serum proteins of the Japanese quail,
Coturnix coturnix japonica. Comp. Biochem. Physiol. 39:789-796.

Schram, A. C., and A. D. Lowe. 1973. Serum proteins of the Texas blue quail,
Callipepla squamata. Comp. Biochem. Physiol. 44B:53-58.

Schram, A. C., and O. B. McNabb. 1975. Serum proteins of the golden pheasant,
Chrysolophus pictus. Comp. Biochem. Physiol. 52B.449-451.

t H

URNATELLA GRACILIS (ENTOPROCTA) FROM
CADDO LAKE, TEXAS AND LOUISIANA

by THOMAS M. CUSAK and JACK D. McCULLOUGH

Department of Biology
Stephen F. Austin State University
Nacogdoches, Texas 15962

ABSTRACT

Urnatella gracilis Leidy occurred in benthic samples collected on 7 November 1982
from two collecting sites on the Louisiana side of Caddo Lake. The species may have
some tolerance of occasional crude oil and brine water contamination. Key words-.
Entoprocta, Urnatella, Caddo Lake.

Urnatella gracilis Leidy is the only freshwater entoproct known from
North America. The species has been reported only five times in the
United States west of the Mississippi River: Lake Dallas, Texas (Weise
1961); Oklahoma (Harrel 1967); Trinity River, Texas (McCullough and
Smith 1975); Angelina River, Texas (Cox and McCullough 1976); and,
Old River Lake, Louisiana (McCullough et al. 1981).

Several specimens of U. gracilis were found in benthic samples
collected on 7 November 1982 at two sites on the Louisiana side of
Caddo Lake (Red River drainage — Harris and Marion counties Texas,
and Caddo Parrish, Louisiana). Station 1 was approximately 1.1 km
south of Oil City, Louisiana, in the open water area known as Big
Lake, and was located near the offshore EMCO oil well number 289.
Water depth was about two meters at this station. Station 2 was
approximately three kilometers north of the first station and about 200
meters west of the mouth of Tiger Branch in the northern part of
James Bayou near Boat Lane Marker E-76. More than 50 producing
offshore oil wells were located near this collection site. Water depth
here also was about two meters.

Given that U. gracilis occurred in the vicinity of a large number of
offshore oil wells, the species may have some tolerance for occasional
crude oil and brine water contamination. On 8 February 1982, near
Station 2, a surface water sample had 2400 milligrams of crude oil per
liter of lake water. Water samples taken at the same time and stations
where U. gracilis was found, had sodium values that ranged from 19
to 35 mg per liter, chemical oxygen demand (COD) of 24 mg per liter,
conductivity of 175 micromhos per centimeter and chloride
concentrations ranging from 31 to 60 mg per liter.

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

142

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

LITERATURE CITED

Cox, D., and J. D. McCullough. 1976. Urnatella gracilis Leidy from the Angelina River,
Texas. Texas J. Sci. 27:489.

Harrell, R. C. , and C. S. Wallis. 1967. Urnatella gracilis Leidy (Entoprocta) new to
Oklahoma. Southwestern Nat. 12:203.

McCullough, J. D., C. W. Reed, and J. L. Jones. 1981. The occurrence of Urnatella
gracilis Leidy in Old River Lake. Louisiana and associated limnological conditions.
Proc. Louisiana Acad. Sci. 44:14-16.

McCullough, J. D., and B. Smith. 1975. Some ecological observations on Urnatella
gracilis Leidy. Southwestern Nat. 20:171-176.

Weise, J. G. 1961. The ecology of Urnatella gracilis Leidy: Phylum Entoprocta. Limnol.
Oceanogr. 6:228-230.

VESICLE CONTRIBUTION TO CYST WALL
FORMATION OF POSTHODIPLOSTOMUM MINIMUM
METACERCARIAE

by THOMAS G. MEADE and JOSE M. GARZA

Division of Life Sciences
Sam Houston State University
Huntsville, Texas 77341

ABSTRACT

Scanning electron microscopy indicates that the primary cyst wall (PCW) of
Posthodiplostomum minimum metacercariae is of parasite origin and the outer fibrous
coat (OFC) of host origin. Fusion of parasite produced vesicles into the (PCW) is shown.
Key words : parasite, metacercariae, cyst wall.

Previous studies have considered the composition of the
metacercarial cyst wall of the diplostome trematode, Posthodiplosto¬
mum minimum (Mitchell 1974; Mitchell and Crang 1976). These cysts,
common in freshwater bluegill , Lepomis macrochirus, consist of an
outer fibrous coat (OFC) and an inner primary cyst wall (PCW) with
an interface zone separating the two layers (Fig. 1). Careful observation
with TEM has shown that these vesicles are produced by the
metacercariae and appear to contribute to the composition of the PCW.
Mitchell and Crang (1976), using SEM, found obvious cystogenous
vesicles on all inner cyst coats observed. Our SEM study revealed
vesicles both at the metacercarial surface (Fig. 1) and at the PCW (Fig.
2). Figure 3 seems to verify the role of the vesicles as major
contributors to the PCW by clearly showing their integration into it.
On the basis of the author’s earlier studies (Harvey and Meade 1969,
Crider and Meade 1975), we believe the OFC to be primarily of host
origin and the PCW of parasite origin. The present micrographs
demonstrate the parasite’s contribution to the PCW. It seems
reasonable that other kinds of metacercariae probably contribute to cyst
wall formation in similar fashion.

ACKNOWLEDGMENTS

This investigation was supported by NSF grant — SPI-8 166367.

The Texas Journal of Science, Vol. XXXVII, Nos. 2&3, September 1985

144

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Figure 1. Fractured cyst with metacercaria (M) in upper left, cyst cavity (C) filled with
a meshwork of proteinacous material, cyst wall (CW), primary cyst wall (PCW) and
outer fibrous coat (OFC), X700.

i few ^ * J90

Figure 2. Another fractured cyst showing the same association of vesicles with the
primary cyst wall, X3100.

METACERCARIAL CYST WALL FORMATION

145

Figure 3. Enlargement of the area indicated by the arrow in Figure 1. The vesicles (V)
appear to have ruptured and fused to the primary cyst wall (PCW). OFC marks the
outer fibrous capsule, X5100.

LITERATURE CITED

Crider, R. and T. G. Meade. 1975. Immunological studies with Posthodiplostomum
minimum. Proc. Helm. Soc. Washington 1:23-28.

Harvey, J. S., Jr. and T. G. Meade. 1969. Observations on the effects of fish serum in
cercarial and metacercarial stages of Posthodiplostomum minimum (Trematoda:
Diplostomidae). Proc. Helm. Soc. Washington 2:211-215.

Mitchell, C. W. 1974. Ultrastructure of the metacercarial cyst of Posthodiplostomum
minimum (MacCallum, 1921). J. Parasit. 60:67-74.

Mitchell, C. W., and R. E. Crang. 1976. Posthodiplostomum minimum : examination of
cyst wall and metacercaria containing calcarious concretions with scanning electron
microscope and X-ray microanalysis. Exper. Parasit. 40:309-313.

ON THE ELEMENTS OF DIFFERENCE EQUATION
MODELING IN SOCIAL SCIENCE

by EVANS W. CURRY and DERALD WALLING

Departments of Sociology and Mathematics
Texas Tech University
Lubbock, Texas 79409

ABSTRACT

The mathematical bases of four well known models are discussed from the view of
difference equations so as to clarify their applications in social science. Details are given
pertaining to the selection of models.

INTRODUCTION

Social science researchers often are concerned with the
correspondence between an initiating variable x and a responding
variable y. This correspondence may be represented by a function y =
f(x) where f maps the independent variable x onto the dependent
variable y. The parametric form of f may lack sound theoretical and
mathematical bases if the researcher fails to recognize the necessity of
achieving isomorphism and reciprocality between substantive and
mathematical theory in quantitative analyses. This paper addresses
some of the mathematical issues that must be considered in choosing
between four of the more common of the two-parameter models. These
models are the linear model (y = a + bx), the logarithmic model (y
= a + b In x), the exponential model (y = aebx), and the power model
(y = axb).

These four models have a “related” mathematical basis. By
illustrating this relationship in terms of difference equations, a method
for selecting the correct model to match the theoretical situation is
presented. Mathematical modeling (when correctly implemented) often
provides theoretical and analytical insight. To achieve this, the models
developed must be both mathematically and substantively sound. This
dual soundness is absolutely important.

MATHEMATICAL PRELIMINARIES

Inasmuch as we are concerned with situations that can be modeled
in terms of difference equations and because we want to be able to
express these difference equations in terms of differential equations, it
is imperative that the variables x and y be continuous variables. We

The Texas Journal of Science, Vol. XXXVII, Nos. 2&3, September 1985

148

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

assume that f is continuous and differentiable in the domain of
interest. We shall designate the number Ax a simple change in x and
Ay the simple change in y that is brought about by Ax.
Correspondingly, we let the differential of x, dx, be the number Ax,
and define the differential of y, dy, by the equation of dy = f’(x)dx.
(For a complete discussion on increments and/or differentials, see
Swokowski, 1979.)

We shall term the ratio the relative change in x and the ratio
the relative change in y. Whereas Ax is simply a change in x,
is a change in x relative to the current value of x at the time of the
change. The use of implies that not only is the change Ax

important but the position of x at the time of the change is equally
important. Likewise, -^L is a change in y relative to the current value
of y at the time of the change. Hence, implies that not only is
the change Ay important but the position of y at the time of the
change Ay is equally important.

THE LINEAR MODEL

The basic assumption upon which the linear model rests is that a
change in y is directly proportional to a change in x. That is, a simple
change in y is related to a simple change in x. In mathematical terms,
this assumption can be stated as

Ay = b Ax, (1)

where b is a constant of proportionality. Corresponding to the
difference equation (1) is the differential equation

dy = b dx, (2)

Equation (2) may be integrated to derive the linear model

y = bx + a, (3)

where a is the constant of integration and all other terms are as
previously defined. The basic assumption of this model implies that
for a unit of change in the independent variable x there are b units
of change in the dependent variable y.

THE LOGARITHMIC MODEL

The basic assumption upon which the logarithmic model rests is
that a change in y is directly proportional to a relative change in x.
That is, a simple change in y is related to a relative change in x. In
mathematical terms, this basic assumption can be stated as

FUNDAMENTAL ISSUES— MODELING

149

Ay =b — , (4)

X

where b is the constant of proportionality. Corresponding to the
difference equation (4) is the differential equation

dy =b 4Z (5)

X

Equation (5) may be integrated to derive the logarithmic model

y = b In x + a, (6)

where a is the constant of integration and all other terms are as
previously defined.

An examination of (4) shows that it is necessary to assume that x
is not zero. This restriction is also in effect for the derived model (6).
The basic assumption of this model as stated in (4) implies that for
a unit of relative change in x there are b units of change in y.

THE EXPONENTIAL MODEL

The basic assumption upon which the exponential model rests is
that a relative change in y is directly proportional to a change in x.
That is, a relative change in y is related to a simple change in x. In
mathematical terms, this basic assumption can be stated as

^=bAx. (7)

y

Corresponding to the difference equation (7) is the differential
equation

= b dx. (8)

y

Equation (8) may be integrated to obtain

lny = bx + k, (9)

where k is the constant of integration and all other terms are as
previously defined. Solving for y in (9), the exponential model is
obtained as

y =

ae

bx

(10)

150

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

where a = exp(k). An examination of (7) shows that it is necessary to
assume that y is not zero. Likewise, y does not take on zero in the
derived model. The basic assumption of this model as stated in (7)
implies that for a unit of change in x there are b units of relative
change in y.

THE POWER MODEL

The basic assumption upon which the power model rests is that a
relative change in y is directly proportional to a relative change in x.
That is, a relative change in y is related to a relative change in x. In
mathematical terms, this basic assumption can be stated as

=b Ax. (11)

y x

Corresponding to the difference equation (11) is the differential
equation

dy =b . (i2)

y x

Equation (12) may be integrated to obtain

In y = b In x + k, (13)

where k is the constant of integration. Solving for y in (13), the power
model is obtained as

_ b

y — ax ,

(14)

where a = exp(k).

An examination of (11) shows that it is necessary to assume that
neither x nor y is zero. The basic assumption of this model as stated
in (11) implies that for a unit of relative change in x there are b units
of relative change in y.

MODEL SELECTION

It appears that too frequently in exploratory analysis, the function
representing a set of data points (x,y) is picked by the researcher based
on an intuitive feeling such as “the curve looks like....” This intuitive
feeling needs to be combined with theoretical implications of a
particular substance to produce models of choice, rather than models
of chance. Even when the model is based on “gut” feeling, the

FUNDAMENTAL ISSUES— MODELING

151

researcher must check to see what the mathematical implications are
in the model of choice. This paper delineates certain issues, that when
taken into account, enable the researcher to better accomplish this
goal.

The most general premise of mathematical modeling is that there
must be a one-to-one correspondence between the assumptions of the
mathematical model and the subject matter to be modeled. In this
vein, the following guidelines for the models discussed herein are
provided. To choose to use the linear model, the researcher should
have some underlying basis that posits a simple change to be
accompanied by simple change. To use the logarithmic model, one
would have to have some underlying basis that posits a simple change
in y to be accompanied by a relative change in x. Additionally, there
should be a basis for arguing that a simple change in x at the right
end of the domain is “less effective” in producing a change in y. To
use the exponential model, the user assumes that relative change in y
is caused by simple change in x. Also the researcher must assume that
at increasingly large values of x a change in x produces increasingly
large changes in y. This is seen most clearly in (10). To use the power
model, one assumes that relative change is caused by relative change.
Finally, because zero is not included in the interval of the domain or
range, one’s theory must not make statements about zero values in
either the independent or dependent variable.

Hence, the first question comes to “what are you explaining in the
dependent variable,” simple change (Ay) implying linear or
logarithmic, or relative change (-^-) implying exponential or power?

The second question is “what are you explaining in the independent
variable,” simple change (Ax) implying linear or exponential, or
relative change (-^-) implying logarithmic or power?

To summarize the “conceptual space” implied in the above
questions, the following figure is provided.

152

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

When either the logarithmic or exponential model would otherwise
seem appropriate, the zero problem can be avoided by the addition of
a constant to the appropriate variable. However, the researcher must
remember that this expedient does change the curve to be estimated,
adds a parameter, and changes the values of the parameter estimates
thereby changing the problem somewhat. If, however, the zero value
of the variable is important theoretically, the logarithmic and
exponential models are simply inappropriate and other models should
be considered.

Zajonc (1968), for example, found that the degree to which students
expressed positive probability of liking a person pictured in a
photograph could be expressed as a logarithmic function of the
number of times they had been previously exposed to the picture.
Although Zajonc’s work involved discrete variables, it could have
benefited from the foregoing discussion as follows. Recall that the
logarithmic function resulting from model (6) disallows the condition
x = 0. By conceiving his independent variable (number of exposures
to a picture) as a proxy for time, he could have examined the
implications of the prohibition of a zero value in the independent
variable. From that perspective the length of exposure has declining
impact on the probability of liking while the lack of exposure (the zero
condition) leads to no prediction. This seems a reasonable argument
and one capable of further testing. That is, variance in liking for no
previous exposure to pictures should be greater than variance in liking
around the predicted curve. Thus, the findings are reenforced and
additional direction for research delineated.

Another example comes from the question of the relationship
between educational achievement and financial reward. A researcher
concerned with such a question might employ the following reasoning
in addressing the question of how education affects raises in salary in
an organization. The dependent variable, salary, would require a
consideration of relative change because simple dollar raises mean
different things at different levels of income. Likewise the importance
of additional educational achievement to the organization would, in
most cases, depend on the level of education from which the increment
occurs. Therefore the independent variable must also be considered as
relative change. Under the additional assumption, that these two
relative changes are directly proportional, the researcher is lead to the
power model.

In conclusion, we feel that it is necessary to point out several facts
regarding the estimation of the parameters of these models. Too many
researchers elect to use the linear model out of the dual factors of
expedience of computation and concern for parsimony. With modern
computer facilities, the issue of ease is often not a valid reason to use

FUNDAMENTAL ISSUES— MODELING

153

the linear model. Estimation of the parameters of the logarithmic
model, for example, is easy because the parameters are linearly
represented. We suspect that in some cases the logarithmic model has
been selected for use simply due to this ease, and the desire for
nonlinear fit, rather than due to a real search for a mathematical basis.
The parameters in the exponential and power models — models (10)
and (14) — are nonlinearly represented. To estimate them and to
compare results between models, it is imperative that the form of the
dependent variable be left intact and that all models have additive error
terms. (For a discussion on these issues, see Walling et al. 1984.)

LITERATURE CITED

Swokowski, E. W. 1979. Calculus with analytical geometry. Prindle, Weber and Schmidt,
Boston, 2nd ed., section 3.4.

Walling, D. D., H. L. Hotchkiss, and E. W. Curry. 1984. Power models and error terms.

Sociological Methods and Research 13:121-126.

Zajonc, R. B. 1968. Attitudinal effects of mere exposure. Personality and Social
Psychology, Monogr. Suppl. 9, no. 2:1-27.

FOODS OF SCALED QUAIL ( CA LLIPEPLA SQUAMATA)
IN SOUTHEASTERN NEW MEXICO

by TROY L. BEST

General College, Department
of Biology, and Museum of
Southwestern Biology
The University of New Mexico
Albuquerque, New Mexico 87131

and RICHARD A. SMARTT

Department of Biology

The University of Texas at El Paso

El Paso, Texas 79968

ABSTRACT

One hundred-twenty scaled quail ( Callipepla squamata) were collected in southeastern
New Mexico to determine the amounts and kinds of food items ingested and to evaluate
sexual and temporal variation in feeding habits. Seeds of Helianthus petiolaris,
Amaranthus, Prosopis glandulosa, Chenopodium, and Croton were the dominant food
items. Occurrence of several uncommon food items differed between the sexes,
suggesting some niche separation between males and females. For both sexes, feeding
habits differed from morning to afternoon. Key words : Callipepla squamata, scaled
quail, food habits, feeding ecology, New Mexico.

INTRODUCTION

Considering the importance of scaled quail (Callipepla squamata) as
a game bird in the Southwest, there are relatively few quantitative data
about the feeding ecology of this species. The early accounts of Judd
(1905) and Kelso (1937) lumped specimens from several southwestern
states; later, detailed studies were conducted in Texas (Lehmann and
Ward 1941; Wallmo 1956; Ault and Stormer 1983), Oklahoma
(Schemnitz 1961), Colorado (Hoffman 1965), and Arizona (Gallizioli
1965). Three previous studies of scaled quail feeding ecology have been
conducted in New Mexico — one near Tucumcari (Russell 1932), one in
Lea County (Campbell 1964; Campbell et al. 1973), and one in western
Lea and eastern Eddy counties (Davis and Banks 1973; Davis et al.
1975). We identified and quantified the food items ingested by scaled
quail in southeastern New Mexico and were the first to investigate
sexual differences and temporal variation in foods selected by this
species.

The Texas Journal of Science, Vol. XXXVII, Nos. 2&3, September 1985

156

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

MATERIALS AND METHODS

The study area was centered at drill hole ERDA 9 (SE corner, sec.
20, T22S, R31E), and extended outward to a radius of eight kilometers.
Most of the area was in eastern Eddy County, but it also extended into
extreme western Lea County. Extensive vegetation analyses have been
conducted on this noncultivated site by W. C. Martin of the University
of New Mexico (see Best and Jackson 1982). Our specimen collections
were restricted to the shinnery oak-mesquite ( Quercus havardii-
Prosopis glandulosa) association to minimize the effects of differing
habitat types. Davis’ study area was a few kilometers southeast of ours,
is similar, and has been described repeatedly (for example, Davis and
Banks 1973; Davis et al. 1974; Davis et al. 1975).

In 1979, 120 scaled quail were collected from 13 to 18 November by
shooting during the day (from 0630 until 1700 MST). For each quail,
the time and sex were recorded. Crop contents were removed, placed
into plastic vials, frozen, and later air dried. Food items were identified
by comparison with plant and arthropod samples collected on the
study site. Food items that were not identifiable to family were listed
as unknowns.

Average weights and measurements of seeds were taken for each food
item found in the crops (Best et al. 1982). The volume of each food
item was determined by multiplying the seed dimensions by the
number of seeds in the crops. Percent volume was calculated for each
food item, using a formula similar to that of Martin et al. (1946) —
volume of each item /total volume X 100. Food items with less than
0.01 percent volume were included as trace occurrences (tr.). Mean and
standard deviation (frequency) also were calculated for each food item.
Discriminant analyses (Nie et al. 1975) were used to evaluate sexual
and temporal variation in food habits; the numbers of seeds of each
food item served as characters. Statistical analyses were conducted
using the IBM computer systems at Eastern New Mexico University
and The University of New Mexico.

RESULTS AND DISCUSSION

Crop Contents

The crop contents for individual scaled quail are listed in Best et al.
(1982). The frequency, mean and standard deviation of frequency, and
percent volume for each food item are presented in Table 1.
Helianthus petiolaris accounted for 28 percent of the total volume and
was present in 75 percent of the crops. The volume of Amaranthus
also was important (26 percent) and was present in 24 percent of the
crops. Seven other food items were present in 28-47 percent of the
crops: Prosopis glandulosa, Chenopodium B, Croton, U128, Paspalum

FEEDING ECOLOGY OF QUAIL

157

Table L— Food items in crops of scaled quail ( Callipepla squamata ) collected in
southeastern New Mexico. Frequency, mean, standard deviation, and percent volume
are listed by sex for each item.

Males (N = 59) Females (N = 56)

Food item' 1 Freq.b Meanc SD % Vol. Freq. Mean SD % Vol.

Amaranthaceae
Amaranthus albus
Amaranthus A
Boraginaceae

Lithospermum multiflo-
rum

Cactaceae

Opuntia phaeacantha
Chenopodiaceae

Chenopodium incanum
Chenopodium A
Chenopodium B
Commelinaceae
Commelina sp.
Compositae
Ambrosia A
Ambrosia B
Helianthus petiolaris
Heterotheca sp.

Verbesina eucleioides
Cruciferae

Dithyrea wislizenii
Cucurbitaceae

Cucurbita foetidissima
Euphorbiaceae
Croton sp.

Euphorbia A
Euphorbia B
Euphorbia C
Euphorbia D
Graminae

Bouteloua gracilis
Panicum obtusum
Paspalum setaceum
Setaria leucopila
Sporobolus cryptandrus
Triplasis purpurea
Labiatae

Monarda punctata
Leguminosae
Astragulus sp.
Hoffmanseggia jamesii
Phaseolus sp.

Prosopis glandulosa

3

60

47

0.05

15

1442

1422

26.60

3

2

1

0.04

1

11

0.23

7

377

351

0.98

8

219

241

1.94

23

313

426

7.20

4

3

2

0.10

5

34

28

2.81

2

6

0

0.08

45

243

278

31.97

0

0

4

6

4

0.14

3

2

2

0.01

0

0

19

10

15

3.07

33

129

359

5.03

16

16

21

0.93

5

4

4

0.01

1

13

tr.

4

8

6

0.01

13

26

68

0.72

28

47

75

2.15

6

13

17

0.14

3

6

2

tr.

3

11

8

0.03

19

76

228

0.52

20

11

17

0.36

4

2

1

0.15

1

6

0.03

23

8

13

8.94

0 0
13 1300 1522 25.73

4

2

2

0.08

2

16

8

0.80

14

1424

1815

9.11

6

81

97

0.67

25

582

739

18.01

3

8

9

0.24

9

37

34

2.30

2

8

4

0.13

38

172

209

23.76

1

8

0.01

0

0

3

2

1

0.01

1

1

0.04

21

11

11

4.64

24

155

209

5.44

12

9

8

0.50

3

11

10

0.03

0

0

1

1

tr.

11

5

4

0.13

26

53

91

2.79

15

9

12

0.29

3

1

0

tr.

4

4

3

0.02

14

17

13

0.10

15

7

10

0.22

3

2

2

0.19

0

0

21

2

2

2.99

158

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 1. — Continued.

Linaceae

Linum aristatum

1

14

0.01

0

0

Loasaceae

Mentzelia sp.

6

11

13

0.06

8

15

23

0.15

Nyctaginaceae

A bornia fragrans

4

14

18

0.20

2

19

21

0.17

Onagraceae

Gaura villosa

3

1

1

0.05

1

1

0.02

Plantaginaceae

Plantago sp.

0

0

1

15

0.02

Polygonaceae

Eriogonum sp.

0

0

1

3

tr.

Rumex sp.

1

1

0.02

0

0

Portulacaeae

Portulaca oleracea

6

24

43

0.02

3

11

11

0.01

Portulaca A

6

12

13

tr.

6

14

18

0.01

Solanaceae

Solanum rostratum

2

53

27

0.12

1

174

0.25

Verbenaceae

Verbena bracteata

3

350

419

0.20

7

227

251

0.36

U6

0

0

2

4

3

0.03

U101

5

4

2

0.02

5

4

2

0.03

U128

19

432

840

5.01

13

70

123

0.69

U135

3

2

2

0.05

1

3

0.03

U136

0

0

1

3

0.02

U138

2

122

99

1

3

U143

1

5

tr.

0

0

U145

0

1

31

U157

1

2

0

U186

0

0

1

1

tr.

U193

4

6

4

tr.

1

6

tr.

U195

0

0

1

1

tr.

U196

0

0

1

23

0.02

U197

1

20

tr.

0

0

Grasshoppers

1

1

0

Insect galls

5

3

2

2

2

1

Mouse feces

1

2

0

aNumber of crops examined = 120; number of crops with contents = 1 15.
bNumber of crops containing each item.

cAverage number of seeds in the crops of those containing the item.

setaceum, Astragalus , Monarda punctata. These seven foods accounted
for 28 percent of the volume. Of the 59 food items found, over one-
half were present in amounts less than 1 percent of the total volume.

Campbell et al. (1973) reported that Acacia, Gutierrezia, Croton,
Euphorbia, and green leaves and stems were most important (58.3
percent of the total volume) in the 227 crops of scaled quail they
examined during 1960-1962; the most frequently encountered items
were green leaves and stems, insects, grit, Croton, Prosopis, and Acacia

FEEDING ECOLOGY OF QUAIL

159

(all at frequencies of more than 40 percent). Acacia was not present in
our study area, but the other major food items reported by Campbell
et al. (1973) were common. Volumes of most food items in their study
were different from those we observed. However, their frequencies of
Amaranthus, Prosopis, and Croton were similar to ours.

Our results also may be compared with cool-season data from Davis
and Banks (1973) and Davis et al. (1975). Prosopis, Euphorbia, Croton,
and Gutierrezia had the greatest mean percent weights in the crops of
scaled quail examined by Davis and Banks (1973). Prosopis,
Euphorbia, and Croton also were among the eight most important
items we found. The food items with the greatest mean volume
reported by Davis et al. (1975) were Gutierrezia, Prosopis, green
vegetation, and insects. Of these, only Prosopis was important in our
study.

Sexual Differences

Campbell and Lee (1956) noted that the number of males in New
Mexico scaled quail populations slightly outnumber females. The sex
ratio they observed for New Mexico in general was 104.3 males per 100
females. This is similar to the sex ratio in the population we sampled:
105.4 males per 100 females.

Discriminant analysis of the crop contents in our sample indicated
there were differences in the feeding habits of males and females (Table
2). Seventy-two percent of the individuals were correctly classified to
sex based upon their crop contents. Females consumed more
Chenopodium incanum and Ambrosia A than males. Conversely,
males ate more Bouteloua gracilis, Amaranthus albus, insect galls,
Chenopodium A, P. glandulosa, and grasshoppers. However, most of
these food items were represented at low frequencies and volumes in
the crops, and there was considerable overlap between sexes for most
major food items. Thus, diets of the sexes were similar in most
respects. Whether the differences reflect significant food-niche
separation of the sexes should be addressed in future studies.

Temporal Differences

Schemnitz (1961) observed that scaled quail in Oklahoma fed from
daybreak until about 10AM and from 4PM until dark. This appeared
to be true on our study area as well. There were some differences in
the feeding habits between morning and afternoon for both sexes
(Table 2). Discriminant analyses were performed separately for the
sexes since some differences were found in their feeding habits. For
males, 95 percent of the birds were correctly classified as to time of
collection (Table 2). Crops of males collected during the afternoon had
more than eight times as much U128 as those collected in the
morning; amounts of U197, Triplasis purpurea, and P. glandulosa

160

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 2. — Discriminant analysis among sexes and times of day based upon crop contents
of scaled quail ( Callipepla squamata ) from southeastern New Mexico.

Actual Group

n

Predicted group membership

Between sexesa,b

Males

Females

Males

59

41(69.5%)

18(30.5%)

Females

56

14(25.0%)

42(75.0%)

Ungrouped

5

5(100%)

0

Between AM and PM for malesc

Morning

Afternoon

Morning

37

34(91.9%)

3(8.1%)

Afternoon

22

0

22(100%)

Between AM and PM for females

d

Morning

Afternoon

Morning

40

40(100%)

0

Afternoon

16

10(62.5%)

6(37.5%)

aThe data below are given as: percent of specimens that were correctly classified; in
decreasing order of importance, the variables accounting for differences.
b72.2%; Bouteloua gracilis, Amaranthus albus, insect galls, Chenopodium A, Prosopis
glandulosa, Ambrosia A, grasshoppers.

c94.9%; Chenopodium A, Croton, Gaura villosa, U197, Portulaca A, Mentzelia, Triplasis
purpurea, Prosopis glandulosa, Amaranthus A, Chenopodium incanum, Euphorbia A,
Chenopodium B, Ambrosia A.

d82.1%; Chenopodium incanum, Chenopodium B, Croton, Cucurbita foetidissima, U186,
Euphorbia B, Bouteloua gracilis.

were also greater in the afternoon. Males collected in the morning had
more Chenopodium A, Croton , Gaura villosa , Portulaca A, Mentzelia,
Amaranthus A, C. incanum, Euphorbia A, Chenopodium B, and
Ambrosia A. The larger amount of U128 and lesser amounts of
Chenopodium A, and Croton consumed by males collected in the
afternoon accounted for most of the difference between morning and
afternoon samples. For females, 82 percent of the birds were correctly
classified to morning or afternoon collection (Table 2). Females
collected in the afternoon had almost four times the C. incanum as
those collected in the morning, but much lesser amounts of
Chenopodium B, Croton, and Euphorbia B. Like males, the lesser
amounts of Chenopodium A and Croton in specimens collected in the
afternoon also contributed to the difference between morning and
afternoon samples.

Many of the food items temporally separating morning and
afternoon quail samples were among the most common items found
in the crops. This supports a claim that there were considerable
differences between morning and afternoon samples for males and
females. We do not beilieve these differences existed simply because of
an accumulation of seeds through the day. If this was true, all or most
of the food items would be represented in the greatest numbers in
specimens collected in the afternoon. This was not the case. Several
food items in both sexes were found in greatest abundance in

FEEDING ECOLOGY OF QUAIL

161

specimens collected in the morning. However, there may have been
some effect of habitat differences from one collection site to another
within our study area. Because specimens were collected as they were
encountered throughout the study area and many sites were revisited
at various times during the day, differences between collecting sites
were probably minimal. It is likely that food items appearing in
greater quantities in either morning or afternoon samples did so
because they were encountered or selected in greater amounts.

ACKNOWLEDGMENTS

D. Clark, C. W. Deihl, B. Hoditschek, K. Leong, P. J. Polechla, and
A. Saiz helped collect quail. The New Mexico Department of Game
and Fish provided the collecting permit. This project received funding
as part of the Los Medanos Waste Isolation Pilot Plant studies from
Sandia National Laboratories (Contract no. 13-2097) and Westinghouse
Electric Corporation (Subcontract no. WFC-53431-50). S. A. Cole
provided assistance in proofreading; D. C. Schmitt helped analyze the
data; and, S. Neuhauser, P. L. Kennedy, and D. J. Hafner critically
reviewed early drafts of the manuscript.

LITERATURE CITED

Ault, S. C., and F. A. Stormer. 1983. Seasonal food selection by scaled quail in northwest
Texas. J. Wildlife Manag. 47:222-228.

Best, T. L. , and D. W. Jackson. 1982. Statistical evaluation of plant density data collected
at the Los Medanos site, New Mexico (1978-1980). Pp. 8-1 through 8-374, in
Ecosystem studies at the Los Medanos site Eddy County, New Mexico (J. Braswell
and J. S. Hart, eds.), U.S. Dept. Energy, Albuquerque, TME 3141, 3 vols., 982+ pp.
Best, T. L., R. A. Smartt, B. Hoditschek, and D. C. Schmitt. 1982. Vertebrate ecology
at the Los Medanos Waste Isolation Pilot Plant, New Mexico: annual report for
FY1981. Pp. 7-1 through 7-336, in Ecosystem studies at the Los Medanos site Eddy
County, New Mexico (J. Braswell and J. S. Hart, eds.), U.S. Dept. Energy,
Albuquerque, TME 3141, 3 vols., 982+ pp.

Campbell, H. 1964. Food habits of scaled quail. Proj. Rpt. W-104-R-4, New Mexico
Dept. Game and Fish, Santa Fe, 14 pp.

Campbell, H., and L. Lee. 1956. Notes on the sex ratio of Gambel’s and scaled quail
in New Mexico. J. Wildlife Manag. 20:93-94.

Campbell, H., D. K. Martin, P. E. Ferkovich, and B. K. Harris. 1973. Effects of hunting
and some other environmental factors on scaled quail in New Mexico. Wildlife
Monogr. 34:1-49.

Davis, C. A., and R. L. Banks. 1973. Some food habits of scaled quail in southeastern
New Mexico. New Mexico State Univ., Agric. Exper. Sta. Res. Rept. 270:1-5.

Davis, C. A., P. E. Sawyer, J. P. Griffing, and B. D. Borden. 1974. Bird populations in
a shrub-grassland area, southeastern New Mexico. New Mexico State Univ., Agric.
Exper. Sta. Bull. 619:1-29.

Davis, C. A., R. C. Barkley, and W. C. Haussamen. 1975. Scaled quail foods in
southeastern New Mexico. J. Wildlife Manag. 39:496-502.

162

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NOS. 2 & 3, 1985

Gallizioli, S. 1965. Quail research in Arizona. Arizona Game and Fish Dept., Phoenix,

12 pp.

Hoffman, D. M. 1965. The scaled quail in Colorado. Colorado Dept. Game, Fish and
Parks, Tech. Publ. 18:1-47.

Judd, S. D. 1905. The bobwhite and other quails of the United States in their economic
relations. U.S.D.A., Biol. Surv. Bull. 21:1-66.

Kelso, L. H. 1937. Food of the scaled quail (preliminary report). U.S.D.A., Bur. Biol.

Surv., Wildlife Res. Manage. Leaflet BS-84.1-9.

Lehmann, V. W. , and H. Ward. 1941. Some plants valuable to quail in southwestern
Texas. J. Wildlife Manag. 5:131-135.

Martin, A. C., R. N. Gensch, and C. P. Brown. 1946. Alternate methods in upland game
bird food analysis. J. Wildlife Manag. 10:8-12.

Nie, H. N., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. H. Bent. 1975. Statistical
package for the social sciences. (SPSS). McGraw-Hill Book Co., New York, 675 pp.
Russell, P. 1932. The scaled quail of New Mexico. Ph.D. thesis, Univ. of New Mexico,
Albuquerque, 143 pp.

Schemnitz, S. D. 1961. Ecology of the scaled quail in the Oklahoma panhandle. Wildlife
Monogr. 8:1-47.

Wallmo, O. C. 1956. Ecology of scaled quail in west Texas. Texas Game and Fish
Comm., Austin, 134 pp.

Present address of Smartt: New Mexico Museum of Natural History, P.O. Box 7010,
Albuquerque, NM 87194.

THYROID-PARATHYROID RESPONSE TO EDTA AND
CaCl2 INFUSIONS IN WHITE-TAILED DEER

by CHUN CHIN CHAO and ROBERT D. BROWN

Caesar Kleberg Wildlife Research Institute
Texas A&I University
Kingsville, Texas 78363

ABSTRACT

EDTA infusion alone or followed by CaCh infusion were used to compare the
efficiency of four commercial parathyroid hormone (PTH) radioimmunoassays and to
determine the responsiveness of the thyroid-parathyroid gland in white-tailed deer. In the
first experiment an eight percent EDTA solution was infused for 20 minutes at a rate
of 2.16 mg/ kg BW/min in five deer. In the second experiment, three deer were infused
with a five percent CaCh solution at a rate of 3 mg/kg BW/min for 20 minutes. In
the third experiment a 20-minute EDTA infusion was followed 30 minutes later by a
20-minute CaCh infusion in three deer. Plasma PTH and calcitonin (CT) were measured
with a competitive protein binding technique, whereas plasma calcium was measured
with a titrator. Similar baseline PTH concentrations were obtained by assays from Iso-
Tex, Nichol’s and Immuno Nuclear Corp. (INC), whereas slightly higher baseline
concentrations were detected by assays from Cambridge. Relative magnitudes of increase
(post-versus preinfusion) in PTH response to EDTA infusion in both experiments 1 and
3 were Cambridge, 15.4; Iso-Tex, 3.4; Nichol’s 4.5 and INC 19 fold. Selection of the assay
from Iso-Tex (TX) for the subsequent determination of deer PTH was based on its low
nonspecific binding, sensitivity, convenience, and the relative correlation coefficients
between plasma calcium and PTH.

Plasma calcium levels responded to both EDTA and CaCh infusions. Circulating
PTH peaked 10 minutes after the start of the EDTA infusion in experiments 1 and 3
and was suppressed 10 minutes after the start of the CaCh infusion in experiments 2
and 3. Circulating calcitonin was elevated to a plateau 10 minutes after the CaCh
infusion in experiments 2 and 3 and was suppressed to low levels 10 minutes after EDTA
infusion in experiments 1 and 3. Further EDTA or CaCh infusion did not affect PTH
or calcitonin concentrations once hormone concentrations were elevated or suppressed.
It is concluded that EDTA /CaCh infusion is a useful means of accessing thyroid-
parathyroid responsiveness in deer. Deer have a rapidly responsive thyroid-parathyroid
similar to that found in other species. Key words: deer, hypocalcemia, hypercalcemia,
parathyroid hormone, calcitonin.

INTRODUCTION

The annual regeneration of antlers by male Cervidae offers a unique
model for the study of human bone metabolism (Cowan et al. 1969).
The severe demand on the mineral reserves of a deer’s skeleton,
especially during the period of rapid antler growth, has received
particular attention (Banks et al. 1968; Stephenson and Brown 1983).

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

164

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Few studies have been done with deer on parathyroid hormone
(PTH) and calcitonin (CT), which are known to control calcium (Ca)
metabolism in other mammals. Grafflin (1942) examined the
parathyroid glands of deer histologically and found no seasonal
variations. Mazur (1969) found accessory parathyroid tissue in 10
percent of 30 road-killed bucks autopsied. Brown et al. (1981) found
no noticeable differences between normal and parathyroidectomized
bucks in their dates of velvet shedding, antler casting or hair coat
changes and no significant difference (P>0.05) between the groups in
blood Ca levels.

Measurements of circulating PTH by radioimmunoassay was first
established 20 years ago (Berson et al. 1963). Parathyroid hormone is
rapidly degraded after it is released; the biologically active N-terminal
portion of the molecule has a shorter half-life, whereas the biologically
inactive C-terminal portion of the molecule has a longer half-life
(Berson and Yalow 1968). In clinical practice radioimmunoassay of
human PTH (hPTH) involving antisera specific for the C-terminal
portion has been widely used (DiBella et al. 1978). To date, circulating
PTH in deer has not been reported. The objectives of this paper were
to compare the relative efficiency of four commercially available PTH
assays and a CT assay (Deftos et al. 1972) to examine the
responsiveness of the deer’s thyroid-parathyroid by challenging it with
infusions of disodium ethylenediaminetetraacetic acid (EDTA) or
calcium chloride (CaCh).

MATERIALS AND METHODS
Animal Experimentation

In the first experiment five adult (one male and four females, 2.5 to
5 years old) white-tailed deer ( Odocoileus virginianus) were maintained
in individual outdoor covered pens and fed a complete pelleted ration
ad libitum. The experiments were conducted during the period April-
June 1982. The male deer had velvet antlers less than three centimeters
in size and the females were not pregnant. One day prior to each
infusion, the animals were fasted to avoid a feeding effect on serum
Ca as reported in rats (Perault-Staub et al. 1975). The animals were
tranquilized with xylazine hydrochloride (Rompun) at a dose of 0.65
milligrams per kilogram of body weight (BW) (Roughton 1975).
Previous studies (Chao et al. 1984) indicated that Rompun had no
effect (P>0.05) on serum Ca, PTH and CT levels from 10 to 60
minutes after the animals were tranquilized.

After tranquilization the animals were infused intravenously with an
eight percent solution of EDTA at a rate of 2.16 mg/ kg BW/min for
20 minutes (Argenzio et al. 1974; Brown et al. 1981). Blood samples

THYROID PARATHYROID RESPONSE IN DEER

165

from the contralateral jugular vein were taken into 10-ml NH4-
heparinized syringes 10 minutes before the infusion, at the start of the
infusion, and then every 10 minutes for another hour. Plasma samples
were separated immediately by centrifugation, and total plasma Ca was
determined on an automatic calcium tritrator (Model 4008, Precision
Systems).

In the second experiment, three adult (two males and one female)
white-tailed deer were infused with a five percent solution of CaCh at
a rate of 3 mg/kg BW/min for 20 minutes. Blood samples were taken
10 minutes pre-infusion and 10, 20, 30, 60, and 90 minutes after the
start of infusion.

The third expriment involved both EDTA and CaCh infusions in
three deer (one male and two females). The animals were tranquilized
as before, and blood samples were again withdrawn 10 minutes before
and at the start of these infusions to establish a baseline. EDTA was
infused at the same dose and rate as previously mentioned, and blood
samples were taken as before. Thirty minutes after completion of the
20-minute EDTA infusion, CaCh was infused at the previous dose and
rate for 20 minutes, and blood was again withdrawn every 10 minutes
for another hour.

Radioimmunoassays of PTH and CT

The four PTH assays tested were commercial kits from Cambridge
Medical Diagnostics, Inc. (Billerica, Massachusetts 01865), Iso-Tex
Diagnostics (Friendwood, Texas 77549), Nichol’s Institute Diagnostics
(San Pedro, California 90731), and Immuno Nuclear Corp. (INC)
(Stillwater, Minnesota 55082). Each of these assays involved use of a
bovine PTH (bPTH) standard, 125I-labeled PTH as the tracer, and a
second-antibody separation technique. The characteristics of each assay
are presented in Table 1. The intra-assay coefficient of variation (CV)
of each assay was determined for reproductibility by using the values
obtained from the standard curves. Infusion of EDTA in experiments
1 and 3 (pre- versus postinfusion) was thus used to test the relative
efficiency of these assays. Plasma PTH was assayed in duplicate for all
samples obtained in experiments 1 and 3. Because of limited samples
in experiment 2, only samples obtained from pre- and 10 minutes
postinfusion were assayed.

Plasma CT was assayed using guinea-pig anti-bovine CT and rabbit
anti-guinea pig antisera (Deftos et al. 1972). The sensitivity of the assay
was 1 pg/ml and the intra-assay CV was within 10 percent of the
mean. The infusion of CaCh in experiments 2 and 3 was used to detect
the relative concentrations of CT in deer at bovine equivalent values.
Plasma CT was assayed in duplicate and only samples from pre- and
10 minute postinfusion of EDTA or CaCli in either experiments 1 or
3 were assayed due to limited samples.

166

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Statistical Analysis

Analysis of variance was used to test effects of time period (Ray
1982). Mean comparisons were based on Tukey’s studentized range test
at a significance level of 0.05. Paired t-test was used to test statistical
differences between PTH or CT concentrations obtained from pre- and
10 minute postinfusion. Pearson product moment correlation
coefficients were calculated to measure the strength of the relationship
between serum Ca and PTH or CT. However, the trends of response
to the infusions of EDTA and/or CaCE or the combination are
physiologically important and meaningful.

RESULTS

Nonspecific binding (NSB), maximum binding (MB) and the intra¬
assay CV of the standard curve from each PTH assay are depicted in
Figure 1. The correlation coefficients between plasma concentrations of
Ca and PTH in samples obtained pre- and 10 minutes postEDTA
infusion from experiments 1 and 3 are summarized in Figure 2. The
baseline PTH level (Mean ± SE, ng / ml) from the eight deer before
the EDTA infusion averaged 1.70 ± 1.06 (Cambridge), 0.53 ±0.17 (Iso-
Tex), 0.61 ± 0.27 (Nichol’s), and 0.55 ± 0.05 (I.N.C.). The numerical
differences among the assays became even more pronounced after the
EDTA infusion. PTH concentrations measured by Iso-Tex, Nichol’s,
and I.N.C. assays remained similar but lower than those of the
Cambridge assay. However, the variation was greatest in the
Cambridge assay. The relative magnitudes of response to EDTA
infusion in terms of increased PTH were Cambridge, 15.4; Iso-Tex,
3.4; Nichol’s, 4.5; and I.N.C. , 1.9 fold. The Iso-Tex assay was chosen
for use in subsequent experiments (see Discussion).

Endocrine Response to the Infusions of EDTA or CaCh

EDTA infusion in experiment 1 induced a drop in plasma Ca levels
of 3-5 mg / dl (Figure 3). The maximum drop in plasma Ca occurred
concurrent with the completion of EDTA infusion, whereas plasma
concentrations of PTH were significantly elevated (P<0.05) only at 10
minutes after the start of the infusion. Plasma concentrations of Ca
returned to near baseline in two hours, whereas plasma concentrations
of PTH returned to near baseline 30 minutes after the peak. Plasma
concentrations of CT averaged 0.44 ± 0.01 ng / ml pre infusion and
were suppressed to 0.14 ± 0.01 ng / ml 10 minutes after the EDTA
infusion.

In experiment 2, plasma concentrations of Ca responded as expected
to CaCE infusion and peaked at the completion of infusion (Figure 4).
Plasma CT peaked 10 minutes after the start of CaCE infusion

PTH

THYROID-PARATHYROID RESPONSE IN DEER

167

cs

43

u

o

c

o

C0

'C

a

£

Q

I

X

h

&H

u

00 CM

< r> — 1

o

CO © ^

9 ^ ^

— ; "5 cm

o •- ©

x ~

• CL,

oo

©

U

X

h

cu

tj

03

+

X

h

cu

U

m

! i *

w ^

d ^ +

■ ^ CM

5 s +

d ^

oo

^ GO

© ,
c- +

d CM

o

o

m

m

^ m
o r-

CM
■ +

w

in ^ ^

o to

ho

c

d

\ s

a

g

he T5

2

<u

N

3

-E c

-d

£

'o?

43

- 0

c * ^

in

rc

\

ju

73

03 03

T3

ho

a

g

C -9

C

fl

S

3

S-I

6 3

X c

□ 4J

H -X X
Dh .y a;
^ 43

■g V e

03 C 8

C 03 |r

^ 3J r

,“5fc

6 <2

X be

£ '2-

Oh

n ^

Is .s

J 3 3

03 0 -Q

bo 9 0

a g "be

os n i

O

X he

fc

Oh 3,

_Q 03

aj -

•3 c .£
= '3 3
^ O -Q
be . o

*“* c

os os

9J

C 03

o

be

O

X -
H -S

Oh 42 C
42 03 -S

T 3

||
w bo

03

C

5°

<u

T3

•a -o — S

03 O S O

n _ _o 2 u

3n3.2 <v

168

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NOS. 2 & 3, 1985

Cambridge

Iso -Tex

PTH (ng/ml)

PTH (ng/ml)

Nichols

I.N.C.

100

100

80

60

NSB 18.90%
MB 26.95%
CV 7.42%

80

60

o

CO

m 40

o

m

m

40

NSB 9.74%
MB 39.21%
CV 3.53%

20

20

0

1 10 100

0

■ . ni.n I. I

1 10 100

PTH (ng/ml) PTH (ng/ml)

Figure 1. Standard curves of four commercial PTH assays. B / Bo (%) indicates percent
of binding. NSB, MB, and CV are non-specific binding, maximum binding, and
coefficients of variations, respectively.

(P<0.01) and remained high for the next 20 minutes. The levels
decreased gradually and had not returned to baseline levels 90 minutes
after the start of CaCh infusion. A five- to eight-fold increase in
plasma CT was found due to the CaCh infusion. The correlation
coefficient between plasma concentrations of Ca and CT was r = 0.69
(P 0.01, n = 18). Plasma PTH levels were 1.43 ± 0.52 ng/ml 10
minutes preinfusion and were suppressed to 0.82 + 0.27 ng/ml 10
minutes after the infusion. The magnitude of suppression of plasma

THYROID-PARATHYROID RESPONSE IN DEER

169

E

O)

c

I

I-

CL

15

10

Cambridge

r = -0.22

O

15

Iso -Tex

Calcium (mg/dl)

Calcium (mg/dl)

15

Nichols

15

I.N.C.

E

cn

c

I

I-

0.

10

r = -0.26

r = -0.43

0

15

0

»V**i

10 15

Calcium (mg/dl)

Calcium (mg/dl)

Figure 2. The relationships between circulating PTH and Ca levels before (o) and after
(A) EDTA infusion (2.16 mg / kg BW / min for 20 minutes) in two male and six
female white-tailed deer.

PTH due to the CaCh infusion was thus smaller than the elevation
due to the EDTA infusion.

When EDTA infusions were followed by CaCh infusions in
experiment 3, plasma Ca responded to both chemicals (Fig. 5).
Maximum doses of EDTA again induced a maximal drop of plasma
Ca, whereas the CaCh caused a peak in plasma Ca of 16.32 mg / dl
in the three deer. Plasma PTH concentrations peaked 10 minutes after
the start of EDTA infusion and were suppressed in response to CaCh
infusion (Fig. 5). Plasma CT levels decreased slightly (P>0.05)
following the EDTA infusion (0.45 ± 0.16 as opposed to 0.15 ± 0.1),
and increased significantly (P< 0.05) in response to CaCh infusion
(0.31 ± 0.12 as opposed to 5.49 ± 1.12).

170 THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Minutes

Figure 3. Plasma Ca and PTH concentrations in response to EDTA infusion (2.16 mg /
kg BW / min for 20 minutes) in one male and four female white-tailed deer (mean
± SE).

DISCUSSION

The Cambridge assay contained a rabbit anti-bPTH antiserum and
seemed to generated the highest magnitude of response to EDTA
infusion and specificity for the deer PTH determination (Fig. 2).
However, we cannot explain the high concentrations of PTH in two
deer that had baseline concentrations close to 6 ng / ml. Similar results
in these two deer were not present in other PTH assays. We thus
suspect the precision of the Cambridge assay. The high NSB may
reflect problems in interference with specific binding. The requirement
of a sample size of 500 ml by the Cambridge assay is 2.5 to 5 times
more than that for the other assays.

The Iso-Tex and Nichol’s assays used guinea-pig anti-bPTH antisera
that are sensitive for human PTH determinations (Gorog et al. 1982).
The characteristics for both assays were similar. The baseline PTH
concentrations measured in the eight deer were similar. The intra-assay
CVs were less than 10 percent of the mean, though higher than for the

THYROID-PARATHYROID RESPONSE IN DEER

171

Figure 4. Plasma Ca and CT concentrations in response to CaCh infusion (3 mg / kg
BW / min for 20 minutes) in one male and two female white-tailed deer (mean +
SE).

other two assays. The CVs may have been associated with small sample
size. Maximum binding in both assays was lower (<30 percent) than
in the other assays, reflecting a weaker affinity between antisera and
bPTH. The NSB was lower in the Iso-Tex assay and higher in the
Nichol’s (Fig. 2), reflecting the degree of interference of the assay. A
significant (P<0.05) correlation was found between plasma Ca and
PTH (r = -0.39, n — 16) in the Iso-Tex assay but was found to be
nonsignificant (r — 026, P>0.05, n = 16) in the Nichol’s assay.

The I.N.C. assay contained chicken anti-human PTH middle
molecule (MM) antiserum which is specific and sensitive for human
PTH determinations. However, it appeared to have less affinity than
other assays in recognizing deer PTH. This could be due to its shorter
incubation time. The significant (P<0.05) correlation between plasma
Ca and PTH levels was r = -0.43 (n = 16); thus it reflected the
strongest linear relationship among the four assays.

Because the sensitivity of all four assays was similar (approximately
0.1 ng / ml), we decided to use PTH assay from Iso-Tex (TX). The
decision was based on the low NSB, sensitivity, the convenience of
methodology and the relative correlation coefficient obtained between
plasma Ca and PTH. However, since none of these assays were
designed specifically for deer PTH determinations, the interpretation

172

18

16

14

— 12

"O

\

O)

io

£

3

O 8

O

6

4

-10 0 10 2 0 30 40 50 60 70 80 90 1 20 150

Minutes

Figure 5. Plasma Ca and PTH concentrations in response to EDTA infusion (2.16 mg /
kg BW / min for 20 minutes) followed 30 minutes later by CaCL infusion (3 mg /
kg BW / min for 20 minutes) in one male and two female white-tailed deer (mean
±SE).

of PTH concentrations must be cautious and interpreted in bovine
equivalents.

EDTA infusion has been previously used as a test for PTH response
in humans (Burckhardt et al. 1975), horses (Argenzio et al. 1974), goats
(Hove 1981), dogs (Fox and Heath 1982), and deer (Brown et al. 1981).
Our results confirm the study of Brown et al. (1981), who found that
it took three hours after an EDTA infusion before Ca levels of normal
bucks returned to near baseline levels. Increased Ca after completion
of our EDTA infusions may have been caused by increased bone
resorption or renal conservation or both, as found in dogs (Fox and
Heath 1982).

The rapid elevation of PTH after only 10 minutes of infusion was
consistent with the studies in goats (Hove 1981) and dogs (Fox and
Heath 1982). It is interesting to note, however, that the decline in
PTH was observed before the end of the EDTA infusion in our study.
A progressive study of acute EDTA infusions in cattle (Blum et al.
1974) also indicated that PTH reached a plateau when the change in
blood Ca was greater than 0.06 to 1.9 mg / dl. There are several
possible explanations for this phenomenon in our study: first, an

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

PTH A.— A

—I EDTA h

I CaCI2l«—

o>

I

I-

QL

THYROID-PARATHYROID RESPONSE IN DEER

173

extremely rapid response of the deer thyroid-parathyroid to maintain
homeostasis; second, the deer may have secreted large amounts of
preformed and stored releasable PTH followed by a sustained secretion
of sub-releasable PTH; third, individual variations may have masked
a specific response. A direct measurement of PTH secretion rate would
have represented more accurately the response of the EDTA challenge.
EDTA infusions, however, also suppressed plasma concentrations of
CT, suggesting an active thyroid response to hypocalcemia.

A five- to eight-fold increase in plasma CT in response to CaCE
infusion in experiment 2 suggests that our bovine CT assay was able
to detect deer CT. An active response of the thyroid in response to
hypo- or hypercalcemia is suggested in our deer group. Decreased PTH
during hypercalcemia was also as expected. The suppression of PTH,
however, was smaller compared to the response to EDTA infusion.

In the third experiment the results from the first 30 minutes
postEDTA infusion were consistent with those from the first
experiment. As blood Ca concentration declined, PTH concentrations
increased dramatically, whereas CT concentrations were only slightly
depressed. In response to the CaCE infusion, the rapid drop of
circulating PTH within 10 minutes again indicated the active function
of the parathyroid glands. This is also consistent with Hove’s (1981)
study in goats in which PTH secretion declined within three to five
minutes in response to CaCE infusion. However, infusion of CaCE
after 10 minutes did not induce a greater suppression of plasma PTH
in our animals. In all three deer, PTH level fell to or below baseline
levels after the CaCE infusion, indicating that the parathyroid can be
depressed by hypercalcemia. That the CT concentrations elevated
within 10 minutes after the start of the CaCE infusion agrees well with
the results of experiment 2.

In conclusion, we feel that that EDTA / CaCE infusion is a useful
means of assessing thyroid-parathyroid responsiveness in white-tailed
deer. The rapid change in circulating PTH and CT in response to
changing plasma Ca levels indicates a highly active system. Further
work is needed to assess the effect of sex, seasons, and antler
development on thyroid-parathyroid responsiveness in deer.

ACKNOWLEDGMENTS

The authors are indebted to Dr. Leonard J. Deftos for the bovine CT
assay. This paper is contribution No. 205 from the Rob and Bessy
Welder Wildlife Foundation, and also was partially funded by NIH
grant No. 506-RR-08107.

174

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

LITERATURE CITED

Argenzio, R.A., J.E. Lowe, H.F. Hintz, and H.F. Schryver. 1974. Calcium and
phosphorus homeostasis in horses. J. Nutr. 104:18-27.

Banks, W.J., Jr., G.R Epling, R.A. Kainer, and R.W. Davis. 1968. Antler growth and
osteoporosis I. Morphological and morphometric changes in the costal campacta
during the antler growth cycle. Anat. Rec. 162:387-398.

Berson, S.A. and R.S. Yalow. 1968. Immunochemical heterogeneity of parathyroid
hormone in plasma. J. Clin. Endocrinal. Metab. 28:1037-1047.

Berson, S.A., R.S. Yalow, G.D. Aurbach, and J.T. Potts, Jr. 1963. Immunoassay of
bovine and human parathyroid hormone. Proc. Natl. Acad. Sci. 49:613-617.

Blum, J.W. , J.A. Fischer, D. Schwoerer, W. Hunziker, and U. Binswanger. 1974. Acute
parathyroid hormone response: sensitivity, relationships to hypocalcemia and
rapidity. Endocrinol. 95:753-759.

Brown, R.D., R.L. Cowan, and J.K. Kavanaugh. 1981. Effect of parathyroidectomy on
white-tailed deer. Texas J. Sci. 33:1 13-120.

Burckhardt, P. , B. Ruedi, and J.P. Felber. 1975. The clinical usefulness of the EDTA
infusion as a stimulation test for parathyroid function. Pp. 75-77, in Calcium
regulating hormones (R.V. Talmage, M. Owen, and J.A. Parsens, eds.), American
Elsiver Publ. Co., Inc., New York.

Chao, C.C., R.D. Brown, and L.J. Deftos. 1985. Effects of xylazine immobilization on
biochemical and endocrine values in white-tailed deer. J. Wildlife Dis. 20:328-332.

Cowan, R.L., E.W. Hartsook, and J.B. Whelan. 1969. Deer antler growth-ideal test for
the study of bone metabolism. Sci. Agr. 17:3-4.

Deftos, L.J., J.F. Habener, G.P. Mayer, A.E. Bury, and J.T. Potts, Jr. 1972.
Radioimmunoassay for bovine calcitonin. J. Lab. Clin. Med. 79:480-490.

Dibella, F.P. , J.B. Gilkinson, J. Flueck, and G.D. Aurbach. 1978. Carboxy-terminal
fragment of human parathyroid hormone in parathyroid tumors: Unique new source
of immunogens for the production of antisera potentially useful in the
radioimmunoassay of parathyroid hormone in human serum. J. Clin. Endocrinol.
46:604-611.

Fox, J., and H. Heath, III. 1982. Parathyroid, renal and skeletal responses to induced
hypocalcemia in the dog. Amer. J. Physiol. 242:287-291.

Gorog, R.H., M.K. Hakin, N.W. Thompson, G.A. Rigg, and D.S. McCann. 1982.
Radioimmunoassay on serum parathyrin: Comparison of five commercial kits. Clin.
Chem. 28:87-91.

Grafflin, A.L. 1942. Further observations upon the parathyroid gland in the Virginia
deer in species taken at intervals throughout the year. Endocrinol. 30:571-573.

Hove, K. 1981. A permanent preparation allowing measurements of secretion of
parathyroid hormone in conscious goats. J. Endocrinol. 90:295-306.

Mazur, P.E. 1969. Location, mapping, and surgical removal of the parathyroid gland in
the white-tailed deer ( Odocoileus virginianus). M.S. thesis, Penn State Univ.,
University Park, Pennsylvania.

Perault-Staub, A.M., J.F. Staub, and G. Milhaud. 1975. Constancy and rhythmicity, a
dual concept for homeostasis exemplified by 24-hour plasma calcium regulation in
the rat. Pp. 229-238, in Calcium regulating hormones (R.V. Talmage, M. Owen, and
J.A. Parsens, eds.), American Elsiver Publ. Co., New York.

Ray, A. A. 1982. SAS user’s guide: statistics. SAS Institute, Inc., Gary, North Carolina,
584 pp.

Roughton, R.D. 1975. Xylazine as an immobilizing agent for captive white-tailed deer.
J. Am. Vet. Med. Assn. 167:574-578.

Stephenson, D., and R.D. Brown. 1983. Calcium kinetics of male white-tailed deer. J.
Nutr. 114:1014-1021.

EFFECTS OF POLLUTION EFFLUENTS
ON TWO SUCCESSIVE TRIBUTARIES AND
VILLAGE CREEK IN SOUTHEASTERN TEXAS

by C . MARC BARCLAY and RICHARD C. HARREL

Department of Biology
Lamar University
Beaumont, Texas 77710

ABSTRACT

Physiochemical conditions and macrobenthos were studied in three successive streams
in southeastern Texas from December 1981 to November 1982.

Physicochemical parameters indicated poor water quality during normal and low flow
conditions in the unnamed stream, which received oil refinery effluent, and Mill Creek,
which received sewage and sawmill effluents. Improved water quality was noted
downstream. During high flow conditions, physicochemical conditions did not show the
effects of the effluents. Vilage Creek was unaffected by the effluents and had good water
quality year-round.

Annually, 18,808 individuals and 155 taxa of macrobenthos were collected.
Oligochaetes (69.6 percent and 29 taxa) and chironomids (19.7 percent and 51 taxa) were
dominant. Species of Limnodrilus and Dero were dominant at all stations except in
Village Creek where chironomids formed more than 50 percent of the number of
individuals. Many pollution intolerant taxa were found only in Village Creek.

Station collection diversity (d) ranged from 1.03 to 4.20. Lower d values, increases in
numbers of oligochaetes and decreases in numbers of intolerant taxa, correlated with
decreases in water quality. Community structure of macrobenthos at individual stations
varied little, regardless of flow conditions. Sorenson’s similarity index indicated greater
faunal overlap between stations with similar water quality. Key words : streams,
pollution, physicochemical, macrobenthos, diversity, Texas.

INTRODUCTION

Village Creek occupies a central position among the nine separate
units and stream corridors of the Big Thicket National Preserve in
southeastern Texas. Additionally, the Roy Larsen Sanctuary of the
Nature Conservancy and the proposed Village Creek State Park site are
on its banks.

Mill Creek has more permitted pollution discharges than any
tributary of Village Creek, yet no published data exist concerning its
water quality. Permits for pollution effluents (daily average) in the
Mill Creek system include: (1) South Hampton Oil Refinery (431,490
liters per day); (2) the city of Silsbee North Sewage Treatment Plant
(473,125 liters per day); and (3) Kirby Forest Industries (sawmill
effluent, 1,835,753 liters per day). The South Hampton Refinery had

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

176

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

reduced production and released only a small fraction of the daily
permitted effluent volume during the entire study period. The
objectives of this study were to determine the effects of the above
effluents on physicochemical conditions and community structure of
macrobenthos of the receiving streams and Village Creek.

The study area was within the coordinates 30°20® to 30°25® north
latitude and 94°10® to 94°15® west longitude, roughly the center of
Hardin County, Texas. The streams studied included: an unnamed
tributary of Mill Creek, Mill Creek, and Village Creek (Fig. 1).

The unnamed tributary receives effluent from South Hampton Oil
Refinery and is intermittent during periods of low rainfall. Stream
length from the refinery to its junction with Mill Creek is 3.2
kilometers. It is a 3rd order stream based on Strahler’s (1957)
modification of Horton’s (1945) stream classification.

Mill Creek is a 4th order stream about 14 kilometers long, including
intermittent branches. From the confluence with the unnamed
tributary to the junction with Village Creek is 4.9 kilometers. It
receives treated effluents from Kirby Forest Industries and the Silsbee
North Sewage Treatment Plant.

Village Creek drainage basin covers 2883 square kilometers and is a
5th order stream. From its source to the Neches River into which it
drains, the distance is 125 kilometers. Many of its tributaries above the
study area drain portions of the Big Thicket National Preserve. The
proposed Village Creek State Park site is below the study area.

Vegetation surrounding the study area is typical of the Beech-
Magnolia-loblolly pine slope forest, which has been altered by housing
developments and frequent lumbering (Watson 1979; Marks and
Harcombe 1981). The lower reaches of Mill Creek and the banks of
Village Creek have the appearance of a flood plain slough. Bald
cypress ( Taxodium distichum), river birch ( Betula nigra), willows
(. Salix sp.) and water tolerant oaks ( Quercus sp.) are the dominant
trees.

Eight sampling stations were established (Fig. 1). Stream channel
width varied from approximately one meter at station 1 on the
unnamed tributary to 35 meters at Stations 7 and 8 in Village Creek.
The substrate ranged from clay and silt at Stations 1 and 2 to fine and
coarse sand at Stations 7 and 8.

METHODS

Physiochemical conditions were measured monthly from December
1981 to November 1982. Stream discharge was described as: (1) low-
channel one-fourth or less full, (2) intermediate-channel over one-
fourth to three-fourths full, and (3) high-channel over three-fourths

SOUTHEASTERN TEXAS STREAM POLLUTION

177

Figure 1. Location of sampling stations (numbered solid symbols) and effluent sources
(X — SH, South Hampton Refinery; KFI, Kirby Forest Industries; and ST, Silsbee
Treatment Plant).

full. Dissolved oxygen concentrations and water temperature were
determined using a Yellow Springs Instrument Model 57 Oxygen
Meter. Specific conductance was determined with a Lab Line Lectro
Mho Meter and pH with a Hellige Comparator. Turbidity was
measured with a Bausch and Lomb Spectronic 20 Spectrophotometer.
All other parameters were determined by standard methods (APHA,
1975) or Hach Chemical Company (1975) procedures.

Macrobenthos were collected seasonally from all stations, except
Station 1. Six petite Ponar grabs (15.24 cm X 15.24 cm) were collected
at each station along stream transects. Samples were washed in a no.
30 mesh wash bucket, preserved with formalin and stained with rose
bengal dye (Mason and Yevich, 1967). In the laboratory, each sample
was washed in a no. 30 U.S. Standard Soil Sieve, organisms were
separated from detritus and stored in 70 percent ethanol.

Shannon’s diversity index, as proposed by Patten (1962) and
modified by Wilhm and Dorris (1968), was calculated by the equation:

d = 2 (ni/n) log2 (nj/n),

178

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

where d is diversity per individual, m is the number of individuals of
species i, and n is the total number of individuals collected. Higher
d values indicate more complex community structure and are generally
associated with better water quality or less stress.

Sorenson’s (1948) index of similarity was used to compare taxa
overlap between pairs of stations by the formula:

c — 2C
A + B

where S is similarity, A and B are the total numbers of taxa at Stations
A and B, and C is the number of taxa common to both stations. This
coefficient can range from zero when no common taxa occur to 1.0
when all taxa occur at both stations.

RESULTS

Intermediate discharge was observed from December 1981 through
July 1982, with exceptions of high discharge in February and May.
Low discharge existed from August through November. Extremes of
physicochemical conditions are presented in Table 1. Water
temperatures varied seasonally from 10°C to 30°C, with minor
fluctuations between stations due to degree of shading and water
volume at sampling stations.

Dissolved oxygen concentrations and biochemical oxygen demand
(BOD) were inversely related and indicated stressed conditions at all
stations, except 7 and 8, at some time during the study. Much of the
BOD was satisfied between upstream and downstream stations in the
unnamed tributary (Stations 2 and 3) and Mill Creek (Stations 4, 5,
and 6). Dissolved oxygen concentrations below 2.0 milligrams per liter,
the minimum acceptable value for flowing Texas surface waters (Texas
Department of Water Resources, 1981) occurred on three occasions in
the unknown tributary and 12 times in Mill Creek. Village Creek
(Stations 7 and 8) had high oxygen concentrations and low BOD year-
round.

Concentrations of alkalinity, carbon dioxide, specific conductance,
orthophosphate, ammonia and nitrate nitrogen, sulfate and total
filtrable residue (TFR) varied directly with concentrations and volumes
of effluents.

Turbidity and total nonfiltrable residue (TNR) fluctuated with
changes in discharge, increasing with higher flow rates. The exception
was Station 1 , which had higher concentrations during no-flow
periods when only pools existed. This station had a colloidal clay
substrate. The pH values of all stations were within the expected range

SOUTHEASTERN TEXAS STREAM POLLUTION

179

for streams in southeastern Texas. Ranges of all parameters at Stations
7 and 8 were considered normal for clean-water streams in this area.

A total of 18,808 individuals distributed among 155 taxa of
macrobenthos were collected (Table 2). Few taxa were common or
abundant at any station. Tolerance limits of common taxa and
occurrence of pollution intolerant taxa correlated well with changes in
water quality.

Oligochaetes (29 taxa) were the dominant major group comprising
69.6 percent of the total annual number of individuals. The pollution
tolerant Limnodrilus sp. (three taxa) and facultati veDero sp. (three
taxa) were collected at all stations. Limnodrilus was dominant or
common at all but Station 7. Dero was dominant or common except
at Station 3. Haemonais waldvogeli was collected at all stations except
2, but was common only at Station 6. Stephensoniana tandyi was
found only in Village Creek where it was common. The only other
oligochaetes that occurred in substantial numbers were Ilyodrilus
templetoni, Tubifex harmani, and Aulodrilus pigueti, each common at
separate stations (Table 2).

Leeches (six taxa) were rare or uncommon in all collections and
were not found at Stations 2 and 3. These stations received oil refinery
effluent and always had relatively high specific conductance values.
Sawyer (1974) stated that many leeches are quite sensitive to salinity.

Dipterans represented the largest number of taxa (69) and were
dominant at Stations 7 and 8. Chironomids accounted for 19.7 percent
of the annual number of individuals and 51 taxa, but few were
represented by large numbers. Goeldichironomus holoprasinus was
abundant at Stations 2 and 3. Chironomus attenuatus was a dominant
taxon at Stations 4 and 6. Polypedilum halterale, P. illinoense, and
Rheotany tarsus exiguus were common at Station 6. Cricotopus
bicinctus, Tanytarsus glabrecens, and T. guerulus were abundant at
Stations 7 and 8. Ceratopogonids were common in Village Creek.

Ephemeropterans formed 2.7 percent of the annual number of
individuals (eight taxa). Only the facultative Caenis sp. occurred at
stations outside of Village Creek. Baetis sp. and Baetisca sp., both
intolerant, were common in Village Creek. Trichopterans (nine taxa)
were found only in Village Creek. Coleopterans (five taxa) were
widespread, but only Stenelus sp. was present throughout the year,
and commonly only at Station 7.

Mollusks (seven taxa) comprised 3.9 percent of the annual number
of individuals and were common at Stations 5 through 8. The
gastropods Hebetancylus excentricus, Laevapex sp., Amnicola sp., and
the fingernail clam, Sphaerium sp. , were common or abundant at
Station 6. The introduced Asiatic clam, Corbicula manilensis, was
restricted to Village Creek and common only at Station 8.

180

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

s

<15

-c

u

8

(/5

a

h

0 "O

QJ C

a o

co (j

t"

CO

CO

O

i4

\

CO

CM

oo

MO

CD

r"

to

O

05

J

tn

oo

tO

CM

ID

p

CO

CM

CM

M"

T1

<

g

04

CM

O-

*—<

'—l

i>-

—

05

CM

CM

CM

—

8 |

C \
O

M £

O W)

Si Q.

o

o o

t" Cj>

S -7

0 — 0

T)-1 O

tO

00 co

to

o

- 05

oo ^

f~ r- ^

04

o
o

— o

— CM , _ .

no CO to ‘•o o

-H CM — '

tO

0^00°°

CM -H —

o

GO ^ CT>
0-0
— co —

tO

r»

CD

tO

CO

CD

oo

CD

iO

CD

tO

CD

CM

OO

CD

o

CD

o

-A

CO

05

CD

-A

<r>

-A

CD

©

CD

p

CD

CO

CD

tO

CD

CD

CO

tO

tO

©

OO

05

to

tO

CM 10 ^ ^

© 2 CO 2 ^ ~

to ■— I 1>1

to _

^ CO 00 CM 00 2

O £ tA - to ^ 6

CM ^ OO

co

oo

CO to o o

I-hOOOiO — 05 — OO —

CM

oo CO

00

OO —
trj 00 CM
©

CO «— i CO — CM

°9CMc^'(D°9CMy?tOt2cMt7'^
^OOCOCM^

CM oo CD ^

00

O © CD — < OO © ^

ciot^lModNtOoOlOO^ff)-1

05 to 05 Tt’ CM tO CO CM CO OO ^ OO OO

-A © oo © © © CO

o

oo ^ oo

CM . CM
© 22 ^
© ©

O to © © © ©

-A ©

©

S ^ ® °°

to ^ to P.
o o

CM

6.0-10.8 71-108 0.8-3. 5 3.0-14.0 5.0-

Table 1. — Continued.

SOUTHEASTERN TEXAS STREAM POLLUTION

181

* \

Z be

h £

& \

b be

H S

O j?

c r> £

O be
Z £

X to

z e

O CD

QO O'

CM CM CD ''t1

i in i

^ oo

^ CD

TJH CO

CM O- ^

xT5 O' 00

CM oo 90
CO CM

CM CM ^

CT> 05 ^ ^ _

O' i O' id GO t''

i tD i iO ' CO

CD •M-1 CD oo 90 (

O' O) TO

O

CO O'

6 2

CM

00

<o _

CO

■ ers

co CM
O' !fl CO ^

eo ■

05

. O §

CxI cm cm cm

0

cd

CM

CM

cd

CM

CO

O'

d

O'

CM

oo

d

00

ir>

O'

CO

d

to-

O

O'

CM

505

CM

O

d

00

CM

cd

i£b

00

00

CM

CM

d

to

CM

d

CO

d

td

CM

CO

d

d

cd

to

r~l

00

to

to

CM

>— <

oo

CD rt; _ -H ^ CM

O O rt N rt ^ oj

o d o o o d

00

CM

O' .

^ CM ^ i — 1

d d o d o

CO ^

CM -H

d 6

CM

o CO o »o>

" eo t . o

"? o 7 w ^ OJ 7

8 - £ o - «

oo

cm O'

^ o cn on

^d®<e°N0-°d

CO CD

X CD ®

i O'

SI °d

TO

d

CD

T oo T

£ 2 Si

d

00 00 CO O' oo

^,7io'vcM7'fl7'o7«o

d cm odd

o CM

■M"1 to CM o CO

■m* o- to

CD ''O iO

t0 00 CO t>- O'
irj^CM^onCMon^co"
*7^70070707

GO to- O'

_ TO

07070

CM CM CM

CO CO CO

182

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 2. — Numbers of macrobenthos collected from the Mill Creek system and Village
Creek.

Taxa

2

3

4

Station

5

6

7

8

Hydrozoa

Hydra sp.

-

-

-

-

32

3

-

Turbellaria

Tricladida (unid.)

Nematoda (unid.)

-

-

5

4

8

4

2

3

1

2

Annelida

Oligochaeta

Bratislavia unidentata

-

-

3

3

43

-

-

Chaetogaster diaphanus

-

-

-

7

25

-

-

*Dero sp.

185

8

160

231

789

31

109

Dero digitata

-

9

225

205

472

-

7

Dero furcata

152

-

11

1

4

2

-

Dero obtusa

67

-

-

-

-

-

-

Haemonais waldvogeli

-

3

208

68

55

5

5

Nais communis

-

-

10

1

13

2

3

Nais variabilis

-

-

-

3

3

10

15

Ophidonais serpentina

-

-

-

-

-

-

1

Pristina breviseta

-

-

-

-

4

-

-

Pristina longidentata

-

-

-

-

1

-

-

Pristina longiseta leidyi

-

-

-

1

3

-

1

Pristina longiseta longiseta

-

-

-

1

4

-

-

Pristina osborni

-

-

-

-

-

1

2

Slavina appendiculata

-

-

-

-

18

2

3

Stephensoniana tandyi

-

-

-

-

-

224

143

Opistocysta tribranchiata

-

-

4

3

13

-

-

Lumbriculidae (Unid.)

-

1

2

1

-

-

-

Kincaidiana hexatheca

-

-

-

-

-

1

-

Lumbriculus variegatus

-

-

-

-

-

-

2

Rhynchelmis sp.

-

-

-

-

-

1

-

Tubificidae (Unid.)

-

-

-

-

1

-

-

Aulodrilus pigueti

-

-

9

32

37

-

87

Ilyodrilus templetoni

-

-

47

4

124

-

-

*Limnodrilus sp.

2658

168

1684

2481

1079

32

108

Limnodrilus cervix

-

7

10

31

34

1

6

Limnodrilus hoffmeisteri

106

87

159

168

92

3

2

Limnodrilus udekemianus

86

1

-

-

11

-

1

Potamothrix vejdovskyi

-

-

-

1

2

8

1

Tubifex harmani

1

-

102

25

12

-

-

Branchiobdellida

Xironodrilus formosus

-

2

-

-

-

-

-

Hirudinea

Actinobdella inequiannulata

-

-

-

-

-

1

1

Batracobdella phalera

-

-

-

2

3

-

-

Helobdella lineata

-

-

-

-

1

-

-

Helobdella papillata

-

-

-

1

-

-

-

Helobdella punctatolineata

-

-

-

-

1

-

-

Helobdella stagnalis

-

-

1

-

9

-

-

Crustacea

Isopoda

Asellus sp.

-

-

-

87

85

-

-

SOUTHEASTERN TEXAS STREAM POLLUTION

183

Table 2.— Continued.

Amphipoda

Crangonyx obliguus richmondensis
Decapoda

**Astacidae (Unid.)

**Cambarellus puer
**Cambarus diminutus
**Procambarus sp.

Palaemonetes kadiakensis
Hydracarina (unid.)

Insecta
Collembola
Isotomurus palustris
Plecoptera
Acroneuria sp.

Ephemeroptera
Baetis sp.

Baetisca sp.

Caenis sp.

Ephemerella trilineata
Hexagenia limbata
Leptophlebia sp.

Stenonema sp.

Tricorythodes sp.

Odonata

Coenagriidae (unid.)

Aphylla sp.

Plathemis sp.

Libellula sp.

Macromia sp.

Orethemis ferruginea
Pachydiplax longipennis
Progomphus obscurus
Somatochlora sp.

Hemiptera
Veliidae (unid.)

Megaloptera
Sialis sp.

Trichoptera
A gray lea sp.

Cheumatopsyche sp.

Chimarra sp.

Hydropsyche sp.

Nectopsyche sp.

Nyctiophylax sp.

Oecetis sp.

Potamyia flava
Smicridea sp.

Coleoptera
Hydroporus sp.

Laccophilus sp.

Dineutus sp.

Dubiraphia sp.

Stenelmis sp.

12 16 1

15 2 1 -

1 1

6

3 .

6 - - - - -

1-1-18 9

5

1

3

1

2

1

2 2 12
4

- 155 54

33 25

1 38 76 107

3

1

3

2 1

10 1

1 - - 1

2

3 1

3

--34
3 1 - -

5

10

6

10

7

2

1

1

4

2

1

6

1

3 118

2

59

15

184

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 2. — Continued.

Diptera
Tipulidae
Helius sp.

Hexatoma sp.

Pilaria sp.

T ipula sp.

Culicidae
Culex sp.

Chaoborus punctipennis
Simulidae
Prosimulium sp.

Chironomidae
Tanypodinae
Ablabesmyia annulata
Ablabesmyia mallochi
Ablabesmyia tarella
Aspectrotanypus sp.
Clinotanypus sp.

Coelotanypus sp.

Conchapelopia sp.

Labrundinia floridana
Larsia sp.

Natarsia Baltimoreus
Pentaneura sp.

Procladius sp.

Psectrotanypus sp.

Tanypus carinatus
Tanypus stellatus
Diamesinae
Potthastia longimanus
Orthocladinae
Coryneura taris
Cricotopus bicinctus gr.
Eukieferiella sp.

Nanocladius distinctus
Orthocladius sp.

Psectrocladius vernalis
Pseudorthocladius sp.
Rheocricotopus sp.
Symbiocladius sp.

T hienemanniella sp.-
Chironomini
Chironomus attenuatus
Cryptochironomus blarina
Cryptochironomus fulvus
Dicrotendipes neomodestus
Dicrotendipes nervosus
Einfeldia sp.

Glyptotendipes sp.
Goeldichironomus holoprasinus
Harnischia curtilamellata
Kiefferulus sp.

Parachironamus schneideri

1

2

1 3

10 8

1 1

1 1 5

1

3 4

2

4

1

9

26

3

1

2 7

1

3

1

1

8

2 3

9

1

3

11

1 2

15 2

2 1

4 14

2 9

2

4 7

12 30

3 5

2

7

1

1 1

1 2 2

1

7 107 89

3

1

1

8 9 1

1 1 1

2

4

2 2

-

-

227

48

79

_

1

_

13

19

-

-

3

-

27

-

1

-

-

1

-

-

3

11

24

-

-

2

2

-

767

27

19

15

-

-

-

4

-

-

-

-

1

-

-

1 1

6

3 17

31 18

3

2 2

SOUTHEASTERN TEXAS STREAM POLLUTION

185

Table 2.— Continued.

Paracladopelma sp.

-

.

-

-

1

-

*Polypedilum sp.

-

-

-

-

143

-

-

Polypedilum convictum

-

-

-

-

-

1

-

Polypedilum halterale

1

6

-

11

73

7

32

Polypedilum illinoense

1

9

2

4

62

19

19

Pseudochironomus sp.

-

-

-

-

-

12

27

Rohackia demeijerei

-

-

•

-

-

3

14

Stenochironomus sp.

-

1

-

-

1

2

1

Stictochironomus devinctus

-

-

1

-

1

10

39

Trihelos jucundus

-

-

-

-

32

1

-

Tanytarsini

Cladotany tarsus sp.

-

-

-

-

-

39

45

Paratany tarsus sp.

-

-

-

1

-

-

-

Rheotanytarsus exiguus

-

3

1

3

84

23

13

Tanytarsus glabrecens

-

-

-

-

4

143

367

Tanytarsus guerulus

-

-

-

5

27

216

343

Ceratopogonidae

Type “A”

-

-

-

-

1

9

25

Type “B”

-

-

-

-

4

5

7

Type “Cu'

-

-

-

-

-

23

31

Type “CT”

-

-

-

1

7

3

50

Culicoides sp.

-

-

-

1

-

-

1

Palpomyia tibialis

-

-

3

9

4

-

1

Tabanidae

Chrysops sp.

■

1

-

-

-

-

-

Haematopota sp.

1

-

-

-

1

--

Canaceidae (unid.)

-

1

-

-

-

-

-

Dolichopodidae (unid.)

-

1

-

-

-

-

-

Empididae (unid.)

-

1

-

-

-

-

-

Mullusca

Gastropoda

#Ancylidae (unid.)

-

-

-

-

-

4

-

Hebetancylus excentricus

-

-

-

71

40

-

-

Laevapex sp.

-

8

-

8

41

1

-

Amnicola sp.

-

-

-

-

88

-

4

Physa sp.

-

1

-

-

-

-

-

Menetus ( Micromenetus ) dilatatus

2

-

-

-

2

1

-

Pelecypoda

Corbicula manilensis

-

-

-

-

-

13

55

Sphaerium sp.

-

-

1

-

385

-

1

Total number of individuals (18,808)

4093

373

2926

3927

4296

1474

2019

Total number of taxa (155)

23

29

33

53

83

79

75

Species diversity d

2.08

2.21

2.15

2.14

3.87

4.49

4.44

#For species diversity, unidentified individuals or subsamples were broken down into
percentages per station based upon identified individuals.

**Not included in species diversity and coefficient of similarity indices.

186

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Table 3. — Collection extremes and annual numbers of taxa and diversities at sampling
stations.

Stream, station,
and effluent

No. of

taxa

d

valves

Unnamed tributary

Oil Refinery

Station 2

5-15

1.03-1.70

(23)

(2.08)

Mill Creek

Lumber Mill
and sewage

Station 4

10-20

1.32-2.44

(33)

(2.15)

Station 5

12-31

1.36-3.18

(53)

(2.14)

Station 6

34-52

3.13-3.92

(83)

(3.87)

Village Creek

Clean

Station 7

18-49

2.87-4.20

(79)

(4.49)

Station 8

27-51

3.34-3.67

(75)

(4.44)

The smallest number of taxa per collection (five taxa) occurred at
Station 2 in April during normal flow conditions (Table 3). The
largest number of taxa (52) occurred at Station 6 in January after high
flow conditions. Generally, no seasonal patterns for number of taxa or
individuals were noted. Annually, numbers of taxa increased from
Stations 2 through 6, then slightly decreased at Stations 7 and 8.
Annually, numbers of taxa ranged from 23 at Station 2 to 83 at Station
6. Stations 7 and 8 had 79 and 75 taxa, respectively.

Seasonally, Shannon’s diversity index (d) ranged from 1.03 at Station
2 to 4.20 at Station 7, both during October and low flood conditions
(Table 3). Annual diversity ranges from 2.08 at Station _2 to 4.49 at
Station 7. Stations 2 through 5 had significantly lower d values and
numbers of taxa than Stations 6 through 8.

Sorenson’s coefficient of similarity for comparing faunal overlap
between stations, ranged from 0.122 between Stations 2 and 8, to 0.688
between Stations 7 and 8 (Table 4). The general trend was higher
similarity between closer stations with similar water quality. Greater
similarity existed between adjacent downstream stations than between
adjacent upstream stations.

SOUTHEASTERN TEXAS STREAM POLLUTION

187

Table 4. — Coefficients of similarity for all pairs of stations.

Station

2

3

4

5

6

7

8

2

.314

.321

.263

.264

,198

.122

3

.361

.346

.288

.187

.212

4

.558

.431

.196

.259

5

.588

.288

.312

6

.506

.506

7

.688

8

CONCLUSIONS

Physicochemical conditions, during low and normal stream flow,
and community structure of macrobenthos year-round indicated poor
water quality in the unnamed tributary and Mill Creek. During high
stream flow, physicochemical conditions did not show the effects of
the effluents due to dilution. However, community structure and
diversity of macrobenthos varied little regardless of flow conditions.

Water quality improved at downstream stations in both streams.
Station 2, in the unnamed tributary, and Stations 4 and 5, in Mill
Creek, were in zones of degradation. Stations 3 and 6 were in recovery
zones.

Physicochemical conditions and community structure of macroben¬
thos indicated that Village Creek was unaffected by the effluents and
had good water quality year-round. However, if the oil refinery
increased production and produced the total permitted volume of
effluent this may effect Village Creek and would certainly cause further
degradation in the unknown tributary and Mill Creek.

LITERATURE CITED

American Public Health Association. 1975. Standard methods for the examination of
water and waste water. Amer. Publ. Health Assoc., New York, 14th ed. 1193 pp.

Hach Chemical Company. 1975. Water and Wastewater Analysis Procedures. Hach
Chemical Company, Ames, Iowa, 3rd ed. 198 pp.

Horton, R.E. 1945. Erosinal development of streams and their drainage basins. Bull.
Geol. Soc. Amer., 56:275-370.

Marks, PL. and PA. Harcombe. 1981. Forest vegetation of the Big Thicket, Southeast
Texas. Ecol. Monogr. 51:287-305.

Mason, W.T. and P.P. Yevich. 1967. The use phloxine B and rose bengal stains to
facilitate sorting benthic samples. Trans. Amer. Micro. Soc. 86:221-223.

Patten, B.C. 1962. Species diversity of net phytoplankton of Raritan Bay. J. Mar. Res.
20:57-75.

Sawyer, R.T. 1974. Leeches (Annelida: Hirudinea). In Pollution ecology of freshwater
invertebrates (C.W. Hart, Jr., and S.L. Fuller, eds.), Academic Press, New York.
Sorenson, T. 1948. A method of establishing groups of equal amplitude in a plant
society based on similarity of species content. K. Danske. Vidensk. Selsk. 5:1-34.

188

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Strahler, A.N. 1957. Quantitative analysis of watershed geomorphology. Trans. Amer.
Geophy. U., 38:913-920.

Texas Department of Water Resources. 1981. Texas surface water quality standards.

Texas Dept. Water Resources, Austin. 108 pp.

Watson, G. 1979. Big thicket plant ecology: an introduction. Big Thicket Museum,
Saratoga, Texas, 2nd ed., 65 pp.

Wilhm, T.L. and T.C. Dorris. 1968. Biological parameter of water quality. Bioscience
78:447-481.

SPECTRUM ANALYSIS OF A TRIPLE-CHANNEL,
PULSE-SLOPE-MODULATED WAVETRAIN:

A COMPARISON OF TWO METHODS

by JOSEPH H. NONNAST

Mission Research Corporation

Colorado Springs, Colorado 80933

and OLAN E. KRUSE

Department of Physics

Texas A&I University

Kingsville, Texas 78363

ABSTRACT

Triple-channel pulse modulation in which one channel is carried by the slope of the
leading edge of the pulse, another channel is carried by the slope of the trailing edge
of the pulse, and still another channel is carried by the amplitude of the pulse has been
developed. Two techniques are presented here for determining the spectrum analysis of
the triple-channel pulse train. The results obtained are then checked experimentally.

INTRODUCTION

Electronic-signal modulation can be defined as “the systematic
alteration of a carrier wave in accordance with the message...”
(Carlson 1968). In general, modulation is required for efficient
transmission of messages. The specific modulation technique often is
chosen so as to take advantage of certain factors such as reduction of
noise and interference, associated time-division multiplexing or
frequency-division multiplexing, and ease of detection.

Two basic types of modulation can be identified, depending on the
type of carrier wave. For continuous wave modulation (CW), the
carrier is a continuous, time-varying waveform, usually a sinusoid.
The carrier wave is at a frequency that is much higher than any of
the frequency components of the message. This results in a frequency
translation, or upshift of the message spectrum to a higher band of
frequencies situated around the carrier frequency. For pulse
modulation, the carrier is a discontinuous train of pulses. Being a
discrete process, pulse modulation is well suited for messages which
are discrete, such as the transmission of binary information.

Pulse modulation also can be used effectively to transmit several
analog messages because of the advantage of time-division
multiplexing. Because pulse modulation is discontinuous, the

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

190

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

frequency spectrum is made up of an infinite number of frequency
components.

Most methods of achieving analog pulse modulation involve the use
of a rectangularly-shaped pulse. In reality, no pulse that is transmitted
and received by practical systems can be perfectly rectangular, but
instead, will have a certain rise time and decay time associated with
these edges. A method for carrying information on the leading edges
of a train of pulses by varying the slopes (pulse-slope modulation) has
been developed (Kruse 1952). Leading-edge pulse-slope modulation has
been extended to trailing-edge pulse-slope modulation (Kruse and
Kincaid 1966). Through the use of trapezoidal pulses, this has been
extended further to double-channel leading- and trailing-edge pulse-
slope modulation (Kruse et al. 1967); finally, a combination of pulse-
amplitude modulation plus leading- and trailing-edge pulse slope
modulation (Kruse and Dobbs 1971) has been devised. In the latter,
three channels of information may be carried on one trapezoidal pulse;
therefore, this is a type of time-division multiplexing. With three
channels of information carried on a single trapezoidal pulse train, it
is then possible to add several other pulse trains in a time-division
mode so as to have three times as many channels of information
transmitted as there are pulse trains transmitted. Of course,
appropriate electronic circuitry must be used to produce and
subsequently separate the various channels of information.

A spectrum analysis has been developed and described completely in
the literature for each of the single-channel systems (Kruse 1954; Kruse
and Montgomery 1966) and for the double-channel system (Kruse et al.
1968). It is the purpose of this paper to describe an investigation that
developed and experimentally checked mathematical analyses of the
spectrum of a triple-channel, pulse-slope-modulated wavetrain.

THEORETICAL DEVELOPMENT

Several approaches for obtaining a spectrum analysis of a triple¬
channel, pulse-slope-modulated waveform can be devised, each based
on different approximations. Two techniques were chosen, the first
based on a method developed by Lewis (1966) for the double-channel
system, the second being a direct Fourier analysis of an unmodulated
trapezoidal pulse with modulation inserted afterwards. The first
method will now be examined in detail.

A trapezoidal waveform can be easily constructed by integrating a
series of alternating positive and negative rectangular pulses, as shown
in Figure 1. Modulation can be inserted by varying the height (positive
or negative) and width of these pulses, where the heights of the pulses
control the slopes of the leading and trailing edges, and the areas

SPECTRUM ANALYSIS OF WAVETRAIN

191

Figure 1. Trapezoidal waveform below obtained by integrating waveform above.

under the pulses determine the height to which the leading edge rises
and from which the trailing edge falls.

Depicted in Figure 2 are a series of rectangularly-shaped positive
pulses of maximum value 1 and a series of rectangularly-shaped
negative pulses of minimum value — 1. For the series of positive pulses
given by

n,nT<t<nT + 2Wi (n = 0,2,4 ...)

1 1 '■O, otherwise (1)

Figure 2. Positive and negative pulse trains.

192

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NOS. 2 & 3, 1985

the Fourier representation is
fi(t) = Wi/T + (2/t r)

• X (l/n)sin(7rnWi/T)cos [(7rn/T)(Wi— t)] (2)

n = 1

where 2Wi is the pulse width, and 2T is the pulse period. For the
series of negative pulses given by

f (-1, nT<t<nT + 2W2 (n = 1,3,5 ...)

2 1 * 0, otherwise (3)

the Fourier representation is

f2(t) = -W2/T - (2/t r)

X (l/n)sin(7rnW2/T)cos [(7rn/T)(W2+T— t)] (4)

n = 1

where 2W2 is the width of the negative pulse. Combining these two
functions so that f(t) = f i( t) + f2(t) and assuming that the area under
a positive pulse will approximately equal the magnitude of the area
under a negative pulse, gives

f(t) = (2/ 7r) I (1/n) [sin(77-nW,/T)cos [(7rn/T)(W,-t)]

+ (-1)"+1 sin(7rnW2/T)cos [(7rn/T)(W2-t)]] (5)

Let Wi and W2 be functions of time such that

2Wi = l/[Ki(H-micos out) (1— macos a>3t)] (6)

2W2 = l/[K2(l+m2COS (02t) (1— m3COS C03t)] (7)

where coi refers to the leading-edge frequency of the trapezoidal pulse,
a>2 to the trailing-edge frequency, m to the amplitude frequency, 1/Ki
and I/K2 are the unmodulated widths of the pulses, and mi, m2, and
m3 are the degrees of modulation. A first order approximation to
equations (6) and (7) gives

SPECTRUM ANALYSIS OF WAVETRAIN

193

2Wi;S:i(l/Ki)(l— micos coit)(l+m3cos co3t) (8)

2W2^(1/K2)(1— m2COs &>2t)(l+m3cos a>3t). (9)

If we let Ki - K2 — K and cn = then

T

f(t) = 2K I (l/n)[X„ + (-l)n+1 Y„] (10)

7T n = 1

where

Xn = (1+micos o>it)sin [cnWi(t)] cos [cn(Wi(t)— t)] (11)

Yn = (l+m2cos w2t)sin [cnW2(t)] cos [cn(W2(t)— t)]. (12)

The end result will be the integral of f(t). Thus,

/f(t)dt =_2K_ £ (l/n) [/x„dt + (— l)n+l/Y„dt], (13)

Jr n = 1

which will be the Fourier representation of the triple-channel pulse-
slope waveform.

A second method for obtaining a Fourier representation involves a
more direct approach. The nonmodulated trapezoidal pulse is shown
in Figure 3, from which it can be seen that

V(t) = V,(t) + V2(t) + V3(t) (14)

where

Vi(t) — kit, 0 < t < xi

or because xi = H/ki

(15)

Vi(t) = kit, 0<t<H/ki.

(16)

Also,

V2(t) = H,H/ki<t<P (17)

where P is the pulse duration of the square wave fed into the
modulator. Finally,

194

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

V(t)

Figure 3. Trapezoidal pulse waveform indicating parameters and variables described in

text.

V3(t) = H - k2(t— P), P<t<P + x3 (18)

but, since x3 = H/k2,

V3(t) = H - k2(t-P), P<t<P + H/k2 (19)

The Fourier series is defined as

00

V(t) = A0 + 2 (Ancos cnt + Bnsin c„t) (20)

n= 1

and the coefficients are found to be

A0 = cPH - (cH2/2) [( 1/ki) - (l/k2)] (21)

An = (kl/27T2n2c) [cos (CnH/kl) — 1] —

(k2/27T2n2c) [cos [cn (P+H/k2)] — cos (cnP)J (22)

Bn = (ki/27r2n2c)sin(cnH/ki) —

(k2/27r2n2c) [sin [cn (P+H/k2)] - sin (cnP)] (23)

SPECTRUM ANALYSIS OF WAVETRAIN

195

Figure 4. Amplitude-modulated waveform (m3=0.0488).

where c = 1/TC.

Now, when the leading and trailing edges and the amplitude of this
trapezoidal pulse train are modulated, ki, k2, and H become functions
of time such that

ki(t) = K0i(l+misin co it)

(24)

k2(t) = K02(l+m2sin out)

(25)

H(t) = H0(l+m3sin cu3t)

(26)

where Koi is the unmodulated leading-edge slope, K02 is the
unmodulated trailing-edge slope, Ho is the unmodulated amplitude,
and mi, m2, and m3 are the degrees of modulation. Substitution of the
above quantities into equations (21), (22), and (23) gives the final
results.

EXPERIMENTAL RESULTS

An experimental check of the two methods of finding the spectrum
of a triple-channel pulse-slope modulated wavetrain was carried out in
the laboratory using a Hewlett Packard Model 310A Wave Analyzer. A
direct check, however, was not possible, as the number of sidebands
produced by the three modulating signals became indistinguishable on
the wave analyzer. Two simplified cases were examined instead, these
being (1) amplitude modulation, and (2) trailing- and leading-edge
modulation.

196

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Vol

age

2.63

Wave height =4. IV -512

..552

.24 9 1

|-2Z*9 Unmod. slopes = .37V/ps |

i

.120 i

i c1 =9.52 xloVad/sec -538 1

i

.100,

1

j u)3= 9.42x10 rad/sec *

i

I

l

r— 1

I

1 1
i i i

iij - Calculated (Direct) i

i

j

.07

1

l

i 1

• j : - — Calculated (Integration) 1

i

i

.06

1

1

l

i

i 1

1 |

j m =0488 1

3 |

i

.05

_ 1

1

i

i

1 1
i i

1 1
i 1

! i

I i

!

i

.04

_ l

1

i

i

j!

i 1

; ;

i

.03

1

1

i

I

! i

1 1

1 1

1 1

! i

i ill

i

i i i

.02

1

1

l

1

1 '
j |

I i

1 1

i i

i i

i i

: i

i i

i

I

.01

_ 1

1

i

i

i

1;

j i i

1:

i !

i ,

1

i

1

L

OLl

JLl

1±

ill! 1 1

i—L

! 1
JLl

l

Figure 5. Frequency spectra of amplitude-modulated waveform showing fundamental
frequency, three harmonics and several sidebands (m3=0.0488).

Voltage

2.63 .538

2.49

.10

2.49

.512

.552

Wave height = 4.1V
Unmod, slopes = .37V/ps
c.| = 9.52 x 10^rad./sec.
w3= 9.42 * 109 rad/sec.
nry .0488

Measured

- Calculated (Direct)

- Calculated (Integration)

.08

.06

.04

I

I

j

.02

C1

__L

c9

c10

Figure 6. Frequency spectra of amplitude-modulated waveform showing fundamental
frequency and nine harmonics (m3=0.0488).

SPECTRUM ANALYSIS OF WAVETRAIN

197

Figure 7. Frequency spectra of leading- and trailing-edge modulated waveform showing
fundamental frequency, two harmonics and numerous sidebands (mi^O.SO and
m2=0,20).

Voltage

Figure 8. Frequency spectra of leading- and trailing-edge modulated waveform showing
fundamental frequency and nine harmonics (mi=0.30 and 012=0.20).

198

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

(/) Amplitude Modulation Only

For the case of amplitude modulation only, co i = a>2 = 0, and,
therefore, mi = m2 = 0. A check was performed on these equations
using a degree of modulation of 0.0488. The waveform used is shown
in Figure 4 and voltage measurements and calculations are plotted in
Figures 5 and 6. The only discrepancies occurred in the even
harmonies of the pulse repetition frequency, where the integration
method predicted no even harmonics, and the direct method predicted
only extremely small voltages.

(2) Trailing- and Leading-Edge Modulation

For both trailing- and leading-edge modulation, C03 = 0 and m3 =
0. An experimental check was performed using degrees of modulation
of mi = 0.30 and m2 — 0.20. Voltage measurements and calculations
are plotted in Figures 7 and 8. The major differences were found in
the second sideband terms of both an and a>2.

CONCLUSIONS

The general agreement between the measured values and the
calculated values of the frequency components seems to indicate that
both derivations are reasonable approximations of the Fourier
representation of a triple-channel, pulse-slope-modulated waveform.
Both methods, however, did not give good results for (1) the first two
even harmonics of the pulse repetition frequency, (2) all even
harmonics of the pulse repetition frequency for the case of amplitude
modulation only, and (3) second sideband terms for the first two
harmonics. The fact that all experimental checks required that P equal
one-half the pulse repetition frequency (that is, the distance between
the fixed points of the leading and trailing edges be equally spaced)
seems to account for the first discrepancy stated above. It was difficult
to get the actual waveform to exactly fit this criterion, and it was noted
that an extremely small deviation produced a large change in these
even harmonics. Both methods failed to predict any even harmonics for
the case of amplitude modulation only, but this was merely because
the modulating terms of the derived equations were not inserted until
after the Fourier analysis had already been performed on the
trapezoidal waveform. Because the frequency spectra of a symmetrical
trapezoidal wavetrain contains no even harmonics, the end result, after
modulation was inserted, also would contain no even harmonics. An
explanation of the third deviation stated above is not immediately
obvious, but most likely it is a combination of the items previously
discussed.

The only way to obtain a more accurate mathematical representation
of the frequency spectra would be to insert the modulating terms

SPECTRUM ANALYSIS OF WAVETRAIN

199

before the Fourier analysis is performed. The extreme complexity
involved probably would overshadow the usefulness for any practical
cases.

LITERATURE CITED

Carlson, A. B. 1968. Communication systems, an introduction to signals and noise in
electrical communication. McGraw-Hill Book Company, New York.

Kruse, O. E. 1952. Variable slope pulse modulation. Proc. I.R.E. 40:1604.

- . 1954. Spectrum analysis of variable slope pulse modulation waves. Texas J. Sci.

6:378-389.

Kruse, O. E., and W. D. Dobbs. 1971. Triple-channel modulation of a single pulse train.
Electronic Engin. 43:36-37.

Kruse, O. E., and J. K. Kincaid. 1966. Trailing edge variable slope pulse modulation.
Electronic Engin. 38:294-295.

Kruse, O. E., D. L. Lewis, and J. K. Jackson. 1967. Combination leading and trailing
edge variable slope pulse modulation. Electronic Engin. 39:174-176.

Kruse, O. E. , D. L. Lewis, and R. L. Metz. 1968. Spectrum analysis of a train of
modulated combination leading- and trailing-edge variable-slope pulses. Electronic
Engin. 40:100-102.

Kruse, O. E., and R. W. Montgomery. 1966. Spectrum analysis of a train of modulated
trailing edge variable-slope pulses. Electronic Engin. 38:593-595.

Lewis, D. L. 1966. An experimental study of leading and trailing edge trapezoidal pulse
modulation. M.S. thesis, Texas A8cl Univ., Kingsville.

STUDIES OF VEGETATIVE PLANT TISSUE
COMPATABILITY-INCOMPATABILITY. VII. INFLUENCES
OF INDIVIDUAL ORGANS ON GRAFT DEVELOPMENT

by RANDY MOORE

Department of Biology
Baylor University
Waco, Texas 76798

ABSTRACT

The contributions of different plant parts to (1) tensile strength of and (2) xylary
redifferentiation in the compatible autograft in Sedum telephoides were estimated by
measuring these features in grafts from which various organs had been removed. The
majority of all grafts cohered, irrespective of the experimental treatment. Grafts between
isolated internodes cohered in the absence of vascular redifferentiation, and were
characterized by tensile strengths averaging 14 g breaking weight per square millimeter
of graft area. By five days after grafting, those grafts characterized by decreased callus
proliferation had tensile strengths 40-50 percent less than those of control (that is,
otherwise untreated) grafts. The contributions of plant organs to the tensile strength of
the graft union were estimated to be as follows: stem tissues adjacent to the graft union,
16-20 percent; axillary buds and immature stem tissue in the scion, 26 percent; scion
leaves, 54 percent. The contributions of organs to the number of xylary strands bridging
the graft union were estimated to be as follows: acropetal movement from the stock, 12
percent; scion leaves, 65 percent; axillary buds and immature stem tissue of the scion,
23 percent. Scion leaves closest to the graft union contribute more to the tensile strength
and amount of xylem bridging the graft union than do other leaves of the scion. Finally,
graft unions having similar tensile strengths do not necessarily have proportional
numbers of xylary strands bridging the graft union.

INTRODUCTION

Graft formation between compatible plant tissues follows a
predictable developmental pattern. In general, the development of
compatible unions can be divided into three stages — (1) initial
cohesion of the stock and scion, (2) proliferation and interdigitation
of callus cells at the graft interface, and (3) redifferentiation of vascular
tissue across the graft interface. Each of these structural events
sequentially contributes to the development of tensile strength of the
graft union (Moore 1982a, 1983a), and is vital to reformation of the
stem at the graft interface.

Recently, several descriptive studies of the various events that occur
during graft development have yielded significant results (Shimomura
and Fujihara 1976; Stoddard and McCully 1979; Moore and Walker
1981a, 1981b, 1981c; Moore 1982a, 1982b, 1982c; Jeffree and Yeoman

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

202

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

1983; see reviews by Moore 1981a, 1984a; McCully 1983). A smaller
number of experimental studies also has provided valuable
information (Yeoman and Brown 1976; Shimomura and Fujihara 1977,
1978; Stoddard and McCully 1980; Parkinson and Yeoman 1982; Moore
1984b; Moore and Walker 1983; also, see review by Moore 1983b). The
most notable of these experimental studies is the investigation by
Stoddard and McCully (1980), who provided evidence that the events
comprising the development of compatible autografts occur
independently.

Previous studies of the development of tensile strength in compatible
plant grafts have suggested that the tensile strength of a graft union
may be used as an indicator of graft development (Moore 1982a, 1983a
and references therein). In this study I have determined the
contributions of various organs to the tensile strengths of and xylary
redifferentiation in compatible autografts of Sedum telephoides
(Crassulaceae). Details regarding the cellular and biochemical aspects
of autograft development in Sedum are presented elsewhere (Moore
and Walker 1981a, 1981b, 1981c; Moore 1982b, 1982c, 1982d).

MATERIALS AND METHODS

A clonal population of Sedum telephoides was grown in a porous
soil mixture in a controlled-environment growth chamber. The plants
were placed under a daily regime of 16 hours of light and 8 hours of
dark with day-night temperatures of 30 and 26° C, respectively. Light
was provided by a combination of incandescent and fluorescent lamps
at an intensity of 40 klux atop the plant canopy. Plants were watered
daily with a modified Hoagland solution.

Grafting procedures were similar to those described previously by
Moore and Walker (1981a). All grafts were made in the middle of the
internode below the third pair of leaves. The diameters of the
internodes at the time the grafts were made ranged from 2.0 to 3.7 mm.
The diameter of the graft union when the experiments were terminated
ranged from 2.0 to 4.4 mm.

Plants were subjected to one of the following treatments immediately
after grafting: A, control (that is, no further manipulation); B, all
leaves removed from the stock; C, all leaves, including the shoot apex,
removed from the scion; D, scion severed approximately a centimeter
above the graft union; E, scion severed approximately a centimeter
above the first node above the graft union; F, stock severed
approximately a centimeter below the graft union; and G, grafted
internodes isolated by severing the stem approximately a centimeter
above and below the graft union. Plants with intact root systems were
maintained in a growth chamber under conditions described above.

PLANT TISSUE COMPATIBILITY-INCOMPATIBILITY

203

Plants subjected to Treatment F were kept with their cut surfaces in
a few mm of distilled water and enclosed in a glass container in the
growth chamber. Grafts between isolated internodes (Treatment G)
were maintained on moist filter paper in a closed petri dish in the
growth chamber.

Xylary strands bridging the graft union were counted in five, 10,
and 15 day-old grafts that had been hand-sectioned longitudinally.
Sections were placed in 70 percent lactic acid at 75° C for 16 hours,
and then stained in a 0.5 percent aqueous basic fuchsin for two to
three minutes. Sections were destained in 0.5 percent sodium
metabisulphite in 0.1 M aqueous HC1. Counts of individual xylary
strands transversing the graft union were made using a light
microscope with brightfield optics. A minimum of seven grafts was
examined for each time point (that is, five, 10, and 15 days after
grafting) of each experimental treatment.

Ten grafts from each treatment and age group were sampled for
cohesion. Grafts were judged to cohere if the stock and scion remained
together after isolating the graft union. The same grafts were used to
determine the tensile strengths of the graft unions. The tensile
strengths of graft unions were measured in a manner similar to that
described previously (Lindsay et al. 1974). Values for the tensile
strength of the graft unions were expressed as the force required to
separate the two graft partners (grams breaking weight) per unit of
contact area (square millimeters of graft area).

Median longitudinal hand-sections were observed with a light
microscope to determine the extent of callus proliferation at the graft
interface. All determinations were subjective evaluations based on the
inspection of at least five different grafts.

The significance of differences in means between different time
points as well as different treatments were assessed using a Student’s
t-test (Snedecor and Cochran 1976). Differences with P > 0.05 under
the null hypothesis were considered insignificant. Comparisons to
determine quantitative contributions of plant parts to graft
development were limited to those treatments in which either the
tensile strengths or numbers of xylary strands bridging the graft union
were significantly different.

RESULTS AND DISCUSSION
Cohesion of Stock and Scion

The majority of all plant grafts examined in this study cohered
irrespective of the experimental treatment (Table 1). However, grafts
subjected to Treatments C, D, and G were characterized by less callus
proliferation at the graft interface than grafts subjected to the other

204

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table I. Percentage cohesion of stock and scion following removal of various organs.

Treatment

5 days

Days After Grafting

10 days

15 days

A.

Control

90

100

100

B.

All leaves removed from stock

100

90

100

C.

All leaves removed from scion

90

100

100

D.

Scion severed just above graft

100

100

100

E.

Scion severed one node above graft

100

90

100

F.

Stock severed just below graft

100

100

90

G.

Isolated internodes

70

60

60

treatments. Cohesion of stock and

scion

in the absence

of xylary

redifferentiation reported here (Table 2, Treatment C and G) and
elsewhere (Muzik 1958; Shimomura and Fujihara 1978; Stoddard and
McCully 1980; Moore and Walker 1981b; Parkinson and Yeoman 1982)
supports the suggestion that cohesion of graft partners is independent
of vascular redifferentiation (Stoddard and McCully 1980).
Furthermore, cohesion of isolated internodes (Table 1, Treatment G)
indicates that neither leaves, axillary buds, nor a large amount of stem
tissue in either the stock or scion are necessary for graft cohesion.
These results are also consistent with the suggestion that graft
cohesion is the result of a cellular wound-response that is not directly
related to graft compatability-incompatability (Moore and Walker
1981a).

A previous study of the development of tensile strength in the
compatible Sedum autograft and the incompatible Sedum-Solanum
heterograft led to the suggestion that the cellular wound-response
responsible for cohesion of the stock and scion contributes
approximately 12 g breaking weight (BW) per square millimeter of
graft area (GA) to the tensile strength of the graft union (Moore
1983a). The tensile strengths of grafts characterized by decreased callus
proliferation and the absence of vascular redifferentiation (Treatment
G) agree well with this value. Significantly, the incompatible
heterograft between Lycopersicon esculentum and Brassica oleraceae ,
in which both interdigitation of callus tissue and vascular
redifferentiation are absent, is characterized by a maximal tensile
strength of approximately 13 g BW/mm2 GA (Moore, unpublished
results). These results provide further support for the suggestion that
the cellular wound response responsible for the initial cohesion of
graft partners contributes approximately 12 g BW/mm2 GA to the
tensile strength of graft unions between herbaceous tissues (Moore
1983a). More comprehensive discussions of the contributions of
structural events that occur during the formation of plant grafts to the

PLANT TISSUE COMPATIBILITY-INCOMPATIBILITY

205

Table 2. Number of xylem strands (± one standard deviation) bridging the graft union
in response to removal of various organs.

5 days

Days After Grafting

10 days

15 days

Treatment

Number of

xylem strands % control

Number of
xylem strands %

control

Number of

xylem strands % control

A. Control

B. All leaves removed

from stock

5 + 6

8 ± 9

a

121 + 46

130 + 41

176 ±61

159 ±53

C. All leaves removed
from scion

0

0

37 + 22

31

61 ±28

35

D. Scion severed just
above graft

2 + 3

12 ±7

10

21 ± 10

12

E. Scion severed one node
above graft

6 + 5

52 + 20

43

91 ±31

52

F. Stock severed just
below graft

3 + 5

32+ 17

26

31 ±20

18

G. Isolated internodes

0

0

0

0

0

0

indicates that mean is not statistically significantly different from that of control.

development of tensile strength are presented elsewhere (Lindsay et al.
1974; Yeoman and Brown 1976; Moore 1982a, 1983a).

Tensile Strength

By five days after grafting, the average tensile strength of untreated
(control) grafts was 28 ± 4 g BW/mm2 GA (Table 3, Treatment A).
The only treatments that resulted in a statistically significant decrease
in the tensile strength of the graft union from those of control grafts
were those involving (1) removal of all leaves from the scion (P <
0.05), (2) severance of the scion just above the graft union ( P < 0.01),
and (3) grafts between isolated internodes (P < 0.05) (Table 3). Because
the number of xylary strands in grafts subjected to each of these
treatments was small at five days after grafting (Table 2), vascular
redifferentiation appears not to contribute significantly to the tensile
strength of the Sedum autograft during the early stages of graft
formation. Furthermore, because the numbers of xylary strands in any
two experimental groups at five days after grafting were not
significantly different, differences in the amounts of vascular tissue
cannot account for the reduced tensile strengths of grafts subjected to
any experimental treatment. I believe that the 40-50 percent reductions
in tensile stengths of grafts subjected to Treatments C, D, and G at
five days after grafting are probably due to the decreased amount of
callus proliferation (and interdigitation) associated with these
treatments. Indeed, decreased callus proliferation at the graft union in
response to similar treatments has been reported previously (Stoddard

206

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 3. Tensile strength (± one standard deviation) of the graft union in response to
removal of various organs.

Treatment

5 days

Days After Grafting

10 days

1 5 days

Tensile
strength2 %

control

Tensile
strength %

control

Tensile

strength % control

A.

Control

18 + 4

59 ±9

69 + 8

B.

All leaves removed

from stock

23 + 6

b

52+ 10

—

63 ±9

—

C.

All leaves removed

from scion

16 ± 7

57

28 + 9

47

32 + 9

46

D.

Scion severed just

above graft

14 + 6

50

14 ±4

24

14 ±8

20

E.

Scion severed one node

above graft

21+6

—

36+11

61

54 ±9

78

F.

Stock severed just

below graft

25 + 8

—

26+ 14

44

31+8

45

G.

Isolated internodes

17 + 6

61

13 ± 8

22

11 ± 7

16

Expressed as g breaking weight per mm2 graft area.

indicates that mean is not statistically significantly different from that of control.

and McCully 1980). Also, failure of pea root autografts has been
positively correlated with decreased amounts of callus proliferation
(Stoddard and McCully 1979).

At 10 and 15 days after graftng, the tensile strengths of grafts from
which all leaves were removed from the stock (Treatment B) were
again statistically indistinguishable from those of controls (Treatment
A). This indicates that leaves of the stock do not contribute
significantly to the development of tensile strength of the graft union
in Sedum autografts. This conclusion is confirmed by the statistically
indistinguishable numbers of xylary strands bridging the graft union
in Treatments A and B (Table 2), and supports the suggestion that
leaves of the stock do not contribute significantly to vascular
redifferentiation during graft formation (Stoddard and McCully 1980).
These results are also consistent with the observation that leaves
located below a wound do not have a promotive effect on vascular
regeneration (Aloni and Jacobs 1977).

The tensile strengths of grafts between isolated internodes were not
significantly different from those of grafts in which the scion was
severed just above the graft union. Therefore, roots, axillary buds, and
the majority of stem tissue of the stock do not contribute significantly
to the initial cohesion between graft partners. These results are
consistent with the previously mentioned suggestion that the initial
cohesion of graft partners results from a localized wound response
(Moore and Walker 1981a).

PLANT TISSUE COMPATIBILITY-INCOMPATIBILITY

207

Comparisons of the tensile strengths of grafts subjected to various
experimental treatments allow for estimations of the relative
contributions of organs above and below the graft union to the total
amount of stimulus responsible for the tensile strength of the graft
union. As a first approximation, I assumed that there is a linear
relationship between the amount of stimulus and the tensile strength
of the graft union. Estimations of contributions of organs to the tensile
strength of the graft union are based on data for 15-day-old grafts,
because it has been shown previously (Moore 1983a) that Sedum
autografts of this age have completed development (that is, have a
tensile strength comparable to that of an ungrafted internode).

When all leaves were removed from the scion (Treatment C), the
tensile strength of the graft union was only 46 percent that of controls
at 15 days after grafting ( P < 0.001). Therefore, leaves of the scion are
responsible for contributing approximately 54 percent of the tensile
strength of the graft union. Also, (1) stem tissue immediately adjacent
to the graft union contributes approximately 16-20 percent of the
tensile strength of the union (Treatments D, G), and (2) the
contributions of axillary buds and immature stem tissue in the scion
to the tensile strength of the graft union must be approximately 26
percent (that is, 46 minus 20). Finally, if (1) the contributions of the
stock and scion to the initial cohesion of graft partners are assumed
to be equal (as suggested by ultrastructural evidence — see Moore and
Walker 1981a), and (2) the initial cohesion does indeed account for
approximately 16 percent of the final tensile strength of the graft
union, then the internodal tissue adjacent to the graft interface in each
partner contributes approximately eight percent (that is, 16/2) of the
tensile strength of the graft union.

At 15 days after grafting, the presence of one pair of leaves and
axillary buds on the scion (Treatment E) resulted in a tensile strength
1.7 times that of a defoliated scion (P < 0.01). That is, although the
leaves and axillary buds of Treatment E represent only 30 percent of
the leaves and buds present in the intact scion, they are responsible for
a 70 percent increase in the tensile strength of the graft union. I believe
that this increase in tensile strength of the graft union is due to the
increased levels of callus interdigitation and xylary redifferentiation
characteristic of this treatment (Treatment E) as compared to grafts
subjected to Treatment C. Interdigitation of callus cells and vascular
redifferentiation contribute significantly to the development of tensile
strength in the Sedum autograft (Moore and Walker 1981a; Moore
1983a).

208

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Xylary Redifferentiation

I did not detect any redifferentiation of xylary elements across the
graft unions between isolated internodes (Table 2, Treatment G). A
similar observation has been made previously for grafted internodes of
Coleus (Stoddard and McCully 1980) and Lycopersicon (Parkinson and
Yeoman 1982), and indicates that internodal tissue adjacent to the graft
union lacks xylem-inducing stimuli. If it is assumed that the number
of xylary strands bridging the graft union is directly proportional to
the amount of xylem-inducing stimulus, then approximately 12
percent of the xylem-inducing stimulus moves acropetally from the
stock into the severed scion by 15 days after grafting (Treatment D).
This finding agrees well with the observations that (1) eight percent
of the xylem-inducing stimulus moves acropetally in Coleus autografts
(Stoddard and McCully 1980), and (2) acropetal flow can account for
0-25 percent of the total stimulus for xylary redifferentiation in
wounded stems (Jacobs 1953; Aloni and Jacobs 1977).

At 15 days after grafting, grafts from which all scion leaves were
removed had 65 percent fewer xylary strands bridging the graft union
than did controls (P < 0.01). Therefore, scion leaves contribute
approximately 65 percent of the xylem-inducing stimuli during graft
formation. Inasmuch as removal of leaves from the stock had no effect
on the number of xylary strands bridging the graft union (see above),
(1) leaves above, but not below, the graft union influence the amount
ox xylem that redifferentiates across the graft union (Aloni and Jacobs
1977), and (2) axillary buds and immature internodal tissue in the
scion contribute approximately 23 percent (that is, 35 minus 12) of the
xylem-inducing stimuli for vascular redifferentiation during formation
of the Sedum autograft. For comparative purposes, Stoddard and
McCully (1980) have determined that immature stem tissue and leaves
of the scion contribute at least 10 and 46 percent, respectively, of the
xylem-inducing stimuli during graft formation in Coleus autografts.

Grafts in which the scion was severed one node above the graft had
4.3 times more xylary strands bridging the graft union at 15 days after
grafting than did grafts in which the scion was severed immediately
above the graft ( P < 0.001). Because leaves and axillary buds of grafts
subjected to Treatment E represented only 30 percent of the total
number of leaves and buds on the scion, I conclude that these organs
contribute a proportionally greater amount of xylem-inducing stimuli
than do other leaves of the scion. Stoddard and McCully (1980) have
suggested that this may be due to the lack of additivity of stimuli from
one type of source, and/or the proximity of the leaves and buds on
grafts subjected to Treatment E to the graft union.

In spite of the disproportionate contribution of xylem-inducing
stimuli by leaves nearest the graft union, the number of xylary strands

PLANT TISSUE COMPATIBILITY-INCOMPATIBILITY

209

in grafts subjected to Treatment E was still only 52 percent that of
controls at 15 days after grafting (P < 5 percent). This indicates that
more distally located leaves and axillary buds contribute a
proportionally smaller, but nevertheless significant, amount of xylem-
inducing stimuli during graft formation in this system.

When the stock was severed just below the graft union, there was
an 82 percent reduction in the number of xylary strands bridging the
graft union by 15 days after grafting (Treatment F) (P < 0.001). A
similar decrease in the number of xylary strands bridging a graft union
has been observed previously in Coleus grafts, and has been attributed
to a reduced basipetal flow of xylem-inducing stimuli (for example,
auxin) resulting from the removal of a sink (that is, the stock).

Development of Compatible Grafts

Although xylary redifferentiation does not contribute to the initial
cohesion of the stock and scion, it does contribute significantly to the
development of tensile strength during the later stages of graft
development (Moore and Walker 1981a; Moore 1982a, 1983a). The
increasing tensile strength characteristic of the development of the
compatible Sedum autograft is typically correlated with increased
numbers of xylary strands bridging the graft union (Tables 2 and 3).
Similarly, there is a good correlation between the mechanical strength
and the total number of wound vessel members near the graft union
in compatible autografts in Lycopersicon (Parkinson and Yeoman
1982). However, the percent tensile strength of a graft union and
number of xylary strands bridging the graft union (as compared to
controls) were not equal. For example, the 22 percent decrease in
tensile strength of grafts in which the scion was severed one node
above the graft union (Treatment E) was correlated with a 48%
reduction in the number of xylary strands bridging the graft union.
Such an observation is consistent with the suggestion that vascular
redifferentiation is not the sole contributor to the tensile strength of
the graft union (Moore and Walker 1981a; Moore 1982a, 1983a).

Separating the contributions of callus interdigitation and vascular
redifferentiation to the tensile strength of a graft union is difficult,
because most treatments that result in reduced numbers of xylary
strands bridging the graft union also are characterized by reduced
proliferation and interdigitation of callus tissue at the graft interface.
Indeed, a similar observation in another grafting system has led to the
suggestion that callus proliferation and xylary redifferentiation in graft
formation are induced by a common stimulus (Stoddard and McCully
1980). The only experimental treatment in which a decreased tensile
strength of the graft union was not associated with reduced production
of callus tissue at the graft interface was Treatment F (severance of the

210

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

stock below the graft). When the contribution of the initial wound
response to the tensile strength of the graft union is corrected for, the
tensile strength of the graft union due to callus interdigitation and
vascular redifferentiation is approximately 20 g BW/mm2 GA (that is,
31 minus 11), a value 34 percent that of a similarly corrected control
graft. In these same grafts, however, there was only 18 percent the
number of xylary strands bridging the graft union as compared to
those of controls. Similar calculations for grafts with comparable
tensile strengths but reduced callus proliferation (Treatment C)
indicate that callus interdigitation and xylary redifferentiation
accounted for 36 percent of the tensile strength and 35 percent of the
number of xylary strands bridging the graft union (as compared to
controls). These results indicate that grafts having similar tensile
strengths may have significantly different numbers of xylary strands
bridging the graft union. Consistent with this observation is the one
of Parkinson and Yeoman (1982) that graft unions having similar
tensile strengths may have markedly different numbers of wound
vessel-members at the graft interface. Therefore, whereas the
contribution of the cellular wound-response to the initial cohesion of
graft partners appears to be rather consistent (that is, approximately
12 g BW/mm2 GA), the contributions of callus interdigitation and
vascular redifferentiation may be quite variable.

The influences of various organs on graft development reported here
are in reasonably good agreement with those reported previously by
Stoddard and McCully (1980). These results suggest that substances
(possibly auxin) derived from the scion (especially scion leaves) are
critical to complete vascular redifferentiation across the graft interface,
and support the suggestion that the development of compatible plant
grafts is actually a composite of several independent events (Stoddard
and McCully 1980). Thorough discussions of these ideas are presented
elsewhere (Stoddard and McCully 1980).

ACKNOWLEDGMENTS

I thank Steve Ransom for critically reviewing the completed
manuscript. Support for this research was provided by Grant ,no. PCM-
8207933 from the National Science Foundation

LITERATURE CITED

Aloni, R., and W. P. Jacobs. 1977. Polarity of tracheary regeneration in young internodes
of Coleus. Amer. J. Bot. 64:395-403.

Jacobs, W. P. 1953. The role of auxin in differentiation of xylem around a wound. Amer.
J. Bot. 39:301-309.

Jeffree, C. E., and M. M. Yeoman. 1983. Development of intercellular connections
between opposing cells in a graft union. New Phytol. 93:491-509.

PLANT TISSUE COMPATIBILITY-INCOMPATIBILITY

211

Lindsay, D. W., M. M. Yeoman, and R. Brown. 1974. An analysis of the development
of the graft union in Lycopersicon esculentum. Ann. Bot. 38:639-646.

McCully, M. E. 1983. Structural aspects of graft formation. Pp. 71-88, in Vegetative
compatability responses in plants (R. Moore, ed.), Baylor Univ. Press, Waco, Texas.

Moore, R. 1981a. Graft compatability and incompatability in higher plants. What’s New
in Plant Physiol. 12:13-16.

- — . 1982a. Graft formation in Kalanchoe blossfeldiana. J. Exp. Bot. 33:533-540.

- . 1982b. Studies of vegetative plant tissue compatability-incompatability. V. A

morphometric analysis of the development of a compatible and an incompatible
graft. Can. J. Bot. 60:2780-2787.

- . 1982c. A SEM study of the early events in graft formation in plants. Scanning

Electron Microscopy/ 1982 3:1103-1107.

- . 1982d. Further evidence for cell wall deposition during graft formation. Ann.

Bot. 50:599-604.

- . 1983a. Studies of vegetative plant tissue compatability-incompatability. IV. The

development of tensile strength in a compatible and an incompatible graft. Amer. J.
Bot. 70:226-231.

- . 1983b. Physiological aspects of graft formation. Pp. 89-105, in Vegetative

compatability responses in plants (R. Moore, ed.), Baylor Univ. Press, Waco, Texas.

- . 1984a. A model for graft compatability-incompatability in higher plants. Amer.

J. Bot. 71:752-758.

- . 1984b. The role of direct cellular contact in the formation of compatible

autografts in Sedum telephoides. Ann. Bot. 54:127-133.

Moore, R., and D. B. Walker. 1981a. Studies of vegetative plant tissue compatability-
incompatability. I. A structural study of a compatible autograft in Sedum telephoides
(Crassulaceae). Amer. J. Bot. 68:820-830.

- . 1981b. Studies of vegetative plant tissue compatability-incompatability. II. A

structural study of an incompatible heterograft between Sedum telephoides
(Crassulaceae) and Solanum pennellii (Solanaceae). Amer. J. Bot. 68:831-842.

- . 1981c. Studies of vegetative plant tissue compatability-incompatability. III. The

involvement of acid phosphatase in the lethal cellular senescence in an incompatible
heterograft. Protoplasma 109:317-334.

- — . 1983. Studies of vegetative plant tissue compatability-incompatability. VI.

Grafting of Sedum and Solanum callus tissue in vitro. Protoplasma 115:114-121.

Muzik, T. J. 1958. Role of parenchyma cells in graft union in Vanilla orchid. Science
127:82.

Parkinson, M., and M. M. Yeoman. 1982. Graft formation in cultured, explanted
internodes. New Phytol. 91:711-719.

Shimomura, T., and K. Fujihara. 1976. Histological observations on graft union
formation in cactus. J. Japan. Soc. Hort. Sci. 44:402-408.

- — . 1977. Physiological study of graft union formation in cactus. II. Role of auxin

on vascular connection between stock and scion. J. Japan. Soc. Hort. Sci. 45:397-406.

- . 1978. Prevention of auxin-induced vascular differentiation in wound callus by

surface-to-surface adhesion between callus of stock and scion in cactus grafts. Plant
Cell Physiol. 19:877-886.

Snedecor, G. W. , and W. G. Cochran. 1976. Statistical methods, 6th edition. Iowa State
Univ. Press, Ames, Iowa, 569 pp.

Stoddard, F. L., and M. E. McCully. 1979. Histology of the development of the graft
union in pea roots. Canadian J. Bot. 57:1486-1501.

- . 1980. Effects of excision of stock and scion organs on the formation of the graft

union in Coleus: a histological study. Bot. Gaz. 141:401-412.

Yeoman, M. M., and R. Brown. 1976. Implications of the formation of the graft union
for organisation in the intact plant. Ann. Bot. 40:1265-1276.

MARINE FUNGI ASSOCIATED WITH SPARTINA
FROM HARBOR ISLAND, TEXAS

by ROBERT D. KOEHN

Department of Biology
Southwest Texas State University
San Marcos, Texas 78666

ABSTRACT

Thirteen species of marine fungi have been identified from Spartina alterniflora in bay
areas near Port Aransas, Texas. All but two of the species are Ascomycetes, with
Leptosphaeria species being the most common.

INTRODUCTION

Spartina alterniflora Loisel., coastal cordgrass, is a dominant
halophyte of the bays and estuaries along the Texas Gulf Coast. It is
an important producer in the coastal ecosystem and its biomass
becomes largely recycled in the estuarine habitats by decomposition
activities of marine heterotrophs. Kohlmeyer and Kohlmeyer (1979)
have assembled a composite listing of marine fungi known to be
saprophytic or parasitic on Spartina species. The most comprehensive
studies of the marine mycoflora associated with 5. alterniflora have
been made on the upper Atlantic Coast (Gessner and Goos 1973;
Gessner 1977). No such studies have been conducted along the Texas
Gulf Coast and, therefore, this study was done to document a
mycoflora associated with S. alterniflora collected from estuarine
habitats near Port Aransas, Texas.

The study covered a period of six years, 1977 to 1983. Monthly
collections were made starting September 1977 and continuing through
May 1980 from the marshes bordering Harbor Island along the
causeway leading from Aransas Pass to Port Aransas, Texas, and from
sites near the Aransas Lighthouse. Plants were uprooted, placed into
plastic bags and transported to a laboratory where direct microscopic
observations were made. Both dead and living material was collected.
Similar collections were conducted each October from 1980 through
1983.

The following fungal organisms were identified from S. alterniflora.

Ascomycetes:

Leptosphaeria albopunctata (West.) Sacc.

L. halima Johnson

L. discors Sacc. et Ellis

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

214

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

L. marina Ellis et Everh.

L. neomaritima Gessner et Kohlm.

L. oraemaris Linder

L. pelagica E.B.G. Jones

Lulworthia medussa (Ellis et Everh.) Cribb et Cribb.

Lindra inflata Wilson

Pleospora sp.

Sphaerulina pedicellata Johnson
Coelomycetes:

Didymella sp.

Septoria sp.

Fungi identified from S. alterniflora collected at the coast of Texas
are mostly ascomycetes that represent five genera. Marine organisms
were identified mainly from senescent leaves and culms, which had
been in contact with estuarine waters. The greatest species richness
occurred in the autumn and winter; more specifically during the
months of October, November, December, and January when S.
alterniflora plants are dying back. There was no correlation between
ambient and water temperature, which were regularly taken during the
first three years of this study, and fungal attack on weakened, chlorotic
plant tissues.

Leptosphaeria species seem to be the most abundantly associated
with Spartina. The taxa listed here agree well with the lists compiled
by Jones (1976) and Gessner (1977). The diversity of Leptosphaeria
species is greater in Texas than in the Rhode Island area studied by
Gessner (1977). The imperfect genera observed here, however, are not
previously listed. Gessner and Seiburth (1972) have shown that
Sphaerulina pedicillata is the first fungus to colonize S. alterniflora
plants and is then followed by species of Leptosphaeria, Pleospora,
and Lulworthia. He believed that other microheterotrophs are
supported by this fungal biomass. Because a similar mycoflora exists
in Texas, this succession probably occurs also on the Texas Gulf
Coast, where these fungi probably initiate the recycling of herbaceous
materials in estuarine habitats.

LITERATURE CITED

Gessner, R. V. 1977. Seasonal occurrence and distribution of fungi associated with
Spartina alterniflora from a Rhode Island estuary. Mycologia 69:477-491.

Gessner, R. V., and R. D. Goos. 1973. Fungi from Spartina alterniflora in Rhode Island.
Mycologia 65:1296-1301.

Gessner, R. V., and J. M. N. Sieburth. 1972. The fungal microcosm of the internodes
of Spartina alterniflora. Marine Biol. 16:269-273.

Jones, E. B. G. 1976. Lignicolous and algicolous fungi. Pp. 1-49, in Recent advances
in aquatic mycology (E. B. G. Jones, ed.), John Wiley and Sons, New York.

Kohlmeyer, J., and E. Kohlmeyer. 1979. Fungi in halophytes of tidal salt marshes. Pp.
79-91, in Marine mycology, the higher fungi, Academic Press, New York

VERTEBRATE USE OF NONTIDAL WETLANDS
ON GALVESTON ISLAND, TEXAS

by ALLAN J. MUELLER

U.S. Fish and Wildlife Service
17629 El Camino Real, Suite 211
Houston, Texas 77058

ABSTRACT

The nontidal wetlands of Galveston Island, Texas, depend on local rainfall for
freshwater, and many dry out during summer. Evaporation and inundation by storm
tides cause salinities to rise; they decline when heavy rains flush out the saltwater.
Aquatic emergents are the dominant vegetation. Nontidal marshes provide important
habitat for many kinds of wildlife, especially birds. In a comparison of two wetlands,
one natural and the other man-made, the natural area received equal or greater use by
all aquatic bird groups except the black-crowned night heron ( Nycticorax nycticorax )
and American coot ( Fulica americana ). Nontidal wetlands are the only available habitat
on Galveston Island for many amphibians and reptiles. Key words : wetlands, barrier
islands, wildlife, Galveston Island.

INTRODUCTION

Galveston Island is a 51 -kilometer-long barrier island on the Texas
coast. Most of the developed east end of the island (60,000 population)
was raised in elevation with fill material after several disastrous
hurricanes in the early 1900s. The west part of the island is
undergoing second-home development, but much of the area still is
devoted to cattle grazing.

Galveston Island has extensive salt marshes and flats that are subject
to regular tidal inundation. However, nontidal wetlands, ranging from
fresh to mesohaline, also are found throughout the island. Although
referred to as nontidal, these wetlands are subject to saltwater flooding
by storm tides. In comparison with salt marshes, nontidal wetlands
have received little attention from the conservation community.
Through an examination of hydrologic and biological characteristics,
this paper describes the wildlife habitat values of natural and man¬
made nontidal wetlands on the island.

One of the characteristic features of the Texas barrier islands and
peninsulae is a series of beach ridges parallel to the existing shoreline.
The ridges have elevations averaging about 1.6 meters above sea level
(Fischer et al. 1972; LeBlanc and Hodgson 1973). The intervening
swales, isolated from regular tidal effect, support most of the
freshwater marshes on Galveston Island. The swales are especially well

The Texas Journal of Science, Vol. XXXVII, Nos. 2&3, September 1985

216

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NOS. 2 & 3, 1985

developed on the central part of the island, where a single swale may
extend for several kilometers; however, most swales frequently are
interrupted by tidal inlets and man-made developments. Swales range
from seven to 25 meters wide with occasional ponds along them.
Bottom elevations vary from near sea level to two meters above sea
level. Other nontidal marshes formed by geologic processes include
isolated remnants of tidal creeks and ponds cut off from tidal action
at the head of inlets.

Wetlands created or modified by man are a second category of
nontidal wetlands. Borrow pits commonly become nontidal wetlands.
Many natural, nontidal wetlands have been modified by highways,
railroads, oil and gas operations, canals, and housing developments.
Modifications sometimes cause the wetlands to become tidal or simply
to be filled and totally lost as a wetland. In other cases, nontidal
wetlands have been enlarged, deepened, partially filled, diked, or had
their drainage patterns changed.

METHODS

Periodic salinity measurements were made with an optical
refractometer at three natural, nontidal wetlands — Long Swale, Deep
Pond, and S Pond (Fig. 1) — and at five man-made, nontidal
wetlands — Pruitt Pond, Triangle Pond, Powerline Pond, Payco Pond,
and Newspaper Marsh (Fig. 2). Salinity measurements were compared
with rainfall and tidal flooding to find a pattern for changes. Two
areas were censused repeatedly for wildlife use. Pruitt Pond, a 2.4
hectare man-made wetland, was censused 237 times from 1977 to 1983;
two natural swales, Long Swale and Deep Pond, were combined as a
2.4-hectare unit and censused 48 times from 1979 to 1983. On each
census, visual observations were made as the perimeter of the area was
walked and shallow areas waded.

The area around Pruitt Pond was sampled for small mammals using
three sizes of snap traps. The traps were placed in runs and other
likely locations and baited with a mixture of uncooked oatmeal,
peanut butter, and bacon grease. The traps were set for 139 trap nights
in January and February 1982.

Vegetation was described from repeated qualitative observations of
the study sites. Many wetlands, in addition to the specific study sites,
were visited and observations made of vegetation, salinity, and wildlife
use.

RESULTS AND DISCUSSION

Hydrology

Nontidal wetlands are flooded with salt water only by storm tides.
A hurricane surge of three meters, sufficient to flood all the wetlands,

VERTEBRATES OF NONTIDAL WETLANDS

217

can be expected four times in every 100 years on Galveston Island
(Bodine 1969). Smaller storms are more frequent, but they flood only
some of the nontidal wetlands.

Most of the time nontidal wetlands receive water only in the form
of rainfall and local runoff. The watersheds of my study areas ranged
from approximately one to seven hectares. The mean annual
precipitation on Galveston Island is 111.4 centimeters, but extended
dry periods are not unusual. With a mean annual temperature of 21.0°
C (NOAA 1980), Galveston Island has high evaporation rates, and
many of its nontidal wetlands dry up without regular rainfall.

Mustang and Mustang Variant are sandy soils that occur in Long
Swale, Deep Pond, and other natural swales. These soils are poorly
and exceedingly poorly drained, respectively, with rapid permeability
above the water table (Crenwelge et al. 1979). This indicates a free
exchange with the water table, whereby the swale contributes to the
water table during wet periods and draws from it during dry periods.
In Pruitt Pond, the soil is Mustang fine sand saline, which has

218

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

characteristics similar to the above-mentioned soils. But, as in many
man-made wetlands, the soil of Pruitt Pond has been disturbed by
excavation. Water-level fluctuations are generally more extreme in the
man-made than in the naturally occuring wetlands.

Salinity

Rainfall and tidal flooding are the dominant factors governing
salinity. In August 1980, Hurricane Allen, a minor storm, caused tides
one to two meters above normal on Galveston Island. All of the man¬
made marshes in which salinity measurements were made, except
Power Line Pond, were tidally flooded. In the four marshes that were
flooded, salinities were relatively high, 8-20 parts per thousand (Fig.
3), when the first salinity measurements were made on 29 November
1980. The trend towards higher salinities as measured on 17 May 1981
reflects the low rainfall (3.5 cm) and high evaporation (30.0 cm) in the
two months after the previous measurement. Normal rainfall during
this period is approximately 13.8 cm (NOAA 1980). Heavy rains in late
May and June (32.0 cm) flooded the man-made marshes and flushed

VERTEBRATES OF NONTIDAL WETLANDS

219

Figure 3. Salinities in Galveston Island nontidal wetlands’ November 1980-February
1982.

out the salt brought in by the storm tides associated with Hurricane
Allen.

The natural swales and Power Line Pond were not flooded by
Hurricane Allen, and salinities remained uniformly low (Fig. 3). A
major storm that would tidally inundate the natural swales could raise
salinities in them for a long period. Because the natural, nontidal
wetlands tend to be better isolated from tidal action than the man¬
made wetlands, they also are more resistant to flushing with rain
water. After a major inundation of Matagorda Island, it takes several
years for salt poisoning to decline in cattle that drink from nontidal
ponds (University of Texas 1973).

Crenwelge et al. (1979) found swale salinities low in the winter (no
value given) and high in the summer (19 ppt). Adjacent to some
swales, they also identified a narrow strip of high soil salinity (96 ppt)
caused by evaporation from the water table, leaving salt at the surface.

Vegetation

Aquatic emergents are the dominant vegetation in most nontidal
wetlands. The most common species are common cattail ( Typha
latifolia ), softstem bulrush ( Scirpus validus), needle rush (J uncus
roemerianus) , and marshy cordgrass ( Spartina patens). These species
usually occur in dense stands, but occasional interspersed pools of
open water greatly improve their value as wildlife habitat (Golet and
Larson 1974).

220

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Table 1. Amphibians, reptiles, and mammals recorded from nontidal wetlands on
Galveston Island. M = this study; E = Espey, Huston and Associates, Inc. (1979).

Cricket frog

Amphibians

Acris crepitans

M,E

Green treefrog

Hyla cinerea

M

Squirrel treefrog

H. squirella

M,E

Gulf coast toad

Bufo valliceps

M,E

Bullfrog

Rana catesbeiana

M,E

Southern leopard frog

R. sphenocephala

M,E

Eastern narrow-mouthed toad

Gastrophryne carolinensis

M

Great plains narrow-mouthed toad

G. olivacea

E

Common snapping turtle

Reptiles

Chelydra serpentina

M

Mississippi mud turtle

Kinosternon subrubrum

M,E

Red-eared turtle

Chrysemys scripta

M,E

American alligator

Alligator mississippiensis

M

Gulf salt marsh snake

Nerodia fasciata

M,E

Western ribbon snake

T hamnophis proximus

M,E

Western cottonmouth

Agkistrodon piscivorus

M,E

Virginia opossum

Mammals

Didelphis virginiana

M,E

Northern rice rat

Oryzomys palustris

E

Hispid cotton rat

Sigmodon hispidus

M,E

Roof rat

Rattus rattus

M

Nutria

Myocaster coy pus

M,E

Eastern cottontail

Sylvilagus floridanus

M

Woody vegetation is generally restricted to drier sites. Rattlebush
( Sesbania drummondii) commonly grows around the edges of swales,
and on drier sites it is the dominant species. French tamarisk (Tamarix
gallica ) occurs in monotypic stands on semi-permanently flooded to
upland sites.

In general, the man-made nontidal wetlands on Galveston Island
have more open water and less vegetation than the naturally occurring
marshes. Pruitt Pond was mostly open water (70 percent) with a
French tamarisk hedge running across its center and a fringe of
aquatic emergents. Both Long Swale and Deep Pond had
approximately 40 percent open water, with the remainder a dense
stand of softstem bulrush and rattlebush.

Wildlife

Nontidal wetlands are an important habitat for many wildlife
species. The fresher water of the nontidal marshes provides habitat not
available in the more abundant salt marshes.

Eight species of frogs and toads were found on Galveston Island
(Table 1). Although the green treefrog ( Hyla cinerea), Gulf coast toad
( Bufo valliceps ), and the southern leopard frog ( Rana sphenocephala)

VERTEBRATES OF NONTIDAL WETLANDS

221

occasionally use brackish marshes (Conant 1975), all require freshwater
habitats for reproduction. In the spring of 1980, Pruitt Pond had a
breeding chorus of green treefrogs, Gulf coast toads, and narrow¬
mouthed toads ( Gastrophryne carolinensis). Since flooding by
Hurricane Allen in August 1980, only the Gulf coast toads remain.

Seven species of reptiles used the nontidal marshes of Galveston
Island (Table 1). Of these, the red-eared turtle ( Chrysemys scripta) and
the western ribbon snake ( Thamnophis proximus) are restricted to
freshwater habitats, and only the Gulf salt marsh snake ( Nerodia
fasciata) prefers brackish areas (Conant 1975). The American alligator
(. Alligator mississippiensis), although common in the early 1900s
(personal communication from B. Sinclair, 3707 11 Mile Road,
Galveston, Texas), is now rare on Galveston Island.

Six species of mammals were recorded in the nontidal marshes of
Galveston Island (Table 1). The nutria ( Myocaster coy pus), which is
the only semiaquatic species, was common in both natural and man¬
made wetlands. The white-tailed deer ( Odocoileus virginianus ) uses
nontidal wetlands on other Texas barrier islands (University of Texas
1973), but has been extirpated from Galveston Island. Although the
raccoon ( Procyon lotor) is mentioned in other studies (Espey, Huston
and Associates Inc. 1979), I found no evidence that it occurred on
Galveston Island.

The small-mammal trapping resulted in the capture of seven hispid
cotton rats (Sigmodon hispidus ) and one roof rat ( Rattus rattus).
Espey, Huston and Associates Inc. (1979), in 180 trap nights in
nontidal marshes near Long Swale, reported the capture of 12 hispid
cotton rats and 12 northern rice rats ( Oryzomys palustris).

I recorded 234 species of birds using the nontidal wetlands of
Galveston Island. Observations indicated that several aquatic species
use the nontidal wetlands more than the salt marshes. These include
the little blue heron ( Egretta caerulea), white-faced ibis (Plegadis
chihi), mottled duck ( Anas fulvigula), common moorhen ( Gallinula
chloropus), purple gallinule ( Porphyrula martinica), American coot
( Fulica americana), and common snipe ( Gallinago gallinago).

In general Long Swale and Deep Pond — which had dense vegetation
that made them difficult to accurately and completely cenus — had
greater use than Pruitt Pond (Table 2). However, the black-crowned
night heron ( Nycticorax nycticorax) and American coot clearly
preferred Pruitt Pond (Table 3). The Fre h tamarisk hedge with open
water on both sides appeared to be an important feature to both
species. Black-crowned night herons used the hedge as a day roost, and
American coots made heavy use of the hedge as an easy access to
adjacent food and cover.

222

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 2. Summary of aquatic bird censuses, 1977-83*.

Pruitt Pond (man-made)

Month

Jan

Feb

Mar

Apr May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

All

Number of

censuses

17

21

20

21

29

17

24

17

19

24

15

13

237

(total)

Species/

census

(average)

5.5

4.3

6.6

8.9

8.7

5.9

5.2

5.1

4.4

4.6

5.9

5.6

6.0

(ave.)

Individuals/

census

(average)

56

38

44

Long

32 23 15 12 10 8

Swale and Deep Pond (natural)

14

29

46

26

(ave.)

Month

Jan

Feb

Mar

Apr May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

All

Number of

censuses

2

3

8

13

8

2

2

1

1

4

2

2

48

(total)

Species/

census

(average)

5.5

10.7

10.8

13.6

11.8

15.0

8.5

13.0

18.0

7.0

8.0

6.0

11.1

(ave.)

Individuals/

census

(average)

10

45

29

46

32

64

20

22

102

13

20

26

35

(ave.)

*Aquatic birds defined as pied-billed grebe ( Podilymbus podiceps ), cormorants
( Phalacrocorax sp.), waders (Ciconiiformes), waterfowl (Anatidae), rails (Rallidae),
shorebirds (Charadriiformes), belted kingfisher ( Ceryle alcyon), “marsh” wrens
( Cistothorus sp.), swamp sparrow ( Melospiza georgiana), and seaside sparrow
(. Ammodramus maritimus ).

All other species or groups made equal or greater use of the natural
swale census area (Table 3). Migratory waterfowl used both areas
equally, but the resident mottled duck made more use of Long Swale
and Deep Pond. Long-legged waders and species normally found in
dense vegetation (bitterns, rails, gallinules, and marsh wrens) also
preferred Long Swale and Deep Pond.

Seasonal use was evident for some species (Table 3). The pied-billed
grebe ( Podilymbus podiceps) is a permanent resident (Mueller 1981),
but was more common at Pruitt Pond in the November-April winter
period. The black-crowned night heron is also a permanent resident
(Mueller 1981), but use of the Pruitt Pond day roost was greatly
reduced during the breeding season. Other seasonal-use patterns either
were not detectable or were related to the migratory habits of the
species. Shorebird use was greatly affected by water level in the ponds,
but small spring and fall peaks still were observed.

I found 12 bird species nesting in Galveston’s nontidal marshes.
Pied-billed grebe, mottled duck, clapper rail ( Rallus longirostris),
common moorhen, and willet (Catoptrophorus semipalmatus) were the
most common aquatic species. A few blue-winged teal ( Anas discors)
nest in similar habitats on other Texas barrier islands (University of

VERTEBRATES OF NONTIDAL WETLANDS

223

Table 3. Aquatic bird censuses3 by taxonomic and habitat-use groups.

Long Swale and Deep Pond
Pruitt Pond (man-made) (natural)

Individuals/ Individuals/

Species/Group Period Species %frequencyb census, ave. Species %frequency census, ave.

Use Greater at Long Swale and Deep Pond

Bitterns

All year

1

4

0.04

2

31

0.4

Botaurini

Herons and egrets

All year

8

67

1.6

8

88

6.5

Ardeini

Ibis and Spoonbill

All year

3

12

0.3

3

73

5.3

Threskiornithidae

Mottled Duck

All year

1

11

0.2

1

65

1.6

Anas fulvigula

Other rallids

All year

6

34

0.7

5

71

3.1

Rallidae

Wetland passerines

All year

3

9

0.2

4

71

3.0

Cistothorus sp.,
Ammodramus maritimus ,
Melospiza georgiana

Use Approx:

imately Equal

Pied-billed Grebe

All year

1

32

0.6

1

48

1.1

Podilymbus podiceps

Nov-Apr

1

62

1.4

1

50

1.3

Other waterfowl

All year

8

43

3.6

8

50

3.5

Aix sponsa, Anas sp,
Aythya sp.

Nov-Apr

7

68

6.6

7

60

5.2

Shorebirds

All year

23

56

3.3

14

48

3.0

Charadriidae

Apr-May

17

78

6.6

10

57

4.2

Recurvirostridae

Aug

8

82

3.7

5

100

10.0

Scolopacidae

Gulls, terns and

All year

11

44

2.0

7

38

2.0

skimmer

Laridae

Use Greater at Pruitt Pond

Night herons

All year

2

42

4.2

2

23

0.4

Nycticaracini

Nov-Feb

1

47

10.9

0

0

0

American Coot

All year

1

68

8.9

1

48

2.8

Fulica americana

Nov-Apr

1

93

17.6

1

60

4.1

“Number of censuses for Pruitt Pond: all year, 237; Nov-Apr,

107; Nov-Feb. 66; Apr-Aug,

108; Apr-May, 50; Aug, 17. Number of censuses

for Long Swale and Deep Pond: all

year, 48; Nov-Apr, 30; Nov-Feb, 9; Apr-Aug, 26; Apr-May, 21; Aug.

1.

% Frequency = number of

censuses

when

at least

one species of

group occurred/total

number of censuses X 100.

Texas 1973), but no recent evidence of its nesting on Galveston Island
was found.

CONCLUSIONS

Man has made a great impact on the nontidal wetlands of Galveston
Island. Many have been filled, deepened, diked, or connected to

224

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

regular tidal action. Still, the remaining marshes — especially the
relatively undisturbed areas — continue to be heavily used by birds and
are the only available habitat on the island for many amphibians and
reptiles.

Several efforts to create new nontidal wetlands on Galveston Island
are underway, but it is too early to judge their success or failure.
Parnell et al. (1978) found that diked spoil disposal areas created new
brackish or freshwater habitats that were heavily used by birds.
However, these areas were subject to rapid drainage when the spoil
material dried and cracked (for example, the Fort Point spoil disposal
area on Galveston Island). In the future it may be possible to design
spoil disposal areas to provide freshwater marsh habitat with greater
permanence and, therefore, of higher quality to wildlife.

The nontidal wetlands of Galveston and other Texas barrier islands
are a valuable wildlife habitat; they deserve more management
attention and protection than they have received in the past.

ACKNOWLEDGMENTS

Dr. James Webb and Bruce Halstead reviewed early versions of this
manuscript. The Mitchell Development Corporation of the Southwest
allowed access to the Long Swale, Deep Pond, and S Pond study sites.
Their help is greatly appreciated.

LITERATURE CITED

Bodine, B. R. 1969. Hurricane frequency estimated for the Gulf coast of Texas. U.S.

Army Corps of Engineers Coastal Engineering Research Center, Tech. Mem. 26:1-32.
Conant, R. 1975. A field guide to reptiles and amphibians of eastern and central North
America. Houghton Mifflin Co., Boston, 429 pp.

Crenwelge, G. W., J. K. Baker, and E. L. Griffin. 1979. Soil survey of Galveston Island.
USD A, Soil Conservation Service, 124 pp.

Espey, Huston and Associates, Inc. 1979. Environmental issues considered for the
development of Pirates Cove section 6. Austin, Texas, Document no. 79205, 125 pp.
Fisher, W. L., J. H. McGowen, L. F. Brown, Jr., and C. G. Groat. 1972. Environmental
geologic atlas of the Texas coastal zone-Galveston/ Houston area. Bureau Econ.
Geol., Univ. Texas, Austin, 91 pp.

Golet, F. C., and J. S. Larson. 1974. Classification of freshwater wetlands in the glaciated
northeast. U.S. Fish and Wildlife Serv. Resource Publ. 116:1-56.

LeBlanc, R. J., and W. D. Hodgson. 1973. Origin and development of the Texas
shoreline. Pp. 197-220, in Barrier islands (M. L. Schwartz, ed.), Dowden, Hutchinson
& Ross, Inc., Stroudsburg, Pennsylvania.

Mueller, A. J. 1981. A checklist to the birds of Galveston. Science Incorporated,
Galveston, Texas, 15 pp.

NOAA (National Oceanic and Atmospheric Administration). 1980. Local climatological
data annual summary with comparative data 1980 Galveston, Texas. USDC,
(Ashville, North Carolina) 4 pp.

VERTEBRATES OF NONTIDAL WETLANDS

225

Parnell, J. F., D. M. DuMond, and R. N. Needham. 1978. A comparison of plant
succession and bird utilization on diked and undiked dredged material islands in
North Carolina estuaries. U.S. Army Engineer Waterways Experiment Station,
Technical Rept. D-78-9, 113 pp.

University of Texas. 1973. Matagorda Island, a natural area survey. Office of Research,
Lyndon B. Johnson School of Public Affairs, Austin, Texas, Part II of IV, 237 pp.

CORRELATION BETWEEN SUSPENDED SEDIMENT
AND OTHER WATER QUALITY PARAMETERS
IN SMALL STREAMS OF FORESTED EAST TEXAS

by ALFREDO B. GRANILLO

Department of Forestry
University of Arkansas at Monticello
Monticello , Arkansas 71665

MINGTEH CHANG

School of Forestry

Stephen F. Austin State University

Nacogdoches, Texas 75962

and EDWARD B. RASHIN

Radian Corporation
P.O. Box 9948
Austin, Texas 78766

ABSTRACT

Sediment load and its relationship with four other water quality parameters were
studied in streams of eight mostly forested watersheds (11-72 square kilometers) in East
Texas between 1977 and 1980. The high correlations of total suspended sediment with
Kjeldahl nitrogen, total phosphorus, and chloride suggest that total suspended sediment
is a reasonable indicator of stream water quality in the study area. Because of gage
density and storm characteristics, streamflow seems to be more useful than rainfall in
predicting stream sediment. Key words : sediment, water quality, streams, Texas.

INTRODUCTION

Flowing water in a stream always carries sediments and other
materials reflecting the geology and land use of the watershed.
Accelerated movement of sediment from the land surface into streams
not only depletes watershed productivity, but also reduces capacity of
any downstream reservoir, impairs water utilization, affects aquatic
life, and usually degrades recreational value of the watershed. Sediment
is a major pollutant of streams; a concentration of 80 parts per million
or more exceeds the Environmental Protection Agency’s (1976)
standard for freshwater aquatic life.

The chemical composition of stream sediment, although closely
resembling the parent material, may be affected by the environment
during transit to the stream. Many substances such as nutrients,

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

228

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

pesticides, heavy metals, and radioactive isotopes may absorb to
sediment in water or leave the watershed with sediment. Many studies
have shown consistent relationships between sediment load and other
water pollutants in runoff from urban, suburban, and agricultural
watersheds (Romkens et al. 1973; Schumann et al. 1973; Ritchie et al.
1975; Kissel et al. 1976). This suggests that sediment load might serve
as an indicator of water quality (Donigian and Crawford 1977), and
could possibly be used to predict runoff water quality when used with
an appropriate multiplier such as potency factor (a ratio of pollutant
to sediment load).

Chang et al. (1983) have shown that in-stream concentrations of
suspended sediment and four dissolved solids are highly correlated
with watershed land use and topography in eight small forested
watersheds in East Texas. In this report, their data are used to
reexamine the relationships between suspended sediment and the four
dissolved solids (nitrate and nitrite nitrogen, Kjeldahl nitrogen, total
phosphorus, and chloride) in the East Texas streams. Because the
previous study was conducted using water samples collected only once
or twice a month, with most of the collections made during low-flow
(runoff recession) periods, fluctuation of sediment concentration was
closely examined using 20 daily observations made in one of the eight
streams (Alazan Watershed).

PROCEDURES

The eight watersheds studied, ranging in size from 11.4 to 71.6
square kilometers, are located in Nacogdoches County, Texas, about
300 kilometers southeast of Dallas and 200 kilometers northeast of
Houston. They are upper tributaries of the Angelina River, which
flows southeasterly into the Neches River and ultimately into the Gulf
of Mexico. Normal annual precipitation and temperature (1941-1970)
of Nacogdoches are 1207 mm and 18.7° C, respectively; most
precipitation occurs in the winter and spring (December-May). The
annual evapotranspiration/ precipitation ratio is about 0.78 (Chang
and Aguilar 1980).

Nacogdoches County is in the Coastal Plains region of Texas, with
Carrizo Sand and Sparta Sand Formations in the north and south,
respectively, and with the Weches Formation covering a small portion
of White Oak and Moss creeks between them. Loblolly pine ( Pinus
taeda), long leaf pine (P. palustris), and short-leaf pine (P. echinata)
are the dominant trees in the area. They are of secondary growth
mixed with scattered hardwood species. Forest cover of the eight
watersheds ranges from 98 to 65 percent with an average of 77 percent.
The majority of the remaining area is used as pasture and for other
agricultural purposes.

STREAM WATER QUALITY IN EAST TEXAS

229

Data on the five water quality parameters— total suspended sediment
(TS), nitrate and nitrite nitrogen (NN), Kjeldahl nitrogen (KN), total
phosphorus (PO4), and chloride (CL) — of the eight small forested
watersheds used in the analysis were previously published by Chang
et al. (1983). Total suspended sediment in the water samples was
filtered using Millipore glass microfiber pre-filters (0.45 microns) and
determined gravimetrically with a Mettler Analytical Balance, model
H-10 (Amer. Public Health Assoc. 1976). The cadmium reduction
colorimetric procedure and the ammonium molybdate procedure
(Amer. Public Health Assoc. 1976) were employed to determine nitrate-
nitrite nitrogen and orthophosphate (POT3) concentrations,
respectively, in water samples. Optical density at each wave-length
specified in the colorimetric procedures was determined with a Bausch
and Lomb spectrophotometer (Model 70). Total Kjeldahl nitrogen was
determined with a Labconco micro-Kjeldahl procedure (Environmental
Protection Agency 1971). Chloride concentrations were determined
titrimetrically using the mercuric nitrate method (Amer. Public Health
Assoc. 1976). All water samples were filtered with Whatman and
Millipore glass microfiber filters before being analyzed for dissolved
solids. These data were weighted to kg*ha_1*yr_1 from 31 composite
water samples collected during a four-year period (1977-1980). They
were employed to develop relationships, through correlation and
regression analysis, between suspended sediment and the other four
water quality parameters in the East Texas area.

Also, suspended sediment (via US DH-48 integrated sediment
sampler) and measurement of streamflow (via Gurley current meter)
were done daily at two stations in the Alazan Watershed (Figure 1)
during the period between 5 and 24 July 1979. The distance between
downstream and upstream sites is about 5.8 kilometers. There was a
recording rain gauge located in the watershed about 1.5 and 4.5
kilometers, respectively, from the downstream and upstream sites. The
20 daily observations, made between 7-8 PM, were used to examine the
fluctuation of sediment concentration during high- and low-flow
periods.

RESULTS AND DISCUSSION

Sediment Versus Other Water Quality Parameters

Total sediment yield (TS) was significantly correlated with KN (total
Kjeldahl nitrogen, r = 0.863; P = 0.003), PO4 (total phosphorus, r =
0.881; P = 0.002), and CL (chloride, r = 0.751; P = 0.016), indicating
a strong association of stream sediment with dissolved solids in the
watersheds studied. Close relationships among total Kjeldahl nitrogen
(KN), total phosphorus (PO4), and sediment concentration have been

230

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Figure 1. The Alazan Creek Watershed, southwest of Nacogdoches, Texas.

reported also by Colston (1974), Donigian and Crawford (1977), and
Zison (1980).

In modeling, the following linear functions are frequently
employed:

Nt = b(TS) (1)

or

Nt = a + b (TS) (2)

where Nt is the mass of nutrient washing off the watershed, TS is the
mass of suspended sediment, and a and b are the intercept and the

STREAM WATER QUALITY IN EAST TEXAS

231

Table I. — Parameters and statistic of equation 3 for estimating Kjeldahl nitrogen (KN),
total phosphorus (PO4), and chloride (CL).

Nt, kg’ha ‘‘yr 1

A

B

R2

SE, %

KN

1.296

0.635

0.88

18.58

PO4

0.943

0.949

0.87

25.18

CL

1.996

0.716

0.72

30.49

Note: R2, coefficient of determination; SE, standard error of estimates.

slope (or so-called potency factor), respectively, of the regression line.
Our analysis showed that the predictability of Nt could be greatly
improved by the following power function:

N, = A (TS)b (3)

where A and B are constants. Values of A and B for estimating KN,
PO4, and CL in kg'ha^-yr-1 along with coefficients of determination
are given in Table 1. The latter indicate that about 88, 87, and 72
percent of the variation in TK, PO4, and CL, respectively, can be
estimated from TS alone. The standard error of estimate for these three
dissolved solids are, in the same order, about 19, 25, and 31 percent
of the observed means.

Inasmuch as the values of B for all nutrients are less than unity,
Equation 3 shows tha Nt yields changed at a greater rate for lower
than for higher TS values. This is probably due to the association of
TS with surface runoff, which dilutes nutrient concentration in the
stream.

Sediment Fluctuation

The fluctuations of sediment concentration along with streamflow
and precipitation for the 20 consecutive daily observations in July 1979
at the upstream and downstream sites on Alazan Creek are plotted in
Figure 2. Both sediment concentration and streamflow were greater at
the downstream station. Suspended sediment varied roughly with
streamflow, and less closely, with precipitation at the Alazan rain
gauge. A one-hour storm of 9.1 mm occurring around 2:00 PM on July
18, increased the sediment load (determined at 8:30 PM) from 5.99
milligrams per liter on the previous day to 19.8 mg/1 at the upstream
station and from 9.6 to 92.6 mg/1 at the downstream station; the latter
was the only observation we made that exceeded the water quality
standard (that is, 80 mg/1) for freshwater aquatic life (Environmental
Protection Agency 1976). Another two-hour storm of much greater size
(27.9 mm) about 10 AM on July 13, 1979, increased sediment from 7.7
mg /I on the previous day to only 14.1 mg /I at the up-stream and
from 27.4 to 56.5 mg /I at the down-stream station. The erratic

232 THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Figure 2. Sediment concentration, streamflow, and precipitation on 20 consecutive days

in Alazan watershed, near Nacogdoches, Texas.

relationship between rainfall and sediment quantities reflect in part
inconsistencies in time between measurements. It also results from the
inadequacy of single-point samples of rainfall to measure
thunderstorm-type rainfall over a watershed. Because spacing of rain
gauges in the United States is about 1.39 gages/ 1000 km2 (Chang
1981), and the most frequently observed and mean diameters of
thunderstorms are about three and eight kilometers, respectively
(Brooks 1946), entire storms on small watersheds may not be recorded
by our present gage network.

In contrast to rainfall, streamflow measurements, especially
continuously recorded, afford an integrated evaluation of watershed

STREAM WATER QUALITY IN EAST TEXAS

233

Table 2. — Some watershed topographic characteristics for Alazan Watershed,
Nacogdoches, Texas.

Topographic Paramter

Upstream

Downstream

Area (km2)

11.0

24.27

Watershed slope factor (length/ width)

1.45

2.34

Main channel length (km)

6.08

11.09

Total channel length (km)

10.18

25.78

Drainage density (total channel length /area)

1.45

1.72

Watershed relief (m)

60.35

83.82

runoff, and a much more effective parameter for estimating stream
sediment. Simple correlation analysis of the 20 daily observations show
that sediment load had higher correlation coefficients (r) with
streamflow than with rainfall at both upstream and down-stream
stations:

Rainfall Streamflow

Upstream sediment 0.311 = r 0.667

Downstream sediment 0.572 0.813

When only the rainy-day data were used in the analysis, r-values were
greater than 0.96 for the streamflow and less than 0.16 for precipitation
at the two stations. The poor correlation for precipitation reflects time
lag between rainfall and runoff.

The greater streamflow and sediment concentrations observed at the
downstream site may be attributed partly to the greater watershed area
and partly to the channel characteristics (Table 2). The drainage
density (total channel length /area) is greater at the downstream
watershed, due to the confluence of two separate stream segments to
form the main channel below the upstream measuring site (Figure 1).
The flows from the upstream watershed coupled with the water from
the two additional stream segments converge into a single channel
through a bottle-neck-like topography before reaching the downstream
site. This relationship caused an increase in flow rate, stream sediment
carrying capacity, channel and bank erosion, and greater sediment
concentration.

CONCLUSIONS

In the streams we studied, high correlation exists between Kjeldahl
nitrogen, total phosphorus, chloride, and total suspended sediment.
Total suspended sediment (TS) is a reasonable indicator of stream
water quality in the study area, and the three other water quality
parameters can be estimated by TS through a power function with R2

234

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

values greater than 72 percent and with standard error of estimates less
than 30 percent of the observed means. Considering all problems
associated with rain gauge representativeness and storm characteristics
in generating surface runoff, streamflow is a more useful parameter for
predicting stream sediment than is rainfall.

LITERATURE CITED

American Public Health Association. 1976. Standard methods for the examination of
water and wastewater, 14th ed., Washington, D.C., 1193 pp.

Brooks, H. B. 1946. A summary of some radar thunderstorm observations. Bull. Amer.
Meteor. Soc. 27:557-564.

Chang, M. 1981. A survey of the U.S. national precipitation network. Water Resour.
Bull. 17:241-3.

Chang, M., and J. R. Aguilar. 1980. Effect of climate and soil on the radial growth of
loblolly pine ( Pinus taeda L.) in a humid environment of southeastern U.S. A. For.
Ecol. and Mgmt. 3:141-150.

Chang, M., J. D. McCullough, and A. B. Granillo. 1983. Effects of land use and
topography on some water quality variables in forested East Texas. Water Resour.
Bull. 19:191-196.

Colston, N. V., Jr. 1974. Characterization and treatment of urban land runoff. EPA-670/
2-74096, Cincinnati, Ohio, 170 pp.

Donigian, A. S., Jr., and N. H. Crawford. 1977. Simulation of nutrient loadings in
surface runoff with the NPS model. EPA-600/3-77-065. Athens, Georgia, 109 pp.
Environmental Protection Agency. 1971. Methods of chemical analysis of water wastes.
Cincinnati, Ohio, 312 pp.

Environmental Protection Agency. 1976. Quality criteria for water. Washington, D.C. ,
501 pp.

Kissel, D. E., C. W. Richardson, and E. Burnett. 1976. Losses of nitrogen in surface
runoff in the Blackland Prairie of Texas. J. Environ. Qual. 5:288-292.

Ritchie, J. C. , A. C. Gill, and J. R. McHenry. 1975. A comparison on nitrogen,
phosphorus, and carbon in sediments and soils of cultivated and noncultivated
watersheds in the north central states. J. Environ. Qual. 413:339-341.

Romkens, M. J. M., D. W. Nelson, and J. V. Mannering. 1973. Nitrogen and phosphorus
compositon of surface runoff as affected by tillage method. J. Environ. Qual. 2:292-
295.

Schumann, G. E., R. E. Burwell, R. F. Piest, and R. G. Spemer. 1973. Nitrogen losses
in surface runoff from agricultural watersheds in Missouri Valley loess. J. Environ.
Qual. 2:299-302.

Zison, S. W. 1980. Sediment-pollutant relationships in runoff from selected agricultural,
suburban and urban watersheds. EPA-Goo/ 3-80-022, Athens, Georgia, 136 pp.

ATMOSPHERIC CARBON MONOXIDE
IN THE EL PASO-CD. JUAREZ AREA

by MANUEL AGUIRRE, JR.

Texas Air Control Board
El Paso, Texas 79925

and HOWARD G. APPLEGATE

Center for Inter-American and Border Studies
University of Texas at El Paso
El Paso, Texas 79968

ABSTRACT

Portions of El Paso, Texas, have been classified by the Environmental Protection
Agency as nonattainment areas for carbon monoxide. This paper reports on carbon
monoxide generated in the city of El Paso, the International bridges, and Cd. Juarez.
It is concluded that unless steps are taken to control emissions from the bridges and Cd.
Juarez, El Paso cannot meet federal standards. Key words: air pollution, Mexico, Texas,
monoxide.

INTRODUCTION

Carbon monoxide (CO) is the single most important and commonly
occurring air pollutant in the El Paso-Cd. Juarez (EPJAZ) airshed.
Most CO is generated by motor vehicles (Applegate 1981). The
Environmental Protection Agency (EPA) has declared the downtown
portion of El Paso to be nonattainment for CO. Three of the four
international bridges feed into this area; it is also adjacent to the
downtown area of Cd. Juarez. This study was undertaken to determine
the relative CO contributions of motor vehicles using the international
bridges and streets of both cities to the CO nonattainment area of El
Paso.

METHODS

All CO measurements were made by Continuous Air Monitoring
Stations (CAMS) using EPA-approved methodology. Data were
reviewed from October 1976 through June 1981. Cases in which eight-
hour average of CO exceeded nine parts per million (ppm) are given
in Table 1. CAMS 6 is located in downtown El Paso approximately
50 meters south of Interstate 10, 1.7 kilometers northeast of the Stanton
and El Paso streets International bridges and 16.9 kilometers north

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

236

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 1. — Carbon monoxide exceedences of 9 ppm averaged over eight hours.

Date

Time

CAMS 6

CAMS 12

Oct 30-31, 1976

2000-0400

1 1.4 ppm

Nov 1

1700-0100

9.4 ppm

Nov 6-7

1800-0200

11.1 ppm

Dec 4-5

1800-0200

10.8 ppm

Dec 8

1600-2400

12.2 ppm

Dec 9

1700-0100

13.5 ppm

Dec 23

1700-0100

10.1 ppm

Nov 25-26, 1977

1800-0200

10.3 ppm

Jan 9, 1978

1700-0100

1 1.0 ppm

Jan 14-15

1800-0200

9.9 ppm

Oct 25, 1979

1600-2400

9.4 ppm

Nov 7

1500-2300

9.6 ppm

Dec 7

1400-2200

16.4 ppm

Dec 11

1400-2200

11.1 ppm

Jan 1, 1980

0100-0900

1 1.0 ppm

Oct 31-Nov 1

1900-0300

11.2 ppm

Nov 1-2

1900-0300

9.3 ppm

Dec 3

1600-2400

9.4 ppm

Dec 22

1600-2400

9.3 ppm

Dec 26

1700-0100

9.6 ppm

Feb 14-15, 1981

2000-0400

9.9 ppm

northwest of the Chamizal International Bridge. Central downtown
Cd. Juarez is 1.8 kilometers directly south of CAMS 6.

At the time of the CO measurements given in Table 1, CAMS 12 was
at Ascarate Park, some eight kilometers east of downtown El Paso and
the Stanton and El Paso streets International bridges and 10.8
kilometers northwest of the Chamizal International Bridge. The
eastern portion of downtown Cd. Juarez is one kilometer directly south
of CAMS 12.

To assist with the evaluation of the times of CO exceedences, wind
roses were plotted. The wind roses associated with CAMS 6 are
composed of hourly wind directions measured at the station during the
eight hours of the exceedence. The area around CAMS 12 in the
United States is not heavily traveled; thus we believe that the CO was
wind transported to it. To establish the general source of the CO the
wind roses associated with the exceedences measured by CAMS 12 were
also hourly but started two hours prior to the beginning of the
exceedence.

Hourly traffic flows into the United States via the El Paso Street and
Chamizal International bridges were obtained from the United States
Customs Service (Fig. 1).

Traffic flows for two intersections in each city are shown in Figures
2 through 5. Calle 16 de Septiembre (Fig. 2) is the most heavily

ATMOSPHERIC POLLUTION IN EL PASO-CD. JUAREZ

237

NORTH BOUND 'TRAFFIC AT EL PASO STREET AND
CHAMIZAL INTERNATIONAL BRIDGES,

Figure 1.

traveled east-west street and Calle Vicente Guerrero (Fig. 3) is one of
the heaviest traveled north-south streets in Cd. Juarez. The intersection
of Paisano and Trowbridge streets (Fig. 4) is in the eastern portion and
the intersection of Alameda and Delta streets (Fig. 5) in the center
portion of the CO non-attainment zone in El Paso.

RESULTS

CAMS 6

Traffic patterns in the central district are shown in Figures 4 and 5.
These are typical of hundreds of such measurements made since 1975.

TRAFIC0 DE 16 DE SEPTIEMBRE T BOLIVIA
21-22 JULIO 1983, JUEVES- V I ERNES

Figure 2.

238

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

TRflFICO DE G M SOLIS T V GUERRERO
28-29 JUNIO 1983, MARTES-M I ERCOLES

Figure 3.

Morning traffic begins to increase around 05:00 hours, peaks around
07:00, and begins to plateau around 09:00. Emissions from this traffic
have never caused a violation of CO standards. Afternoon traffic begins
to rise around 15:00 hours, peaks around 17:00 and 18:00 and drops
around 20:00. The onset of the 14 exceedences at CAMS 6 was at 14:00
hours twice, 15:00 once, 16:00 and 17:00 four times each, and 18:00
thrice. Why did the essentially same amount of traffic cause
exceedences in the afternoon and not in the morning?

The exceedences of November 1 and December 19, 1976, January 19,
1978, December 11, 1979, and December 3 and 22, 1980, all had a

TRAFFIC AT PAISAN0 AND TROWBRIDGE
24 JULY 1980, THURSDAY

Figure 4.

ATMOSPHERIC POLLUTION IN EL PASO-CD. JUAREZ

239

TRAFFIC AT ALAMEDA AND DELTA
31 AUGUST 1982, TUESDAY

Figure 5.

similar characteristic — two CO peaks that jointly caused the eight-hour
CO averages to exceed 9 ppm. Data for December 9, 1976, are given
to illustrate the group (Fig. 6). The CO peak began at 16:00 hours and
coincided with the traffic increase. By 19:00 hours, the peak had

J I I I I I I I l - 1 - ! I I I - 1_ 1 - 1 - 1 - 1 - 1 - 1 - 1 - L

3.6 2.2 3.0 2.8 2.5 1.8 3.0 2.9 1.5 1.9 1.8 0.3 0.8 0.8 0.6 1.3 2.5 2.8 5.2 5.7 3.6 3.2 2.7 5.5 4.4

WIND SPEED. MPH

307 I 245 I 245 I 238 I 242 I 97 I i'll I 84 I 250 I 269 I 259

290 278 218 246 113 149 107 73 300 328 !

113

WIND DIRECTION. DEGREES

318

280

Figure 6.

240

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NOS. 2 & 3, 1985

DEC 9, 1976

Figure 7.

CO NO

NMHC

2.9 1.9 2.3 2.3 3.3 1.5 1.6 1.6 1.3 3.1 2.8 2.6 2.8 3.6 1.0 1.8 1.5 l.l 3.1 29 3.1 0.2 6.8 7.4 7.4

WIND SPEED. MPH

56 I 257 I 334 I 312 I 299

267 267 312

102 I 109 I 116 I 211 "T 150 I 85 I 110 I 86

157 81 153 178 122 102 119 286 223

WIND DIRECTION. DEGREES

Figure 8.

CONCENTRATION „ PPM

ATMOSPHERIC POLLUTION IN EL PASO-CD. JUAREZ

241

CAMS 6
DEC 23. 1976

Figure 9.

CO NO
25 r 0.5

CO

CAMS 12- EAST
OCT 31 -NOVI. 1980
CO VIOLATION - I 1.2 PPM
TIME 1900-0200

NMHC

10

L. - _i__

l£

14 16 18

20 22 24 02 04 06

HOUR

08

10

— 1 — 1 — 1—

12

50 3.0 20 1.0 6,0 5.0 2.0 0.0

1.0 0.0 20 20 ID 1.0 1.0 2.0 20 20 70 90 90 9.0 60 20

WIND SPEED. MPH

30

243 ,,'7

13 ^7

293 ! 183 1 181 1 325 1 245,1 346

249 157 15 338 305

WIND DIRECTION. DEGREES

1 41

19

1 25 1

50 72

238

Figure 10.

242

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Figure 11.

ended — as had downtown traffic. The second CO peak started at 20:00
hours, peaked at 21:00 hours, and did not disappear until 05:00 hours
the following day. Note that in the wind rose for this exceedence (Fig.
7) during the first peak, the wind was coming to the station from the
southeast. During the second peak, the wind was from the east to
southeast. Certainly some of the CO in the second peak plus the wind
direction makes it evident that the traffic across the international
bridges and in Cd. Juarez also contributed to the exceedence. Neither
CO peak alone would have caused an exceedence, but together they
did.

Exceedences for December 4-5, 8, 23, 1976, January 14-15, 1978, and
December 7, 1979, also shared a common characteristic — winds were
from the southeast and south during the times of the exceedences. Data
from December 23, 1976, are shown to illustrate these eventss (Fig. 8).
Once again, while some of the CO previously measured was being
blown back across the monitor, the height and duration of the
exceedence peak plus the wind rose (Fig. 9) makes it evident that
international traffic plus that in Cd. Juarez was a contributing factor.

CAMS 12

Of the seven exceedences, all but one occurred after 17:00 hours and
four of the seven occurred after 19:00 and 20:00 hours — long after

ATMOSPHERIC POLLUTION IN EL PASO-CD. JUAREZ

243

homeward bound traffic had peaked. Obviously CO was transported to
the site from a distance. A typical exceedence is shown in Figure 10
for October 31-November 1, 1980. The wind rose, starting two hours
prior to the onset of the exceedence is shown in Figure 11. There can
be little doubt the exceedence is due to CO from downtown El Paso,
the international bridges and Cd. Juarez.

The odd reading of Jnuary 1, 1980, is a mystery and can only be
speculated upon as having been caused by the mobility of late New
Year party celebrants.

DISCUSSION

A comparison of traffic patterns between the two cities shows two
striking differences. El Paso traffic peaks around 07:00 and 17:00 hours;
in Cd. Juarez, both peaks occur two to three hours later. In no case
did the morning traffic in either city or the bridges lead to a CO
exceedence. In 57 percent of the exceedences, the onset was 17:00 hours
or later. Clearly, something other than traffic in El Paso was
contributing to the exceedences. Wind directions and traffic patterns in
Cd. Juarez suggest these are a contributing factor.

Traffic across the international bridges is also implicated. The
mayor of El Paso formed an international task force to study the CO
problem within EPJAZ when the EPA threatened economic sanctions
for failure to meet the 1982 Ambient CO standards. Exhausts from 722
vehicles crossing the international bridges were monitored using Texas
Air Control Board methodology. (Data are being prepared for
transmittal to the EPA via the Office of the Mayor of El Paso and will
be published thereafter.)

United States registered vehicles averaged 4.3 percent and Mexican
registered vehicles averaged 5.0 percent normal idle CO; five percent of
United States registered vehicles and eight percent of Mexican
registered vehicles registered off scale (10 percent). Any reading of more
than two percent is considered excessive.

The average wait to pass through customs is a matter of dispute.
During peak traffic, delays of more than one hour are common. Using
a waiting period of 20 minutes, it was estimated that three percent of
the total CO measured in 1977 in EPJAZ came from the bridges and
nine percent from Cd. Juarez (Applegate 1981). Aguirre (1981) using
different assumptions, estimated emissions from Cd. Juarez to be 10
times greater. He did not estimate emissions from traffic from the
bridges.

Northbound traffic across the international bridges peaks around
18:00 and 19:00 hours. These are later than the afternoon traffic peak
in El Paso but coincide well with the onset of CO exceedences.

244

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

CONCLUSION

Vehicles are the major source of CO in EPJAZ. The three area
sources are downtown El Paso, central Cd. Juarez, and the
international bridges. The EPA and the Texas Air Control Board have
jurisdiction over sources in El Paso only. Control of those sources will
not bring El Paso into compliance with the 1982 ambient air standard
for CO.

LITERATURE CITED

Aguirre, M. 1981. A study of carbon monoxide in the El Paso-Cd. Juarez area. M.S.

thesis, Univ. Texas at El Paso, El Paso, Texas, 158 pp.

Applegate, H. G. 1981. Allocation of vehicular emissions of carbon monoxide in El
Paso, Texas and Ciudad Juarez, Chihuahua. Envir. Sci. Tech. 15:963-966.

FACTORS INFLUENCING CELLULOLYTIC ACTIVITY OF
THE SOIL FUNGUS, ASPERGILLUS CANDID US

by J. ORTEGA and E. J. BACA

School of Sciences and Mathematics
Pan American University
Edinburg, Texas 78539

ABSTRACT

Effects of pH, incubation time, nitrogen source and carbohydrate source upon
production of carboxymethyl cellulase by Aspergillus candidus grown in liquid medium
were studied. Maximal cellulase activity per unit of protein occurred in culture fluids
with a pH of 5.5, after 10 days of cultivation of the fungus. Ammonium nitrate
supported the highest production of the enzyme. However, in the nitrogen-source test,
highest accumulation of extracellular protein was found in fluids from cultures with
urea as the nitrogen source. Cotton fibers produced the greatest gain in cellulase activity
when the carbohydrate source was incremented from 0.5 to 1.0 percent. The highest
cellulase activity was measured in fluids whose carbohydrate source was carboxymethyl
cellulase (CMC). In the carbohydrate-source test, highest accumulation of extracellular
protein was found in fluids from cultures with powdered cellulose as the carbohydrate
source, at 1.0 percent concentration.

INTRODUCTION

Studies on the cellulolytic activities of fungi have concentrated
mostly on species of Trichoderma (see Mandels and Reese 1957;
Gritzali and Brown 1979; Sternberg and Mandels 1979; Ratnam et al.
1982; Sternberg and Mandels 1982). However, other species of fungi
may be capable of cellulolytic activities similar or superior to those of
Trichoderma. Several studies (Stutzenberger 1972; Coutts and Smith
1976; Andreotti et al. 1977; Fennington et al. 1982) have shown that
different species of fungi have different requirements of nitrogen,
carbohydrate, pH, and incubation time for maximum production of
cellulases.

The objective of this work was to study some factors that influence
the production of cellulases in Aspergillus candidus.

MATERIALS AND METHODS

The fungus Aspergillus candidus was isolated from soil taken from
a sorghum field near Edinburg, Texas, in October of 1982. Stock
cultures of the test fungus were maintained at 4° C in petri plates of
potato-dextrose-agar, PDA (Difco).

The Texas Journal of Science, Vol. XXXVII, Nos. 28c3, September 1985

246

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

The test fungus was grown in basal liquid medium of the following
composition: glucose, 2.5 grams; powdered cellulose (Sigma, type 50),
5.0 grams; urea, 1.075 grams; KH2PO4, 2.722 grams; MgS04.7H20, 0.8
grams; Ca(N03)2.4H20, 0.5 grams; Fe(N03)3.9H20, 1.44 mg; ZnS04,
0.88 mg; MgS04.4H20, 0.40 mg; distilled water, 1000 milliliters. The
pH of the basal medium was adjusted to 5.0 with 0.02 M KH2PO4
buffer. The medium was dispensed in 250-milliliter flasks, 120
milliliter per flask, and sterilized for 15 minutes at 121° C and 15 psi.

To test the effect of incubation time on production of cellulase,
samples of culture fluids were taken as described below, after four,
eight, 10 and 15 days of growth. The effect of the pH on cellulase
production was tested in culture fluids with pH adjusted to 3.5, 4.0,
4.5, 5.0, 5.5, and 6.0, and with sterile 0.02 M KH2PO4 or KH2PO4
buffer. To test the effect of the nitrogen source on cellulase production,
the urea of the basal medium was substituted separately in the test
media for the following compounds: NaN03, (NH4)2S04, and
NH4H2P04. The amount of the nitrogen source in each test medium
was the necessary to provide 0.5 grams of nitrogen per liter of medium.
After the best nitrogen source was identified (NH4N03), the media for
all subsequent tests had this compound as the source of nitrogen. The
effect of the carbohydrate source on production of cellulase by the
fungus was tested similarly. The following carbohydrate sources were
substituted separately in the test media for the carbohydrates of the
basal medium: glucose, powdered cellulose (Sigma, type 50),
carboxymethyl cellulose (sodium salt, CMC, type 7HF, by Hercules,
Inc.), and absorbent cotton fibers (surgical grade, by Johnson and
Johnson Co.). Each carbohydrate source was tested at two
concentrations, 0.5 and 1.0 percent.

Seven-day-old cultures of the test fungus grown in petri plates of
solid medium were used as sources of inoculum. Each culture flask
was inoculated by aseptically transferring two 5-mm disks carrying
hyphae and spores (Ortega 1980). All cultures were incubated at 25° C.
Except for the incubation-time tests, samples from the liquid cultures
of the test fungus were taken after 10 days of growth by pipetting 12
milliliters of the fluid into separate sterile test tubes. The fluids were
centrifuged at 6000 g for 15 minutes at 20° C. Then, the clear
supernatant was pipetted into sterile test tubes and frozen until the
cellulase assays were made. Extracellular protein was determined by
the method of Bradford (1976).

Cellulase activity (endoglucanase EC 3.2. 1.4) was determined by
measuring the change in viscosity of a buffered test solution of sodium
CMC, when the fluid from the test fungus was mixed with it and the
reaction mixture was kept at constant temperature. The test solution
consisted of 8.0 grams of sodium CMC (type 7HF with a D.S. of 0.7,

CELLULOLYTIC ACTIVITY OF SOIL FUNGUS

247

by Hercules, Inc.) dissolved in 1000 milliliters of 0.05 M sodium citrate
buffer, pH 5.0. The reaction mixture consisted of nine milliliters of
CMC test solution and one milliliter of a 50 percent dilution of the
fungus fluid in 0.05 M sodium citrate buffer.

Changes in viscosity of the reaction mixture were determined with
Cannon-Fenske routine viscometers. Efflux time of the viscometers was
determined in seconds. All tests were made at 40° C. The specific
viscosity (Nsp) of the substrate was determined as described before
(Ortega 1980), using the data obtained from the viscosity tests. All tests
were replicated three times.

Activity of the enzyme was determined with the exponential curve
model described before (Ortega and Baca 1983). Each unit of cellulase
activity is that amount of enzyme that causes a reduction of 0.1
milligram per milliliter in the concentration of the substrate per
minute, under the conditions described for the assay. Specific activity
was expressed in units of cellulase activity per milligram of
extracellular protein.

RESULTS AND DISCUSSION

The pH for the greatest production of extracellular protein was 4.5
(Fig. 1). However, the highest cellulase activity per unit protein
occurred at a pH of 5.5. Mandels and Weber (1969) observed maximum
cellulase activity of Trichoderma viride, a mesophilic fungus, at pH
5.0. Thermoactinomyces gave maximum CMCase activity at pH of 5.5
to 6.0 (Hagerdal et al. 1979). Saddler (1982) reported maximum
cellulase activity of Trichoderma reesei mutants at pH of 6.0 to 6.6 and
highest production of soluble protein at pH 6.0.

Maximal cellulase activity, greatest concentration of extracellular
protein, and greatest cellulase activity per unit protein were measured
in fluids taken from cultures grown for 10 days in the basal medium
(Fig. 2). T. viride, grown in cellulase-containing media developed
maximal cellulase activity after 12 days of growth (Mandels and Weber
1969). Thermophilic fungi have reached maximal cellulase activities in
shorter periods. Sporotrichum thermophile developed maximal
cellulase activity in two to four days of growth at 50° C in wood
cellulose (Solka-Floc) medium (Coutts and Smith 1976). Strain YH-78
of Humicola grisea var. thermoidea developed maximal cellulase
activity after four growing days in wheat bran medium (Yoshioka et
al. 1982).

Effects of nitrogen from various sources on cellulase activity and
production of extracellular protein by A. candidus grown in test media
are shown in Table 1. In this test, maximal cellulase activity (167 units
per milliliter) was measured in fluids from cultures whose source of

248

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Figure 1. Influence of pH on specific activity and production of extracellular protein
by A. candidus. Each point is the average of three determinations.

added nitrogen was NH4NO3. Saddler (1982) has reported similar
results obtained from studies of the cellulase activities of T. reesei
mutants. S. thermophile gave the highest cellulase yields in media
containing urea as the nitrogen source (Coutts and Smith 1976). In our
work, NH4NO3 induced a production of cellulase 4.6 times higher
than that induced by KNO3, which induced the lowest cellulase

CELLULOLYTIC ACTIVITY OF SOIL FUNGUS

249

Figure 2. Time of growth, specific cellulase activity and extracellular protein of A.
candidus. Each point is the average of three determinations.

activity measured. The largest amount of extracellular protein (12
microgram per milliliter) measured in this test was accumulated in
fluids from cultures whose nitrogen source was NH2CONH2 (Table 1).

Table 2 shows that cellulase activity of the test fungus was
influenced not only by the concentration but also by the chemical form
of the carbohydrate added to the growing medium. Of the

Protein, micrograms/ml.

250

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Table 1. — Effect of nitrogen source on the cellulase activity and protein production of
A. candidus.

Nitrogen source

Cellulase activity3
(units/ml)

Protein

(micrograms / ml)

Specific

activity5

X 1 0-3

NH4N03

167

2

83.5

NH2CONH2

93

12

7.8

NH4H2P04

78

4

19.5

KN03

36

2

18.0

aAverage of three replications.

bSpecific activity = units of cellulase activity per milligram of extracellular protein.

carbohydrates tested, CMC at a concentration of one percent induced
the highest cellulase activity. The cellulase activity induced by CMC
at this concentration (219 units per milliliter) was 4.6 times more than
the cellulase activity induced by glucose at the same concentration of
carbohydrate.

Increase in the concentration of added carbohydrate from 0.5 percent
to 1.0 percent increased the cellulase activity of the test fungus in three
of the four carbohydrate-source tests (Table 2). The largest increment
(3.9 times more) in cellulase activity was measured in fluids from
cultures where cotton fibers were the carbohydrate source. The smallest
increment (1.7 times more) in cellulase activity was measured in fluids
from cultures with glucose as the source of carbohydrate.
Thermomonospora curvata gave maximum yields of cellulase in media
containing 0.8 percent cotton fibers as the carbohydrate source. When
the concentration of this carbohydrate was raised to 1.0 percent, the
production of cellulase was inhibited (Stutzenber 1972).

In this test of carbohydrate sources, fluids with 1.0 percent powdered
cellulose accumulated the greatest amount of extracellular protein
(Table 2). The lowest amount of extrcellular protein was found in
fluids from cultures with glucose as the source of carbohydrate.

The specific cellulase activity measured in fluids where the
carbohydrate source was CMC was about the same as when glucose
was the carbohydrate source (Table 2). It is possible that cellulase
produced by this strain of A. candidus is a constitutive enzyme.

Our results are comparable to those reported for wild strains of other
mesophilic fungi grown under similar conditions (Ortega 1980,
Saddler 1982). A. candidus gave maximum yields of extracellular
cellulase in the same incubation time as several mutants of T. reesei.
These mutants are widely recognized as some of the better cellulase
producers known. Results reported by Saddler (1982), show that some
mutants of T. reesei produce higher amounts of extracellular protein
than our wild strain of A. candidus. However, higher amounts of

CELLULOLYTIC ACTIVITY OF SOIL FUNGUS

251

Table 2.-— Effect of carbohydrate source on the cellulase activity and protein production
of A. candidus.

Carbohydrate

Cellulase

activity3

units/ml

Protein

micrograms / ml

Specific

activity

X 1 0-3

CMC

0.5%

219

-

-

1.0%

223

16

13.9

Cotton fibers

0.5%

39

-

-

1.0%

155

16

9.7

Powdered cellulose

0.5%

67

-

-

1.0%

109

22

5.0

Glucose

0.5%

27

-

-

1.5%

47

4

11.8

aAverage of three replications.

extracellular protein or units of cellulase activity, or both, do not
necessarily mean higher rates of cellulose decomposition. Mandels et
al. (1981) showed that diluted enzyme of T. reesei mutants is more
efficient than concentrated enzyme in the decomposition of cellulose.

LITERATURE CITED

Andreotti, R. E., M. Mandels, and C. Roche. 1977. Effect of some fermentation
variables on growth and cellulase production by Trichoderma QM9414, Pp. 249-267,
in Bioconversion of cellulosic substances into energy, chemicals and microbial
protein, (T. K. Ghose, ed.), Proc. Bioconversion Symp., IIT, Delhi.

Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem.
72:248-254.

Coutts, A. D., and R. R. Smith. 1976. Factors influencing the production of cellulases
by Sporotrichum thermophile. Appl. Environ. Microbiol. 31:819-825.

Fennington, G., D. Lupo, and F. Stutzenberger. 1982. Enhanced cellulase production in
mutants of Thermomonospora curvata. Biotech. Bioeng. 24:2487.

Gritzali, M., and R. D. Brown. 1979. The cellulase system of Trichoderma : Relationships
between purified extracellular enzymes from induced or cellulose grown cells. Pp.
237-260, in Hydrolysis of cellulose: Mechanisms of enzymatic and acid catalysis. (R.
D. Brown, and L. Jurasek, eds.). Advances in Chemistry Series 181, Amer. Chem. Soc.
Washington, D.C.

Hagerdal, B., J. Ferchak, E. K. Pye, and J. R. Forro. 1979. The cellulolytic enzyme
system of Thermoactinomyces. Pp. 331-345, in Hydrolysis of cellulose: Mechanisms
of enzymatic and acid catalysis. (R. D. Brown, and L. Jurasek, eds.). Advances in
Chemistry Series 181, Amer. Chem. Soc. Washington, D.C.

Mandels, M., and E. R. Reese. 1957. Induction of cellulase in Trichoderma viride as
influenced by carbon sources and metals. J. Bacteriol. 73:269-278.

Mandels, M., and J. Weber. 1969. The production of cellulases. Pp. 391-413, in
Cellulases and their applications. (R. F. Gould, ed.), Advances in Chemistry Series
95, Amer. Chem. Soc., Washington, D.C.

Mandels, M., J. E. Medeiros, R. E. Andreotti, and F.H. Bisset. 1981. Enzymatic
hydrolysis of cellulose. Evaluation of cellulase culture filtrates under use conditions.
Biotechnol. Bioeng. 23:2009-2026.

252

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 & 3, 1985

Ortega, J. 1980. Cellulase activities of soil fungi. Tex. J. Sci. 32:241-246.

Ortega, J., and E. J. Baca. 1983. Viscometric measurement of the cellulase activity of
a soil fungus. Tex. J. Sci. 35:261-267.

Ratnam, D. A., S. Pavlov, and A. G. Fredrickson. 1982. The effect of oxygen transfer
rate (OSR) on the formation of cellulases by Trichoderma viride in submersion
culture. Biotech. Bioeng. 24:2675.

Saddler, J. N. 1982. Screening of highly cellulolytic fungi and the action of their
cellulase enzyme systems. Enzyme Microb. Technol. 4:414-418.

Sternberg, D., and G. R. Mandels. 1979. Induction of cellulolytic enzymes in
Trichoderma reesei by sophorose. J. Bacteriol. 139:761-769.

Sternberg, D., and G. R. Mandels. 1982. B-Glucosidase induction and repression in the
cellulolytic fungus Trichoderma reesei. Experimental Mycology 6:115.

Stutzenberger, F. J. 1972. Cellulolytic activity of Thermomonospora curvata: Nutritional
requirements for cellulase production. Appl. Environ. Microbiol. 24:77-82.

Yoshioka, H., S-I. Anraku, and S. Hayashida. 1982. Production and purification of a
novel type of carboxymethyl cellulase from Humicola grisea var. thermodidea YH-
78. Agric. Biol. Chem. 46:75-82.

COMPARATIVE BEHAVIOR AND EXTERNAL

COLOR PATTERNS OF TWO SYMPATRIC CENTIPEDES

(CHILOPODA: SCOLOPENDRA) FROM CENTRAL TEXAS

by RAYMOND W. NECK

Texas Parks and Wildlife Department
4200 Smith School Road
Austin , Texas 78744

ABSTRACT

The external coloration and defensive behavior of two congeneric species of
centipedes, Scolopendra heros and Scolopendra viridis, which occur sympatrically in
rocky sites in central Texas are contrasted. Key words : centipedes, Scolopendra viridis ,
Scolopendra heros, defensive behavior, aposematic coloration, diel behavior.

INTRODUCTION

Limestone uplands, slopes and canyons of the Texas Hill Country
are inhabited by two species of large centipedes, Scolopendra viridis
Say 1821, and Scolopendra heros Wood 1863. External color patterns
of these two species differ quite strikingly. Laboratory and field
observations revealed differing defensive behavioral regimes which
appear to be correlated with the difference in brightness of external
color patterns.

All animals observed in this study were collected on an east-facing
slope of Edwards Limestone immediately west of the Comal Springs
Fault. The site is located approximately 3.3 kilometers west-southwest
of New Braunfels, Comal Co., Texas. Laboratory observations were
made at the Brackenridge Field Laboratory of The University of Texas
at Austin. Centipedes were kept in plastic shoe boxes with soil and
limestone rocks. Animals were fed various insects, especially field
crickets, Gryllus sp. They were kept under 14 hours light-10 hours
darkness diel conditions. Examinations during dark periods were made
with short light exposure. No duplicate observations were same on the
same day.

Scolopendra viridis is a moderate-sized centipede (average adult
length, 70.5 mm) of relatively drab appearance. The head, first body
segment and antennae are dull reddish. Remaining body segments are
dull yellow with bluish posterior margins. Also present is a vague
middorsal blue line, which is thicker at the posterior margin and thins
anteriorad. The anal segment and anal legs are dull orange-brown.

The Texas Journal of Science, Vol. XXXVII, Nos. 2&3, September 1985

254

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NOS. 2 8c 3, 1985

Body legs are pale blue to dull cream-colored. Colors tend to be weakly
differentiated.

Scolopendra heros is a large centipede (average adult length, 163.8
mm) that is brightly colored. The head and first two body segments
are reddish-orange. All other body segments and the three basal
segments of the anal legs are bluish black. All body legs and the apical
segment of the anal legs are yellow. Color of antennae grades from
reddish-orange (basal segments) to yellow (apical segments). All colors
are strongly differentiated.

RESULTS

Diel activity patterns of the two species contrast greatly. Systematic
checks of activity period in the laboratory revealed that S. vidiris was
active only during the dark period (active none of 10 diurnal checks
and six of 10 nocturnal), whereas S. heros is active exclusively during
the light period (active eight of 10 diurnal checks and none of 10
nocturnal). These results agree with daylight observations of S. heros
active above ground. S. vidiris is found only under rocks and wood
during daylight hours. Laboratory observations indicate that S. heros,
which begins to feed during daylight hours, continues to feed after
dark.

These two species of Scolopendra also differ in reaction to
disturbance and subsequent defensive behavior. When the container
with a S. vidiris in a soil burrow is disturbed, the centipede remains
in the burrow. When brought to the surface, it moves close to the soil
surface until it locates an area of soft soil whereupon the animal
crawls back into the soil. When it is almost totally underground, the
anal legs are spread apart but remain on the soil surface. A human
impression of this last activity is one of a threat to grasp any pursuing
object. Movement of the anal legs in such a manner exposes those
portions of the legs that are paler in color.

When a container with S. heros is disturbed, the centipede becomes
active and runs around the container for a longer time period and at
a faster rate than S. vidiris. When entering the soil burrow, S. heros
moves more rapidly than 5. vidiris. The anal legs are raised and waved
back and forth; they are then slowly pulled into the burrow. The legs
are particularly visible because of the bright yellow color of the apical
segment of the anal leg. When a subterranean S. heros is prodded in
the posterior portion of its body, the animal backs out of the burrow
while displaying the brightly-colored anal legs. Active defense by S.
heros involves attempts to grab the offending object with either the
front legs (in order to bite) or the anal legs (to hold). When charging
forward, the head is moved rapidly from side to side as the body is

CENTIPEDE BEHAVIOR AND COLOR PATTERNS

255

moved in snake-like undulation. The passive /active dichotomy in
defensive behavior of 5. vidiris and S. heros is apparent when
specimens are handled for measurement.

DISCUSSION

Scolopendra viridis is characterized by dull coloration, nocturnal
activity, and avoidance defensive behavior. Scolopendra heros is
characterized by bright coloration, diurnal activity, and confrontation
defensive behavior. These contrasts indicate that the coloration of S.
heros should be considered aposematic.

Comparative seriousness of discomfort and complications associated
with bites of these species was not investigated, but limited
information is available. Baerg (1924) induced a S. heros from
Arkansas to bite him, producing ‘ ‘sharp and strictly local” pain that
decreased dramatically after 15 minutes; after three hours pain could
be felt only by pressing the punctures. Norman (1897) reported on the
effects of Scolopendra “morsitans” (S. heros) on mice but did not
report effects on humans. I know of no references on the effect of bites
of S. vidiris on humans. Severity of bite might be greater in diurnal,
aposematic species than in a nocturnal, nonaposematic species.

S. vidiris and S. heros are able to coexist sypatrically, in part,
because they are not synchronic. Lack of temporal overlap probably
reduces or even eliminates potential competition for food. The
differences in external coloration and defense behavior reflect the
varied adaptive features of nocturnal and diurnal species.

I thank Gerald Summers for supplying a taxonomic key used in
identification of these species.

LITERATURE CITED

Baerg, W. J. 1924. The effect of the venom of some supposedly poisonous arthropods.

An. Ent. Soc. Amer. 17:343-352.

Norman, W. W. 1897. The poison of centipedes ( Scolopendra morsitans). Proc. Texas

Acad. Sci. 1:118-119.

THE TEXAS ACADEMY OF SCIENCE, 1984-85
OFFICERS

President :

President-Elect:

Vice-President:

Immediate Past President:
Secretary - T reasurer:

Editor:

AAAS Council Representative :

William J. Clark, Texas A&M University
Billy J. Franklin, Lamar University
Lamar Johanson, Tarleton State University
Michael J. Carlo, Angelo State University
Fred S. Hendricks, Texas A&M University
J. Knox Jones, Jr., Texas Tech University
Ann Benham, University of Texas at Arlington

DIRECTORS

1983 D. Lane Hartsock, Austin
Katherine Mays, Bay City

1984 E. D. McCune, Stephen F. Austin State University
Jim Neal, U.S. Fish and Wildlife Service

1985 George B. McClung, San Angelo
Barbara Schreur, Texas A&I University

SECTIONAL CHAIRPERSONS

I — Mathematical Sciences : John Seaman, Baylor University

II — Physical Sciences : John Hubisy, College of the Mainland

III — Earth Sciences : James O. Jones, University of Texas at San Antonio

IV — Biological Sciences: Randy Moore, Baylor University

V — Social Sciences: Thomas Gray, Southwest Texas State University

VI — Environmental Sciences: Dean V. Ferguson, Austin

VII — Chemistry: Lynn Melton, University of Texas at Dallas

VIII— Science Education: Thomas Koballa, University of Texas at Austin

IX — Computer Sciences: H. R Haiduk, Amarillo College

X— Aquatic Sciences: Darrell S. Vodopich, Baylor University

COUNSELORS

Collegiate Academy: Shirley Handler, East Texas Baptist College

Helen Oujesky, University of Texas at San Antonio
Junior Academy: Ruth Spear, San Marcos

Peggy Carnahan, San Antonio

COVER PHOTO

Map of Village Creek area, Hardin County, Texas

from Barclay and Harrel, pp. 175-188

2nd CLASS POSTAGE
PAID AT LUBBOCK
TEXAS 79401

3024 I 85

Library Acquisitions
Smithsonian Institute
Washington DC 20560

ISSN 0040-4403

December 1985

me XXXVII, Number 4

PUBLISHED QUARTERLY BY
THE TEXAS ACADEMY OF SCIENCE

r-’Z-A

ALBERT ZLATKIS
Distinguished Texas Scientist, 1985

SECTION I

MATHEMATICAL SCIENCES
Mathematics, Statistics,
Operations Research

AFFILIATED ORGANIZATIONS
Texas Section, American Association of Physics Teachers
Texas Section, Mathematical Association of America
Texas Section, National Association of Geology Teachers
American Association for the Advancement of Science

GENERAL INFORMATION

MEMBERSHIP. Any person or group engaged in scientific work or interested in the pro¬
motion of science is eligible for membership in The Texas Academy of Science. Dues for
members are $20.00 annually; student members, $12.00 annually; sustaining members, at
least $30.00 in addition to annual dues; life members, at least $400.00 in one payment;
patrons, at least $500.00 in one payment; corporate members, $250.00 annually; corporate
life members, $2000.00 in one payment. Library subscription rate is $45.00 annually. Pay¬
ments should be sent to Dr. Fred S. Hendricks, TAS Secretary-Treasurer, Drawer H6,
College Station, TX 77844.

The Journal is a quarterly publication of The Texas Academy of Science and is sent
to all members and subscribers. Inquiries regarding back issues should be sent to the
Secretary-T reasurer.

The Texas Journal of Science is published quarterly at Lubbock, Texas U.S.A. Second
class postage paid at Post Office, Lubbock, TX 79401. Please send form 3579 and
returned copies to Texas Tech Press, Box 4240, Lubbock, TX 79409.

THE TEXAS JOURNAL OF SCIENCE

Volume XXXVII, No. 4 December 1985

CONTENTS

Recent developments in gas chromatographic trace analysis using new
concentration techniques. By Albert Zlatkis, Shary Weisner, and

Labib Ghaoui. . . . . . 259

Strengths and weaknesses of damage assessment programs: the Ixtoc-I

and Burmah Agate oil spills, and the benthic macroinfauna of the Texas

continental shelf. By George S. Lewbel . . . 269

A record of Symbos (ArtiodactylarBovidae) from Kaufman County,

Texas. By Jerry N. McDonald . . . . . 311

Reproduction data on Dionda episcopa from Fessenden Spring, Texas.

By Leslie M. Wayne and B. G. Whiteside . 321

Fecal and intestinal bacteria of swine maintained on atherogenic

diets. By D. H. Lewis . . . 329

Effects upon selected marine organisms of explosives used for sound
production in geophysical exploration. By Thomas L. Linton, Andre M.

Landry, Jr., James E. Buckner, Jr., and Robert L. Berry . 341

Records of the spotted skunk and long-tailed weasel from the Llano
Estacado of Texas. By J. Knox Jones, Jr., Robert R. Hollander, and

David A. McCullough . 355

Fine structure of the secondary walls of sclereids of Rauwolfia

serpentina. By A. J. Mia . 359

Studies on the use of boiled chicken egg yolk as a feed for rearing

penaeid shrimp larvae. By David M. Fuze, Joshua S. Wilkenfeld, and

Addison L. Lawrence. . . 371

Sea urchins from the Brazos Santiago Pass jetty, South Padre Island,

Texas. By Richard R. Fairchild and L. O. Sorensen. . . . . 383

Occurrence of mussels in the orbits of a blue crab. By George N.

Greene, David C. Me Aden, and William B. Baker, Jr . . . 387

An instance of a largemouth bass, Micropterus salmoides, feeding on
a water snake, Nerodia erythrogaster transversa. By Dennis Parmley

and Charles Mulford. . . 389

Index (including list of authors and reviewers). . . . . . 391

Instructions to authors. . . 401

Dates of publication of volumes 34-36

402

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Editor:

J. Knox Jones, Jr., Texas Tech University
Assistant to the Editor:

Marijane R. Davis, Texas Tech University
Associate Editor for Botany:

Randy Moore, Baylor University
Associate Editor for Chemistry:

Marvin W. Rowe, Texas A8cM University
Associate Editor for Computer Science:

Ronald K. Chesser, Texas Tech University
Associate Editor for Mathematics and Statistics:

George R. Terrell, Rice University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Scholarly papers in any field of science, natural history, or
technology will be considered for publication in The Texas Journal of
Science. Instructions to authors are published one or more times each
year in the Journal on a space-available basis, and also are available
from the Editor (The Museum, Box 4499, Texas Tech University,
Lubbock, Texas 79409).

RECENT DEVELOPMENTS IN

GAS CHROMATOGRAPHIC TRACE ANALYSIS

USING NEW CONCENTRATION TECHNIQUES

by ALBERT ZLATKIS, SHARY WEISNER,
and LABIB GHAOUI

Chemistry Department

University of Houston-U niversity Park

Houston, Texas 77004

ABSTRACT

Trace analysis has become increasingly important when dealing with environmental
samples and biological fluids. Organic compounds often are present in such low
concentrations that it has been necessary to develop sophisticated techniques in order to
be able to analyze them. Two such concentration techniques are discussed here: the use
of Tenax-GC as an effective adsorbent to trap the trace organics, and direct on-column
injection of large samples in capillary columns. Key words: gas chromatography, trace
analysis, concentration techniques, capillary columns.

Difficulties frequently are encountered in attempting to analyze
directly organic compounds of interest, which are often below the part
per billion level. Despite the use of highly sensitive instruments,
detection of trace amounts of substances in this range presents a
technical challenge. A good example is the isolation and identification
of compounds that have profound physiological effects at extremely
low concentrations. Another area of interest is the analysis of volatiles
from biological fluids (urine, for example) to seek distinctive
differences between “normals” and those afflicted by disease. More
recently, the awareness has grown that minute concentrations of
chemical pollutants can have far-reaching effects as health hazards,
further underscoring the need for reliable analytical techniques. Flavor
and odorant studies are also areas of interest in this analytical region.

At the part per billion level or lower, it has almost always been
necessary to use some cumulative or concentrating technique to obtain
measurable amounts of sought-after compounds. Ideally, we wish to
eliminate as much as possible of unwanted “background” compounds
(usually water or air) while accumulating the desired substances
quantitatively. For most techniques, the result is a compromise of
these two goals. Several approaches come to mind readily— fractional
distillation, freeze concentration, zone melting, solvent extraction,
chromatographic techniques, and adsorption.

Another approach, which seemed to be analytically unattainable
until recently, is the use of direct on-column injection of

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

260

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

extraordinarily large sample volumes in capillary gas chromatographic
columns. Some of the recent work in this area also will be presented
in this survey.

SELECTIVE ADSORPTION USING TENAX-GC POLYMER

A wide variety of adsorbents has been investigated as selective
trapping materials for trace organics. These range from strong
adsorbents like activated charcoal to relatively weak polymeric
adsorbents. Desorption from strong adsorbents involves either solvent
extraction, or heating to temperatures that often result in chemical
changes. Weak adsorbents, on the other hand, will allow many desired
substances to escape.

For most analytical situations it is convenient to sample at ambient
temperatures, without the necessity for cold traps. It is preferable to
use adsorbents that are efficient collectors for many compounds of
interest at room temperature, and that do not introduce chemical
changes or artifacts during thermal desorption.

Tenax-GC, a porous polymer of 2,6-diphenyl-p-phenylene oxide,
has proven to be a versatile selective adsorbent. It has a modest surface
area of 18 m2 g_1 and has a low retention for low molecular weight
polar substances, especially water. Higher molecular weight
compounds having relatively low polarity are trapped and desorbed
thermally with high efficiency. This is illustrated in Table 1, which
lists the trapping capacity and desorption characteristics of Texas-GC
for various organic compounds. This is evaluated by injecting a fixed
amount of each compound on the Texax-GC and eluting the trap with
0.5, 1.5, and 5.0 liters of pure nitrogen. The amount retained is
determined chromatographically by desorption at 300°C. At this
temperature Tenax-GC does not contribute detectable artifacts, due to
its unusual thermal stability.

Tenax-GC is like any other chromatographic stationary phase and
must be evaluated from the point of view of partitioning of a
compound between adsorbent and carrier gas. As a consequence, the
results in Table 1 apply to the specific amount of Tenax-GC employed
in these experiments (0.28 g). That is to say, the “breakthrough”
volume is directly proportional to the amount of adsorbent. Table 1
shows that for molecules with molecular weights as high as that of n-
octadecane, adsorption and desorption are reversible and complete.
Low molecular weight polar compounds (for example, methanol,
ethanol, and acetone) have low retention and relatively small
breakthrough volumes. Water is not noticeably adsorbed.

Figure 1 illustrates a typical system for trapping compounds from
ambient air or any other source of volatile organics. The air or other

GAS CHROMATOGRAPHIC TRACE ANALYSIS

261

Table L Recovery (percent) of organic compounds from Tenax-GC after adsorption at
room temperature.

Volume of N2 passed through trap (liters)

Compound

0.5

1,5

5.0

Menthol

1

0

0

Ethanol

1

0

0

Methyl chloride

3

1

0

Acetone

68

2

0

Chloroform

100

84

5

Diethylamine

80

50

1

Isobutanol

100

95

16

n-Pentane

100

50

9

Cyclohexane

100

50

9

n-Hexane

100

100

20

Ethyl acetate

100

100

35

n-Butanol

100

100

35

Toluene

100

100

100

C7-C12 alkanes, alkenes

100

100

100

Styrene, ethylbenzene

100

100

100

Xylenes

100

100

100

Pyridines

100

100

100

Chlorophenols

n-Ci3,Ci4,Ci5,Ci6,Ci7,Ci8 alkanes

100

100

100

carrier gas is drawn through the trapping tubes, which may be
arranged in parallel or in series. Typically, a sampling tube will be
of about 4 mm in diameter and 12 cm long. One-quarter gram of
Tenax-GC occupies a length of about 9 cm, and is held in place by
two plugs of glass wool. The tubes containing the adsorbent are
preconditioned at 350°C in a stream of pure nitrogen, then allowed to
cool in screw-capped test tubes. Flow rates are maintained at about 30
ml min-1 to allow satisfactory contact time with the Tenax-GC. For
many applications a total of 5 liters of air or carrier gas is adequate
for 0.25 g of adsorbent, where substances are present in the gas below
the part per million level. Desorption is carried out at 300°C by rapidly

Source of
Organic —

Volatiles

Flowmeter

Vacuum

Source

TENAX GC SAMPLING
TUBES (SERIES OR PARALLEL)

Figure 1. Schematic diagram of typical system for trapping organic volatiles with
Tenax-GC.

262

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

heating the Tenax-GC trap while passing an inert gas through it, then
cold-trapping the volatiles ahead of a chromatographic column.
Subsequently, the cold trap is flash-heated and the contents analyzed
by chromatographic techniques.

Another area of research to which Tenax-GC has been applied is
that of determining the trace contaminants in water for a pollution
study. Here again the hydrophobicity of Tenax-GC works to great
advantage. As usual, significant adsorption occurs for compounds
above C6.

ON-COLUMN, LARGE VOLUME SAMPLE INJECTION
IN CAPILLARY COLUMNS

Trace level analysis in conventional capillary gas chromatography
generally has been limited by the sample size that would still maintain
satisfactory resolution. In a relatively few instances, it has been
possible to utilize element-specific detection devices, but wide spectrum
detectors such as the flame ionization detector (FID) have been limited
typically to the part per million range due to sample size restrictions.

The development of bonded phase (nonextractable or immobilized)
fused silica open tubular columns (FSOT) has made it possible to
employ liquid on-column injection techniques without loss of column
performance. Blomberg et al. (1982) have shown that the immobilized
stationary phase is not displaced from the capillary surface by solvent
or even water. Sandra et al. (1981) have reported that film thickness,
film homogeneity, and resolution were not affected in such bonded
phase FSOT columns by extensive rinsing with both polar and
nonpolar solvents. Work recently carried out in our laboratory showed
that peak symmetry of aromatics was preserved in this column with
the solvent n-hexane in amounts as large as 100 /zl, using an on-
column injector constructed for this purpose (Wang et al. 1982).

The successful injection of large samples would significantly
improve trace component analysis. However, the solvent tends to
swamp out component peaks. Grossly distorted chromatograms result
with sample sizes of this order even when the “trace” components are
present in relatively large amounts.

The work discussed here utilizes our on-column injector to introduce
sample volumes of up to 100 /ul into bonded phase FSOT columns.
The first step causes the liquid sample to pass through the column in
order to adsorb the trace components of interest at relatively low
temperature, permitting the stripping solvent to leave the column. The
second step recovers these trace substances in a liquid nitrogen trap;
finally, analysis is carried out in the same column by operating it in
reverse. In essence, this procedure amounts to a modified “heart¬
cutting” technique.

GAS CHROMATOGRAPHIC TRACE ANALYSIS

263

Table 2. Analytical data for 1 /d on-column injection of n-hexane containing 5 ng of
n-nonane, n-decane, and n-undecane.

n-

-nonane

n-decane

n-

undecane

area

count

retention
time, min

area

count

retention
time, min

area

count

retention
time, min

3646

21.68

3691

29.67

3893

37.84

3737

21.71

3779

29.69

3986

37.85

3750

21.71

3798

29.69

4012

37.85

Mean

3711

21.70

3756

29.68

3964

37.85

Std. dev.

56.7

0.017

57.1

0.012

62.6

0.0058

Rel. std. dev.

(%)

1.53

0.08

1.52

0.04

1.56

0.02

A stock solution of n-nonane, n-decane, and n-undecane was
prepared in n-hexane to a concentration of 5 ng//ul for each
component. This was used to obtain retention time data, and as a
stock solution for dilution to approximately ppb for other
experiments. Solutions of the same concentrations in n-hexane were
also prepared for 2-octanone, 5-nonanone, and 2-decanone. The
bonded phase FSOT column used was a DB-1 60 m x 0.32 mm i.d. ,
1.0 jam film, which is a nonpolar, hydrophobic SE-30 cross-linked
polymeric phase.

Table 2 summarizes the analytical data for three runs with a 1 /td
sample of the stock solution of n-nonane, n-decane, and n-undecane
(5 ng/jul each) using on-column injection. This provided retention
time and area count data for these components.

This stock solution was then diluted one hundredfold with n-hexane
(0.05 ng/jul for each component) and 100 /d of this solution injected.
During this operation, the column was not connected to the detector.
Liquid was observed bubbling from the column outlet at 6.0 minutes,
and it ceased at 16.2 minutes. At the 15.1 minute mark, the column
outlet flow was directed into a stainless steel trap immersed in liquid
nitrogen. Then the oven temperature was raised to 150°C to desorb
trace substances from the column. After 20 minutes, the oven was
cooled, and the column was reversed, that is the outlet of the trap was
connected to the injector, and the other end of the column to the
detector. Table 2 summarizes the analytical data for three runs.

The relative standard deviations observed in the 100 /jl\ injection
experiments were substantially larger than in the 1 fi\ data (Table 3).
Nevertheless, the recovery was more than 90 percent for each of the
standards.

Table 4 lists the analytical data obtained for three runs using the
stock solution containing 5 ng each of 2-octanone, 5-nonanone, and
2-decanone (1 /xl on-column injection). This was used to establish
retention time and area count data.

264

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 4, 1985

Table 3. Analytical data for 100 yul on-column injection of n-hexane containing 5 ng
of n-nonane, n-decane, and n-undecane.

n-

-nonane

n-decane

n-

undecane

area

count

retention
time, min

area

count

retention
time, min

area

count

retention
time, min

4017

21.88

4402

29.78

4473

37.87

2602

21.65

4096

29.73

3562

37.89

3843

21.90

3582

29.86

3516

37.85

Mean

3487

21.81

4027

29.79

3850

37.87

Std. dev.

Rel. std. dev.

772

0.14

414

0.065

540

0.020

(%)

Recovery (%)

22.1

0.64

94.2

10.3

0.22

107

14.0

0.053

97.1

Table 5 lists the analytical data for three runs of the same solution
diluted 100-fold, with 100 fi\ injected (5 ng of each component total).
Recovery exceeded 90 percent for each of the standards used.

The data reported for the hydrocarbons and ketones are at a
concentration level of approximately 50 parts per billion. However, by
operating the flame ionization detector at its maximum sensitivity
(with two to one signal to noise ratio), it would be feasible to analyze
for these substances below the part per billion level. With this
technique, cumbersome (and often nonquantitative) preconcentration
steps can be avoided in many analyses, for example, analysis of
priority pollutants. Alternatively, the use of concentration steps can
achieve sensitivity at the part per trillion level or even lower. The
sensitivity of this method also could be enhanced by the use of more
selective detection systems including electron capture, flame
photometry, or photoionization. We also felt that this work established
the potential feasibility of analyzing the trace impurities for one

Table 4. Analytical data for 1 /d on-column injection of n-hexane containing 5 ng of
2-octanone, 5-nonanone, and 2-decanone.

2-octanone 5-nonanone 2-decanone

area

count

retention
time, min

area

count

retention
time, min

area

count

retention

time, min

3110

27.02

3372

33.93

3023

43.43

3120

27.02

3415

33.93

3036

43.41

3134

26.99

3416

33.90

3051

43.38

Mean

3121

27.01

3401

33.92

3037

43.41

Std. dev.

12.1

0.017

25.1

0.017

14.0

0.025

Rel. std. dev.

(%)

0.39

0.063

0.74

0.050

0.46

0.058

GAS CHROMATOGRAPHIC TRACE ANALYSIS

265

Table 5. Analytical data for 100 jul on-column injection of n-hexane containing 5 ng
of 2-octanone, 5-nonanone, and 2-decanone.

2-octanone

5-nonanone

2-decanone

area

count

retention
time, min

area

count

retention
time, min

area

count

retention
time, min

4521

27.22

3836

34.10

3251

43.57

4103

27.17

3140

34.06

2714

43.54

3359

27.08

3223

33.99

2786

43.47

Mean

3994

27.16

3400

34.05

2917

43.53

Std. dev.

Rel. std. dev.

589

0.071

380

0.056

291

0.051

(%)

Recovery (%)

14.7

0.26

128

11.2

0.16

100

9.98

0.12

96

substance dissolved in another solvent. This would involve a dual
elimination — one for the solvent and the other for the principal solute.

In a second series of experiments, on-column injection of up to 250
/jl\ of n-pentane solutions of halogenated hydrocarbons was carried out
successfully utilizing an electron capture detector (ECD). Because n-
pentane does not respond to an ECD, it was eliminated from the
chromatogram (except for solvent impurities responsive to ECD). For
this work we sought to analyze both low-boiling and high-boiling
halogenated hydrocarbons using two stock solutions. The first
contained the low-boiling compounds dichloromethane (60 ppm),
chloroform (1.0 ppm), and the high-boiling compounds lindane (4 x
10 3 ppm), heptachlor (4 x 10 3 ppm) and endrin (4 x 10-3 ppm). This
was used to further prepare three solutions diluted with n-pentane by
factors of 50, 100 and 250. A second n-pentane stock solution,
containing only dichloromethane (60 ppm) and chloroform (1.0 ppm)
was prepared, from which three additional solutions were prepared,
diluted with n-pentane by factors of 102, 104, and 106.

Two bonded phase, fused silica capillary columns (FSOT) were
joined in series with a stainless steel trap using low dead-volume
unions. The first column (upstream) was 22 m x 0.32 mm i.d. , 5 jum
cross-linked methylsilicone film. The second column, also FSOT,
(downstream) was 25 m x 0.32 mm i.d., 0.30 jum cross-linked OV-17
film.

One set of experiments was conducted based on the first stock
solution, to observe the effect of sample volume injected. For this
purpose, 50, 100, and 250 /d of the three diluted solutions (50-, 100- ,
and 250-fold, respectively) were injected. This provided a constant
amount of solutions. Instrument conditions were: 40°C oven
temperature for 5 minutes followed by programming up at 2°C/min
(for 17 minutes in the case of the 50 and 100 /d samples, for 27

266

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 6. Analytical data for three runs: 1.0 /d of undiluted n-pentane stock solution
(area counts).

Dichloro-

methane

Chloroform

Lindane

Haptachlor

Endrin

Mean

4841

10,158

3525

3728

4166

Std. dev.

123

644

52.7

130

46.5

Rel. std. dev. %

2.5

6.3

1.5

3.5

1.1

Solution ppm

60

1.0

4 x 10~3

4 x 10 3

4 x 10'3

Table 7. Analytical data for three runs: 50 /d of 50-fold diluted stock solution (area
counts).

Dichloro-

methane

Chloroform

Lindane

Heptachlor

Endrin

Mean

3756

9340

4107

2752

2378

Std. dev.

421

257

20.0

103

187

Rel. std. dev.

11

2.7

0.49

3.7

7.8

Solute ppb

1200

20

0.08

0.08

0.08

Table 8. Analytical data for three runs: 100 /d of 100-fold diluted stock
counts).

solution (area

Dichloro-

methane

Chloroform

Lindane

Heptachlor

Endrin

Mean

4117

8510

4241

2579

3163

Std. dev.

182

175

145

245

70.5

Rel. std. dev.

4.4

2.1

3.4

9.5

2.2

Solute ppb

600

10

0.04

0.04

0.04

Table 9. Analytical data for three runs: 250 /d of 250-fold diluted stock solution (area
counts).

Dichloro-

methane

Chloroform

Lindane

Heptachlor

Endrin

Mean

4312

8903

3411

2570

2175

Std. dev.

1043

753

149

114

98.8

Rel. std. dev.

24

8.4

4.4

4.4

4.5

Solute ppt

240,000

4000

16

16

16

GAS CHROMATOGRAPHIC TRACE ANALYSIS

267

Table 10. Concentrations and absolute amounts of dichloromethane and chloroform
(100 m1)-

Dichloromethane Chloroform

Dilution ratio

cone.

abs. amt.

cone.

abs. amt.

102

600 ppb

60 ng

10 ppb

1.0 ng

104

6.0 ppb

600 pg

0.10 ppb

10 pg

106

60 ppt

6 pg

1.0 ppt

0.10 pg

minutes in the case of the 250 /d sample). The helium carrier gas flow
rate was 1.5 ml /min. Following these respective periods, the stainless
steel interconnecting trap was immersed in liquid nitrogen, and the
temperature raised to 190°C for a period of 27 min to thermally desorb
(and trap) the high-boiling compounds from the first column (thicker
bonded phase). At the end of that period, the oven temperature was
reduced to 70°C, the liquid nitrogen removed and the analysis resumed
programming the temperature up at 5°C/min to a maximum of
220°C. Carrier gas flow rate was maintained at 1.5 ml /min.

A second set of experiments was based on the second stock solution,
injecting 100 /d of the solutions diluted by factors of 102, 104, and 106.
Instrument conditions were as previously noted, except that the carrier
gas flow rate was increased to 7 ml /min. These experiments were run
to provide differing concentrations and information on solvent
impurities.

A third set of experiments consisted of the analysis of 1.0 /d of the
first stock solution (1.0 /id on-column injection of undiluted solution),
which were used to establish retention times and area counts as shown
in Table 6. Elution of n-pentane was monitored through the use of the
flame ionization detector.

Table 7 summarizes data for the 50-fold diluted stock solution (of
Table 6). Here, 50 /d were injected, on-column, providing the same
absolute amounts of each compound. Table 8 summarizes data for a
100 /d on-column injection of 100-fold diluted stock solution. Table 9
summarizes data for a 250 yul on-column injection of 250-fold diluted
stock solution. This dilution brings the concentration of lindane,
heptachlor and endrin into the part per trillion range.

Chromatograms were obtained for 100 /d on-column injected
samples of the diluted second stock solution (102, 104, and 106 fold),
which contained the low-boiling dichloromethane and chloroform
only in the n-pentane solvent (see Table 10)

We concluded that the analysis of both low-boiling and high-boiling
halogenated hydrocarbons is feasible at the part per trillion level using
this procedure. The use of n-pentane as solvent permits detection of
these solutes with ECD. We noted, however, that solvent purity

268

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

becomes highly important at such trace levels, and for more exacting
work, it would be necessary to improve the purity of the solvent,
probably through a chromatographic procedure.

LITERATURE CITED

Blomberg, L. , J. Buitjen, K. Markides, and T. Wannman. 1982. Peroxide-initiated in situ
curing of silicone gums for capillary column gas chromatography. J. Chromatogr.
239:51-60.

Sandra, P. , G. Redant, E. Schacht, and M. Verzele. 1981. In situ cross-linking of the
stationary phase in capillary columns. J. High Resolut. Chromatogr. 4:411-412.

Wang, F.-S., H. Shanfield, and A. Zlatkis. 1982. On-column injector for capillary gas
chromatography. An. Chem. 54:1886-1888.

STRENGTHS AND WEAKNESSES OF
DAMAGE ASSESSMENT PROGRAMS: THE IXTOC-I AND
BURMAH AGATE OIL SPILLS, AND THE BENTHIC
MACROINFAUNA OF THE TEXAS CONTINENTAL SHELF

by GEORGE S. LEWBEL

LGL Ecological Research Associates, Inc.

1410 Cavitt Street
Bryan, Texas 77801

ABSTRACT

The uncontrolled oil leak and fire at Ixtoc-I, a well in the Gulf of Campeche, Mexico,
released more than 475,000 metric tons of crude oil into the ocean in 1979-80. While
Ixtoc-I was still flowing, a collision and fire near Galveston Bay, Texas, released an
additional 7000-9000 metric tons of crude oil into the ocean from the oil tanker Burmah
Agate . Currents transported residues from both spills through an area of the Texas
continental shelf that had been thoroughly studied in previous years by researchers from
the South Texas Outer Continental Shelf (STOCS) program.

This paper discusses the subtidal macroinfaunal communities at 12 STOCS stations
sampled in 1979 and 1980, utilizing data from the earlier baseline study. In addition,
communities are described at 26 new stations within the STOCS study area, a station
at the Burmah Agate spill site, and another station 40 kilometers down-current from the
wreck, all sampled in 1980.

Temporal changes at the 12 STOCS sites and areal differences between the 38 stations
samped in 1980 within the STOCS area were not due to either spill, because no Ixtoc-
I or Burmah Agate residues were found in benthic samples. Major biological findings
at the STOCS sites included: 1) reductions in numbers of individuals and numbers of
taxa, and restrictions in depth range of taxa in 1979 and 1980 samples, compared to
1975-1977 samples; 2) positive correlation between sediment grain size and both
abundance and numbers of taxa; 3) minimal temporal changes in sediment grain size;
4) three clusters of stations arrayed parallel to shore; 5) a distinctive set of species
associated with coarser sediment; and 6) relatively constant proportions through time of
many numerically dominant taxa and groups of taxa, with polychaetes most abundant,
followed by amphipods, molluscs, sipunculids, and nemerteans.

The 1980 samples from all 38 stations within the STOCS area revealed 1) more
individuals and taxa at shallow, sandy stations than at deeper stations with finer
sediment; 2) four clusters of stations parallel to shore; and 3) similarities in large groups
of taxa, with polychaetes and gastropods most abundant in all four clusters.

Oil from the Burmah Agate was found in benthic samples adjacent to the wreck, and
tentatively identified at the station 40 kilometers away. Two of the 38 uncontaminated
stations sampled in 1980 served as a posteriori “control” sites for community
comparisons. However, because the accident occurred in a highly disturbed area,
differences between these four stations cannot be attributed unequivocally to the Burmah
Agate spill. Major biological findings included: 1) polychaetes and sipunculids most
abundant at both sets of stations; 2) the lowest numbers of individuals and taxa near
the spill; and 3) few taxa common to both sets of stations.

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

270

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

A number of problems in the damage assessment program are addressed in this paper.
Even if oil had affected the STOCS stations, this program might not have been able to
assign temporal changes to ‘their proper causes. The new stations within the STOCS
area might have been useful in mapping the extent of spills, and several of them were
valuable as a posteriori “controls.” Recommendations for improving oil spill damage
assessment programs include increasing the frequency of sampling; formation of
permanent, accessible voucher collections; acquiring life history and toxicological
information; and sequencing sample analysis so that chemical results are available before
biological sample analysis begins. Key words : oil spill, damage assessment, Ixtoc,
Burmah Agate, benthic, macroinfauna, continental shelf, Texas.

INTRODUCTION

Between 3 June, 1979 and 23 March, 1980, an uncontrolled leak
from Ixtoc-I, an exploratory well in the Gulf of Campeche, Mexico,
released more than 475,000 metric tons (140,000,000 gallons) of crude
oil into the gulf (Jernelov and Linden 1981). The oil apparently
weathered rapidly in the fire near the wellhead; about 30,000 metric
tons were deposited on Mexican beaches, but currents carried some
northward into United States waters. About 3900 metric tons of
weathered oil washed ashore on the Texas coast south of Port Aransas
from 6 August through the beginning of September 1979. Most of this
oil was then removed in mid-September by a tropical storm, although
considerable amounts of petroleum remained on the beach as “tar
mats” (Gundlach et al. 1981; Rabalais and Flint 1983). The quantity
of oil reaching the benthos is unknown, though Jernelov and Linden
estimated it by subtraction to be about 120,000 metric tons.

On 1 November, 1979, while Ixtoc-I was still pouring oil into the
Gulf of Mexico, the oil tanker Burmah Agate collided with the
freighter Mimosa seven kilometers from the mouth of Galveston Bay,
Texas. The Burmah Agate became grounded at the site with its decks
awash and its cargo of 57,000 metric tons (16,700,000 gallons) of crude
oil on fire. Approximately half the cargo was consumed by fire, but
7000-9000 metric tons spilled into the ocean and was carried mainly
toward the south and southwest by surface currents. About 300 metric
tons of oil washed ashore on the South Texas coast in November 1979,
reaching locations from Galveston to Mansfield Pass, 470 kilometers
from the wreck (Thebeau and Kana 1981). More than 6000 metric tons
were estimated (by subtraction) to have been dispersed offshore (Kana
etal. 1981).

Fortuitously, part of the area within United States waters believed
to be exposed to petroleum from the spills had been extensively studied
by researchers from the University of Texas between 1975 and 1977.
The South Texas Outer Continental Shelf (STOCS) program under the
sponsorship of the Bureau of Land Management (BLM)-— now the
Minerals Management Service (MMS)— -included a variety of ecological
measurements at a large number of subtidal sites, in order to meet

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

271

BLM’s needs for future impact assessment (Flint and Rabalais 1980a,
1981).

Practically, the circumstances could not have been expected to be
better for a damage assessment study. Both oil spills were massive.
Based upon the paths their slicks took, and because they deposited oil
on beaches inshore of the STOCS study area, it was considered likely
that they might have contacted STOCS sites, for which there was a
substantial amount of baseline data available. In one sense, a study of
the ecological effects of these spills could have served as the first test
of the utility of the baseline concept in the United States, because no
similar catastrophe had occurred in an area previously included in a
comprehensive baseline program.

Among the subtidal biological communities surveyed by the STOCS
investigators, the macroinfauna (defined as soft-bottom organisms
retained on a 0.5 mm screen) was characterized as showing stability of
species groupings and a lack of statistically significant seasonal
fluctuations within stations grouped by depth or by transect location,
although the variability in parameters measured (for example,
numbers of individuals) within stations was sometimes high (Holland
et al. 1980). In light of this apparent stability, and to capitalize on the
use by STOCS of repeatable, standardized methods for benthic
sampling, BLM designed and sponsored a study of macroinfaunal
communities during the Ixtoc-I and Burmah Agate spills (November
1979) and a year later (December 1980) to determine if any effects of
either spill on macroinfauna could be detected.

The 1979 samples were collected at 12 of the 34 STOCS sites by a
multi-agency oil spill response team (Woods and Hannah 1981). The
1980 samples were collected under contract to BLM by LGL Ecological
Research Associates, Inc. (LGL) and Energy Resources Company, Inc.
(ERGO), at the same 12 STOCS sites visited in 1979; at 26 new stations
in the STOCS sample area; and at two stations near the Burmah Agate
spill.

This paper describes the results of biological analyses of the 1979
and 1980 samples by LGL. The biological results are interpreted in the
light of petroleum hydrocarbon analyses by ERGO, and sediment
texture analyses by Geomet Technologies, Inc. The paper provides an
area-wide overview and general conclusions of the study; data for
individual stations may be found in Boehm et al. (1982), and Lewbel
et al. (1982a, 1982b, and 1982c). The paper is divided into four major
sections. The first section discusses findings for the 12 STOCS stations,
and compares LGL’s results to those of the STOCS program. The
second section discusses expanded sampling in 1980 within the STOCS
area. The third section concerns the two Burmah Agate stations. The
fourth section reviews the damage assessment process itself, and

272

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

specific recommendations are made for improving the quality of future
damage assessments.

STOCS STATIONS REVISITED

Methods

The 12 stations selected for sampling in November 1979 and
December 1980 were arrayed along four transects roughly
perpendicular to shore (Fig. 1). Transect and station numbers used in
the STOCS program were retained. At each station, six benthic
samples were collected with a 0.1 square meter Smith-Mclntyre grab
for biological and sedimentological analyses; an additional grab
furnished samples for chemical analyses. Biological samples were
sieved through a 0.5 mm screen on board. Biota were dyed with rose
bengal and preserved in five to 10 percent neutrally buffered formalin.
Chemical methods are summarized in Boehm et al. (1983). Sediment
texture was determined as described by Folk (1980) and represented
following Buchanan and Kain (1971).

Laboratory analyses included identification and enumeration of most
individuals to the lowest possible taxon. Due to limitations on
resources and inherent taxonomic difficulties with some groups of
organisms, not all taxa received equal attention. Most STOCS
taxonomic voucher specimens were not available to LGL. All species
identifications were independently verified by specialists, including
some personnel who had identified STOCS samples, to ensure
comparability with the baseline research. Whenever possible, taxa were
referred to STOCS identifications to avoid spurious appearances or
disappearances in the data set caused by changes in taxonomy.
Organisms collected but known not to be macroinfauna (for example,
planktonic taxa and fishes) were not included in statistical analyses. A
reference collection of 1979 and 1980 specimens was deposited in the
United States National Museum. STOCS data were supplied to LGL
on magnetic tape. Data from this program were stored in standardized
format at the Environmental Data and Information Service, National
Oceanic and Atmospheric Administration, Washington, D.C.

It was necessary to reduce the number of taxa during data analysis
to statistically tractable and conceptually manageable proportions. The
majority of the 576 macroinfaunal taxa identified were rare; for
example, 249 taxa were represented by five or fewer individuals.
Statistical analyses focused on numerically dominant taxa. For analyses
of the entire data set, numerically dominant taxa were arbitrarily
defined as those represented by at least 0.2 percent of the total number
of animals (65,166) — 130 individuals. Use of this cutoff retained 72
taxa and 56,584 individuals (87 percent of the total). A one-percent

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

273

N-4 IV-5 1-2

1-4 Ml -4 1-1 11-1 11-1 IV— 1 11-4 HI-5 11-2

I - i - * - g - - 1 - 3 - 1

10 15 20 25 30 35 40 45 50

STATION DEPTHS I ml

Figure L Locations of 12 STOCS stations surveyed in this program. Depths are shown
along bottom scale.

cutoff was used in analyses for individual sampling periods or stations,
a variable restriction requiring that a taxon be represented by at least
one percent of the individuals in each data subset. Cutoff levels are
denoted in figure legends when appropriate.

Due to changes in sampling schedules and numbers of replicates in
the first year of the STOCS program, statistical analyses were restricted

274

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

to 1976 and 1977 winter (January-February) and fall (September-
October) STOCS samples, and November 1979 and December 1980
samples. The 1979 and 1980 samples thus were taken in months falling
between STOCS fall and winter samples. The six periods compared
were (in chronological order) winter 1976, fall 1976, winter 1977, fall
1977, November 1979, and December 1980.

Statistical analyses included comparisons between sampling periods
within stations, and comparisons between stations and groups of
stations within sampling periods. Replicates were not combined for
any analyses. Cluster analyses using the Czekanowski Quantitative
Index .(= Bray-Curtis) (Bloom 1981) with complete linkage (Boesch
1977) were used to elucidate groupings of taxa, stations and time
periods, and sediment types. The biological data were non-normally
distributed and highly heterogeneous, leading to significant interaction
terms in parametric analyses of variance (ANOVA). Consequently, tests
for central tendency were restricted to nonparametric procedures such
as the Friedman and Kruskall-Wallis ANOVA’s (Friedman 1937,
Kruskal and Wallis 1952) followed by Nemenyi’s multiple range test
(Nemenyi 1963). Kendall’s Tau (Kendall 1938) was used to test for
correlations between animal abundances, sediment texture indices, and
total organic carbon (TOC). Data from all six sampling periods were
included in correlation analyses. Community summary statistics
included the maximum likelihood estimate of diversity as H’ and a
scaled measure of evenness as V’ (Fager 1972; Pielou 1977). Numerical
abundance data are presented as actual numbers of individuals
collected, rather than as individuals per square meter.

Results

Because none of the sediment samples contained detectable traces of
Ixtoc-I or Burmah Agate oil (Boehm et al. 1982), the following results
should not be interpreted as pre-, mid-, or post-spill findings.

The numbers of taxa identified at all 12 stations taken together
changed markedly with time (Fig. 2). Many taxa were present in more
than one sampling period (Fig. 3), though most were collected only
once (201 taxa) or twice (116 taxa). Statistically significant differences
(. P < 0.05) in numbers of taxa separated 1979 and 1980 samples (few
taxa) from fall 1976 and winter and fall 1977 samples (many taxa).

The numbers of individuals changed sharply from one sampling
period to the next, with statistically significant differences ( P < 0.05)
separating 1979 and 1980 (low abundances) from fall 1976 and winter
and fall 1977 samples (high abundances). Winter 1976 samples were
intermediate between the two groups and did not differ significantly
from either.

NO. TAXA NO. INDIVIDUALS

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

275

igpoo

18,000

17000

16,000

15,000

14,000

9,000

8,000

7000

6,000

5,000

4,000

3,000

0

360

320

280

240

200

160

120

0

.72

.70

.68

.64 _
•62 _
4.0 _
3.8 _

I

3.6 _
3.4 _
3.2 _

3.0 J

WINTER FALL WINTER FALL NOV DEC

1976 1976 1977 1977 1979 1980

SAMPLING PERIOD

Figure 2. Community summary statistics for all 12 STOCS stations together, by
sampling period.

276

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 3. Number of macroinfaunal taxa occurring in one or more sampling periods

at 12 STOCS stations.

Neither H’ nor V’ showed clear, consistent changes with either
numbers of species or individuals for all stations taken together. For
example, H’ was highest when both numbers of species and
individuals were at intermediate levels (fall 1977), and V’ was highest
when both numbers of species and individuals were at their second-
lowest values (1979). H’ and V’ were at their lowest in 1980, though,
when both numbers of species and individuals were also at their
lowest.

Substrate type was an important correlate of species abundance.
Within stations, sediment texture indices were fairly constant from one
time period to the next (Fig. 4). Although there was some variability —
notably for Stations IV-5 and II-4 in 1980 — the differences were not
statistically significant. The abundances of most numerically dominant
taxa were positively correlated ( P < 0.05) with relative proportions of
larger particles (Table 1). Of 72 numerically dominant taxa, the
abundances of 41 were positively correlated with mean grain size; 49
with the ratio of sand to mud; and 49 to percentage sand. The
abundances of 55 taxa were negatively correlated with percentage silt

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

277

•-DEC 1980
OWOV 1979
*fall 1977
-^Winter 1977
♦ fall 1976
$ -Winter 1976

Figure 4. Sediment texture at 12 STOCS stations, by sampling period within stations.

278

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 1. Results of correlation analysis of abundance of numerically dominant taxa (0.2
percent cutoff) with sediment parameters. S:M = ratio sandimud (mud = silt + clay
+ fines); fines = particles < 0.001 mm; TOC = total organic carbon; + or _ =
statistically significant positive or negative correlation (P < 0.05); no symbol = no
statistically significant correlation.

Taxon

Correlate

Mean

grain

size

Percent

sand

Percent

silt

Percent

clay

Percent

fines

S:M

TOC

Abra aequalis

+

+

-

-

+

-

Aedicira belgicae

+

+

+

—

+

Aglaophamus circinata

+

+

—

-

+

Aglaophamus verrilli

+

+

—

-

—

+

—

Ampelisca abdita

+

Ampelisca agassizi

Ampelisca cf. cristata

+

+

—

—

+

Ampelisca sp.

+

+

-

-

+

Ampelisca verrilli

+

+

—

Anadara transversa

+

+

-

-

+

Apoprionospio pygmaea

+

+

-

—

-

+

Apseudes sp. A

+

Aricidea jeffreysii

+

+

—

—

+

Aricidea taylori

+

+

-

-

-

+

Aricidea ivassi

+

+

-

-

+

Armandia maculata

Asychis elongata

+

-

-

+

Asychis sp.

+

—

+

-

+

Caecum pulchellum

-

+

—

+

Clymenella torquata

+

+

—

-

+

Corbula szviftiana

+

—

Cossura delta

+

Diopatra cuprea

+

+

-

—

-

+

-

Diplodonta cf. soror

+

+

-

-

+

Eudorella monodon

-

-

+

+

+

—

Hyala sp. A

-

-

+

+

lsolda pulchella

+

+

—

—

+

Listriella barnardi

+

+

-

-

+

—

Litocorsa stremma

+

+

-

-

+

-

Lucina amiatus

+

+

-

-

+

-

Lumbrineris cruzensis

-

-

+

Lumbrineris sp. nov.

+

+

-

-

+

-

Lumbrineris ernes ti

+

+

-

-

+

Mage Iona longicornis

-

-

+

+

+

-

Magelona pettiboneae

+

+

—

-

+

Magelona phyllisae

+

+

—

—

—

+

—

Magelona rosea

+

+

+

—

Maldanidae (misc. unid.)

+

+

—

-

+

Mediomastus calijorniensis

+

+

-

-

+

Minuspio cirrifera

+

-

Natica pusilla

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

279

Table 1. Continued.

Nemertinea (misc. unid.)

+

+

-

-

+

-

Nephtys incisa

-

-

+

+

+

—

+

Nereis micromma

+

+

—

—

+

—

Ninoe nigripes

-

-

+

+

+

-

+

Notomastus cf. latericeus

+

+

Nuculana acuta

+

+

-

-

+

Onuphis sp.

+

+

-

-

+

Ophiuroidea (misc. unid.)

+

+

-

—

+

Ostracoda (misc. unid.)

+

+

—

-

-

+

Paleanotus heteroseta

+

+

-

-

+

Paraonidae (misc. unid.)

+

-

+

-

+

Paraonides lyra

+

+

-

-

+

Paraonis gracilis

+

—

Paraonis sp. A

+

Paraprionospio pinnata

+

—

—

Phoronida (misc. unid.)

+

+

-

-

+

Photis macromanus

+

-

Prionospio cristata

+

+

—

—

—

+

Prionospio steenstrupi

+

+

-

-

+

Protankyra benedeni

+

Sigambra tentaculata

+

+

-

+

—

Sipuncula (misc. unid.)

+

+

—

—

—

+

-

Spionidae (misc. unid.)

+

+

—

Spiophanes bombyx

+

+

—

—

+

Tellina versicolor

+

+

-

—

+

Terebellides stroemii

+

+

-

-

+

Tharyx annulosus

-

+

-

+

Tharyx marioni

+

+

-

-

+

-

Vitrinella floridana

Xenanthura brevitelson

—

—

+

—

Zoantharia (misc. unid.)

-

-

+

or clay, or both; 57 with percentage of fines (particles < 0.001 mm in
size); and 15 with TOC.

A few taxa showed an inverse of this pattern, their abundances being
positively correlated with fine particles and negatively correlated with
coarser particles. Four taxa (the cumacean, Eudorella monodon, the
magelonid polychaete, Magelona longicornis, the nephtyid polychaete,
Nephtys incisa, and the lumbrinerid polychaete, Ninoe nigripes)
showed negative correlations with mean grain size, with the ratio of
sand to mud, and with percentage sand; and positive correlations with
percentages of clay, silt, and fines. The gastropod Hyala sp. A was
negatively correlated with mean grain size and percentage sand, and
positively correlated with percentage clay. Nephtys incisa and Ninoe
nigripes were the only taxa in which abundance was positively
correlated with TOC.

Most taxa were present throughout the study region (Table 2).
Ubiquitous taxa included surface deposit-feeding polychaetes

280

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 4, 1985

Table 2. Presence (+) of numerically dominant macroinfaunal taxa at 12 STOCS
stations, by station and sampling period (0.2 percent cutoff). Taxa are listed vertically
in order of increasing average depth of collection. Stations are listed from left to right
in order of increasing depth. Within each station, sampling periods are shown in

STATION

TAXON

1-4

RCLROPHRMUS VERE ILL I
XENflNTHURR BREVITELSOH
PH0R0N I DR (.fllSC. )

LUC I NR RNIRNTUS
ISOLDR PULCHELLR
PRLERNOTUS HETEROSET R
NRT_ICfl PUSILLfl
PRQTRNKYRfl BENEDENI
NRGELONR PETTIBONERE
CLYHENELLfl TORQURTR
RGLROPHRMUS CIRCIHRTR
RBRR REQURLIS
NRLDflN I DRE (NISC.)

RNPEL I SCR CF. CRISTRTR
RMPEL I SCR SP.

L I T0C0RSR STRENMR
SP I OPHRNES BOHBYX
LUNBRINERIS CRUZENSIS
DIPL0P0HTR CF. SOROR
CRECUN PULCHELLUM
TELL I NR VERSICOLOR
RP0PRI0N0SPI0 PYGNRER
TEREBELL I DES STROEMI I
PRI0N0SPI0 CRISTRTR
PHOT I S NRCRONRNUS
RNRDRRR TRRNSVERSR
RRICIDER TRYLORI
LUNBRINERIS ERNESTI
0STRRC0DR (NISC. )
LISTRIELLR BARNARD I
ZOANTHRR I R (NISC.)

DIOPRTRR CUPRER
NUCULRNR RCUTR
NRGELONR PHYLLISRE
PRRRON I DES LYRR
PRI0N0SPI0 STEENSTRUP I
RSYCHIS SP.

NINUSPIO CIRRIFERR
RRICIDER JEFFREYSI I
RRICIDER URSSI
NEREIS NICRONNfl
LUNBRINERIS SP. NOV.
RRNRNDIR HRCULRTR
VITRINELLR FLORIDRNR
RSYCHIS ELONGRTR
ONUPHIS SP.

S I PUNCULfl (NISC. )

SIGRNBRR TENTRCULRTR
NED I ONRSTUS CRL I FORM I ENS 1 S
RNPEL I SCR RGRSSIZI
THRRYX NRRIONI
REDICIRR BELGICRE
THRRYX RNNULOSUS
OPHIUROIDER ( N I SC . )
NEMERTINER (NISC. )

SP I ON I DRE (NISC. )
PRRRPRIONOSPIO PINNRTR
NRGELONR ROSER
PRRRON I DRE (NISC. ;

PRRRON 1 S GRRCILIS
COSSURR DELTR
RNPEL I SCR VERRILL1
NOTONRSTUS CF. LRTERICEUS
NINOE N1GRIPES
RNPEL I SCR RBDITR
NEPHTYS INCISR
NRGELONR L0NGIC0RN1S
RPSEUDES SP. fl
CORBULR SMI FT I RNR
HYRLR SP. fl
PRRRON I S SP. fl
EUDORELLR N0N0D0N

♦ 4 4 4 4 4
444444
4 44 444
4 44 4 4

444 4
444 + -+ +

III - 4

4 4 4444

444

IV - 4 |-1 ||-1 HI-1

44 4444
44 44

4 4 4 4 4

4444
444
444 4

4 4 4 4

44444

4 4 4 4 4-

4- 4- 4- 4- 4-
444444

444444
4- 444

444444

4444

44444

444444

44+4+4

444444

4-4-4 4

44444

444444

4 4

4 4

444444
444 4

44 4

444444

444 4

444444

444444

444444

44

444444
4 4 4 4
444444
44444

4 44

44 44

444

444444

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

281

chronological order as vertical columns (for example, Aglaophamus verrilli was
collected at the shallowest stations, and was at 1-4 in the first five sampling periods,
at III-4 and IV-4 in all sampling periods, and at IT only in the sixth sampling
period).

STATION

TAXON

IV- 1

RGLRQPHRMUS VERRILL I
XENRNTHURR BREVITELSON
PHORON I DR (MISC. )

LUC INR RMIRNTUS
ISOLDS PULCHELLR
PRLERNOTUS HETEROSETR
NRTICR PUSILLR
PRQTRNK YRR BENEDENI
MRGELQNR PETTI BONERE
CLYMINILLR TQRQURTR
RGLROPHRMUS CIRCIHRTR
RBRfi REQURLIS
MRLDRN I DRE (MISC. >

RMPEL I SCR CP. CRISTRTR
RMPEL I SCR SP.

LITOCQRSR STREMMR
SP I QPHRNES BOMBYX
LUMBR I HER I S CRUZENSIS
D I PLODONTR CF. SOROR
CRECUM PULCHELLUM
TELL I HR VERSICOLOR
RPOPR I QNOSP I 0 PYGMRER
TEREBELL IDES STROEMI I
PRIQNOSPIO CRISTRTR
PHOT I S MRCROMRNUS
RNRDRRR TRRNSVERSR
HR I Cl PER TRYLORI
LUMBR I HER I S ERHESTI
OSTRRCODR (MISC. )
LISTRIELLR 1RRNRRD I
ZORHTHRR I R (RISC. )

DIOPRTRR CUPRER
HUCULRHR RCUTR
MRGELQNR PHYLLISRE
PRRRONIDES LYRR
PRI0H0SPI0 STEENSTRUP I
RSYCHIS SP.

MINUSPIO CIRRIFERR
RRICIDER JEFFREYS I I
RRIC1DER HRSSI
NEREIS MICROMMR
LUMBR I NER I S SP. NOV.
RRMRNDIR MRCULRTR
VITRINELLR FLOP I DRHR
RSYCHIS ELQNGRTR
ONUPHIS SP.

S I PUNCULR (MISC. )

SIGRMBRR TENT RCULRT R

MED I OMRSTUS CRL I FORM I ENS I S

RMPEL S SCR RGRSSI2I

THRRYX MRRIQNI

REDICIRR BELGICRE

THRRYX RNNULOSUS

OPHIUROIDER (MISC.)

NEMERT I NER (MISC.)

SP I ON I DRI (MISC. )
PRRRPRIONOSPIO PINNRTR
MRGELQNR ROSER
PRRRON I DRI (MISC. )

PRRRONIS GRRC1LIS
COSSURR DELTR
RMPEL I SCR VERRILLI
NOTOMRSTUS CF. LRTERICEUS
NINOE NIGRIPES
RMPEL I SCR RBDITR
NEPHTYS INCISR
MRGELQNR LONGICORNIS
RPSEUDES SP. R
CORBULR SHIFT! RNR
HYRLR SP. R
PRRRONIS SP. R
EUDORELLR MONODON

4444 4

4444

4 4 4* 4- 4*
44444

444

44 4 4

4*4 4* 4

44

444444
44 444

4.4.4.

444444
444444
4-4' 4

4444

4. 4. 4. 4. 4. 4.

44 4

4 4 4 4s
444
4444s
4444

11-4 IV-5 111-5

44444
444 4

4 4 4

444444
4444 4

444444

444444

4444

444 44

444444

444444

*44444
f444
» 444 4

°44444

4444

4444 4

444444
4444
44444
4 4 4

44

444444

444444

1-2 11-2

4 4 4 4 H
4444

4444

4444

282

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

(Magelona phyllisae, M. longicornis, M. rosea, Tharyx marioni, and
numerous spionids, especially Paraprionospio pinnata ); subsurface
deposit-feeding polychaetes ( Paraonis gracilis, Aricidea jeffreysi, A.
taylori, Mediomastus californiensis); omnivorous and carnivorous
polychaetes ( Nephtys incisa, Lumbrineris ernesti, Lumbrineris sp.
nov.); tubicolous amphipods ( Ampelisca agassizi, A. abdita, A. verrilli);
and sipunculids and nemerteans.

A relatively large suite of numerically dominant taxa was restricted
to the shallowest stations, with an apparent faunal break between the
stations at 15 meters and those at 18 meters. It is thought that many
such cases were due to differences in preferences in sediment type,
which generally varied with depth. One of the deeper stations (IV- 1,
depth 27 meters) — which had anomalously coarse sediment — had many
species in common with the three shallowest stations (1-4, III-4, and
IV-5, depth 10-15 meters). Fifteen numerically dominant taxa were
found only at these four stations, all of which had fairly coarse, sandy
sediment.

At the other end of the scale, three taxa — the gastropod, Hyala sp.
A, the paraonid polychaete, Paraonis sp. A, and the cumacean,
Eudorella monodon — found primarily at the muddy, deeper stations
were rare or absent from the shallowest stations. Other than these, no
clearly defined set of taxa was associated with the deeper stations. The
majority of taxa common at the deepest stations also were present at
the shallowest stations.

There was an apparent relationship between overall abundance and
the areal distribution of numerically dominant taxa. During the three
sampling periods when total abundance and numbers of taxa were
highest, most numerically dominant taxa spanned the entire depth
range. In fall 1976, 40 taxa were present at the shallowest stations (10-
15 meters) as well as at the deepest stations (42-49 meters), and nine
taxa were collected at all 12 stations. Comparable figures for winter
and fall 1977 were 33 and 38 taxa spanning the depth range, with 11
and five taxa at all 12 stations, respectively.

During the three sampling periods with fewest taxa and individuals,
restrictions in distribution were evident. In winter 1976, 25 numerically
dominant taxa spanned the depth range and three taxa were present
at all 12 stations. The 1979 and 1980 samples appeared markedly
different. In 1979 and 1980, only 17 and 11 taxa, respectively, spanned
the depth range, and in both years only a single taxon was present at
all 12 stations.

The relative proportions of individuals of numerically dominant
taxa remained fairly constant over time in the entire study area (Fig.
5). The polychaetes Magelona phyllisae, Lumbrineris sp. nov., and
Paraprionospio pinnata, and miscellaneous unidentified sipunculids

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

283

%

Figure 5. Relative proportions (in percent) of numbers of individuals of numerically
dominant macroinfaunal taxa for all 12 STOCS stations together, by sampling
period ( one percent cutoff).

and nemerteans were important species. The relative proportions of
each major group of taxa were also rather stable (Fig. 6). Deposit¬
feeding and omnivorous polychaetes were most important, followed by
amphipods, molluscs, sipunculids, and nemerteans. Polychaetes were
numerically dominant in every sampling period, with deposit feeders
ranked first, followed by errant omnivores and carnivores, suspension
feeders, and those of undetermined feeding type. Amphipods were the
largest group of crustaceans, and pelecypods outnumbered gastropods
in all periods but one.

However, large fluctuations in the abundance of any given taxon
and even of the major groups of taxa were common at most stations
from one time period to the next. Notable examples are the
lumbrinerid polychaete, Lumbrineris cruzensis, collected in large
numbers only in 1979; the gastropod, Natica pusilla, which
“bloomed” in 1980; and the holothuroid, Protankyra benedeni,
collected only in 1980.

284

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

%

40

DEC 1980 20

40

NOV 1979 20

40

FALL 1977 20

40

WINTER 1977 20

40

FALL 1976 20

40

WINTER 1976 20

. I- 1 .

41

r

. . r-, . . m .

•

n . m . r —

49

. n . r-. . r— , . . .

■

i — i . i i . 1 1 . i — i .

51

,1

n

1 . — . . i — i . — , , — . .

■

i — i . i — i . i i . , — . .

53

. i i . . i — i .

i

50

. n .

[ 1 L 1 - 1 ± I 1

57

.1

n

n.

/

¥///////////^

<P A & A 40

cf qv & Jy

<<? # S &

£ £

J? £ £ £•

'///

£ <P

Figure 6. Relative proportions (in percent) of numbers of individuals of major groups
of macroinfaunal taxa for all 12 STOCS stations together, by sampling period.

Relatively clear patterns of association between stations were
produced for each sampling period using cluster analysis based upon
abundance of numerically dominant taxa, and then inverted to group
stations. Stations were generally clustered into three major groups by
use of a minimum distance measure of 0.80 (Fig. 7). The nearshore
cluster included stations having many broadly-distributed taxa as well
as those taxa restricted to sandy, nearshore sites. The offshore cluster
included stations having broadly-distributed taxa, and those few taxa
associated with deep water. The intermediate cluster included stations

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

285

WINTER 1976

TRANSECT

TRANSECT

TRANSECT

TRANSECT

TRANSECT

TRANSECT

TRANSECT

TRANSECT

Figure 7. Schematic map of station groupings derived from cluster analysis based on
abundance of numerically dominant macroinfaunal taxa at 12 STOCS stations (0.2
percent cutoff), by sampling period.

286

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

having broadly-distributed taxa and a reduced number of nearshore-
associated taxa not found at the deeper stations.

Discussion

The reasons for revisiting the STOCS stations during and after the
Ixtoc-I and Burmah Agate spills were to (1) compare pre-spill with
mid-spill and post-spill biological communities; and, should the data
from the chemical analyses permit, (2) determine if observed biological
differences were correlated with the presence of Ixtoc-I or Burmah
Agate residues.

With regard to the first objective, there is no question that major
changes have occurred through time at the 12 study stations. Three
successive sampling periods characterized by high infaunal abundance
and numbers of taxa followed one with lower abundance and numbers
of taxa within a two-year time span, and were again followed by two
sampling periods that resembled the first one. The changes were area¬
wide, affecting all 12 stations; whatever factors influenced these aspects
of community structure thus acted without regard to depth or location.

The changes were not only area-wide but also pan-phyletic. Most
taxa showed parallel changes in abundance while retaining their
relative numerical importance (percentage of total individuals per
sampling period), indicating that the differences between sampling
periods were not due simply to changing abundances of a few
particularly common organisms. In addition, larger groups of taxa
more or less retained their relative positions with respect to one
another over time.

With regard to the second objective, although both slicks are known
to have transited South Texas continental shelf waters before being
beached, no oil from either spill was detected in samples from the 12
STOCS sites. Ixtoc-I oil was identified in samples collected with towed
sorbent pads inshore of the southernmost transect (IV) in November
1979, and was thought to be associated with suspended particulate
matter (Boehm et al. 1982). However, the nature and source of the
particulate matter were not determined, and it was not possible to
assess the concentration of Ixtoc-I oil in the water. There is no
evidence that this matter was derived from, or associated with, the
benthic community. It is possible that the contaminated material
might have been suspended in either the nepheloid layer (a persistent
high-turbidity layer of water near the bottom in the northern Gulf of
Mexico) or in flocculent surficial sediment from the bottom. Under
some conditions, spilled crude oil can be preferentially adsorbed and
reconcentrated on fine sediment (Beslier et al. 1980).

If flocculent material overlying surficial sediments was contami¬
nated, it would be necessary to assume that the Smith-Mclntyre grab

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

287

did not adequately sample the upper benthic layers for newly-deposited
oil. The Smith-Mclntyre grab is not the ideal tool for benthic chemical
sampling (Boehm and Lewbel 1982). The weather on both 1979 and
1980 sampling cruises was frequently rough and the winch operators
were not always able to control precisely the rate of fall of the grab.
It is possible that a pressure wave from the grab displaced flocculent
materials from the bottom before striking, although it is unlikely that
large amounts of hydrocarbons from the spill would have left no
detectable traces in the sediments.

Even if one were to assume — without sufficient evidence, in this
case— that oil from one or both spills did contact benthic communities
at the study sites, it is by no means certain that permanent alterations
would have resulted, and it is likely that any impacts would have
affected various taxa differently. For example, in the Amoco Cadiz
spill, the benthic communities in fine sediments were more seriously
affected than those in coarse sediments, with some taxa (ampeliscid
amphipods, for example) much more vulnerable than others (Cabioch
et al. 1980). In the Antonio Gramsci spill, a large benthic area was
exposed to crude oil; several groups of organisms were unaffected (for
example, polychaetes and nematodes), and no major changes in
community structure were observed (Bonsdorff 1981).

The frequency of sampling in this program was inadequate to
delineate short-term community changes. Several previous studies of
oiled subtidal sites have shown that although impacts on infaunal
communities may occur rapidly, offshore sites may not show obvious
signs of major damage only one year after a spill. For instance, in the
barge Florida spill, faunal changes in both the oiled and unoiled
offshore stations were “relatively slight and seasonal in nature,” with
recovery in density, number of species, and diversity after one year at
oiled stations (Sanders et al. 1981). On the other hand, it has been
argued that long-term, permanent community modifications should be
the focus of damage assessment studies, inasmuch as resilience
probably is of greater concern than passing fluctuations (Lippincott
and Wolfson 1981). If so, the sampling frequency in this program
should have been adequate to document such changes.

EXPANDED SAMPLING WITHIN THE STOCS AREA

Methods

Benthic samples were collected during the December 1980 cruise at
26 stations not previously sampled for macroinfauna. These “new”
stations were scattered throughout the region in which the 12 STOCS
stations were located (Fig. 8). Fifteen of the new stations lay shoreward
of the shallowest of the 12 STOCS stations, providing collections from

288

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 8. Locations of 12 STOCS stations (large dots, capital letters) and 26 “new”
stations sampled in 1980 (small dots, lower case letters). Key to station numbers at
right.

a depth range (4.5-9 meters) not sampled in earlier programs. The new
stations were intended to make mapping of areas impacted by oil more
complete. The samples were therefore viewed as a synoptic collection
from 38 stations. Field and laboratory methods for the new stations
were identical to those described for the STOCS stations in the
preceding section of this paper.

To define a subset of numerically dominant taxa for statistical
procedures involving all 38 stations, an arbitrary minimum cutoff of
one percent of the total number of individuals was used. This cutoff

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

289

level reduced the number of taxa to 18, and was selected in the
interests of consistency with previous analyses of the STOCS stations,
which also retained 18 taxa with a one-percent cutoff (see Fig. 5).
Presence /absence descriptions benefited from the inclusion of a larger
number of taxa; a 0.1 percent cutoff retained 88 taxa, a group size
similar to the subset with a 0.2 percent cutoff (which included 72
numerically dominant taxa) used for presence /absence analyses of the
STOCS stations (see Table 2). Methods for statistical analyses were as
described for the STOCS stations. Replicates were not combined for
any analyses. Numerical abundance data are presented as acutal
numbers of individuals collected.

Results

Because none of the sediment samples contained detectable Ixtoc-I or
Burmah Agate oil (Boehm et al. 1982), the following results should not
be interpreted as post-spill findings. Samples from all 38 stations
included 222 taxa comprised of 15,646 individuals. Most taxa were
rare. For example, 87 were represented by five or fewer individuals,
and 41 by only a single individual. There were far more individuals
collected at stations shallower than 20 meters than at deeper stations.
Among the shallower stations the largest numbers of individuals were
collected at southerly locations (Fig. 9). More individuals were
collected per station among the 26 new stations than among the 12
STOCS stations (473 as opposed to 279), primarily because the new
stations included many locations in shallow water and to the south.

Cluster analysis based on relative abundance of taxa showed four
distinct groups of stations at a similarity level of 0.85 (Fig. 10). Three
of the clusters matched the nearshore, intermediate, and offshore sets
of stations recognized for the 12 STOCS stations (see Fig. 8), adding
13 new stations to clusters formed by old stations. A fourth cluster of
13 stations (all new) was located in shallower water.

Faunal differences between clusters were evident (Table 3). The 13
shallow stations and the 13 nearshore stations both had large numbers
of taxa and individuals, whereas the eight intermediate stations and
four offshore stations showed progressively fewer taxa and numbers of
individuals. The decrease with distance from shore was not strictly due
to including fewer stations in the clusters, because the number of
individuals per station in intermediate and offshore clusters also
decreased to one-fifth and one-tenth (respectively) of the values in the
shallow and nearshore clusters. Diversity and evenness were highest in
shallow and offshore clusters. The most abundant organisms in all
clusters were carnivorous and omnivorous polychaetes, deposit-feeding
polychaetes, and gastropods (Fig. 11).

290

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 9. Schematic map of macroinfaunal abundance within STOCS area in 1980.
Solid dots indicate 12 STOCS stations; stippled dots indicate 26 “new” stations.

The shallow cluster of stations (4.5-9 meters deep) had sandy
sediment, for the most part (Fig. 12). The numerically dominant
organism was the spionid polychaete, Apoprionospio pygmaea,
followed by the lumbrinerid polychaete, Lumbrineris sp. nov., and the
naticid gastropod, Natica pusilla (Fig. 13). Seven numerically
dominant taxa were found exclusively in the shallow cluster (Table 4).
However, most taxa at the shallow stations also were present at the
nearshore stations.

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

291

TRANSECT

Figure 10. Schematic map of station groupings derived from cluster analysis based on
abundance of macroinfauna within STOCS area in 1980.

The stations in the nearshore cluster (9-27 meters) had sediment
ranging from sand to roughly even mixtures of sand, silt, and clay.
Seven numerically dominant taxa were found only at stations in the
nearshore cluster. The most abundant organism was Natica pusilla,
followed closely by the spionid polychaete, Paraprionospio pinnata,
the magelonid polychaete, Magelona phyllisae, Lumbrineris sp. nov.,
and the holothuroid, Protankyra benedeni. Protankyra was found in

292

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 3. Community summary statistics by station cluster.

Cluster

Shallow

Nearshore

Intermediate

Offshore

No. of stations

13

13

8

4

No. of individuals

7437

7227

742

227

No. of individuals /station

572

556

93

57

No. of taxa

144

177

59

49

Mean no. of taxa /station

11

14

7

12

Mean no. of ind. /taxon

52

41

13

5

Diversity (H’)

3.27

3.16

2.65

3.21

Evenness (V’)

0.64

0.59

0.59

0.73

large numbers at only two stations (N-18 and 1-1). Approximately half
the numerically dominant taxa were present only at shallow and
nearshore stations.

The stations in the intermediate cluster (25-55 meters) had silty-clay
sediment. The most abundant organisms were Paraprionospio pinnata
and the nephtyid polychaete, Nephtys incisa. All of the numerically
dominant taxa in the intermediate cluster also were found in other
clusters.

Fine sediment with approximately equal proportions of silt and clay
characterized the stations in the offshore cluster (42-55 meters), with
the exception of the shallowest (N-39), which had an added proportion
of sand. The most abundant organism was Nephtys incisa, followed
by miscellaneous unidentified nemerteans and by Paraprionospio
pinnata. All of the numerically dominant taxa present at offshore
stations also were found at stations in other clusters; however, several
taxa present in samples from other clusters were not found in the
offshore cluster. The most abundant of these was the capitellid
polychaete, Mediomastus californiensis. Four taxa found in the
offshore samples were absent from the shallow cluster, though they
were found in nearshore and intermediate clusters; these were the
alpheid shrimp, Alpheus sp. A., the magelonid, Magelona longicornis,
the lumbrinerid, Ninoe nigripes, and the cossurid polychaete, Cossura
delta. The largest fraction of uncommon or rare taxa also was
identified in samples from the offshore cluster of stations; 71 percent
of the individuals are grouped in Figure 13 as “other,” indicating that
the taxa to which they belonged did not meet the minimum cutoff of
one percent.

Discussion

Expanding the sampling in 1980 to include 26 new stations revealed
patterns that would not have been evident at just the 12 STOCS
stations. Of course, the 26 new stations had more taxa (208 as opposed

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

293

20

uj 40

t-

<

Q

£ 20
cc

40

40

20

. „ . m .n .

, , f - 1

. — ■ . . — ■ . 1 — 1 .

• i - 1 . i - 1 . - - - . . i - 1 ,

- ^ — 1. .r~~l 1 I~T~1 1 l . I ■ ,

, n. 1 1 .

/ *

$

£ / ,

/ /

/

/

Sf

i *

/ £
£ §

*> $

/

/

✓

/y

/

/

/

Figure 11. Relative proportions (in percent) of numbers of individuals of major groups
of macroinfaunal taxa within STOCS area in 1980, by cluster.

to 127), and more individuals (12,300 as opposed to 3346). Sampling
26 new stations thus provided information on 79 taxa that would not
have been collected otherwise, in addition to broadening the areal
coverage of the study.

There were apparent north-south differences among shallow
stations; the cluster analysis separated the southern shallow stations
from the northern shallow stations at a slightly lower similarity index
level (0.82). This may reflect a transition between two zoogeographic
provinces thought to meet in the STOCS region (Gallaway 1981).

The shallow assemblage of stations was characterized by high
average abundance, by the presence of a variety of numerically
dominant taxa not found at other stations, and by many taxa also in
the nearshore cluster group. The taxa of the shallow cluster were not
beach-associated, however, despite their proximity to shore. Samples

294

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 12. Sediment texture by station within STOCS area in 1980, within clusters.

collected in September 1979 in the intertidal and shallow subtidal
zones of the barrier islands between the Mexican border and Corpus
Christi, Texas, had almost no species in common with shallow cluster
samples, and were dominated by haustoriid amphipods, bivalves
(various species of Donax), and the lumbrinerid polychaete,
Lumhrineris impatiens (Tunnell et al. 1981).

Nearly all of the numerically dominant taxa of the intermediate and
offshore clusters were present also at the shallow and nearshore
stations. These taxa were ubiquitous in the study area. The deepest
stations, by contrast, were not characterized by a particularly distinct
group of organisms limited to those sites, although a few taxa were
found only at the offshore stations. There was a striking drop in
numbers of individuals per station with distance offshore, suggesting
that the fine-sediment habitat does not favor high abundances even for
taxa that can tolerate environmental conditions over the full spectrum

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

295

Figure 13. Relative proportions (in percent) of numbers of individuals of numerically
dominant macroinfaunal taxa in 1980, by cluster (one percent cutoff).

from shallow to offshore stations. The numbers of taxa per station
showed no clear trend with increasing distance from shore, but the
average number of individuals per taxon decreased by an order of
magnitude with increasing depth.

It was not possible to assign faunal differences between clusters of
stations to any particular factor, since only limited benthic
measurements (sediment texture, TOC, and hydrocarbons) were made
during the 1980 cruise. A detailed treatment of the relationships
between physical variables and many of the taxa collected in this study
and the STOCS study may be found in Flint and Holland (1980), Flint
and Rabalais (1980b), and Flint (1981). In general, the same species
that STOCS investigators identified as abundant or ubiquitous also
were abundant in this study at similar depths and in corresponding
clusters of stations.

BURMAH AGATE STATIONS

Methods

Field and laboratory methods used were identical to those previously
described in this paper. Benthic samples were collected on the

296

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 4. Presence (+) of numerically dominant macroinfaunal taxa in 1980 by cluster
(0. 1 percent cutoff). Taxa are listed in order of increasing average depth of collection.

Cluster

Shallow

Nearshore Intermediate Offshore

Acanthohaustorius millsi
Ancinus depressus
Kalliapseudes sp.

Magelona cf. sacculata

Proto haustorius bousfieldi
Scolelepis sp.

Virgularia mirabilis

Armandia agilis

+

+

+

+

+

+

+

+

+

Chone jilicaudata

+

+

Glycera americana

+

+

Grubeulepis mexicana

+

+

Haploscoloplos foliosus

+

+

Haploscoloplos fragilis

+

+

Isolda pulchella

+

+

Albunea paretii

+

+

Litocorsa stremma

+

+

Lovenella grandis

+

+

Lucina amiantus

+

+

Macoma tenta

+

+

Magelona cincta

+

+

Magelona pettiboneae

+

+

Ampelisca sp.

+

+

Monoculodes nyei

+

+

Nassarius acutus

+

+

Nereis succinea

+

+

Nucula aegeensis

+

+

Ogy rides limicola

+

+

Onuphis sp. B

+

+

Orbiniidae

+

+

Pagurus bullisi

+

+

Trichophoxus floridanus

+

+

Photis melanicus

+

+

Platyischnopus sp.

+

+

Prionospio cristata

+

+

Anachis obesa

+

+

Renilla mulleri

+

+

Anadara transversa

+

+

Spiophanes bombyx

+

+

Synchelidium americanum

+

+

Tharyx marioni

+

+

Aglaophamus verrilli

+

+

Abra aequalis

+

+ +

Heterospio longissima

+

+ +

Hexapanopeus angustifrons
Protankyra cf. benedeni
Goniada littorea

+

+

+

+ +

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

297

Table 4. Continued.

Pseudeurythoe ambigua

+

+

+

Ampelisca agassizi

+

+

+

Onuphis sp. A.

+

+

+

Sipunculids

+

+

+

Lepidasthenia maculata

+

Clymenella torquata

+

Maldanidae

+

+

+

Mediomastus californiensis

+

+

+

Xenanthura brevitelson

+

Aricidea sp.

+

+

+

Cyclaspis sp. B.

+

+

+

Corbula caribaea

+

+

+

+

Ampelisca verrilli

+

+

+

+

Diopatra cuprea

+

+

+

+

Aricidea taylori

+

+

+

+

Ostracoda

+

+

+

+

Magelona phyllisae

+

+

+

+

Paraonis gracilis

+

+

+

+

Paraprionospio pinnata

+

+

+

+

Armandia maculata

+

+

+

+

Apoprionospio pygmaea

+

+

+

+

Micropholis atra

+

+

+

+

Apseudes sp. A

+

+

+

+

Lumbrineris ernesti

+

+

+

+

Natica pusilla

+

+

+

+

Nemerteans

+

+

+

+

Nephtys incisa

+

+

+

+

Sigambra tentaculata

+

+

+

+

Nereis micromma

+

+

+

+

Speocarcinus lobatus

+

+

+

+

Nereis sp. D.

+

+

+

+

Lumbrineris januarii

+

+

Notomastus cf. latericeus

+

+

+

+

Lumbrineris sp. nov.

+

+

+

+

Vitrinella floridana

+

+

Nuculana acuta

+

+

Cossura delta

+

+

+

Alpheus sp. A.

+

+

+

Ninoe nigripes

+

+

+

Magelona longicornis

+

+

+

December 1980 cruise at

the Burmah Agate spill

site

(Station G-l,

depth 12 meters) and 40 kilometers southwest (Station G-3, depth 10

meters) (Fig. 14).

In order to assess the possible effects of spilled oil on the benthos,

the statistical evaluation

of biological

data from

G-l

and G-3 was

delayed until hydrocarbon analyses were complete. This permitted the

a posteriori selection of two uncontaminated stations sampled on the
same cruise within the STOCS area (N-9, depth 9 meters, and 1-4,

298

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 14. Locations of Burmah Agate stations.

depth 10 meters) to act as “control” sites (Fig. 8). No oil from either
Ixtoc-I or Burmah Agate was found at either 1-4 or N-9. These two
stations were geographically nearest to the Burmah Agate, and most
closely matched G-l and G-3 in depth. Furthermore, 1-4 and N-9 were
members of the same cluster (Fig. 10), suggesting similarity to one
another in community structure.

Methods for statistical analyses were as described for the STOCS
stations. Replicates were not combined for any analyses. To define a
set of numerically dominant taxa, an arbitrary mininum cutoff of one

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

299

percent of the total individuals at each station was used. This cutoff
level reduced the number of taxa to 15. Numerical abundance data are
presented as actual numbers of individuals collected.

Results

Burmah Agate residues were found in benthic samples at G-l, and
tentatively identified at G-3. Ixtoc-I oil was not found at G-l or G-3
(Boehm et al. 1982).

Samples from the Burmah Agate stations included 51 taxa and 495
individuals. By comparison, the two “control” stations had a total of
52 taxa and 649 individuals. The complete set of samples from all four
stations included 80 taxa. Diversity indices (IT) for G-l and G-3 were
2.07 and 2.47, respectively; corresponding values for N-9 and 1-4 were
2.67 and 2.28. Evenness indices (V’) for G-l, G-3, N-9 and 1-4 were
0.64, 0.56, 0.67, and 0.59, respectively.

Though similar numbers of taxa were collected at the Burmah Agate
sites and the “control” sites, relatively few taxa were present at both
sets of stations (Table 5). Only 23 taxa (29 percent of 80) were common
to both sets of stations, and only seven taxa were present at all four
stations. Three taxa at the Burmah Agate sites were not collected at
any of the other 38 sites surveyed in 1980. Samples from G-3 included
two callianassid mud shrimps ( Callianassa acanthochirus and C.
latispina) and the cumacean, Oxyurostylis salinoi. The sediment at G-
3 was considerably sandier than that at G-l, N-9, or 1-4 (Fig. 15).

The numerical dominant at all four stations was the magelonid
polychaete, Magelona phyllisae (Fig. 16). M. phyllisae accounted for
one-third of the individuals at G-l and G-3, one-fourth of the
individuals at 1-4, and one-fifth of the individuals at N-9. Other
prominent taxa included the nereid polychaete, Nereis micromma ; the
spionid polychaetes, Paraprionospio pinnata and Apoprionospio
pygmaea ; miscellaneous unidentified sipunculids and nemerteans; the
gastropod, Natica pusilla; the xanthid crab, Hexapanopeus
angustifrons; the lumbrinerid, Lumbrineris sp. nov.; and the
ophiuroid, Micropholis atra. The most abundant groups of organisms
at G-l and G-3 were deposit-feeding polychaetes, carnivorous and
omnivorous polychaetes, sipunculids, and echinoderms (mainly
ophiuroids) (Fig. 17). The most important groups of organisms at N-
9 and 1-4 were deposit-feeding polychaetes, gastropods, decapods,
omnivorous and carnivorous polychaetes, and sipunculids.

There were important dissimilarities between the two Burmah Agate
stations. Station G-l had far fewer taxa than did G-3 (17 as opposed
to 46) and fewer individuals (136 as opposed to 359). The number of
individuals at G-l was also low compared to the two “control”
stations. The differences were statistically significant ( P < 0.05). Only

300

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 5. Taxonomic checklist for Burmah Agate and “control” stations.

Burmah Agate

“Control”

Taxon

G-l

G-3

N-9

1-4

A bra equalis

+

+

Aglaophamus verrilli

Alpheus sp. A.

+

+

Alpheus sp. B.

Ampelisca sp. B

+

+

Anemones (misc. unid.)

Albunea paretii

T

+

+

Anachis obesa

+

Anadara ovalis

+

Anadara transversa

+

Apoprionospio pygmaea

Armandia maculata

+

+ '

Asychis carolinae

Cabira incerta

+

+

Calappidae (misc. unid.)

Callianassa acanthochirus

+

+

Callianassa latispina

Ceratonereis irritabilis

+

+

Ceriantharian (unid.)
Chasmocarcinus mississippiensis
Chione clenchi

+

+

+

+

Corbula caribaea

Cossura delta

+

+

+

+

Diopatra cuprea

+

+

+

Glycera americana

+

+

Gyptis brevipalpa

Haploscoloplos foliosus

+

+

Hemipholis elongata

+

Hexapanopeus angustifrons

+

+

Lepidasthenia maculata

+

Linopherus ambigua

+

Lucina amiantus

+

+

Lumbrineris ernesti

+

+

+

+

Lumbrineris sp. nov.

+

+

+

+

Macoma tenta

+

Magelona cincta

Magelona longicornis

+

+

+

+

Magelona phyllisae

+

+

+

+

Magelona cf. sacculata

Maldanidae (misc. unid.)

Marphysa sp.

+

+

+

Micropholis atra

Mediomastus calijorniensis

+

+

+

+

Minuspio cirrifera

+

Monoculodes sp. B

+

Mysidopsis bigelowi

+

Nassarius acutus

+

+

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

301

Table 5. Continued.

Natica pusilla

+

+

Nemerteans (misc. unid.)

+

+

+

Nereis micromma

+

+

+

+

Nephtys incisa

+

+

Nereis succinea

+

+

Ninoe ni gripes

+

+

+

+

Notomastus cf. latericeus

+

+

Nucula aegeensis

+

Nuculana acuta

+

Nuculana concentrica

+

Ogyrides limicola

+

+

+

Onuphis sp. A

T

Owenia fusiformis

+

Oxyurostylis salinoi

+

Paraprionospio pinnata

+

+

+

+

Paguridae (misc. unid.)

+

Pagurus bullisi

+

+

Petricola pholadiformis

+

Phascolion sp.

+

Phyllodoce mucosa

+

Pinnixa sp.

+

Polyodontes lupina

+

Scolelepis sp.

+

Sigambra tentaculata

+

+

Sipunculida (misc. unid.)

+

+

+

+

Speocarcinus lobatus

+

+

Squilla empusa

+

Sthenelais limicola

+

Terebra protexta

+

.+

Tharyx marioni

+

+

Thy one mexicana

+

Upogebia affinis

+

Vitrinella floridana

+

12 taxa were common to both Burmah Agate stations, despite their
proximity to one another. The taxa at G-l were for the most part a
subset of the taxa at G-3. Only five of the 17 taxa at G-l were not
collected also at G-3, and these were represented by only one
individual per taxon.

The “control” stations were quite similar to one another in terms
of numbers of taxa (32 at N-9, 34 at 1-4) and numbers of individuals
(396 and 253, respectively). Nonetheless, considerable heterogeneity
existed between the two stations in that only 14 taxa were present at
both.

Discussion

The most striking difference between the four stations was that G-
1 (immediately adjacent to the spill site) was depauperate in terms of

302

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 15. Sediment texture at Burmah Agate and “control” stations in 1980.

numbers of individuals and taxa. The differences could not be
attributed to atypically high values at N-9 and 1-4. For example, two
other stations in shallow and nearshore clusters on Transect I (N-37
and 1-1), had 458 and 493 individuals, more than did either N-9 or I-
4. Similarly, there were fewer taxa at G-l than at any of the four
shallow or nearshore stations on Transect I (range, 32-45).

No clear pattern was obvious in either diversity or evenness indices,
although diversity at G-l was lower than at N-9, 1-4, or G-3. Low
diversity was not uncommon, however, for stations near the northern
inshore end of the study area. For example, stations M-14 and M-24,
both in the same cluster as N-9 and 1-4, had H’ values of 1.58 and 1.77,
respectively. Evenness for G-l was intermediate between the values for
the two “control” stations.

Community summary values for Station G-3 were not depressed
relative to the “control” stations, or outside the range spanned by
other nearby nonimpacted stations. The number of individuals at G-
3 was intermediate between the numbers of individuals at N-9 and I-
4, although more taxa were collected at G-3 than at either N-9 or I-

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

303

BURMAH

AGATE

"CONTROL*

G1

G3

N9

1-4

ANEMONES IMISC.I

ASYCHIS CAR0LINAE

HEMIPHQLIS elongata

]

NEPHTYS INCISA

]

OWENIA FUSIFORMIS

3

PINNIXA SP.

DIOPATRA CUPREA

]

ABRA AEQUALIS

ALPHEUS SP. A

APOPRIONOSPIO PYGMAEA

I

HEXAPANOPEUS ANGUSTIFRONS

=3

MARPHYSA SP.

3

NASSARIUS ACUTUS

3

PAGURUS BULLISI

]

PETRICOLA PHOLADIFORMIS

SPEOCARCINUS LOBATUS

]

NATICA PUSILLA

J

TEREBRA PR0TEXTA

CERATONEREIS IRRITABILIS

3

LUMBRINERIS ERNESTI

□

LUMBRINERIS SP. NOV.

3

MAGELONA PHYLLISAE

1

1

1

1

MICROPHOLIS ATRA

n

3

NEMERTEANS IMISC.I

:□

]

NEREIS MICROMMA

□

3

]

NINOE NIGRIPES

□

PARAPRIONOSPIO PINNATA

□

1

1

SIPUNCULIDS IMISC.I

OTHER

i

rVn i i

:.tti

□

“1

33

1

riii

1 l 1 l

1 l

0 20 40 0 20 40 0 20 40 0 20 40

PERCENTAGE OF TOTAL INDIVIDUALS

Figure 16. Relative proportions (in percent) of numbers of individuals of numerically
dominant macroinfaunal taxa at Burmah Agate and “control” stations in 1980 (one
percent cutoff), by station.

4. Diversity at G-3 was intermediate between that at N-9 and 1-4.
Evenness at G-3 was slightly lower than at N-9 or 1-4, but still higher
than evenness at two of the four shallow and nearshore cluster stations
in Transect I. Based on data previously described, one might expect
greater numbers of individuals and taxa at G-3 simply because its
sediment was sandier.

Results suggest the possibility — but not the probability — that oil
from the Burmah Agate caused the reduced numbers of taxa and
individuals at the spill site. Determination of the actual causes would
require much more information than is presently available. For

304

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

PELECYPODS

GASTROPODS

DECAPODS

ECHINODERMS

NEMERTEANS

SIPUNCULIDS

POLYCHAETES3

POLYCHAETES b

Figure 17. Relative proportions (in percent) of numbers of indivduals of major groups

of macroinfaunal taxa at Burmah Agate and “control” stations in 1980, by station.

example, the relative quantity of petroleum that contacted the benthos
is not known for either G-l or G-3. The chemical results were
somewhat ambiguous. There is no doubt that G-l (at the spill site)
was exposed to Burmah Agate residues; however, it was not possible
for ERCO to positively identify Burmah Agate residues at G-3,
probably due to the amount of time that passed between sample
collection and sample analysis (personal communication 1983, P.D.
Boehm, Batelle New England Marine Research Laboratory, Duxbury,
Massachusetts).

Furthermore, the designation of “control” stations to facilitate
comparisons after the samples had already been taken was a
compromise at best. Probability levels derived from statistical tests
comparing such “controls” to the Burmah Agate are certainly suspect,
especially because there were so few taxa common to all stations.
However, the “control” stations seemed most appropriate based on
their distance, location, and depth. There were also some obvious
biological similarities between stations, although species lists were
deliberately not compared before selection of “controls.” For example,
the most abundant organism at all four stations was the magelonid
polychaete, Magelona phyllisae. The nereid polychaete, Nereis
micromma, frequently found with Magelona phyllisae (Flint and
Rabalais 1980b), was abundant at three of the four stations.
Nemerteans and sipunculids also were common at all four stations.

In practical terms, one rarely has adequate environmental
information even to confirm that benthic stations selected a priori are

BURMAH AGATE "CONTROL*

G1

]

□

G3

D

]

]

1

N9

D

zz

ZD

]

]

1-4

ZD

□

ZD

1

1

l

1 J 1 1 1

~ 1

r i iiii

n

ri rr— m

r i i ii

0 20 40 0 20 40 0 20 40 0 20 40 60

PERCENTAGE OF TOTAL INDIVIDUALS

a DEPOSIT FEEDERS
b ERRANT OMNIVORES/CARNIVORES

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

305

realistic controls. This is especially true near the South Texas coast,
where environments change considerably over short distances parallel
to shore. The Burmah Agate and ‘‘control” stations lay near a
zoogeographic discontinuity (in the area of Matagorda Bay) marking
the boundary between faunal assemblages influenced by hyposaline
estuaries and heavy river flow, and those associated with minimal river
flow and more saline estuaries (Gallaway 1981). The Burmah Agate
spill site (G-l) was adjacent to the mouth of Galveston Bay, a
comparatively large estuarine system with an annual average river
inflow of 1.2 x 1010 m3, whereas the uncontaminated stations were
located near the mouth of Matagorda Bay, an estuarine system with an
annual average river inflow of less than a third of that reaching
Galveston Bay (Natural Resources Division 1980). Some of the
differences between stations may be partially attributable to the degree
of physical and biological estuarine influence experienced by them.
For example, G-l may be exposed to lowered salinities, organic
loading, and higher turbidity during peak periods of river flow
compared to the other stations.

However, there is a more serious problem related to the specific
location of the spill and the lack of either pre-spill baseline data or
of other data from nearby comparative stations not exposed to Burmah
Agate oil. Station G-l was within the entrance channel to Galveston
Bay at the intersection of four navigational fairways for large vessels,
an area that has been highly disturbed in recent years. Most ship traffic
bound for Houston, Texas, travels through the Galveston entrance to
the Houston Ship Channel. Vessels passing the area of G-l have been
observed raising benthic sediment to the surface with their propellors
(personal communication, 1982, G. D. Dennis, III, Texas A&M
University, College Station).

In addition, the entrance and shipping channels immediately
inshore of G-l are heavily dredged. The dredge spoil is dumped several
kilometers from G-l. However, spoil deposits from other dredging
surround the station. It would not be unreasonable to expect this
station to be subject to some physical effects from dredging, such as
increased turbidity and suspension of fine sediments (see Allen and
Hardy 1980, for a review). Finally, G-l probably has been exposed to
previous high-level petroleum contamination. Although there is a
variable background of petroleum hydrocarbons throughout the Gulf
of Mexico due to natural seeps and operational ship discharges
(Gallaway 1981; Ehler et al. 1983), this particular area may have been
especially vulnerable to pollution from the Houston Ship Channel.
The channel serves the third busiest seaport in the United States, and
is adjacent to the greatest concentration of petrochemical industries in
the world (as of 1978, the most recent year for which tabulated data

306

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

were available). There are more than 350 waste discharges into the
channel, and, ultimately, into Galveston Bay and the adjacent ocean
(Bates 1980). During periods of net outflow from the bay, G-l would
have been directly in the path of effluent water. Also, exposure to
wastes from ships cannot be discounted. Vessels entering the channel
to load or unload cargo currently are prohibited from discharging
wastes overboard within coastal waters although this has not always
been required. Nonetheless, illegal discharge of ballast water to save
the expense and time of pumping it into shore-based facilities is
almost certain to have occurred in the vicinity of G-l, due to its
convenient location near the harbor entrance.

Industrial and chemical wastes near harbors may have profound
effects on benthic organisms, greatly reducing populations and
altering species composition (Reish et al. 1980). It is possible either
that the fauna of Station G-l might have been relatively sparse (both
in terms of numbers of individuals and taxa) before the Burmah Agate
spill, or that G-l might have been heavily impacted by the spill.
Unfortunately, no pre-spill or post-spill macroinfaunal samples were
available, and the issue must therefore remain unresolved.

DAMAGE ASSESSMENT STUDIES: PROBLEMS AND RECOMMENDATIONS

Virtually all ecological research programs are forced to strike a
balance between the ideal and the possible with respect to time,
money, expertise, and other resources. “Applied” studies such as the
one described in this paper usually are funded by government agencies
using competitive bidding procedures. The number of stations and
samples, the sampling schedule, and the types of analyses permitted
are specified in advance. Any changes made during such studies
require contractual modifications that often are difficult to obtain. The
following comments should be viewed in the light of these practical
realities rather than as a plea for the ultimate study design, keeping
in mind that hindsight is relatively easy to have.

With regard to the 12 STOCS stations, it is impossible to assign any
particular cause to the pronounced differences in the macroinfaunal
community from one sampling period to the next. The gaps in time
between sampling periods after the conclusion of the STOCS program
were lengthy, and intervening events left no clearly interpretable record
to be inferred from the data. Future damage assessment programs
could benefit greatly from increased sampling frequency, and from
post-spill studies at sites that have been contaminated — such as the
Burmah Agate stations — and at uncontaminated sites as near by as
possible. In addition, the choice of sampling equipment should be
carefully considered. The Smith-Mclntyre grab, while excellent for

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

307

benthic biological and sediment samples, should be supplemented
with other gear for benthic chemical samples.

Significant taxonomic difficulties occurred due to lack of access to
a complete reference collection of STOCS specimens. Changes in the
abundances of some species may be artificial, resulting from using
different names for the same species and thus causing spurious
appearances or disappearances in the data set. A complete, accessible
voucher collection should be maintained in a central location for any
future damage assessment programs, because later taxonomic revisions
are always possible.

A serious concern was that all of the attention was focused on the
structure of the macroinfaunal community (for example, numbers of
organisms) rather than on community dynamics. Many uncommon
organisms, such as predators, may be important in forming and
shaping a community in a way that far outweighs their numerical
abundance. This program was designed to assess large-scale changes in
the most common taxa. While some might argue that any truly
important modification in community function also would result in
alteration of the abundance of conspicuous animals, the time required
for this to occur cannot be predicted, and the argument itself is
circular. Unfortunately, there is no easy solution to the problem, but
the accumulation of life history information and toxicological
response data would go a long way toward improving the situation.

One of the most controversial issues was whether or not the addition
of 26 new stations in 1980 increased the chances of detecting the
impacts of Ixtoc-I or Burmah Agate oil on the subtidal community.
Because there already were 12 stations for which years of baseline data
existed, what was the value of increasing the size of the synoptic post¬
spill sample, especially since the results from the 26 new stations could
not be compared to earlier information? The answer is related to the
sample analysis schedule. The results of sediment hydrocarbon
analyses were not available until the end of the study. It seemed likely
that either spill could have contacted the benthos within the study
area, especially because slicks had been tracked through the study area
and onto the beach. Consequently, BLM based its program design on
the assumption that the data would elucidate pre-, mid-, and post-spill
conditions.

If all 12 STOCS stations had been contaminated by oil, it might
have been particularly difficult to assign biological changes to the
effects of the oil. Had there been major changes in the infaunal
community at all stations, with or without oil, they could have been
within the range of natural variability. By adding a large number of
stations in hopes that at least some might not have been contacted by
oil, it might have been possible to compare oiled stations to unoiled

308

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

stations, and to relate benthic oil concentrations to macroinfaunal
community structure. In other words, taking a large synoptic post-spill
sample was considered insurance in case all of the STOCS stations
were contaminated and new a posteriori control stations were
necessary. No reasonable alternative could have provided mapping
information, either.

Finally, on a practical note, the effort and cost of evaluating the
biological samples could and should have been deferred until oiled
sediments confirmed that a spill had impacted the benthos. The most
efficient approach to future damage assessment studies of this type
would be to collect all of the necessary samples, and then archive the
biological samples until the chemical samples are analyzed. If spill-
related petroleum residues were identified, then there would an
obvious need to analyze the biological samples. If no spill residues
were detected, there would be no necessity to analyze the biological
samples. Nonetheless, it is unlikely that such decisions will be made
upon strictly scientific bases, given the enormous political, economic,
and sociological impacts of any oil spill of comparable magnitude.

DISCLAIMER

This report has been reviewed by MMS and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of MMS, nor does mention of trade
names or commercial products constitute endorsement or recommenda¬
tion for use.

ACKNOWLEDGMENTS

The author acknowledges assistance from the following LGL
personnel: Randall L. Howard, Benny J. Gallaway, John G. Cole,
Donald W. Plitt, Russell O. McMillan, Lynn Maritzen, Jean E. Erwin,
G. Fain Hubbard, Bonnie Bower-Dennis, and Robert E. Dillinger. The
late Scott W. Anderson performed the statistical analyses. Paul D.
Boehm and David L. Fiest of ERGO were responsible for hydrocarbon
analyses and contract administration. Research and publication funds
were provided by the Outer Continental Shelf Environmental Studies
Program of MMS under contracts no. AA851 -CTO-71 to ERGO and
no. 14-12-0001-29147 to LGL. The author also thanks Bob Avent,
Murray Brown, Bob Rogers, and Carroll Day of MMS.

LITERATURE CITED

Allen, K. O., and J. W. Hardy. 1980. Impacts of navigational dredging on fish and
wildlife: A literature review. U.S. Fish and Wildlife Serv., Office Biol. Serv.
(Washington, D.C.), Rept. no. FWS/OBS-80-07, 81 pp.

STRENGTHS AND WEAKNESSES OF DAMAGE ASSESSMENTS

309

Bates, B. H., Jr. (ed.). 1980. Houston Ship Channel annual monitoring program, 1973-
1978. Texas Dept. Water Resources, Austin, Rept. no. LP-122, 65 pp.

Beslier, A., J. L. Birrien, L. Cabioch, C. Larsonneur, and L. Le Borgne. 1980. La
pollution des Baies de Morlaix et de Lannion par les hydrocarbures de 1’ “ Amoco
Cadiz”: repartition sur les fonds et evolution. Helgolander Meeresunters 33:209-224.
Bloom, S. A. 1981. Similarity indices in community studies: potential pitfalls. Mar. Ecol.
Prog. Ser. 5:125-128.

Boehm, P. D., and G. S. Lewbel. 1982. Evaluation of damage assessment program. Pp.
291-298, in Ixtoc oil spill assessment, final report. Energy Resources Co. Inc.,
Cambridge, Massachusetts, Rept. for Bur. Land Manage. (Washington, D.C.)

Boehm, P. D., A. D. Wait, D. L. Fiest, and D. Pilson. 1982. Chemical assessment —
hydrocarbon analyses. Pp. 23-152, in Ixtoc oil spill assessment, final report. Energy
Resources Co. Inc., Cambridge, Massachusetts, Rept. for Bur. Land Manage.
(Washington, D.C.)

Boehm, P. D. , D. L. Fiest, I. Kaplan, P. Mankiewicz, and G. S. Lewbel. 1983. A natural
resources damage assessment study: the Ixtoc I blowout. Pp. 507-515, in Proc. 1983
Oil Spill Conf. (Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst.,
Washington, D.C.

Boesch, D. F. 1977. Application of numerical classification in ecological investigations
of water pollution. U.S. Environ. Protection Agency, Environ. Res. Lab., Corvallis,
Oregon, Rept. no. EPA-660/ 3-77-033, 115 pp.

Bonsdorff, E. 1981. The Antonio Gramsci oil spill impact on the littoral and benthic
ecosystems. Mar. Pollut. Bull. 12:301-305.

Buchanan, J. B., and J. M. Kain. 1971. Measurement of the physical and chemical
environment. Pp. 30-58, in (N. A. Holme and A. D. McIntyre, eds.), Methods for the
study of marine benthos. Internat. Biol. Prog. Handbook, no. 16, Blackwell Sci.
Publ., Oxford, England.

Cabioch, L. , J. C. Dauvin, J. Mora Bermudez, and C. Rodriquez Babio. 1980. Effets de
la maree noire de 1’ “ Amoco Cadiz” sur le benthos sublittoral de nord de la Bretagne.
Helgolander Meeresunters 33:192-208.

Ehler, C. N., D. J. Basta, and T. F. LaPoint. 1983. Analyzing the potential effects of
operational discharges of oil from ships in the Gulf of Mexico. Pp. 323-330, in Proc.
1983 Oil Spill Conf. (Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst.,
Washington, D.C.

Eager, E. W. 1972. Diversity: a sampling study. Amer. Nat. 106:293-310.

Flint, R. W. 1981. Gulf of Mexico outer continental shelf benthos: macroinfaunal-
environmental relationships. Biol. Oceanogr. 1:135-155.

Flint, R. W. , and J. S. Holland. 1980. Benthic infaunal variability on a transect in the
Gulf of Mexico. Estuar. Coast. Mar. Sci. 10:1-14.

Flint, R. W. , and N. N. Rabalais (eds.). 1980a. Environmental studies, South Texas outer
continental shelf, 1975-1977. Vol. I, Ecosystem description, 648 pp.; Vol. II, Data
mariagement, 337 pp; Vol. Ill, Study area final reports, 329 pp. Univ. Texas Mar.
Sci. Inst., Port Aransas, Rept. for Bur. Land Manage. (Washington, D.C.).

- . 1980b. Polychaete ecology and niche patterns: Texas continental shelf. Mar. Ecol.

Prog. Ser. 3:193-202.

- (eds.). 1981. Environmental studies of a marine ecosystem, South Texas outer

continental shelf. Univ. Texas Press, Austin, 240 pp.

Folk, R. L. 1980. Petrology of sedimentary rocks. Hemphill Publ. Comp., Austin, 182

pp.

Friedman, M. 1937. The use of ranks to avoid the assumption of normality implicit in
the analysis of variance. J. Amer. Stat. Assoc. 32:675-701.

Gallaway, B. J. 1981. An ecosystem analysis of oil and gas development on the Texas-
Louisiana continental shelf. U.S. Fish and Wildlife Serv., Office of Biol. Serv.
(Washington, D.C.) Rept. no. FWS/OBS-81 /27, 89 pp.

310

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Gundlach, E. R., K. J. Finkelstein, and J. L. Sadd. 1981. Impact and persistence of Ixtoc
I oil on the South Texas coast. Pp. 477-485, in Proc. 1981 Oil Spill Conf.
(Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst., Washington, D.C.

Holland, J. S., J. Holt, S. Holt, R. Kalke, and N. N. Rabalais. 1980. Benthic
invertebrates: macroinfauna and epifauna. Pp. 515-589, in Environmental studies,
South Texas outer continental shelf, 1975-1977, (R. W. Flint and N. N. Rabalais,
eds.), Vol. Ill, Study area final reports. Univ. Texas Mar. Sci. Inst., Port Aransas,
Rept. for Bur. Land Manage. (Washington, D.C.).

Jernelov, A., and O. Linden. 1981. Ixtoc I: a case study of the world’s largest oil spill.
Ambio 10:299-306.

Kana, T. W. , E. P. Thompson, and R. Pavia. 1981. Burmah Agate chronology and
containment operations. Pp. 131-138, in Proc. 1981 Oil Spill Conf. (Prevention,
Behavior, Control, Cleanup). Amer. Petrol. Inst., Washington, D.C.

Kendall, M. G. 1938. A new measure of rank correlation. Biometrika 30:231-248.

Kruskal, W. H., and W. A. Wallis. 1952. Use of ranks on one-criterion variance analysis.
J. Amer. Stat. Assoc. 47:583-621.

Lewbel, G. S., R. L. Howard, and S. W. Anderson. 1982a. Biological assessment. Pp.
173-290, in Ixtoc oil spill assessment, final report. Energy Resources Co. Inc.,
Cambridge, Massachusetts, Rept. for Bur. Land Manage. (Washington, D.C.)

- . 1982b. Biological assessment: 1980 baseline samples. Pp. 43-71, in Ixtoc oil spill

assessment, final report, appendix 9. Energy Resources Co. Inc., Cambridge,
Massachusetts, Rept. for Bur. Land Manage. (Washington, D.C.)

- . 1982c. Burmah Agate oil spill assessment. Pp. 72-83, in Ixtoc oil spill assessment,

final report, appendix 9. Energy Resources Co. Inc., Cambridge, Massachusetts,
Rept. for Bur. Land Manage. (Washington, D.C.)

Lippincott, W. H., and A. A. Wolfson. 1981. A method for identifying areas of biological
concern. Internat. Biol. Consultants (Encinitas, California), Rept. for Environ.
Protection Agency, Oceans Programs Branch, Oil and Special Materials Control
Division (Washington, D.C.), 42 pp.

Natural Resources Division. 1980. Management of Texas coastal resources. Pp. Ill- 1 - HI-
86, in Texas coastal program state hearing draft, September 1980, Nat. Res. Div.,
Texas Energy Nat. Res. Adv. Council, Austin.

Nemenyi, P. 1963. Distribution-free multiple comparisons. Ph.D. thesis, Princeton
Univ., Princeton, New Jersey.

Pielou, E. C. 1977. Mathematical ecology. Wiley-Interscience, New York, 285 pp.

Rabalais, S. C., and R. W. Flint. 1983. Ixtoc-1 effects on intertidal and subtidal infauna
of South Texas gulf beaches. Corn. Mar. Sci. 26:23-35.

Reish, D. J., D. F. Soule, and J. D. Soule. 1980. The benthic biological conditions of
Los Angeles-Long Beach Harbors: results of 28 years of investigations and
monitoring. Helgolander Meeresunters 34:193-205.

Sanders, H. L. , J. F. Grassle, G. R. Hampson, L. S. Morse, S. Garner-Price, and C. C.
Jones. 1981. Long-term effects of the barge Florida oil spill, project summary.
Environ. Protection Agency, Municipal Environ. Res. Lab., Cincinnati, Ohio, Rept.
no. EPA-600/S2-81-012, 3 pp.

Thebeau, L. C., and T. W. Kana. 1981. Onshore impacts and cleanup during the
Burmah Agate oil spill — November 1979. Pp. 139-145, in Proc. 1981 Oil Spill Conf.
(Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst., Washington, D.C.

Tunnell, J. W., Jr., L. C. Thebeau, Q. R. Dokken, and M. E. Kindinger. 1981. Effects
of the Ixtoc I oil spill on the intertidal and subtidal infaunal populations along
lower Texas coast barrier island beaches. Pp. 467-475, in Proc. 1981 Oil Spill Conf.
(Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst., Washington, D.C.

Woods, E. G. , and P. R. Hannah. 1981. Ixtoc I oil spill — the damage assessment
program and ecological impact. Pp. 439-443, in Proc. 1981 Oil Spill Conf.
(Prevention, Behavior, Control, Cleanup). Amer. Petrol. Inst., Washington, D.C.

A RECORD OF SYMBOS (ARTIODACTYLA: BOVIDAE)
FROM KAUFMAN COUNTY, TEXAS

by JERRY N. McDONALD

Department of Geography
Radford University
Radford, Virginia 24142

ABSTRACT

The first unequivocal record of Symbos from Texas is described and illustrated. This
record extends the known range of the genus some 315 kilometers south and 260
kilometers west of the previous boundary. The orientation of the parieto-temporal
suture, the degree of transverse constriction of the frontals, and the supraoccipital-
occipital depth ratio in this specimen differ from the pattern typically seen in Symbos
crania and thus raise questions about the traditionally accepted monospecific status of
the genus. Key words: Symbos, musk ox, Texas, paleozoology.

Identifiable remains of the extinct Nearctic musk ox genus Symbos
are distributed widely across North America (Fig. 1). Most known
specimens are from Alaska-northwestern Canada, the middle
Mississippi Valley-southern Great Lakes, and the northeastern Great
Basin (Hay 1923, 1924, 1927; Frick 1937; Harington 1975, 1978;
McDonald 1985; C. E. Ray and J. N. McDonald, unpublished data).
Records of Symbos from other areas are uncommon, but specimens do
appear occasionally and thereby contribute to the better documentation
of the spatial and temporal distribution of this genus. Described here
is one such record from Kaufman County, Texas, that was found by
a college student at Southern Methodist University in 1969, and
identified at that time as Symbos by Bob H. Slaughter of the Shuler
Museum of Paleontology at Southern Methodist. This specimen is of
interest because (1) it is the first unequivocal record of Symbos
reported from Texas, (2) it extends the range of the genus onto the
northwestern Gulf Coastal Plain, and (3) it raises questions about the
currently recognized monospecific status of the genus Symbos.

The specimen described here (SMU-SMP 69127) is the property of
the Shuler Museum of Paleontology, Southern Methodist University,
Dallas, Texas. The specimen was found in a “shallow deposit of a
Trinity (River) tributary . . . not a part of the better known terrace
sequence” (letter from B. H. Slaughter to C. E. Ray received 6 August
1969). The site was later examined by Slaughter but no additional
skeletal material was found. Slaughter considered the “shallow
deposit” to be of Wisconsin age — probably dating from somewhere

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

312

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 1. The provenience of SMU-SMP 69127 and its relationship to the previously
documented distribution of Symbos. Records identified in text as forming the
southern boundary of the range are indicated.

between 21,000 and 24,000 years ago, but possibly from as much as
about 75,000 years ago (personal communication, B. H. Slaughter, 27
October 1983). The specimen was found along the east side of the
channeled lower segment of Little Brushy Creek, about half a
kilometer south of where U.S. Route 175 crosses Little Brushy Creek
and about 2.24 kilometers WSW (about 261-265° from True North) of
City Hall in Kaufman, Kaufman County, Texas (32° 35' 10" N, 96° 20'
W; 750296 E, 3608307 N, 14, N; Kaufman, Texas, Quadrangle,
U.S.G.S. 7.5' series, 1963 edition).

DESCRIPTION OF THE SPECIMEN

SMU-SMP 69127 includes most of the postorbital part of the
cranium. It has been extensively damaged by both abrasion and
weathering, and it is friable and chalky. Most of the horn cores are
missing, as are the rostral part of the presphenoid, the jugular

SYMBOS IN TEXAS

313

Figure 2. View of caudal surface of SMU-SMP 69127.

processes, the left occipital condyle, the zygomatic processes, and much
of the external occipital protuberance and crest, the nuchal line, and
the temporal crest (Figs. 2-5).

The dorsal surface of the parietals and remaining frontals is concave
transversely. Most of this surface, although abraded, is still covered
with exostosis, but a shallow longitudinal groove over the parietal
midline appears to have been without exostosis (Figs. 2, 4). There is
no evidence that a longitudinal crest of exostosis existed on the median
plane. Much of the horn core bases remain. The dorsal surface of the
horn core bases is concave rostrocaudally, but this might be a result
of asbrasive damage; the ventral surface is convex rostrocaudally (Fig.

314

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 3. View of right lateral surface of SMU-SMP 69127. Note the high supraoccipital
(B-C):occipital (A-B) ratio and the oblique orientation of the parieto-temporal suture.

3). The frontal sinus region is constricted transversely to an unusual
degree where the ventral surface of the horn sheath related to the
frontal and parietal bones (Fig. 2). The parieto-temporal suture is
oriented ventrocaudally-dorsorostrally, not horizontally as in most
specimens referred to Symbos (Fig. 3) (Harington 1975).

The caudal surface of the cranium is abraded and badly weathered.
Most of the surface bone is missing from the supraoccipital surface of
the parietals and from the dorsal and lateral edges of the occipital. The
deeper reaches of the insertion surfaces for the M. semispinalis capitis
still can be recognized, and the position of the nuchal line can be
approximated by referring to the insertion surfaces for M. semispinalis
capitis. The supraoccipital-occipital ratio is greater in this specimen

SYMBOS IN TEXAS

315

0 tOO mm

Figure 4. View of dorsal surface of SMU-SMP 69127 (caudal edge at top).

(Fig. 3) than in most crania that have been referred to Symbos (see,
for example, Osgood 1905a; Allen 1913; Lyon and Hall 1937;
Harington 1975; McDonald 1985).

The basioccipital is broad, especially near the caudal end where it
approximates one-half the breadth of the occipital condyles (Fig. 5).
Rostrally, its lateral margins first trend gradually, then markedly,
toward the median plane. A discontinuous median groove on the
ventral surface partly divides the basioccipital into right and left
halves. Measurements for the specimen are given in Table 1.

IDENTITY OF THE SPECIMEN

SMU-SMP 69127 is referred to Symbos (Osgood 1905a, 1905b) on the
basis of (1) the exostosis covering most of the dorsal surface of the
parietals and the remaining frontals; (2) the lateral direction of the
proximal horn cores; (3) the great supraoccipital-occipital depth ratio
of the caudal cranial surface; and (4) the morphology of the ventral
surface of the basioccipital bone, including the presence of a median
groove, the shape of a “V” or “shield” resulting from the convergence
of its lateral edges rostrally toward the median, and the relatively great
transverse breadth of the caudal portion, which approximates half the
breadth of the occipital condyles.

316

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 4, 1985

i - - 1

0 100mm

Figure 5. View of ventral surface of SMU-SMP 69127.

Informally, Symbos is considered to be a monospecific genus
containing only 5. cavifrons (Kurten and Anderson 1980), although
five other species (including Gidleya zuniensis , which probably
belongs in Symbos) have been recognized (Leidy 1852; Osgood 1905a;
Gidley 1906; Cossmann 1907; Brown 1908; Allen 1913; Hay 1920;
Barbour 1934). Most workers traditionally have recognized only S.
cavifrons, but the fact that distinct differences in some cranial
characters exist among specimens referred to Symbos leaves open the
possibility that one or more species in addition to S. cavifrons might
be represented. Among the distinct character differences are: (1) the
height of the frontal sinuses; (2) the presence and degree of transverse
constriction at the level of the frontal sinuses; (3) the extent to which
exostosis covers the dorsal surface of the cranium; (4) the shape of the
basioccipital; and (5) the course of the parieto-temporal suture (and,
hence, the shape of these two bones). Considerable size differences also
occur among crania referred to Symbos. The frequency, cause, and
significance of these differences are not yet clear; they might represent
individual (including choroclinal or chronoclinal) variation, they
might be age- or sex-related, or they might be taxonomic differences.
Until an examination of all available Symbos crania is completed, and

SYMBOS IN TEXAS

317

Table 1. Measurements of characters in SMU-SMP 69127 (in mm). Values in parenthesis
are estimates based upon nearly complete characters. Numbers in parenthesis
following the written descriptions of some measurements refer to numbered
measurements for Bos given by von den Driesch (1976).

Greatest breadth of basioccipital . . .

Greatest breadth, occipital condyles, primary (26) .

Greatest breadth, occipital condyles, with auxiliary surface. .

Greatest breadth, foramen magnum (28) .

Height of foramen magnurmbasion-opisthion (29) .

Greatest height of occipital region:basion-nuchal line .

Minimum height, occipital: opisthion-nuchal line .

Least breadth of parietals .

Least breadth of frontals (32) . . .

Horn cores: greatest length at base .

Horn cores: greatest dorsoventral diameter .

Transverse width of cranium at base of horn cores .

Angle, foramen magnum plane with occipital plane .

Angle, basioccipital plane with foramen magnum plane. . . .

. 89.9

. (135.8)*

. (161.6)*

. 29.4

. (44.2)

. (135.7)

. (110.7)

. 144.5

. 136.3

R 115.1, L 123.4
. . . .R 84.7, L—

. (147.9)

. . 124°

. 146°

*Dimensions based upon the breadth of the right half of the occipital condyles.

determination is made of the presence or absence of ontogenetic,
temporal, and spatial clustering of these differing characters, it seems
best to leave open the possibility of multiple species rather than to
refer automatically all individual specimens identified as Symbos to S.
cavifrons. This is particularly true for specimens that differ
conspicuously from the type specimen and description of S. cavifrons,
although, obviously, the spurious, casual erection of new taxa also
should be avoided.

SMU-SMP 69127 differs from the type of S. cavifrons as well as from
the holotypes of the other three nominal species in the genus
(including Gidleya) that have been founded on cranial material (two
species, S. promptus and S. australis, were founded on teeth). In
particular, the height of the frontal sinus region relative to the
occipital is much greater in SMU-SMP 69127 than in the type
specimens of S. cavifrons (Leidy 1852; Osgood 1905a), S. tyrrelli
(Osgood 1905a), or Gidleya zuniensis (Gidley 1906). The shape of the
parieto- temporal suture in the Kaufman County specimen differs from
the same character in all these type specimens. SMU-SMP 69127 also
differs from the type specimen of S. tyrrelli in the shape of the ventral
surface of the basioccipital. SMU-SMP 69127 is set apart from S.
convexifrons, which was founded on a convexly contoured dorsal
cranium with most of the right horn core attached (Barbour 1934), by
the concave surface of its dorsal cranium. At this time, therefore, SMU-
SMP 69127 is referred only to genus.

318

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

RANGE EXTENSION

The southern boundary of the known range of Symbos, based on
published records of identifiable skull material, previously was located
along a line extending from 40 miles southeast of Atlantic City, New
Jersey (Whitmore et al, 1967), through Saltville, Virginia (Ray et al.
1967), New Madrid, Missouri (De Kay 1828), Tunica Hills, Louisiana
(Lowery 1974), Ft. Gibson, Oklahoma (Leidy 1852), Chickasha,
Oklahoma (Stovall and Self 1936), Black Rocks, New Mexico (Gidley
1906), Grand Mesa, Colorado (McDonald 1985), Santaquin, Utah
(Bissell 1963), and Saanich Peninsula, British Columbia (Harington
1975). Undescribed crania from Montezuma County, Colorado, and
Modoc County, California, also contribute to the formation of this
southern boundary.

Although Kurten and Anderson (1980) placed Symbos in Texas, this
assignment was based on their informal synonymy of all species of
Symbos and most species of Bootherium with S. cavifrons. The
holotype of B. brazosis ( =B . sargenti) is from Brazos County, Texas
(Hesse 1942; Ray 1966); this is the only cranial material of musk ox
described to date from Texas, and positive identification of musk ox
remains to the level of genus or below must be based upon cranial
material because the dental and postcranial characters of Bootherium
are still unknown. The validity of Bootherium and Symbos as distinct
genera has been questioned for many years; some authors have felt that
individuals assigned to one or the other genus are actually male
( Symbos ) and female ( Bootherium ) of the same taxon (Osgood 1905a;
Allen 1913; Semken et al. 1964). Until the available fossil material
assigned to each genus is examined in detail, however, this problem
cannot be resolved satisfactorily and, in the meantime (to preclude
unnecessary confusion), both genera should be treated as being at least
nominally distinct. Other than the type specimen of B. brazosis, little
ovibovine material is known from Texas and none of this is known
to be cranial (personal communications: A. H. Harris, May, 1979; W.
W. Dalquest, 17 February 1981; B. H. Slaughter, 18 February 1981; E.
L. Lundelius, Jr., 25 February 1981; R. J. Stanton, 10 March 1981).
SMU-SMP 69127 is, therefore, the first definite Symbos record for
Texas, it is the first Symbos record from the western Gulf Coastal
Plain of the United States, and it extends the range of Symbos in the
trans-Mississippi West approximately 315 kilometers south and 260
kilometers west of the previously recognized southwestern boundary.

ACKNOWLEDGMENTS

The author wishes to thank Clayton E. Ray and Bob H. Slaughter
for providing unpublished information and granting permission to

SYMBOS IN TEXAS

319

study this specimen. The research for this paper was completed while
the author was a postdoctoral fellow at the Smithsonian Institution.

LITERATURE CITED

Allen, J. A. 1913. Ontogenetic and other variations in muskoxen with a systematic
review of the muskox group, Recent and extinct. Mem. Amer. Mus. Nat. Hist., n.s.
1:103-226.

Barbour, E. H. 1934. A new ovibovine, Symbos convexifrons, sp. nov. Bull. Nebraska
State Mus. 1:295-298.

Bissell, H. J. 1963. Lake Bonneville: geology of southern Utah Valley, Utah. U.S. Geol.
Survey Prof. Paper 257-B: 1 01 -130.

Brown, B. 1908. The Conard Fissure, a Pleistocene bone deposit in northern Arkansas:
with descriptions of two new genera and twenty new species of mammals. Mem.
Amer. Mus. Nat. Hist. 9:155-208.

Cossmann, M. 1907. (Proposal to replace Liops with Gidleya). Rev. Crit. Paleozool. 11th
year:64.

De Kay, J. E. 1828. “Notes on a fossil skull in the cabinet of the Lyceum. . . An.
Lyceum Nat. Hist. New York 2:280-291.

Frick, C. 1937. Horned ruminants of North America. Bull. Amer. Mus. Nat. Hist. 69:1-
669.

Gidley, J. W. 1906. A new ruminant from the Pleistocene of New Mexico. Proc. U.S.
Natl. Mus. 30:165-167.

Harington, C. R. 1975. Pleistocene muskoxen ( Symbos ) from Alberta and British
Columbia. Canadian J. Earth Sci. 12:903-919.

- . 1978. Quaternary vertebrate faunas of Canada and Alaska and their suggested

chronological sequence. Syllogeus 15:1-105.

Hay, O. P. 1920. Descriptions of some Pleistocene vertebrates found in the United States.
Proc. U.S. Natl. Mus. 58:83-146.

- . 1923. The Pleistocene of North America and its vertebrated animals from states

east of the Mississippi River and from the Canadian provinces east of Longitude 95°.
Publ. Carnegie Inst. Washington 322:1-499.

- . 1924. The Pleistocene of the middle region of North America and its vertebrated

animals. Publ. Carnegie Inst. Washington 322a: 1-385.

- . 1927. The Pleistocene of the western region of North America and its vertebrated

animals. Publ. Carnegie Inst. Washington 322b: 1-346.

Hesse, C. J. 1942. The genus Bootherium, with a new record of its occurrence. Bull.
Texas Arch. Paleo. Soc. 14:77-87.

Kurten, B., and E. Anderson. 1980. Pleistocene mammals of North America. Columbia
Univ. Press, New York, 1-442.

Leidy, J. 1852. Memoir on the extinct species of American ox. Smithsonian Contrib.
Knowledge 5:1-20.

Lowery, G. H., Jr. 1974. The mammals of Louisiana and its adjacent waters. Louisiana
State Univ. Press, Baton Rouge, 1-565.

Lyon, M. W., Jr., and F. T. Hall. 1937. Skull of musk-ox, genus Symbos, from
Montgomery County, Indiana. Amer. Midland Nat. 18:608-609.

McDonald, J.N. 1985. Symbos cavifrons (Artiodactyla:Bovidae) from Delta County,
Colorado. Great Basin Nat., in press.

Osgood, W. H. 1905a. Scaphoceros tyrrelli, an extinct ruminant from the Klondike
Gravels. Smithsonian Misc. Coll. 48:173-185.

— - . 1905b. Symbos, a substitute for Scaphoceros. Proc. Biol. Soc. Washington 18:223-

224.

320

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Ray, C. E. 1966. The status of Bootherium brazosis. Texas Mem. Mus. Pearce-Sellards
Ser. 5:1-7.

Ray, C. E., B. N. Cooper, and W. S. Benninghoff. 1967. Fossil mammals and pollen in
a late Pleistocene deposit at Saltville, Virginia. J. Paleontol. 41:608-622.

Semken, H. A., Jr., B. B. Miller, and J. B. Stevens. 1964. Late Wisconsin woodland musk
oxen in association with pollen and invertebrates from Michigan. J. Paleo. 38:823-
835.

Stovall, J. W. , and J. T. Self. 1936. A new specimen of Symbos from Chickasha,
Oklahoma. J. Mammal. 17:422.

von den Driesh, A. 1976. A guide to the measurement of animal bones from
archaeological sites. Bull. Peabody Mus. Arch. Ethnol. 1:1-137 (2nd printing).

Whitmore, F. C., Jr., K. O. Emery, H. B. S. Cooke, and D. J. P. Swift. 1967. Elephant
teeth from the Atlantic continental shelf. Science 156:1477-1481.

REPRODUCTION DATA ON DIONDA EPISCOPA
FROM FESSENDEN SPRING, TEXAS

by LESLIE M. WAYNE and B. G. WHITESIDE

Aquatic Station, Department of Biology
Southwest Texas State University
San Marcos, Texas 78666

ABSTRACT

The roundnose minnow, Dionda episcopa, is primarily restricted to clear spring-fed
waters that have slight temperature variations. Ova diameters, numbers of separated ova,
and gonadosomatic index indicate that spawning occurred from January thru August
with a peak occurring in April and May and a second peak in July and August.
Spawning peaks occurred when the number of daylight hours were approximately the
same (14 hours). There was a significant positive correlation between the number of
separated mature ova and the standard length of the fish. D. episcopa showed no
significant deviation from a one-to-one sex ratio. Key words : reproduction, Dionda,
thermal stability, Texas.

The roundnose minnow, Dionda episcopa, is primarily restricted to
clear spring-fed waters that have slight temperature variations (Brown
1953; Hubbs 1951; Hubbs et al. 1953; Jurgens 1951; Kuehne 1955;
Tilton 1961; and Trevino 1955).

There is a paucity of information available concerning reproduction
in D. episcopa. Hubbs (1951) observed two breeding populations in the
Nueces River, Texas, in April. The fish were in water with a
temperature of 17-18° C, and several breeding individuals were found
buried more than 2.5 centimeters in the gravel as far as 0.3 meters from
the bank. This gravel substrate was filled with ground water; and
when disturbed, the fish became quite active. However, they soon
returned to the same spot and resumed spawning. At another locality
in the Nueces River, he observed spawning in 2.5 centimeters of water.
The eggs were described as heavy, nonadhesive and lodged in gravel.
Tilton (1961), based on personal communication from Dr. Hubbs,
reported that D. episcopa in central Texas spawned from March
through November. Koster (1957) stated that in New Mexico, D.
episcopa spawned over gravel in spring-fed streams in summer. Hubbs
and Miller (1977) noted that several other species of Dionda in Mexico
were sexually mature or had spawned from mid-December to May,
depending on species. D. nubila has been observed spawning in late
spring (Eddy and Underhill 1974) and in May and June (Cross and
Collins 1975).

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

322

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

The purpose of this study was to determine the reproductive cycle,
fecundity, sex ratio, and sexual dimorphism of D. episcopa.

MATERIALS AND METHODS

Study area. — The study area consisted of the headwaters of Fessenden
Spring, which flows into Johnson Creek and then into the Guadalupe
River. Fessenden Spring is located approximately 16 kilometers NW
Ingram, Kerr Co., Texas, and serves as the primary water source for
the Heart of the Hills Research Station. The collecting area was a pool
with a surface area of approximately 300 squate meters. This pool was
the second in a series of three formed by the damming of the spring.
The water was clear with a mean depth of 0.5 meter, and a maximum
depth of 1.2 meters. The substrate consisted primarily of silt except in
the areas where the spring water surfaced. In these areas, gravel was
the main component of the substrate. Spirogyra sp. was the dominant
vegetation in the pool during the study period with small amounts of
Ludwigia sp. and Rorippa sp. also present. Numerous pecan trees were
in the area but, for the most part, were located a considerable distance
away from the shoreline so that little shading occurred. The mean
water temperature was 21.2° C with a low of 20.0° C in December 1977
and January 1978 and a high of 22.5° C in August 1977.

Collection of specimens. — Monthly samples of 20 fish were taken
from February 1977 through January 1978 (total of 240 fish). For
catches that included more than 20 specimens, 20 representative fish
were selected from the catch. Collections were made using a 0.5-
centimeter-square mesh common sense seine. The majority of fish were
taken within three meters of the bank. Fish were preserved in 10
percent formalin.

Reproduction. — Gonads were carefully removed and each fish was
sexed. If it was impossible to determine the sex of an individual
accurately, it was noted as “unable to sex.” For each male, the greatest
length and width of the left testis was measured to the nearest 0.01
millimeter using an ocular micrometer. If the left testis was absent or
had been damaged upon removal, measurements were taken on the
right testis.

Total ovarian weight for each female were determined by weighing
the left and right ovaries to the nearest 0.00001 gram with a Mettler
H54 single-pan balance. If only one ovary was obtained, its weight was
doubled. The left ovary was cut into several pieces and placed into a
vial containing 10 percent formalin. The vial was shaken by hand for
10 seconds to free as many eggs as possible. The contents were poured
into a petri dish 10 centimeters in diameter, and all free eggs were
counted and recorded as mature or immature. Mature ova were yellow

REPRODUCTION IN DIONDA

323

and almost completely filled with yolk. Immature ova were slightly
yellow with a small amount of yolk, or were white and contained no
yolk. Counts obtained from the left ovary were doubled to represent
the total number of separated ova. All free eggs that fell into five grid
areas (each 1.25 centimeters square) on the petri dish were measured
for ova diameter to the nearest 0.01 millimeter with an ocular
micrometer. If the left ovary was not available for egg counts and
measurements, data were taken from the right ovary.

Only mature females, greater than 30 mm standard length, were
used in analyzing ova count and measurement data and for
determining the gonadosomatic index (GSI). GSI was expressed as
mean percent ovary wet weight per total body wet weight.

Physical methods . — Water temperature was measured with a
standard mercury thermometer placed approximately 20 centimeters
below the surface.

Computer programs and statistical analysis. — The data obtained in
this study were analyzed with the DEC system- 10 computer using
BASIC programs, programs of the Biodat Series (written by Dr. D. G.
Huffman, Southwest Texas State University), and the Statistical
Package for the Social Sciences (Nei et al. 1975).

A Chi-Square Goodness of Fit test was run to test for significant
deviations from a one-to-one sex ratio. Other statistics used were those
associated with simple linear regression, including the Pearson
product moment correlation coefficient (r) and coefficient of
determination (r2) (Zar 1974).

RESULTS AND DISCUSSION

Reproductive cycle. — Only mature females (94) were used in analysis
of the reproductive cycle data because immature females, (less than 30
mm standard length) never contained mature ova. Hickling and
Rutenberg (1936) stated that duration of the spawning period in fishes
can be determined by recording the diameter of ova as they develop
within the ovary. The mean separated ova diameter showed a peak in
April and May (1.09 mm) and another peak in July and August (0.99
mm), followed by a drastic decrease in September (0.27 mm), which
was the lowest mean monthly ova diameter during the sampling
period (Fig. 1). Mature ova were present from February to September
and did not appear again until January. Thus, ova diameters indicated
that spawning did not occur from September through December.

The GSI also indicated similar spawning peaks to that of ova
diameter (Fig. 1). The mean number of separated ova also showed a
similar pattern. During May there was a major peak of 433 separated
ova per mature female with a second peak occurring in July with 172
separated over per mature female (Fig. 1).

324

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 1. Mean number of separated ova, mean separated ova diameter (mm), and mean
gonadosomatic index for Dionda episcopa 30.0 mm or greater in standard length each
month. Collections from Fessenden Spring, Texas. Number of specimens examined
each month is given in parentheses. The amount of daylight hours each month is
given (A. H. Belo Corporation 1975, 1976).

Based on 80 males from Fessenden Spring, the mean testis length
and width showed a pattern (Fig. 2) similar to the female reproductive
cycle data (Fig. 1). A peak in mean testis length (19.70 mm) and mean
testis width (3.53 mm) occurred in May followed by a decrease in their
size until January (Fig. 2).

The above data indicate that some spawning occurred from January
through August (eight months) with spawning peaks occurring in
April and May and in July and August. An extended spawning period
is not unusual for organisms found in spring-fed systems where water
temperature varies slightly year round (Hubbs and Strawn 1957; Hubbs
et al. 1968).

Photoperiod and temperature have been cited as important
environmental factors in the regulation of reproductive cycles (Hoar
1957, 1969). Harrington (1950) stated that fish such as minnows may
be induced to spawn by an increase in temperature or an increase in
illumination, or both. At Fessenden Spring, the water temperature
varied only slightly during the sampling year: therefore, water
temperature probably was not a significant spawning stimulus. An
increase in the ovarian development of D. episcopa occurred as the
number of daylight hours increased until the first spawning peak was
reached in May (Fig. 1). The second spawning peak that followed in
July occurred when the number of daylight hours (14) was
approximately the same as that of the first spawning peak.

REPRODUCTION IN DIONDA

325

1977 1978

MONTH

Figure 2. Mean testis length and width (mm) for D. episcopa each month. Collections
from Fessenden Spring, Texas. Number of specimens examined each month is given
in parentheses.

Fecundity. — In this study fecundity is defined as the number of
separated mature ova in a female that contained mature ova. Based on
the 50 females containing mature ova, fecundity ranged from two to
540 with a mean of 164 ova.

There was a positive correlation between the number of separated
mature ova and the standard length (SL) of fish examined that
contained mature ova (P=0. 00001, r=0.69132, r2=0.47793). It has been
shown in several species of fishes that there is an increase in the
number of mature ova with an increase in body size (Hoar 1957;
Rounsefel 1957; Vladykov 1956). Explanations other than larger body
size have been proposed as answers to this phenomenon. Hubbs et al.
(1968) stated that fecundity is a product of the frequency of spawnings
and the number of eggs produced per spawn. Also, fish that spawn
over an extended period may not be able to store the necessary
metabolites in preparation for the next spawn; therefore, the quantity
of absorbed nutrients or the amount of alimentary absorptive surface
area or both, could be limiting factors. Williams (1959) indicated that
food may be the limiting factor for ova production up to a certain
point, but beyond that point the mechanical restrictions on the
amount of space in the coelom and the necessity of maintaining
locomotor functions at a reasonable level of efficiency probably become
the limiting factors.

326

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Variation in the number of separated mature ova in any 1.00 mm
interval of SL for fish that contained mature ova from Fessenden
Spring was evident. Inasmuch as D. episcopa has an extended
spawning period, some of the variation could have been due to
collecting fish immediately before or after spawning. It is not
uncommon for there to be a considerable amount of variability in
fecundity between fish of the same length, weight, or age (Bagenal and
Braum 1970; Hoar 1957).

There was only a slight positive correlation between the mean
diameter (mm) of separated mature ova and the SL of fish examined
that contained separated mature ova (P=0. 00081, r=0. 43870,

r2=0. 19246). Most of the variation in separated mature ova diameter
occurred in smaller mature fish. This correlation is not as strong as
might have been expected because Hoar (1957) reported that the final
size of the ova of fishes depends both on the size of the female and
on her level of nutrition during the period preceding spawning.

Sex ratio and sexual dimorphism. — Sex determination of D. episcopa
from Fessenden Spring showed no significant deviations (P=0.05) from
a one-to-one sex ratio. Eighty males and 98 females were collected with
a 0.82:1.00 female-to-male ratio.

The normal color pattern for D. episcopa is reddish brown dorsally,
silvery ventrally, with a dark lateral band and a caudal spot (Eddy and
Underhill 1978; Hubbs and Brown (1956). During spawning, this fish
takes on a bright yellow-orange color, which is superimposed on the
normal color pattern. This bright coloration is present on the
proximal two-thirds of all the fins and forms a streak from the tip of
the snout to the origin of the anal fin (Hubbs 1951).

Breeding males were observed to have the bright yellow-orange color
as described above. However, some of the larger males collected during
the nonspawning period were yellow in color. Mature females during
the spawning season had a slight yellow color but were not as bright
as males. Similar color patterns have been documented in other species
of Dionda (Cross and Collins 1975; Hubbs and Miller 1977; Miller and
Robinson 1973).

Breeding males had distinct tubercles present on their head during
the spawning season. Tubercles also were present during other times
of the year but were not so well developed. Two of the larger females
were observed to have a few small tubercles on the head. Tubercles
have been reported from other species of Dionda (Branson 1962; Hubbs
and Miller 1977; Hubbs and Brown 1956); however, in some of these
species, females with tubercles have not been collected (Branson 1962;
Hubbs and Miller 1977).

REPRODUCTION IN DIONDA

327

ACKNOWLEDGMENTS

We gratefully acknowledge W. C. Young and D. G. Huffman for
their advice and constructive criticism during the course of this
research. Thanks also are expressed to the following people for their
help in the field collections: John William; Richard Heath; David
Morris; and Rob Wayne.

LITERATURE CITED

Bagenal, T. B., and E. Braum. 1970. Eggs and early life history. Pp. 159-181, in,
Methods for assessment of fish production in fresh waters (W. E. Ricker, ed.),
Blackwell Scientific Publ., London.

Branson, B. A. 1962. Observations on the breeding tubercles of some Ozarkian minnows
with notes on the barbels of Hybopsis. Copeia 1962:532-539.

Brown, W. H. 1953. Introduced fish species of the Guadalupe River Basin. Texas J. Sci.
5:245-251.

Cross, F. B., and J. T. Collins. 1975. Fishes in Kansas. Public Ed. Ser., Mus. Nat. Hist.,
Univ. Kansas.

Eddy, S., and J. C. Underhill. 1974. Northern Fishes. Univ. Minnesota Press, 414 pp.

Harrington, R. W. 1950. Preseasonal breeding by the bridled shiner Notropis bifrenatus,
induced under light-temperature control. Copeia 1950:304-311.

Hickling, C. F. , and E. Rutenberg. 1936. The ovary as an indicator of the spawning
period in fishes. J. Mar. Biol. Assoc. U. K. 21:311-317.

Hoar, W. S. 1957. The gonads and reproduction. Pp. 287-317, in The physiology of
fishes, (M. E. Brown, ed.), vol. 1, Academic Press, New York.

- . 1969. Reproduction. Pp. 1-72, in Fish physiology, (W. S. Hoar and D. J. Randall,

eds.), vol. 3, Academic Press, New York.

Hubbs, C. 1951. Observations on the breeding of Dionda episcopa serena in the Nueces
River, Texas. Texas J. Sci., 3:490-492.

Hubbs, C., and W. H. Brown. 1956. Dionda diaboli (Cyprinidae), a new minnow from
Texas. Southwestern Nat. 1:69-77.

Hubbs, C., R. A. Kuehne, and J. C. Ball. 1953. The fishes of the upper Guadalupe
River, Texas. Texas J. Sci. 5:216-244.

Hubbs, C. L. , and R. R. Miller. 1977. Six distinctive cyprinid fish species referred to
Dionda inhabiting segments of the Tampico embayment drainage of Mexico. Trans.
San Diego Soc. Nat. Hist. 18:267-336.

Hubbs, C., M. M. Stevenson, and A. E. Peden. 1968. Fecundity and egg size in two
central Texas darter populations. Southwestern Nat. 13:301-324.

Hubbs, C., and K. Strawn. 1957. The effects of light and temperature on the fecundity
of the greenthroat darter, Etheostoma lepidum. Ecology 38:596-602.

Jurgens, K. C. 1951. The distribution and ecology of the fishes of the San Marcos River.
M.A. thesis, Univ. Texas, Austin.

Kuehne, R. A. 1955. Stream surveys of the Guadalupe and San Antonio Rivers. Texas
Game and Fish Com., IF Rept. Ser. 1:1-56.

Miller, R. J., and H. W. Robison. 1973. The fishes of Oklahoma. Oklahoma State Univ.
Press, 246 pp.

Nei, N. H., D. H. Brent, C. H. Hill, J. G. Jenkins, and K. Steinbrenner. 1975. Statistical
package for the social sciences. McGraw-Hill Pub. Co., New York, 2nd ed.

Rounsefel, G. A. 1957. Fecundity of North American Salmonidae. Bull. U.S: Fish and
Wildlife Serv.57:451 -468.

328

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Tilton, J. E. 1961. Ichthyological survey of the Colorado River of Texas. M.A. thesis,
Univ. Texas, Austin.

Trevino, D. B. 1955. The ichthyofauna of the lower Rio Grande River, from the mouth
of the Pecos to the Gulf of Mexico. M.A. thesis, Univ. Texas, Austin.

Vladydov, V. D. 1956. Fecundity of the wild speckled trout ( Salvelinus fontinalis) in
Quebec lakes. J. Fish. Res. Board Canada 13:799-841.

Williams, G. C. 1959. Ovary weights of darters: a test of the alleged association of
parental care with reduced fecundity in fishes. Copeia 1959:18-24.

Zar, J. H. 1974. Biostatistical analysis. Prentice-Hall, Inc., Englewood Cliffs, New
Jersey.

FECAL AND INTESTINAL BACTERIA OF
SWINE MAINTAINED ON ATHEROGENIC DIETS

by D. H. LEWIS

Department of Veterinary Microbiology
Texas Agricultural Experiment Station and
Texas Ah’ M University
College Station , Texas 77843

ABSTRACT

Combinations of carbohydrate, lipid, and cholesterol dietary constituents previously
have been shown to induce aortic intimal lipidosis in swine. A study was conducted to
assess the effects of those dietary constituents on the intestinal microflora of swine.

Swine were maintained on standard rations during the first eight weeks of life.
Afterwards, the animals were divided into four groups and maintained on formulated
rations for 60 weeks. For this purpose, basal rations were supplemented with 33 percent
sucrose (sucrose diet), 33 percent sucrose and 0.9 percent cholesterol (sucrose-cholesterol
diet), 20 percent cottonseed oil (cottonseed oil diet), and 20 percent cottonseed oil and
1.0 percent cholesterol (cottonseed oil-cholesterol diet).

For as long as 72 hours after birth, feces from the hysterectomy-derived swine
possessed fewer organisms than could be cultured by the described techniques. By the
end of the first week however, large numbers of fecal bacteria (1011 organisms per gram
of feces), primarily lactobacilli developed. Bacterial groups constituting what later
became predominant fecal microflora of adult pigs ( Bacteroides sp., coliforms,
enterococci, and lactobacilli) were established during the second week of life. The
bacterial flora appeared to have stabilized in terms of numbers and kinds of bacteria by
the eighth week of life, and the dietary treatments used in the study appeared to have
minimal quantitative effects on that flora. However, rations supplemented with
cottonseed oil appeared to enhance development of coliform and enterococci, which were
capable of deconjugating glycine and taurine from cholic and deoxycholic conjugates.
Similar bacteria were not recovered from swine maintained on other rations.

Factors that influence the microecology of intestinal bacteria are
likely to indirectly effect a wide range of host metabolic processes.
Studies over the last few years have demonstrated that certain dietary
constituents can have profound effects upon the numbers and kinds of
intestinal bacteria in various animal species. Feeding large quantities
of lactose to pigs resulted in increased numbers of fecal lactobacilli
(Wilbur 1959). Increased numbers of celluloltyic bacteria were
recovered from intestines of pigs maintained on a high fiber diet (Varel
et al. 1982). The feeding of protein-rich diets to mice and rats resulted
in fewer lactobacilli and increased numbers of enterococci and
coliforms in their feces (Johansson and Sarles 1949; Porter and Rettger
1940). Clostridium perfringens occurred in greater numbers in feces of

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

330

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

rats when they were maintained on a diet supplemented with butter
and sodium cholate (Graber et al. 1966b).

Within the last few years, interest has focused upon the potential
role of diet as a factor in various vascular diseases. Aortic lesions
resembling those associated with human atherosclerosis can be induced
in swine when the animals are maintained on diets containing elevated
levels of lipids and carbohydrates (Suzuki et al. 1969). The synthesis,
inter-conversion, and excretion of steriod compounds believed to be
involved in atherogenesis are effected both by dietary constituents and
intestinal bacteria (Drasar et al. 1966; Eyssen 1973; Kellog 1971;
Kritchevsky 1979; Kritchevesky 1964; Medwedt 1974; Midwest and
Norman 1967); however, data relative to effects of atherogenic dietary
factors upon intestinal bacteria appears to be lacking.

The present study was thus initiated to examine the effects of certain
atherogenic dietary regima upon swine intestinal bacteria.

MATERIALS AND METHODS

Swine used in this study were Duroc males within four weeks of the
same age and derived from breeding of a single sire with line-bred
females. Experimental animals were surgically derived according to
Young et al. (1965). Until eight weeks of age, the animals were
individually fed a sterilized diet consisting of homogenized cows milk
fortified with minerals, dextrose, eggs, yeast extract, vitamin D, cod
liver oil, and agar (Meyer et al. 1963). At eight weeks of age, the swine
were ear-notched for identification, transferred from swine animal
housing facilities to concrete pens, and placed upon the dietary regima
described below. The swine were distributed five to each pen such that
equal numbers of pigs of the same age were contained in the various
pens. The pens had slotted floors to minimize coprophagy and solid
concrete walls to minimize vermin infestations. Routine care could be
accomplished without entering the pens. Four dietary treatments were
used in the study.

One group of animals was fed a standard basal ration (Morrison
1959) enriched with 33 percent sucrose (sucrose diet), another group
was fed the ration supplented with 33 percent sucrose and 0.9 percent
cholesterol (sucrose-cholesterol diet) the third group was fed the ration
supplemented with 20 percent cottonseed oil (cottonseed oil diet), and
the fourth group was fed the ration enriched with 20 percent
cottonseed oil and 1.0 percent cholesterol (cottonseed oil-cholesterol
diet).

Sampling and Processing Methods

Two types of samples were analyzed in the described study: 1) fecal
samples derived after anal stimulation with sterile wooden tongue

BACTERIA OF SWINE

331

depressor blades, and 2) samples derived from various portions of the
intestinal tract as the animals were necropsied at the end of the
experiment. Just prior to processing the samples, 0.05 milliliter of two
percent sodium bicarbonate and 0.05 milliliter of one percent sodium
bicarbonate — one percent cysteine were added to bring total volume up
to 9.0 milliliter for each aliquot of diluting fluid (buffered saline,
cysteine, and sodium carbonate) and tryptone yeast extract (anaerobic
broth). The samples were processed 30 minutes or less after collection.
Processing involved weighing the sample to the nearest 0.01 gram,
diluting the samples in a series of 10-fold dilutions in reduced diluting
fluid, and reduced anaerobic broth and distributing 0.1 millileter of
the various sample dilutions on each of three plates of the various
media. Anaerobic broth was incubated three days at 37°C, and the
highest dilutions wherein turbidity was observed were streaked on
selective anaerobic media. The media used for aerobic enumeration
consisted of SS agar, Staphylococcus medium no. 110 (Staph 110),
Phenylethanol agar, MacConkey agar, LBS agar, Enterococci agar,
and five percent sheep blood agar. The selective anaerobic media used
for anaerobic enumeration and for recovering organisms from turbid
anaerobic broth were prepared within three hours of use and consisted
of five percent sheep blood agar (fresh pour plate), kanamycin (50
milligram per milliliter) vancomycin (17.5 milligram per milliliter)
laked blood agar (KVLB), Naglers agar with neomycin (100 milligram
per milliliter), and reinforced clostridial medium with 5.0 percent
blood and neomycin (100 /xg per milliliter) (RCM) (Graber et al.
1966b). The fluids were spread over the surface of the media with an
alcohol-flamed glass rod, incubated 37°C for 72 hours, and the
colonies counted. Plates for anaerobic culture were incubated in
anaerobic jars containing palladium catalyst and a gaseous atmosphere
of 10 percent carbon dioxide in hydrogen.

During the first week of life, fecal samples were obtained from each
animal six hours, 18-24 hours, 48 hours, and 72 hours after birth.
Afterwards, the samples were collected at weekly intervals.

At 14 months of age, the animals were euthanized. Immediately
following the death of each animal, samples were taken by opening
the abdominal cavity and ligating various segments of the intestinal
tract. Specimens for bacteriologic analysis were obtained from the
duodenum, the ileum (three feet proximal to the ileocecal junction),
the cecum, and the rectum. These segments were removed and the
contents extruded into sterile containers for bacteriological analysis.
The entire process took no more than 10 minutes per pig and the
samples were immediately taken to the bacteriology laboratory for
analysis.

Bacteria isolated from animals on the various experimental regima
were identified by conventional means and tested for their ability to

332

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

deconjugate bile salts in a manner similar to that described by
Shimada et al. (1969).

The sodium salts of cholic and deoxycholic acids conjugated with
taurine or glycine were incorporated at a final concentration of two
milligrams per milliliters in Thioglycollate medium (assay medium).
The medium was steamed for 15 minutes, cooled and inoculated with
a loopful of organisms from a plate or slant. After 48 hours incubation
at 37°C, cultures were filtered thru a bacteria retaining filter (0.25 /x)
and 20 jxl of the filtrate were applied to a thin layer (400 n) plate of
silica Gel G. At the same time, standards consisting of 20 /xl samples
of a mixture of the free bile acids ( two milligrams each bile acid per
milliliter), uninoculated Thioglycollate medium containing the bile
acid conjugates and inoculated and uninoculated Thioglycollate
medium less the bile acid conjugates were applied to the plates. A total
of 12 samples was applied to each plate. The mobile phase for free bile
acids consisted of chloroform:acetone:acetic acid (70:20:10) and
development time at ambient temperature was three to four hours. The
spots were visualized after spraying the plates with 10 percent
phosphomolybdic acid in ethanol and heating 100°C for 10 minutes.
Chemical changes in the bile acid conjugates were inferred by
comparing the spots of the cultures with those of the standards.

RESULTS

Throughout the study, the general condition and performance of the
experimental animals were considered good.

Fecal Bacterial Flora

For as long as 72 hours after birth, feces from the hysterectomy-
derived swine possessed fewer organisms than could be cultured by the
described techniques. By the end of the first week however, large
numbers of fecal bacteria (1011 organisms per gram feces), primarily
lactobacilli, developed.

Bacteroides sp. , enterococci, and coliforms were first observed when
the pigs were two weeks of age. Greatest numbers of coliforms (109
organisms per gram of feces) observed in the experiment were
recovered from the two-week-old pigs. The population density of the
coliforms gradually decreased until the pigs were one month old. The
numbers of these organisms remained at a constant level (105
organisms per gram of feces) until the animals were transferred to the
concrete pens at eight weeks of age. Staphylococci and Bacillus sp. first
were observed in feces after the animals had been transported to their
pens and fed experimental rations. These organisms were present
sporadically in all groups of animals at low numbers (104 or less

BACTERIA OF SWINE

333

Figure 1. Fecal flora of experimental swine first eight weeks of life (logio number
organisms per gram feces).

organisms per gram feces) throughout the study. A sharp increase in
numbers of coliforms (107 organisms per gram feces) and enterococci
(108 organism per gram feces) also was observed in all animals at this
time (eight weeks) (Fig. 1). The numbers of coliforms and enterococci
declined during the 9th and 10th weeks to a concentration of
approximately 106 organisms per gram of feces. After the ninth week,
the anaerobic organisms were present at concentrations of 10 10 to 1011
organisms per gram of feces and remained at this level for the duration
of the study. Bacteroides sp. (other than B. nigrescens) were the
predominating anaerobes recovered from those samples. Although
Veillonella sp. and Bacteroides nigrescens (melaninogenicus) were
recovered in large numbers from certain portions of the intestinal
tracts of all animals, these organisms were rarely recovered from fecal
samples. Clostridium sp. were rarely recovered from fecal samples but,
when recovered, were present at concentrations of 104 or less organisms
per gram of feces. Except for sporadic appearance of staphylococci,
Bacillus sp., B. nigrescens, Veillonella sp., and Clostridium sp., the
fecal bacterial flora of all the pigs was similar (Figs. 2-3).

Bacterial Flora of the Intestinal Tracts

The anterior portions of the intestinal tracts contained fewer
cultivatable bacteria (105 to 107 organisms per gram of material) than
the posterior portions (1010 to 1012 organisms per gram of material).

334

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

12
1 1
10
9
B
7
6

Sample Interval (Months)
Sucrose Diet

Lactobaci Hi
Col i -forms

Enterococci
Bacteroi des

Figure 2. Fecal flora of swine maintained on sucrose diet and on sucrose-cholesterol diet
(logio number organisms per gram feces). Monthly average of weekly samples.

Lactobacilli were the predominant organisms cultivated from the
duodenum (105 ogranisms per gram material). Direct smears of the first
dilution (10_1) of the duodenal contents revealed large numbers of
Gram-positive streptococci (presumably Peptostreptococcus sp.) in
close association with cellular debris, but efforts to cultivate these

BACTERIA OF SWINE

335

5 6 7 B 9

Sample Interval (Months)
Cottonseed Oil Diet

12

11

10

9

8

7

6

5

4

Sample Interval (Months)
Cottonseed Oi 1 -Choi esterol Diet

I

Lactobaci 1 1 i
Col i -forms

Enterococci
Bacteroi des

Figure 3. Fecal flora of swine maintained on cottonseed oil diet and on cottonseed oil-
cholesterol diet (Logio number organisms per gram feces). Monthly average of weekly
samples.

organisms failed. Coliforms, enterococci, staphylococci, and Bacillus
sp. were not recovered from this portion of the intestinal tracts in any
of the pigs.

Gram-positive streptococci similar to those observed in the
duodenum also were observed in preparations from the ileum, but

336

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

could not be cultivated. Lactobacilli were the predominant aerobic
organisms (107 organisms per gram) in this region of the intestinal
tracts, although coliforms and enterococci were present in low
numbers (103 to 104 organisms per gram).

Gram-positive rods ( Bifidobacterium sp.) were the predominant
anaerobic bacteria recovered from the cecum. Fusobacterium sp. also
could be recovered from 107 to 109 dilutions of the cecal samples.

B. nigrescens, Fusobacteria sp., and Verillonella sp. were the
predominant recoverable anaerobes of the rectum (1010 organisms per
gram of feces). The numbers and kinds of bacteria from various
intestinal segments appeared to be similar in all experimental animals.

Preliminary studies indicated that the anaerobic broth used in the
present study contained compounds that interfered with analysis of
bile acid degradation products; thus Thioglycollate medium was
utilized as the basal medium for assay of bile acid degradation. All
aerobic isolates grew well, but many anaerobic isolates failed to grow
or grew poorly in the assay media. Rabbit serum and menadione when
added to the medium increased the viability of many of the isolates in
the assay medium, but these compounds gave rise to products that
interfered with the assay. Therefore, only those cultures that were
capable of growing in the unsupplemented assay media were tested for
ability to degrade bile acid conjugates.

Most of the Bacteroides sp. tested were capable of hydrolyzing
glycine and taurine from their cholic and deoxycholic acid conjugates.
None of the Bacteroides sp. isolates tested were capable of further
degrading the bile acid conjugates under the conditions of the analysis.
Exclusive of the Clostridia sp., none of the anaerobic Gram-positive
organisms grew in the assay medium. Clostridia sp. was infrequently
recovered, and thus was believed not to be a major constituent of the
swine microflora.

Few of the Veillonella sp. and Fusobacterium sp. grew in the assay
media, but of those that did grow, none was capable of structurally
altering bile acids under the test conditions. There appeared to be no
relationship between diet and distribution of those anaerobes capable
of degrading conjugated bile acids, because organisms with the
capacity were isolated from animals representing all dietary groups.

Of the aerobes tested, only enterococci and E. coli isolated from
animals on the cottonseed oil diets were capable of deconjugating the
bile acids. The enterococci (Streptococcus faecalis and Streptococcus
durans) from these animals removed glycine and taurine from both
cholic and deoxycholic acid conjugates. E. coli isolates were capable
of cleaving taurine from cholic acid and glycine and taurine from
deoxycholic acid (Fig. 4).

BACTERIA OF SWINE

337

TC

GC

GD

Cottonseed Oil Diet

TC= Sodium
GC= Sodium
TD= Sodium
GD= Sodium

Taurochol ate
G1 ycochol ate
Deoxychol ate
Deoxychol ate

Cottonseed Oi 1 -Choi esterol
Diet

HU Bacteroi des 0 Col i -forms
HI Enterococci

Figure 4. Percent bacteroides, coliform and enterococci isolates from animals on dietary
treatments capable of deconjugating bile acids (90-123 isolates per species per
treatment group).

DISCUSSION

Graber et al. (1966a) observed that the intestinal bacterial flora of
swine was similar to that of man. The species distribution of intestinal
microflora of the swine used in their study were not altered during a

338

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

nine-month period by the manner in which the animals were fed (ad
libitum or restricted), by sterilizing the feed or by incorporating fat (1.0
per cent lipid) or sucrose (72 percent carbohydrate) in the rations. The
present study confirms and extends those observations to include
animals maintained for 14 months on rations supplemented with
sucrose, sucrose-cholesterol, cottonseed oil, and cottonseed oil-
cholesterol.

While the numbers and kinds of organisms do not appear to have
been altered by various dietary factors of this study, the activities of
certain bacterial groups appear to have been influenced by those
factors. Intestinal bacteria may play a role in the metabolism of bile
acids and their precursor, cholesterol, by either favoring the resorption
of those compounds after deconjugation or by facilitating their
excretion in the feces after being further metabolized to poorly
absorbed products.

The present study suggests that rather than influencing overall
composition of the intestinal microflora, certain dietary factors may
influence activities of various groups of that microflora. Most of the
studies that have considered the role of intestinal bacteria upon sterol
metabolism have emphasized the activities of the anaerobic intestinal
bacteria. The present study suggests not only that certain facultative
anaerobes are capable of participating in bile acid deconjugation, but
that dietary factors may influence the development of those intestinal
bacteria.

ACKNOWLEDGMENTS

I wish to express appreciation to Dr. Glen Mott, formerly of the
Department of Biophysics and Biochemistry, Texas A & M University,
for his assistance in analysing samples for bile acid deconjugation and
to Ms. Willie Mae Charanza for microbiologic technical assistance.

This research was conducted by the Texas Agricultural Experiment
Station under the auspices of the Expanded Research Program for
Swine Health project 6706.

LITERATURE CITED

Drasar, B. S., M. J. Hill, and M. Shiner. 1966. The deconjugation of bile salts by human
intestinal bacteria. Lancet 1:1237-1238.

Eyssen, H. 1973. Role of the gut microflora in metabolism of lipid and sterols. Proc.
Nutr. Soc. 32:59-63.

Graber, C. D., R. W. Moore, M. Suzuki, H. E. Redmond, R. M. O’Neal, and B. M.
Lockhart. 1966a. Autochthonous intestinal bacterial flora and cholesterol levels in
specific pathogen free swine fed high-lipid and high sucrose diets. J. Bacteriol.
92:1290-1297.

BACTERIA OF SWINE

339

Graber, C. D., R. M. O’Neal, and E. R. Rabin. 1966b. Effect of dietary sodium cholate
and lactose on the fecal microflora and blood cholesterol of rats. Gastroenterology
51:357-363.

Johansson, K. R., and W. B. Sarles. 1949. Some considerations of the biological
importance of intestinal microorganisms. Bacteriol. Revs. 13:25-45.

Kellog, T. F. 1971. Microbiological aspects of enterohepatic neutral sterol and bile acid
metabolism. Fed. Proc. 30:1808-1814.

Kritchevsky, D. 1979. Diet, lipid metabolism and aging. Fed. Proc. 38:2001-2006.

- . 1964. Experimental atherosclerosis. Pp. 63-130, in Medical chemistry, vol. 2 (R.

Paolet, ed.), Academic Press, New York.

Meyer, R. C., E. H. Bohl, R. D. Henthorn, V. L. Tharp, and D. E. Baldwin. 1963. The
procurement and rearing of gnotobiotic swine. Lab. Animal Care. 13:655-663.

Midwedt, T. 1974. Microbial bile acid transformation. Amer. J. Clin. Nutr. 27:1341-1347.

Midwedt, T., and A. Norman. 1967. Bile acid transfomrations by microbial strains
belonging the genera found in intestinal contents. Acta Path. Microbiol. Scan. 71:629-
638.

Morrison, F. B. 1959. Feeds and feeding. Morrison Publishing Co., Ithica, New York,
22nd ed.

Porter, J. R., and L. F. Rettger. 1940. Influence of diet on distribution of bacteria in
the stomach, small intestine and cecum of the white rat. J. Inf. Dis. 66:104-110.

Shimada, K., K. S. Bricknell, and S. M. Finegold. 1969. Deconjugation of bile acids by
intestinal bacteria: Review of literature and additional studies. J. Inf. Dis. 119:273-
281.

Suzuki, M., R. W. Moore, H. E. Redmond, D. H. Lewis, Y. Hosoda, R. Reiser, and R.
M. O’Neal. 1969. Arterial lesions in SPF and Conventional Pigs. Fed. Proc. 28:448.

Varel, W. H., W. G. Pond, J. C. Pekas, and J. T. Yen. 1982. Influence of high-fiber diet
on bacterial populations in gastrointestinal tracts of obese- and lean- genotype pigs.
Appl. Evniron. Microbiol. 44:107-112.

Wilbur, R. D. 1959. The intestinal flora of the pig as influenced by diet and age. Ph.D.
dissertation, Iowa State College, Ames, Iowa.

Young, G., N. Underdahl, and R. W. Hinz. 1965. Procurement of baby pigs by
hysterectomy. Amer. J. Vet. Res. 16:123-125.

EFFECTS UPON SELECTED MARINE ORGANISMS
OF EXPLOSIVES USED FOR SOUND PRODUCTION
IN GEOPHYSICAL EXPLORATION

by THOMAS L. LINTON

Department of Wildlife and Fisheries Sciences

Texas A&M University

College Station, Texas 77843

ANDRE M. LANDRY, JR.

Department of Marine Biology
Texas A&M University at Galveston
Galveston, Texas 77553

JAMES E. BUCKNER, JR.

County Extension Marine Agent
P.O. Box 699
Anahuac, Texas 77514

and ROBERT L. BERRY
Texas Parks and Wildlife Department
4200 Smith School Road
Austin, Texas 78744

ABSTRACT

Survival rate and extent and nature of injury were monitored for red drum ( Sciaenops
ocellatus), black drum ( Pogonias cromis), blue crab ( Callinectes sapidus ), white shrimp
(. Penaeus setiferus), and American oyster ( Crassostrea virginica ) held in cages at
logarithmic distances (one to 46 meters) from where a strand of commercially available
explosive (Primacord with 100 grams of powder per 33 centimeters), commonly used to
produce sound waves for seismic exploration, was detonated in a shallow water
environment. Survival of test organisms varied with species, depth of cage, and distance
from detonation site. Fish held at the surface exhibited low mortality, whereas those in
bottom cages closest to site of detonation (one and 23 meters away) exhibited mortality
rates between 40 and 100 percent. The swimbladder, kidney, and peritoneum were the
most frequently damaged organs in fish. Shrimp exhibited modest mortality rates at all
stations and water depths. Survival of shrimp did not appear to be related to distance
from detonation. Blue crab survival appeared to be directly related to distance from
detonation site. Survival of oysters was high at all stations and inversely proportional
to distance from sound source. Varying results among test organisms were attributed to
pressure wave characteristics associated with charge detonation. Comparable testing is
needed during summer months to determine effects under “worse case” conditions when
greater numbers and life stages of organisms are present and ambient conditions more
stressful in these shallow water environments.

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

342

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

INTRODUCTION

“In 1842 David Milne invented an instrument for recording and measuring the
movements of the ground during an earthquake and called it a seismometer, the
earliest seismological term. A few years later the name seismograph was given to
an instrument erected in 1855 by L. Palmieri in the observatory on Versuvius.

“The first practical use of the seismograph for anything except recording
earthquakes happened during World War I when German scientist, Dr. L. Mintrop,
invented a portable seismograph for the German army to use for locating Allied
artillery. He would set up three seismographs in known positions along the
battlefront opposite where there was an Allied gun bombarding them. When a gun
fired, a record would be made of the earth vibrations and the exact location of the
gun could be calculated so accurately that often the first shot from a German gun
would make a direct hit.

“The Germans found that errors were introduced into their distance calculations
because velocities varied with the geological formations through which the
vibrations passed, and certain assumptions about geology had to be made to
compute the distances. After the war, Dr. Mintrop decided to reverse the process. He
would set off a charge of dynamite and record the vibrations produced in the earth
on his same portable seismographs, but this time he would measure the distances
and compute the geology. And that was the birth of the present day seismograph
contracting industry.

“In 1924, the Gulf Production Company brought one of Dr. Mintrop’s crews to
Texas to hunt shallow salt domes” (Petty 1976).

This new technique introduced by Gulf Production Company was
quickly adopted by industry. Since that time, the increased frequency
of use of explosives in exploration for oil and gas deposits has been
substantial (Gowanlock and McDougall 1944). Because of concern
regarding the side effects explosives used in seismic exploration might
have upon aquatic organisms, field testing was advocated. Numerous
studies have been conducted to assess these side effects. Field
observations indicate explosives cause mortality or injury to aquatic
organisms in the vicinity of the detonation (Gowanloch and
McDougall 1944; Aplin 1947; Fitch and Young 1948; Coker and Hollis
1950; Hubbs and Rechnitzer 1952; Sieling 1954; Kemp 1956; Spears
1980). A review of these studies is given in Linton et al. (1984).
Attempts have been made to better understand these effects through
conducting experiments designed to determine their extent and
magnitude upon aquatic organisms at selected depths and distances
from the detonation site. However, these studies were limited by lack
of standardization, small number of replications, or differences in
study sites. Nevertheless, they described the general effects explosives
have upon aquatic organisms.

Governmental agencies that permit geophysical exploration have
authored regulations intended to minimize detrimental effects on
marine organisms. The study reported herein was conducted to provide
additional information for judging the adequacy of those regulatory
programs in effect or those proposed.

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

343

STUDY AREA AND METHODS

Tests to determine effects of explosives used in geophysical
exploration on aquatic oganisms were conducted in conjunction with
a commercial seismic survey that was being conducted in East Bay,
Texas, on 8 and 9 December 1981 (Fig. 1). The study site, located
approximately three kilometers southwest of Smith Point, Texas,
exhibited substrate, depth, and salinity conditions typical of the
shallow water bay systems of the upper Texas coast. Water depth
averaged 2.5 meters, whereas substrate consisted of soft mud and oyster
shell reefs. Water temperature and salinity averaged 6° C and 12 parts
per thousand, respectively.

344

THE TEXAS JOURNAL OF SCIENCE-VOL. XXXVII, NO. 4, 1985

Red drum ( Sciaenops ocellatus), black drum ( Pogonias cromis),
white shrimp ( Penaeus setiferus), blue crab (Callinectes sapidus), and
American oyster ( Crassostrea virginica ) were used as test organisms; all
were collected in Trinity Bay, north of Smith Point, within one
kilometer of the detonation site. Red and black drum were captured
with trammel nets, blue crabs with commercial traps, oysters with a
commercial dredge, and white shrimp with an otter trawl. Fish, crabs,
and oysters were transported in aerated tanks to open-water holding
pens immediately after capture. They were held there for at least 24
hours prior to the experiment to monitor injuries and mortalities
resulting from capture and handling. Dead or injured individuals were
discarded. Only individuals that had no visible external damage and
exhibited normal behavior (that is, active swimming, up-right
orientation) were used as test animals. No attempt was made to
determine shrimp mortality or injury resulting from capture or
handling. Shrimp were captured the day of the experiment and
transferred directly to test cages. All organisms were acclimated to test-
cage conditions for at least one hour prior to detonation.

Test animals were selected for uniform size within species. Range
and average total lengths were: red drum, 30-39 centimeters, 32
centimeters; black drum, 21-30, 23; white shrimp, 7-11, 8.5; blue crab
(carapace width), 14-18, 17. Oysters were not measured individually,
but none was less than 15 centimeters in total shell length.

Cylindrical holding cages, 90 by 75 centimeters, and enclosed with
1.8 centimeter nylon mesh webbing housed test organisms during the
experiment. Cages holding shrimp also contained a 0.5 centimeter
mesh nylon liner. In order to make more manageable the number of
cages used, compatible organisms (that is, blue crabs and oysters or red
drum and black drum) shared cages. Consequently, 10 crabs and 10
oysters were caged together and five red drum were caged with five
black drum. White shrimp was the only species caged alone (10
individuals per cage).

Surface and bottom cages were deployed at five stations arranged
perpendicular to, and at logarithmic distances of, one, 11, 23, and 46
meters from the detonation line (Fig. 2). A control array, using the
same set-up of organisms and cages as at the other four sites, was
established at a distance of 136 meters from the detonation site. Fish
and shrimp were held in paired cages at surface and bottom strata of
each station (four cages per strata), whereas crabs and oysters were only
deployed in paired bottom cages. Control organisms received the same
treatment (that is, method of capture, holding time, and so forth) as
test organisms, with the exception that they were not in the water at
the time of the explosion.

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

345

Figure 2. Test array of cages at surface and bottom in relation to detonation cord.

A 33-meter strand of 100 gram/33 centimeter Primacord detonation
cord was laid perpendicular to the transect of test cages. It was
positioned to form the top of the letter “T” and the line of cages
forming the base (Fig. 2). Both ends of the Primacord were weighed
to hold the cord on the bottom during detonation. A blasting cap
attached to one end of the Primacord was wired to a detonator on
board the seismic boat.

All personnel and boats were removed from the test site prior to
detonation. Control cages at the 136-meter station were removed from
the water seconds before detonation and returned shortly thereafter.
Observers reentered the test site immediately after detonation, raised
test cages, and recorded visible changes (behavioral and physical) and
mortality among test animals. Criteria used to denote death were as
follows: fish — cessation of opercular movement; oysters — shell
permanently agape; shrimp — cessation of gill movement; and crabs —
cessation of movement of chela, appendages, and mouth parts.

Dead organisms were removed from cages, iced in labeled plastic
bags, and held for laboratory analyses. Cages with living organisms
were returned to the position they occupied at the time of detonation
and observed 24 hours later. After the second set of observations was
completed, all organisms were removed from the cages and separated
as to living or dead. With the exception of live oysters, they were
placed on ice in plastic bags and returned to the laboratory for internal
examination. Live oysters were transferred to the Texas A&M
University at Galveston boat basin in moistened jute sacks and
monitored thrice weekly for two weeks.

Test organisms, except oysters, were measured to standard length
(fish), total length (shrimp), or carapace width (crabs). Caudal fin

346

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 1. Mortality rate (percent) for test species as a function of cage depth, lapsed time,
and distance from detonation site.

Species

Number

per cage

Cage

depth

Lapsed
time (hr)

Distance (m) from detonation site

1

11

23

46

Control

Black Drum

10

surface

0

10

0

0

0

0

24

0

0

0

0

0

10

bottom

0

90

60

20

0

0

24

10

10

20

90

20

Red drum

10

surface

0

0

0

0

0

0

24

0

0

0

0

0

10

bottom

0

90

0

0

0

0

24

0

0

0

0

0

White shrimp

20

surface

0

0

5

0

20

5

24

5

5

25

0

0

bottom

0

5

30

5

0

0

24

0

5

5

0

Blue crab

20

bottom

0

40

35

35

10

0

24

20

15

0

0

0

American oyster

20

bottom

0

5

5

10

10

0

24

0

0

5

5

0

damage due to handling, caging, or Primacord detonation (or some
combination of these factors) necessitated that standard length, instead
of total length, be recorded for fish. Although fish, shrimp, and crabs
were examined for external abrasions or wounds, no attempt was made
to quantify external injury. The coelomic cavities of fish were dissected
and examined for internal damage.

RESULTS

Survival of black drum was greatest among those held in cages at
the surface (Table 1). Nine of the 10 fish held in one-meter surface
cages were alive immediately after detonation (10 percent mortality)
and eight of those nine were alive and outwardly appeared to be
healthy after 24 hours (total mortality after 24 hours, 20 percent). Black
drum were alive at all surface cage locations immediately and 24 hours
after detonation. All exhibited normal behavior and no visible external
damage was observed.

Mortality occurred among black drum held in cages at all bottom
locations. Except for those at the 46-meter station, mortality rate was
generally highest at bottom stations closest to the detonation zone, and
partially time dependent. All black drum in the one-meter bottom
cages died, with 90 percent dying immediately after detonation. All but
six from 11-, 23-, and 46-meter bottom cages survived the initial
impact of detonation. However, survival rates at these stations 24
hours after detonation were 30, 60, and 10 percent, respectively. Black

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

347

Table 2. Internal injury rate (percent) for black drum and red drum as a function of
cage depth and distance from detonation (S and B denote surface and bottom cage,
respectively).

Distance from detonation

lm

11m

23m

46m

Control

Species

S

B

S

B

s

B

S

B

S

B

Black Drum

Swimbladder

0

22

0

20

0

20

0

50

0

60

Kidney

40

100

40

90

70

100

50

100

20

0

Liver

30

0

0

20

40

0

10

0

0

0

Gallbladder

10

0

0

0

0

10

0

0

0

0

Stomach

0

11

0

0

0

0

0

0

0

0

Peritoneum

10

78

0

80

0

50

0

0

0

0

Red Drum

Swimbladder

10

91

0

60

0

90

0

0

0

0

Kidney

10

91

0

40

0

50

0

42

0

0

Liver

10

46

0

20

40

10

50

0

0

0

Gallbladder

0

0

0

0

0

0

0

0

0

0

Stomach

0

0

0

0

0

0

0

0

0

0

Peritoneum

0

18

0

0

0

0

0

18

0

0

drum in control cages survived initial effects of detonation but
exhibited an overall survival rate of only 80 percent.

External injury to black drum was minor, whereas internal injury
was substantial. Loss of opercular scales as the only visible external
injury. Internal injuries were diverse and most numerous in drum
from bottom cages (Table 2 and Fig. 3). Drum exhibited injury in six
coelomic cavity sites — swimbladder, kidney, liver, peritoneum,
gallbladder, and stomach. The specific nature of injury to coelomic
cavity sites is given in Table 3. Kidney damage was the most frequently
recorded injury (Table 2 and Fig. 3) and was prevalent in surface- (42
percent) and bottom-caged (82 percent) drum. At least 50 percent of all
black drum in bottom cages at one-, 11- and 23-meter stations
experienced peritoneum-related injuries. Swimbladder damage was
present in drum from all bottom cages, and was greatest at the 46-
meter station (50 percent) and among controls (60 percent). Other
injuries were less frequently observed (Fig. 3) and exhibited no
discernable pattern of occurrence (Table 2). Most of the black drum
that survived detonation exhibited internal damage, with injury rates
highest among those in bottom cages (Table 4).

All red drum except those in bottom cages at the one-meter station
survived detonation (Table 1). Several in the one-meter bottom cages
were alive but in a stunned, disoriented (belly up) state following
detonation and died shortly thereafter.

348

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

INJURY

Figure 3. Anatomical injuries in six internal structures in black drum ( Pogonias cromis)
and red drum ( Sciaenops ocellatus ) from surface and bottom tests. (Ten animals in
each cage.)

Red drum, exhibited no visible external injuries. Internal injuries
were limited to the swimbladder, kidney, liver, and peritoneum (Fig.
3 and Table 3) and were mostly evident in drum from bottom cages
at one-, 11- and 23-meter stations (Table 2). Many red drum alive 24
hours after detonation exhibited some sort of internal injury (Table 4).
These injury rates were always highest in bottom-caged fish.

In addition to observations of fish held in cages, other fish species
appeared to be influenced by the detonation. Disoriented bay anchovy
(. Anchoa mitchilli) and Atlantic croaker ( Micropogonias undulatus )
were observed floating near the surface well beyond the 136-meter
control station.

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

349

Table 3. Nature of internal injuries in black drum and red drum exposed to detonation.

Peritoneum

Swimbladder

Kidney

Liver and gall bladder

Stomach and intestine

a. parietal peritoneum torn away from coelomic cavity

b. visceral peritoneum ruptured

a. rupture and stretching of swimbladder wall

b. capillary hemorrhage within swimbladder wall

a. renal portal vein hemorrhage

b. rupture of kidney tubules, mesovarium, and mesorchium

a. rupture of liver

b. rupture of gall bladder wall

c. rupture of cystic duct

a. rupture of stomach wall

b. rupture of intestine wall

White shrimp suffered mortalities in all test cages and depths but
exhibited no well-defined pattern relative to survival and distance from
detonation site (Table 1). Shrimp in surface cages exhibited relatively
high survival (>95 percent) immediately after detonation, but
experienced mortality rates ranging from five percent one-meter
station) to 15 percent (23-meter station) thereafter. The unexpected
mortality (five percent) among shrimp in control cages and 95 percent
overall survival rate for shrimp at the one-meter station reflect the lack
of discernable trends for surface-caged individuals. Results for bottom-
caged shrimp were also mixed. Highest and lowest survival rates were
recorded at stations closest to the site of detonation (one-meter station,
95 percent; 11 -meter station, 65 percent). Survival in more distant
bottom cages fluctuated between 88 percent at 23 meters and 100
percent at 46 meters. No mortality occurred among shrimp in control
cages. In contrast to surface trends, most shrimp mortality in bottom
cages occurred immediately after detonation.

Survival of blue crabs in bottom cages was directly related to
distance from the site of detonation (Table 1). Mortality, ranging from
40 percent at one meter to 10 percent at 46 meters, was greatest
immediately after detonation. Blue crab deaths 24 hours later were few
and confined to one- and 11 -meter stations where survival rates were
40 and 50 percent, respectively. No mortality was observed in control
cages.

Survival of American oysters, which were tested only in bottom
cages, was high and inversely related to distance from the site of
detonation (Table 1). Survival rates ranged from 95 percent at the one-
and 11 -meter stations to 85 percent at the 23- and 46-meter stations.
Most oyster deaths occurred immediately after detonation. No mortality
occurred in oysters originally held in control cages during the two-
week period they were observed following the experiment.

350

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 4. Number of black drum and red drum survivors and percent (in parentheses)
of those exhibiting internal injury as a function of cage depth and distance from
detonation.

Species

Cage

depth

1

Distance (m) from detonation

11 23 46

Control

Black drum

surface

9(33)

10(40)

10(80)

10(60)

10(0)

bottom

0

3(100)

6(100)

1(100)

8(20)

Red drum

surface

10(20)

10(0)

10(40)

10(50)

10(0)

bottom

1(100)

10(70)

10(100)

10(82)

10(0)

DISCUSSION

Survival rates of organisms in this study varied with species, depth,
and distance from the site of detonation. This variability is similar to
that reported by other workers (Yelverton et al. 1975; Sakagucki et al.
1976; Hill 1978; Wright 1980) who found mortality is a function of
size, shape, species, orientation of the organism to the shock wave,
amount, and type of explosive, detonation depth, water depth, and
bottom type. Roguski and Nagata (1970) reported that unknown
factors, possibly orientation of organisms at time of detonation, caused
marked variation in mortality at any given depth and distance.

Autopsy results from fish in this study agree with those of other
investigators who found the swimbladder the most frequently damaged
organ when exposed to explosive detonations (Kearns and Boyd 1965;
Christian 1973; Falk and Lawrence 1973; Yelverton et al. 1975).
Approximately 23 percent of the fish in this study exhibited
swimbladder damage. Other frequently reported injuries caused by
explosives include rupture and hemorrhage of the kidney, liver,
gonads, spleen, and sinus venosus, and torn adipose tissue in the body
wall (Tyler 1960; Yelverton et al. 1975; Sakagucki et al. 1976). Kidney
and liver damage was observed in 46 and 22 percent, respectively, of
the two drum species tested in this study.

Certain findings of the present study differ from results of other
studies in Texas of the effects of seismic activities upon marine
organisms. Kemp (1956) tested effects of explosives (20 kilograms of
dynamite) on surface- and bottom-caged fish, shrimp, oysters, and blue
crabs in studies conducted in Corpus Christi Bay and reported: 1)
shrimp and crabs completely immune to explosives; 2) fish subject to
a mortality rate as high as 70 percent (especially at the point of
explosion and near the bottom), and no impact beyond a 15-meter
radius from the detonation zone; and 3) oyster mortality varying from
10 to 30 percent, with damage most severe within a 7.5-meter radius
of detonation zone. Conversely, Spears (1980) found in studies
conducted in Aransas Bay, Texas, that most fish, shrimp, and crabs
caged near the surface and bottom suffered 100 percent mortality

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

351

within 15 and 22.5 meters of the detonation zones. Our results differed
from those discussed above in that 1) blue crab mortality rates
diminished with distance from the site of detonation; 2) oyster
mortality was greatest in cages furthest from the detonation site; and
3) mortality of most surface-caged organisms was relatively minor.

Varying results among our test species probably were due to several
factors. One primary factor is the nature and impact of pressure waves
produced by Primacord detonation. Underwater explosions produce
two types of positive pressure waves: an initial compression or shock
wave followed by a bubble pulse wave (Wright 1980). The initial
compression wave has been found to be most harmful to fishes (Hubbs
et al. 1960). High mortality rates associated with black and red drum
in bottom cages nearest (one-meter and 11 -meter stations) the
detonation site probably were due to the initial compression wave
produced by the blast. The lethality of this compression wave is
evident by the fact that more than 64 percent of all fish mortalities
occurred immediately after detonation. In addition, the swimbladder is
a major site of damage resulting from these shock waves (Wright 1980).
Correspondingly, 23 percent of fish exposed (both living and dead) in
this study exhibited swimbladder damage.

The tendency among black drum and oysters for mortality rates to
increase with distance from the detonation site may have been due to
the pattern of the shock wave transmitted through the water column.
Trasky (1976) reported shock waves from charges detonated in the
seabed travel in a well-defined cone expanding toward the surface. At
a given distance from the source, the arrival of the shock wave causes
water pressure to rise almost instantly and then to decrease
exponentially to a negative pressure (rarefraction). Hubbs et al. (1960)
found this rapid compression and rarefraction killed some species of
fish. Mortalities and internal injuries among organisms caged at 23-
and 46-meter stations may have been due to lethal effects of the
conelike shock wave. The disoriented bay anchovy and Atlantic croaker
observed floating near the surface well beyond the 136-meter control
station give further indication that a conelike shock wave eminated
from the explosive.

Mortality rates calculated during this study may have been
influenced by one or more variables. Deaths among control organisms
indicated that handling influenced mortality rates. Retrieval of control
cages prior to detonation and their immediate redeployment may have
produced sufficient stress to account for the mortality or injury
observed among black drum and white shrimp. Handling may have
been responsible for deaths of black drum and shrimp in control cages,
whereas the combination of handling, bagging, and transport of black
drum targeted for autopsy may have produced the high occurrence (30

352

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

percent) of swimbladder injury in this species. Further research is
needed to determine the role handling and transport plays in injury
and mortality of test organisms. Conversely, the 24-hour observation
period may have been insufficient to obtain total mortality rates. Wiley
and Wilson (1974) found that fish with broken swimbladders show
remarkable recuperative powers but the loss of buoyancy control
increases their vulnerability to predation. We suspect that, had testing
been conducted for a 96-hour period, mortality rates in fishes would
have been greater. Our tests were conducted in winter (water
temperature 6° C) when ambient stress to nektonic organisms is less
than that in warmer seasons. Summer months exhibiting water
temperatures typically between 25° and 30° C produce increased
physiological stress in aquatic organisms because of increased
metabolic demand and reduced dissolved oxygen in the water.
Therefore, the season in which our study was conducted possibly
influenced mortality rates; mortality rates would be greater if ambient
conditions already are stressing to resident organisms.

ACKNOWLEDGMENTS

The authors wish to thank the Marine Advisory Service of the Texas
Agricultural Extension Service and the Texas A&M University Sea
Grant College Program for providing the funds for this project. We
also thank Mr. Leland Roberts, Chief, Resource Protection Branch,
Fisheries Division, Texas Parks and Wildlife Department, for his
paticipation, and that of his staff, in this project. The assistance
provided by the commercial geophysical contractors, working East
Bay, also is gratefully acknowledged.

We are especially indebted to Mr. Joe Nelson, Chairman, Chambers
County Marine Advisory Committee, for coordinating the efforts of the
local commercial fishermen that participated in the project. Mr. Ben
Nelson, member, Chambers County Marine Advisory Committee,
donated time to operate his outboard boat for transport of materials
used in the project. Mr. Robert Amason captained the vessel and
Messers Jay and Charlie Woody assisted in handling the test organisms
used in the project.

LITERATURE CITED

Aplin, J. A. 1947. The effect of explosives on marine life. California Fish and Game
33:23-30.

Christian, E. 1973. The effects of underwater explosions on swimbladder fish. Naval
Ordnance Lab, no. 73-103, 41 pp.

Coker, C. M., and E. H. Hollis. 1950. Fish mortality caused by a series of heavy
explosions in Chesapeake Bay. J. Wildlife Manage. 14:435-444.

EFFECTS OF EXPLOSIVES ON MARINE ORGANISMS

353

Falk, M. R., and M. J. Lawrence. 1973. Seismic exploration, its nature and effect on
fish. Environment Canada Tech. Rep. CENT-73-9:l-51.

Fitch, J. E., and P. H. Young. 1948. Use and effect of explosives in California coastal
waters. California Fish and Game 34:53-73.

Gowanlock, J. N., and J. E. McDougall. 1944. Louisiana experiments pave way for
expanded oil research. Louisiana Conservationist 3(1):3, 6.

Hill, S. H. 1978. A guide to the effects of underwater shock waves on Artie marine
mammals and fish. Unpub. manuscript, Pacific Bay, Sydney, British Columbia, 50

pp.

Hubbs, C. L., and A. B. Rechnitzer. 1952. Report on experiments designed to determine
effects of underwater explosives on fish life. California Fish and Game 38:333-366.

Hubbs, C. L., E. P. Schultz, and R. Wisner. 1960. Preliminary report on investigations
of effects on caged fishes from underwater Nitro-Carbo-Nitrate explosions. Univ.
California Scripps Inst. Ocean, Unpub. manuscript, 21 pp.

Kearns, R. K. , and F. C. Boyd. 1965. The effects of a marine seismic exploration on fish
populations in British Columbia coastal waters. J. Canadian Soc. Expl. Geophy.
12:83-106.

Kemp, R. J. 1956. Do seismographic explosions effect marine life? Texas Game and
Fish 1 4(9): 1 1-13.

Linton, T. L. , A. M. Landry, N. Hall, and D. LaBomascus. 1984. Report to the
International Association of Geophysical Contractors; data base development for
geophysical exploration guidelines, an annotated bibliography and literature review,
64 pp.

Petty, O. S. 1976. Seismic Reflections. Geosources, Inc., Houston, Texas, 79 p.

Roguski, E. A., and T. H. Nagata. 1970. Observations on the lethal effect of under ice
detonations on fish. Alaska Dept. Fish and Game Rept. 13 pp.

Sakagucki, S., O. Fukuhara, S. Umezawa, M. Fujiya, and T. Ogawa. 1976. The
influence of underwater explosion on fishes. Bull. Nansei Reg. Fish. Res. Lab. 9:33-
65.

Sieling, F. W. 1954. Experiments on the effects of seismographic exploration on oysters.
Proc. Nat. Shellfish Assoc. 1953:93-104.

Spears, R. W. 1980. The effect of Primacord on selected marine organisms. Texas Parks
and Wildlife Dept. Unpub. rept. 8 pp.

Trasky, L. L. 1976. Environmental impact of seismic exploration and blasting in the
aquatic environment. Alaska Dept. Fish and Game Rept. 13 pp.

Tyler, R. W. 1960. Use of dynamite to recover tagged salmon. U.S. Fish and Wildlife
Service, Sp. Sci. Rep. Fisheries 353:1-9.

Wiley, M. L., and J. S. Wilson. 1974. Environmental effects of explosive testing.
Chesapeake Biol. Lab. Ref. No. 74-9, 120 pp.

Wright, D. G. 1980. A discussion paper on the use of explosives in the marine waters
of the North West Territories, (draft). Guidelines Fish Habitat Secion, Dept, of
Fisheries and Oceans, Govt, of Canada, Winnepeg, Manitoba.

Yelverton, J. T., D. R. Richmond, W. Hicks, K. Sanders, and E. R. Fletcher. 1975. The
relationship between fish size and their response to underwater blast. Defense Nuclear
Agency (DNA), Washington, D.C., Topical Rep. DNA 3677T, 42 pp.

)

V

RECORDS OF THE SPOTTED SKUNK AND
LONG-TAILED WEASEL FROM THE LLANO
ESTACADO OF TEXAS

by J. KNOX JONES, JR., ROBERT R. HOLLANDER,
and DAVID A. McCULLOUGH

The Museum and Department of Biological Sciences
Texas Tech University
Lubbock, Texas 79409

ABSTRACT

There is but one published record of the eastern spotted skunk, Spilogale putorius,
from the Llano Estacado of western Texas and no previous reports of the long-tailed
weasel, Mustela frenata, from that region. Specimens of both species are reported,
documenting occurrence of these two taxa on the Llano. Key words : Mustela, Spilogale,
Texas Llano Estacado.

In the most recent summary of the distribution of Texas mammals
(Davis 1974), neither the long-tailed weasel (Mustela frenata) nor the
eastern spotted skunk ( Spilogale putorius ) were mapped as occurring
on the Llano Estacado of the northwestern part of the state. Similarly,
Schmidly (1984) also excluded the Llano Estacado from the known
distribution of the two species in a publication on Texas furbearers.
Hall (1981), however, included western Texas in the range of both
taxa, on supposition in the case of M. frenata, and based on a single
published record overlooked by both Davis and Schmidly in the case
of S. putorius.

Specimens of these two mustelids in The Museum of Texas Tech
University (TTU) confirm that both occur over much of the Llano
region (see Fig. 1), and are discussed below. Measurements are given
in millimeters.

Mustela frenata neomexicana (Barber and Cockerell, 1898). — The
long-tailed weasel evidently is widely distributed on the Llano
Estacado, albeit uncommon. We have examined eight specimens as
follows (listed from north to south): an adult male captured in an
unused section of irrigation pipe in a grassy area near a playa 7.4 mi.
NW Hart, Castro County, on 19 November 1978 (TTU 34774); an
adult female taken 3.3 mi. N Fieldton, Lamb County, on 5 September
1968 (TTU 7764); a young adult female killed by a dog 3 mi. S and
3 mi. E Shallowater, Lubbock County, on 18 July 1983 (TTU 41316);
four adults, two of each sex, taken between 1962 and 1970 within what

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

356

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 1. Localities from which Mustela frenata (half circles) and Spilogale putorius
(solid circles) have been taken on the Llano Estacado of Texas or in adjacent areas.
The diamond represents a general place in Lubbock County where both species have
been recorded. The two small solid symbols to the north represent previously
reported specimens of S. putorius (Van Gelder 1959). The two small open circles to
the west in New Mexico represent previously reported records of M. frenata (Aday
and Gennaro 1973).

SPILOGALE AND MUSTELA ON LLANO ESTACADO

357

now are the city limits of Lubbock, Lubbock County (TTU 63, 2675,
7520, 17489); and a young male that was captured while carrying a
cotton rat ( Sigmodon hispidus) on the city golf course at Sundown,
Hockley County, in March 1983 (TTU 41314).

These specimens are assigned to the subspecies M. /. neomexicana
following Hall (1951). External and cranial dimensions of three adult
males (2675, 7520, 34774) and two adult females (7764, 17489) are,
respectively, as follows: total length, 470, 445, 482, 424, 430; length of
tail, 170, 164, 170, 165, 168; length of hind foot, 48, 44, 50, 40, 42;
length of ear, 26, 26, 28, 23, 22; condylobasal length, 53.1, 52.8, 56.4,
47.5, 48.4; zygomatic breadth, 31.6, 30.3, 32.2, 26.0, 26.8; mastoid
breadth, 26.0, 27.1, 27.4, 23.4, 24.2; interorbital constriction, 11.4, 11.9,

12.1, 9.5, 9.2; postorbital constriction, 9.6, 8.0, 9.4, 8.9, 7.7; length of
maxillary toothrow, 14.8, 15.2, 15.7, 12.9, 13.6.

Spilogale putorius interrupta (Rafinesque, 1820). — Although
apparently rare in West Texas, this spotted skunk was originally
reported from the region on the basis of an adult female (TTU 793)
live-trapped on 29 April 1963 along a highway 5 mi. W and H mi. N
Lubbock, Lubbock County (Packard and Garner 1964).

On 23 April 1970, an adult male (TTU 17492, testes 28 mm long)
was killed on Texas Highway 114 at a place 4.5 mi. W Carlyle, in
Hockley County, providing the second record from the Llano. We have
also examined an adult female (TTU 17491) taken 1 mi. S Post, Garza
County, just east of the Llano, on 22 February 1969. Additionally a
specimen represented by a skull alone (TTU 1848) is available from
even farther eastward, 10 mi. SE Haskell, Haskell County (not
mapped).

Our specimens clearly are assignable to the eastern species of spotted
skunk rather than to S. gracilis (specimens of the latter from Pecos
County, Texas, used in comparisons) because they have much less
extensive white markings dorsally (see Schmidly 1984:fig. 2), only a
small white spot on the head, lack the characteristic white-tipped tail
of gracilis, and have a somewhat narrower braincase and much less
flared mastoid region. External and cranial measurements of the male,
followed by those of the two females, are: total length, 502, 395, 425;
length of tail, 194, 145, 169; length of hind foot, 49, 46, 42; length of
ear, 28, 29, 28; condylobasal length, 57.5, 50.6, 49.6; zygomatic
breadth, 35.4, 32.0, 31.8; interorbital constriction, 16.2, 14.1, 14.2;
mastoid breadth, 31.5, 28.0, 27.8; length of maxillary toothrow, 18.1,

16.2, 15.7. The male weighed 634.5 grams, one female (Lubbock
County), 279.1. The frontal sinuses of the cranium of the male have
deep lesions resulting from infection by the roundworm ( Skrjabingylus
sp.) and one of the females has a small lesion.

358

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

LITERATURE CITED

Aday, B.J., Jr., and A. L. Gennaro. 1973. Mammals (excluding bats) of the New Mexican
Llano Estacado and its adjacent river valleys. Stud. Nat. Sci., Eastern New Mexico
Univ. 1(5): 1-33.

Davis, W. B. 1974. The mammals of Texas. Bull. Texas Parks and Wildlife Dept. 4:1-
294.

Hall, E. R. 1951. American weasels. Univ. Kansas Publ., Mus. Nat. Hist. 4:1-446.

- . 1981. The mammals of North America. John Wiley and Sons, New York, 2nd

ed. 2: vi+60 1-1181 +90.

Packard, R. L. , and H. W. Garner. 1964. Records of some mammals from the Texas
High Plains. Texas J. Sci. 16:387-390.

Schmidly, D. J. 1984. The furbearers of Texas. Bull. Texas Parks and Wildlife Dept.
1 1 l:vii+l-55.

Van Gelder, R. G. 1959. A taxonomic revision of the spotted skunks (genus Spilogale ).
Bull. Amer. Mus. Nat. Hist. 117:229-392.

FINE STRUCTURE OF THE SECONDARY WALLS
OF SCLEREIDS OF RAUWOLFIA SERPENTINA

by A. J. MIA

Department of Life Sciences
Bishop College
Dallas, Texas 75241

ABSTRACT

A study of the fine structure of Rauwolfia sclereids indicated that the secondary walls
comprise a complex, multi-layered structure. The massive sclereid is composed of outer,
middle, and inner regions analogous to the Si, S2, and S3 layers of xylem cells. The
outermost region consists of a single layer, is usually thicker than the layers of the
middle and inner regions, and contains cellulose microfibrils that lack obvious
concentric order. Each inner layer within the middle and inner regions consists of two
parallel rows of two parallel rows of cellulose microfibrils intercalated with amorphous
wall matrix. The cellulose microfibrils in the inner layers appear to have steep
orientation to the vertical axis of the sclereid. The regular concentric layering pattern
of microfibrils often is less obvious or inconspicuous near the cell lumen. Key words :
sclereids, sclereid walls, secondary walls, fine structure of secondary walls.

Sclereids have been known for more than a century (Mirbel and
Payne 1849) to occur in plant organs, but little is known about their
fine structure (Mia and Setterfield 1965, 1966a, 1966b; Parameswaran
1975; Boyd et al. 1982; Harris 1983). The use of bright-field microscopy
for studying these specialized cells has generated a considerable
amount of information, ranging from the general organization and
anatomical peculiarities of sclereids to their mode of differentiation
(Foster 1944, 1945, 1946, 1955, 1956; Bloch 1946; Sterling 1947, 1954;
Rao 1951, 1957; Arzee 1953a, 1953b; Foard 1959; Gaudet 1960; Mia and
Pathak 1965, 1968). Comparable knowledge is yet to be achieved on the
fine structure of sclereids, especially with regard to the thick,
lamellated secondary walls that distinguish them from their
neighboring thin-walled parenchymatous cells. The fine structure of
sclereids from several plants have been investigated using the
techniques of autoradiography with tritiated glucose (Setterfield and
Bayley 1957, 1959), thin sections and transmission electron microscopy
(Pease 1964; Mia 1968, 1969; Mia and Pathak 1968), and shadowing
with palladium-gold (Beer and Setterfield 1958; Mia 1968). The present
article describes the anatomy and fine structure of the secondary walls
of sclereids in Rauwolfia serpentina using palladium-gold shadow
casting of thin sections.

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

360

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

MATERIALS AND METHODS

Approximately three-year-old Rauwolfia serpentina (L.) Benth.
plants were grown in a thermostatically-controlled greenhouse. Nodal
segments 2-4 mm long either were fixed in 70 percent ethanol for light
microscopy or extracted of lignin, pectin, and other noncellulosic wall
substances before fixation for electron microscopy. All cortical tissues
including the phloem were removed from the nodal segments by
peeling off the cambial layer prior to further tissue preparation. Tissue
samples for light microscopy were dehydrated in graded ethanol and
xylene and embedded in Paraplast (Fisher Scientific). Serial microtome
sections (8-12 jum) were stained in safranin and mounted on
Kleermount (Carolina Biology). Observations of free-hand sections
from fresh nodal tissue and whole sclereids separated by maceration
(Mia 1962) also were made with bright-field and polarizing
microscopes. The sections and the macerated sclereids were stained in
dilute aqueous safranin or in phloroglucinol and HC1. The extraction
procedure for removing the noncellulosic wall substances from the
highly-thickened secondary walls of Rauwolfia sclereids was long and
tedious, but no alternate method was found suitable for preparing
ultrathin sections for electron microscopy. Unfixed nodal segments
were extracted for three hours at 98° C in an oven using four percent
aqueous NaOH and one-to-one solution of 30 percent H2O2 and
glacial acetic acid, respectively. The tissue segments were washed
thoroughly in glass distilled water, transferred to 10 percent pectinase
at pH 4.0, and left overnight. The tissues then were washed and
extracted alternately with 0.1N HC1 and 0.1N NaOH at 60° C six
times for 30 minutes each. Extracted and unextracted stem segments
were dehydrated in a graded series of ethanol and methacrylate,
embedded in a mixture of two-to-one butyl and methyl methacrylate,
and polymerized at 60° C for six hours.

Ultrathin (800A°) sections were cut with a Reichert ultramicrotome
and mounted on formvar-coated copper grids. Embedding medium was
removed from the sections by treatment with amyl alcohol and amyl
acetate, and sections were then shadowed with a palladium-gold alloy
at 15° angle in a vacuum evaporator. The grids were then examined
in the transmission electron microscope. One /im sections from
unextracted sclereids were mounted in a drop of water on a glass slide,
covered with a cover slip and were observed with bright-field and
polarizing objectives.

FINE STRUCTURE OF RAUWOLFIA

361

RESULTS

Optical Microscopy

Many free-hand and serial sections of both young and mature
sclereids, which occur at the nodes (Fig. 1, arrow), were made from
fresh and chemically-fixed stems. These sections were studied under
bright-field and polarizing optics. The anatomical features of
Rauwolfia sclereids (Figs. 2, 3) were similar to those reported in R.
vomitoria (Mia 1964) with regard to distribution, differentiation,
shape, size, and massive wall construction. Sclereids in R. serpentina
occurred most frequently as idioblasts (Fig. 4) and occasionally in
longitudinal files (Fig. 5) as in R. vomitoria. Sclereids in R. serpentina
occasionally occurred in pairs and in groups. Such a group or cluster
of sclereids may develop with a single central sclereid (Fig. 4, arrow).
Calcium oxalate crystals occurred in the pith cells often adjacent to
sclereids but not within them.

An interesting feature of the sclereids of R. serpentina was
subdivision of the protoplast to form compartments of walls within a
single sclereid (Fig. 3, arrows). The most striking feature visible under
the light microscope was the massive secondary wall containing many
dark and light concentric lamellae. These were clearly visible in
transverse and longitudinal sections stained with dilute aqueous
safranin or phloroglucinol and HC1 (Figs. 2, 3; Mia 1964; Mia and
Setterfield 1966b). The alternate light and dark bands, which contain
crystalline cellulose and amorphous matrix, respectively, appeared
more pronounced when viewed in the polarizing and transmission
electron microscopes (Figs. 5-8). Mature sclereid walls contained 30 or
more light and dark bands (Figs. 7, 8), although there was

considerable variation in the number of such concentric lamellae. This
variation depended primarily upon the size and maturity of the
sclereids.

The lamellated wall structure (unextracted) when observed with the
bright-field and electron microscopes showed no apparent crystalline
structure. This amorphous appearance of the wall was altered when
observed with the polarizing microscope. Under polarizing
illumination, secondary walls revealed a complex pattern of
birefringence indicating that within this amorphous image of the wall
there contained a much ordered substructure composed of cellulose
(Figs. 5, 6). In polarized light, a fully mature sclereid frequently
exhibited three distinct regions, outer, middle and inner (Figs. 5, 6),
which were optically anisotropic. These regions were demarcated by
layers containing noncellulosic, amorphous substances that appeared
optically isotropic under polarized light.

362

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figures 1-4. 1. Shoot of Rauwolfia serpentina showing the position of the node (arrow)
in which sclereids occur, Xone-fourth natural size. 2. Macerated sclereids showing
variable shape and size, X30. 3. A sclereid showing thick lamellated secondary wall
and two transverse partitions (arrows), X260. 4. Transverse section of a node showing
the distribution of single and grouped sclereids in the pith interior to protoxylem
(px) ring above, X75.

FINE STRUCTURE OF RAUWOLFIA

363

Figures 5-6. 5. Longitudinal section (1 /xm) of grouped sclereids cut from the
methacrylate-embedded block to show broad outer (o, arrow), middle (m), and inner
(i) regions, and the narrow alternating light and dark bands under polarized optics.
Note the presence of two sclereids on either side of the middle sclereid, X300. 6.
Transverse section (1 /im) of grouped sclereids cut from the methacrylate-embedded
block to show under polarized light broad outer (o, arrow), middle (m), and inner
(i) regions, and the narrow alternating light and dark bands in the secondary walls,
X300.

Electron Microscopy

As indicated earlier, sclereid walls contained concentric lamellae that
were pronounced with polarizing optics (Figs. 5, 6). The lamellar
sclereid wall, corresponding to the alternate light and dark bands in
Figures 5 and 6, also was seen in transverse sections in the electron
microscope. Figures 7 and 8 are low magnification electron
micrographs and show many obvious lamellae. Extracted walls (Figs.
9, 10) indicated that the crystalline component of the wall was
confined to the light bands and composed of many cellulose
microfibrils as revealed under polarized optics. The dark bands are
empty spaces once occupied by noncellulosic wall substances prior to
chemical extraction.

As indicated, the secondary wall may be divided into three regions —
outer, middle and inner. These divisions may be regarded as analogous
to the Si, S2, and S3 layers in secondary walls of vessels, tracheids and
fibers (Bailey and Kerr 1935; Bailey 1938; Wardrop 1964a). The
secondary wall of the sclereid is composed of uniform layers
containing light and dark bands except that the outer layer is generally

364

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figures 7-10. 7. Electron micrograph of a sclereid wall showing primary wall (p), broad
outer (o), middle (m), and inner (i) regions surrounding the cell lumen. The inner
region in some sclereids as in this figure tends to shrink and separate from the
middle region during the tissue preparation indicating weak bonds between regions.
This layer where separation occurred is optically isotropic, X800. 8. Electron
micrograph of a transverse section of a portion of paired sclereids showing broad
outer (o) and middle (m) regions in the lamellated thick secondary walls, X2470. 9.
Electron micrograph of a sclereid wall after extraction of the amorphous substances
showing microfibrils in the primary wall (arrow), and in the broad outer (o) and
middle (m) regions of the secondary wall, X13,500. 10. Electron micrograph showing
transverse section of a sclereid wall with microfibrils in a chemically-extracted wall.
Note the broad outer (o) region with microfibrils having none of the concentric
lamellae evident in the middle (m) region. Both regions contain cellulose microfibrils
in the light bands and have dark bands from which amorphous substances were
extracted, XI 2, 000.

FINE STRUCTURE OF RAUWOLFIA

365

Figures 11-12. 11. Electron micrograph of a tangential longitudinal section through the
middle region of a sclereid showing the multinet structure in the primary wall (p),
and the steep angular dispersion of the microfibrils in the secondary wall (s). Arrow
indicates the approximate long axis of the sclereid, X9750. 12. Electron micrograph
showing a pit canal in the thick secondary wall of a young sclereid and a remnant
of the protoplast (circular structure), X4500.

thickest (Figs. 7, 8). The thick outer wall of the chemically-extracted
sclereids contained numerous microfibrils with an anastomosing
network pattern (Figs. 9, 10). The middle region of the sclereid wall
consisted of many narrow concentric layers with each layer commonly
containing two parallel rows of cellulose microfibrils (Figs. 9, 10),
whereas the inner region exterior to the cell lumen showed less
conspicuous concentricity in layering (Fig. 7). External to the
outermost layer of the secondary wall is the primary wall of the
sclereid, which contains a multinet structure of microfibrils (Fig. 11).
Figure 11 is a tangential section of a sclereid showing steep
microfibrillar dispersion to its long axis.

The layering of the secondary wall of sclereids commenced at an
early stage of growth and proceeded progressively toward the cell
lumen. Young sclereid walls with only a few layers of cellulose
microfibrils alternating with noncellulosic wall layers frequently were
seen (Fig. 12). This mode of layering seemed to persist throughout the
growth of the wall until the lumen was considerably narrowed by the
build up of the wall in successive order toward the lumen (Fig. 7).

366

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

DISCUSSION

The central point that emerges from the study of Rauwolfia sclereids
is the regular concentric layering in the secondary wall. The layered
structure in the secondary walls of fibers, vessels and trachieds (three
major layers — Si, Si and S3) first was described by Bailey and Kerr
(1935). Since then, this wall construction has been recognized in
investigations of cell walls, of a wide range of gymnosperm and
angiosperm woody plants (Wardrop 1964a, 1964b; Wardrop and
Harada 1965). The layered structure described here for sclereids may be
divided into outer, middle, and inner regions, which seem analogous
to the three layers of secondary wall described for fibers, trachieds, and
vessel elements. In the sclereids, the outer wall is generally prominent
and is distinguished by a single broad layer. This outer layer (region)
contains cellulose microfibrils, which unlike the middle and inner
regions show no obvious concentric lamellae. The middle region of the
sclereid wall, which constitutes the major portion of the secondary
wall, is, on the other hand, composed of many alternatively arranged
light and dark areas. This middle region of the wall can be considered
as a layer analogous to that of S2 layer of fibers and other xylem
elements (Wardrop 1964a). The light and dark areas represent areas
with differences in chemical composition and microfibrillar
orientation as described in other secondary walls (Wardrop 1964a).
Each light area contains two juxtaposed rows of parallel microfibrils.
The dark structureless areas of the wall seen in the electron
micrographs presumably were occupied by lignin and other
noncellulosic wall components, but lost with chemical extraction,
leaving empty spaces. The inner region of the wall, which corresponds
to the S3 layer of fiber, occurs around the lumen of the sclereid. This
inner region of the wall lacks the distinct layering pattern seen in the
middle region.

A strikingly similar lamellar organization is known to occur in the
secondary wall of cotton fibers described by Balls (1919) and Anderson
and Kerr (1938). They described this wall as a firmly coherent structure
composed of a continuous phase of alternately light and dense areas
of cellulose. Electron micrographs of extracted secondary walls of
Rauwolfia sclereids reveal that the crystalline phase of the cellulose
microfibrils is deposited only in the light bands, which alternate with
dark bands composed of lignin, pectin and hemicellulose. Beer and
Setterfield (1958) described lamellated collenchyma wall of celery
petiole as possessing cellulose microfibrils in the light bands and non¬
cellulosic wall substances in the dark bands.

Balls (1919) first attempted to correlate the formation of concentric
lamellae in cotton hairs with the daily light and temperature

FINE STRUCTURE OF RAUWOLFIA

367

periodicity. Anderson and Kerr (1938) demonstrated experimentally
that these lamellae (growth rings) in cotton hairs were obligatorily
correlated with daily rhythms. They found that continuous light and
constant temperature inhibited lamellae formation. In sclereids, the
formation of lamellae is not associated with 24-hour light and dark
periods; rather a lamella forms every 48 to 72 hours as determined by
pulse-labelling autoradiographic studies, regardless whether the plants
receive continuous light or alternate light and dark daily rhythms
(Mia, unpublished data). Autoradiographic studies also reveal that the
enormously thick secondary wall of sclereids is built by the process of
apposition (Mia and Setterfield 1966b) as tritiated glucose molecules
are found to be incorporated into the inner layer of the newly-formed
wall surrounding the lumen, that is, on the outer surface of the
cytoplasm. The deposition of the wall substances in sclereids is
possibly brought about by golgi vesicles as the sclereid cytoplasm
contains an abundance of membrane-bound vesicles frequently
associated with golgi bodies (Mia and Setterfield 1966a, 1966b; Mia and
Pathak 1968). Boyd et al. (1982) and Harris (1983) also reported the
presence of numberous golgi bodies associated with vesicles in the
maturing sclereids of Camellia petiole and in macrosclereids of Pisum
seed coat, respectively. Similar vesicular incorporation has also been
observed by Wardrop (1965), Wooding and Northcote (1964), Esau et
al. (1966), and Pickett-Heaps and Northcote (1966) in the secondary
walls of tracheary elements of several plants.

LITERATURE CITED

Anderson, D. B., and T. Kerr. 1938. Growth and structure of cotton fiber. Indus. Engin.
Chem. 30:48-54.

Arzee, T. 1953a. Morphology and ontogeny of foliar sclereids in Olea europa. I.
Distribution and structure. Amer. J. Bot. 40:680-687.

- •. 1953b. Morphology and ontogeny of foliar sclereids in Olea europa. II. Otogeny.

Amer. J. Bot. 40:745-752.

Bailey, I. W. 1938. Cell wall structure of higher plants. Indus. Engin. Chem. 30:40-47.
Bailey, I. W., and T. Kerr. 1935. The visible structure of secondary wall and its
significance in physical and chemical investigations of tracheary cells and fibers. J.
Arnold Ar. 16:273-30.

Balls, W. L. 1919. The existence of daily growth rings in the cell wall of cotton. Proc.
Royal Soc. London 690:542-555.

Beer, M., and G. Setterfield. 1958. Fine structure of thickened primary walls of
collenchyma cells in celery petiole. Amer. J. Bot. 45:571-580.

Bloch, R. 1946. Differentiation and pattern in Monstera deliciosa. The idioblastic
development of the trichosclereids in the air root. Amer. J. Bot. 33:544-551.

Boyd, D. W., W. M. Harris, and L. E. Murry. 1982. Sclereid development in Camellia
petiole. Amer. J. Bot. 69:339-347.

Esau, K., V. I. Cheadle, and R. A. Gill. 1966. Cytology of differentiating tracheary
elements. I. Organelles and membrane systems. Amer. J. Bot. 53:756-764.

368

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Foard, D. E. 1959. Pattern and control of sclereid formation in the leaf of Camellia
japonica. Nature 184:1663-1664.

Foster, A. S. 1944. Structure and development of sclereids in the petiole of Camellia
japonica L. Bull. Torrey Bot. Club 71:302-326.

- . 1945. Origin and development of sclereids in the foliage leaf of Trochodendron

araliodes Sieb. and Zucc. Amer. J. Bot. 32:456-468.

- . 1946. Comparative morphology of the foliar sclereids in the genus Mouriria

Aubl. J. Arnold Ar. 27:253-271.

- . 1955. Structure and ontogeny of terminal sclereids in Boronia serrulata. Amer.

J. Bot. 42:551-560.

- . 1956. Plant idioblasts: Remarkable examples of cell specialization. Protoplasma

46:184-193.

Gaudet, J. 1960. Ontogeny of foliar sclereids in Nymphea odorata . Amer. J. Bot. 47:525-
532.

Harris, W. M. 1983. On the development of macrosclereids in seed coats of Pisum
sativum L. Amer. J. Bot. 70:1528-1535.

Mia, A. J. 1962. Polymorphic parenchymatous cells of Rauwolfia vomitoria Afzl. Texas
J. Sci. 14:305-318.

- . 1964. Ontogeny and differentiation of sclereids in Rauwolfia. Amer. J. Bot. 51:78-

87.

- . 1968. Organization of tension wood fibers with special reference to gelatinous

layer in Populus tremuloides Michx. Wood Sci. 1:105-115.

- . 1969. Study of cell walls in angiospermous plants using light and electron

microscopes. Wood Sci. 2:1-10.

Mia, A. J., and S. M. Pathak. 1965. Histochemical studies of sclereid induction in the
shoot of Rauwolfia species. J. Expt. Bot. 16:177-181.

- . 1968. A histochemical study of the shoot apical meristem of Rauwolfia with

reference to differentiation of sclereids. Canadian J. Bot. 46:115-120.

Mia, A. J., and G. Setterfield. 1965. Structure and deposition of secondary wall in
sclereids of Rauwolfia. Proc. Canadian Soc. Plant Physioilogists 6:23.

- . 1966a. Structure and deposition of secondary wall of Rauwolfia sclereids. Amer.

J. Bot. 53:617.

- . 1966b. Organization of cellulose in plant cell wall. Chemistry in Canada 18:36,

37, 63.

Mirbel, C. F. , and A. Payne. 1849. Organographic et physiologie vegetale. Memoire sur
la composition et al structure de plusiers organismes des plantes. Mem. Acad. Sci.
Inst. France 20:497-531.

Parameswaren, N. 1975. On the fine structure of the wall of sclereids in some tree barks.
Protoplasma 85:305-314.

Pease, D. C. 1964. Histochemical techniques for electron microscopy. Academic Press.
New York.

Pickett-Heaps, J. D., and D. H. Northcote. 1966. Relationships of cellular organelles to
the formation and development of plant cell wall. J. Expt. Bot. 17:20-26.

Rao, T. A. 1951. Studies of foliar sclereids in dicotyledons. V. Structure and development
of the terminal sclereids in the leaf of Memecylon heyneanum Benth. Proc. Indiana
Acad. Sci. 34:329-334.

- . 1957. Comparative morphology and ontogeny of foliar sclereids in seed plants.

I. Memecylon L. Phytomorphology 7:306-330.

Setterfield, G., and S. T. Bay ley. 1957. Studies on the mechanism of deposition and
extension of primary cell walls. Canadian J. Bot. 35:435-444.

- . 1959. Deposition of cell walls in oat coleopdles. Canadian J. Bot. 37:861-870.

Sterling, C. 1947. Sclereid development and the texture of Bartlet pears. Food Res.
19:433-443.

FINE STRUCTURE OF RAUWOLFIA

369

Wardrop, A. B. 1964a. The structure and formation of the cell wall in xylem. Pp. 87-
134, in Formation of wood in forest trees (M. H. Zimmerman, ed.), Academic Press,
New York.

- . 1964b. The reaction of arborescent angiosperms. Pp. 405-456, in Formation of

wood in forest trees (M. H. Zimmerman, ed.), Academic Press, New York.

- . 1965. Cellular differentiation in xylem. Pp. 61-97, in Cellular ultrastructure of

woody plants (W. A. Cote, Jr., ed), Syracuse Univ. Press, Syracuse, New York.

Wardrop, A. B., and H. Harada. 1965. The formation and structure of the cell wall in
fibers and tracheids. J. Expt. Prod. 16:356-371.

Wooding, F. B. P., and D. H. Northcote. 1964. The development of secondary wall of
the xylem in Acer pseudotplatanus. J. Cell Biol. 23:327-337.

STUDIES ON THE USE OF BOILED CHICKEN EGG YOLK
AS A FEED FOR REARING PENAEID SHRIMP LARVAE

by DAVID M. FUZE, JOSHUA S. WILKENFELD,
and ADDISON L. LAWRENCE

Texas Agricultural Experiment Station
Texas A&M University System
P.O. Drawer Q
Port Aransas, Texas 78373

ABSTRACT

The effectiveness of hard-boiled chicken-egg yolk as a food source for larval penaeid
shrimp was examined. A small-scale experimental system was employed consisting of
one-liter Imhoff cones and a synthetic seawater media. Larvae ( Penaeus setiferus
Linnaeus, P. aztecus Ives, or P. vannamei Boone) were stocked at a density of 100 per
liter and reared from the protozoea-one (Pi) substage to one-day-old postlarvae (PLi).
The animals were fed various combinations of algal ( Chaetoceros gracilis and
Skeletonema costatum, or Tetraselmis chuii) and animal ( Artemia , rotifers, and/or
boiled chicken-egg yolk) foods, added at the Pi, P2, P3, or mysis-one (Mi) substages.

Survival of larvae fed algae and egg yolk was equal to, or slightly worse than, survival
of larvae fed algae only or algae plus Artemia. The more advanced the substage of
shrimp larvae when egg yolk was initially added to the algae diet, the faster was the
rate of metamorphosis. In most cases, animals fed egg yolk and algae grew to the same
mean dry-weights of PLi as those fed algae alone. For all treatments tested, larvae fed
algae and Artemia, starting at the P2 substage, grew to the largest mean dry-weights at
the PLi substage. Larvae did not survive or metamorphose when fed egg yolk alone. Egg
yolk did not improve survival, metamorphosis, or growth of animals fed T. chuii and
rotifers.

Overall, there was little evidence that egg yolk was nutritionally beneficial to the
larval of the three species of shrimp using any of the different larval diets and
experimental conditions evaluated in this study. Key words : egg yolk, shrimp larvae,
feed.

The larvae of penaeid shrimp have been reared successfully in
hatcheries all over the world (Hudinaga and Kittaka 1966; Cook and
Murphy 1969, 1971; Mock 1971; Heinen 1976; Platon 1978). The usual
approach is to provide combinations of algal and animal foods (the
latter normally in the form of freshly hatched nauplii of the brine
shrimp, Artemia, or rotifers, Brachionus), which require additional
rearing facilities and expense (Sorgeloos and Persoone 1975; Heinen
1976; Sorgeloos 1980; Kuban et al. 1983; Quinitio and Reyes 1983).

Some success has been achieved rearing Penaeus sp. larvae on diets
of nonliving or artificial ingredients (Hirata et al. 1975; Loon 1976;
Jones et al. 1979a, 1979b; Mock et al. 1980). However, because of price,

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

372

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

availability, and water quality problems associated with most prepared
diets, only live phytoplankton is fed to the shrimp protozoea larvae at
the majority of hatcheries (Heinen 1976). To reduce space, cost, and
labor requirements in the hatchery, research efforts have been made to
develop acceptable inert diets for shrimp larvae.

One food source of particular interest to aquaculturists is the hard-
boiled yolk of chicken eggs. Chicken eggs are a cosmopolitan,
relatively inexpensive food, which can be stored for about 10 days at
18°C, and as long as two months with refrigeration (Reader’s Digest,
1976). These basic attributes make it attractive for consideration as an
animal food source for crustacean larvae, which often are reared in
relatively inaccessible locations. Ling (1962) found steamed hen’s egg
to be a suitable food material for larval Macrobrachium rosenbergii as
it could be mixed with other ingredients, such as yeast, before cooking.
Chow (1978) determined that the fat content in egg yolk (62.2 percent
dry weight) gave physical stability to the egg yolk particles allowing
them to remain suspended in water for long periods. Quinitio et al.
(1983) observed that diets of Tetraselmis sp. and boiled chicken egg
yolk gave the best growth of P. monodon larvae (protozoea to mysis)
when compared with diets of Tetraselmis sp. in combination with soy
cake, whole boiled egg, or brown mussel meat. Quinitio and Reyes
(1983) showed a 45 percent mean survival of P. monodon protozoea to
postlarvae using a combination of Tetraselmis sp. , rotifers ( Brachionus
sp.), and boiled chicken egg yolk.

At this time, there are no published reports evaluating the use of
chicken egg yolk as a substitute for Artemia nauplii, which are used
worldwide as an animal food source for penaeid shrimp larvae (Cook
and Murphy 1969; Heinen 1976; Sorgeloos 1980; Kuban et al. 1983).

Chicken egg yolk was examined for its effectiveness as a food for
larval shrimp by introducing it to the algae diet at various larval
substages. Comparisons were made of survival, metamorphosis, and
growth between penaeid larvae fed egg yolk and algae, algae only, egg
yolk only, or a “standard” diet of diatoms with Artemia added starting
at the P2 substage (Kuban et al. 1983, 1985; Wilkenfeld et al. 1984).
Additional experiments were performed to test various algae and egg
yolk parameters (type, density, size, and so forth), and the addition of
rotifers.

MATERIALS AND METHODS

Sources of Animals

Experiments performed in this study involve four trials with the
larvae of penaeid shrimp. In Experiment 1 , Penaeus setiferus L. larvae
were obtained from a mixture of three spawns by adult shrimp from

EGG YOLK AS SHRIMP LARVAL FEED

373

off the coast of Corpus Christi, Texas, and spawned at the Texas A&M
University Shrimp Mariculture Facility, Corpus Christi. In
Experiments 2, and 3, P. aztecus Ives larvae were obtained from mixed
spawns of three and two adult female shrimp, respectively, from off
the coast of Freeport, Texas, and spawned at the Texas A&M
University Larviculture Facility, Galveston, Texas. P. vannamei Boone
larvae, used in Experiment 4, were received from mixed spawns of two
adult females induced to mature and reproduce in captivity at the
Texas A&M University Maturation/Reproduction Facility, Corpus
Christi. All experiments were performed at the Galveston Laboratory,
National Marine Fisheries Service.

Larval Food Sources

Phytoplankton was mass cultured in 300-liter tanks using filtered
seawater and Guillards f/2 media (Guillard 1975) with vigorous
aeration and 24 hours of fluorescent illumination. Algae were used
during the exponential growth phase, or less than five days old in
mass culture, in an attempt to utilize algal cells having a more
constant nutritional value. Cell densities of algal cultures were
determined using replicated hemacytometer counts. During
experiments, appropriate volumes of algae from the mass cultures were
added to the larval growing media in order to achieve desired feeding
levels.

Diatoms, as a feeding regime, consisted of both Skeletonema
costatum and Chaetoceros gracilis , at 70,000 cells per milliliter each.
Tetraselmis chuii also was used as an algal food source for more direct
comparisons to work performed by Quinitio et al. (1983) and Quinitio
and Reyes (1983). Feeding densities ranged from 5000 to 50,000 cells
per milliliter.

Dried cysts of Bio-Marine® (Bio-Marine®/Aquafauna, P.O. Box 5,
Hawthorne, California 90250) brand Artemia were hydrated for 24
hours in seawater with vigorous aeration and illumination. Newly-
hatched nauplii were then separated from the hatched and unhatched
cysts, and fed at a level of three per milliliter (or 3000 per one-liter
cone). These were added to the algal diet of the shrimp starting at the
protozoea-2 (P2) substage.

Brachionus plicatilis was cultured in 12-liter inverted carboys fed
Tetraselmis chuii and Isochrysis sp. Filtered seawater was used as the
medium (28-30 parts per thousand salinity, 22-24°C) with mild
aeration and 24 hours illumination. Densities of 70-240 rotifers per
milliliter were maintained with a daily harvest and algae replacement
of 20 percent or less of the volume of the carboy. Populations were
estimated using aliquot samples and calculated volumes were removed
and screened to concentrate the animals for feeding to shrimp larvae.

374

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Feeding levels of 10 animals per milliliter were established after
Quinitio and Reyes (1983). Rotifers were added starting when larvae
reached the P2 substage. Animals fed rotifers only were given to the
shrimp larvae at 10 per milliliter starting at the Pi substage.

A wet weight of egg yolk equivalent in protein content to a
previously established standard feeding level of three Artemia nauplii
per milliliter as prescribed by Wilkenfeld et al. (1983) was fed to the
shrimp larvae. The amount of egg yolk fed to the larvae was
determined by assuming the dry weight and protein content of a 0.4
mm Artemia nauplii to be 0.02 milligrams and 42.5 percent of dry
weight, respectively (Sorgeloos and Persoone 1975), and the protein
level and the moisture content for hard-boiled egg yolk to be 32.8
percent of dry weight (Chow 1978) and 46.9 percent, respectively. The
wet- weight feeding level of egg yolk, which would represent a similar
protein level in the larval diet as three Artemia nauplii per milliliter,
was estimated to be 16.5 mg per liter. Two other feeding levels (8.25
and 4.0 mg per liter) were chosen for testing as they represent densities
of 50 percent and 25 percent, respectively, of the original feeding level.

Though several egg yolk particle sizes were tested, 102 /u. m was most
frequently used for several reasons. Artemia nauplii range in size from
200-400 jum (Sorgeloos and Persoone 1975; Sorgeloos 1980) to 500 jum
(Jones et al., 1979a). Quinitio et al (1983) and Quinitio and Reyes
(1983) used a size of 40 fxm or smaller to feed P. monodon larvae egg
yolk and other inert foods Jones et al. (1979a) found that the setose
appendages of P. japonicus larvae became entangled with fine 15-25
jum microencapsulated particles fed at high concentrations (0.88 per
larva per day). They found that prepared feeds strained through a 100-
H m mesh were broken into smaller particles on contact with seawater.
A screen of 102 jum was selected for use in this study as it gave an
acceptable range of particle sizes (less than 102 /xm) and fell into an
average range between Artemia nauplii and previously tested particle
sizes.

Kroger (Kroger Foods Company, Cincinnati, Ohio 45201) brand
eggs, U.S.D.A. Grade A large (white shelled) or Farm Fresh (Farm
Fresh or Strictly Fresh eggs, Texas license #4425, 2615 Airline Drive,
Houston, Texas 77009) brand eggs, U.S.D.A. Grade A large (brown
shelled, fertile) were submerged in boiling water and boiled for 14-15
minutes to produce a solid yolk. The shell and albumen were then
removed and weighed portions (1.65, 0.83, and 0.40 gram) of yolk,
respective to calculated feeding levels, were forced through screens of
various preselected sizes, and suspended in 100 milliliters of growing
medium. One milliliter of the suspension was then added to each one-
liter cone at the appropriate larval substage. This resulted in feeding
levels of 16.5, 8.25, and 4.0 mg per liter wet weight.

EGG YOLK AS SHRIMP LARVAL FEED

375

Table 1. Survival, rate of metamorphosis, and growth to the postlarvae-one (PLi)
substage of Penaeus setiferus larvae fed diatoms and egg yolk beginning at the
protozoea-one (Pi), P2, P3 and mysis-one (Mi) substages, respectively, or diatoms
alone or with Artemia (Experiment 1).

Treatment3

Mean survival
(% ± S.D.)

Mean meta. to PLi
(% + S.D.)

Mean weight
(jum ± S.D.)

Diatoms/egg yolk Pi

84.0+7.6 (BC)b

45. 8+21. 2(D)

63.4+1. 4(B)

Diatoms/egg yolk P2

77.8±5.4(C)

72.6+17. 3(C)

66.2+3. 1(B)

Diatoms/egg yolk P3

83.2+6. 1(BC)

76.4±8.3(BC)

66.2+3. 9(B)

Diatoms/egg yolk Mi

84.8±9.6(ABC)

79.2±6.5(BC)

63. 9+3. 0(B)

Egg yolk alone Pi

0%

—

—

Diatoms alone

91.8+4. 1(AB)

90.4±4.0(AB)

59.4±6.8(B)

Diatoms /Artemia P2

92.2+4. 2(AB)

95. 8+3. 3(A)

108.9±0.6(A)

aTreatments include: diatoms composed of Chaetoceros gracilis and Skeletonema
costatum at 70,000 cells/ml each; egg yolk from hard-boiled chicken eggs (commercial
Grade A, white shell), screened to a 102 /rm particle size and fed at 16.5 mg/1 wet
weight; Artemia nauplii fed at three animals/ml starting at the P2 substage.
bLetters in parenthesis represent rankings acording to Duncan’s multiple range test.
Numbers followed by the same letter(s) are not significantly different (P<0.05).

Experimental System

In all experiments, shrimp nauplii were acclimated to the synthetic
sea salt growing medium, Marinemix® plus Bioelements (Marinemix
plus Bioelements by Wiegandt GMBH and Company, KG, D4150,
Krefeld, W. Germany; distributed locally). The saltwater was prepared
using deionized water at least 24 hours prior to use. Environmental
parameters were kept at a salinity, temperature, and pH of 30-32 parts
per thousand, 28°C, and 7. 8-8. 2, respectively. EDTA (ethylenedinitrote-
traacetic acid disodium salt, dehydrate) was added at a rate of 10 mg
per liter. Nauplii were then counted and stocked at a density of 200
larvae per liter in 12-liter inverted glass carboys, each containing a
different diet. When larvae reached the Pi substage, actively swimming
animals were selected and stocked in plastic Imhoff settling cones
(Wilkenfeld et al. 1983) at a density of 100 animals per liter. Five
replicates (cones) of each treatment were started, and respective cones
contained the same diet as the carboys from which the animals were
removed. The preliminary carboy phase was established so that larvae
were in contact with the experimental diet as soon as they
metamorphosed to the Pi substage. The animals were stocked into
cones at the Pi substage to avoid nonexperiment-related mortality
normally associated with the transition from the nauplius to the
protozoea stage.

Individual diets (listed in Tables 1-4) were continued, with daily
water exchanges of 100 percent, until one treatment contained 90-100
percent one-day-old postlarva (PLi). Animals were visually inspected

376

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 2. Survival, rate of metamorphosis, and growth to the postlarvae-one (PLi)
substage of Penaeus aztecus larvae fed diatoms and egg yolk beginning at the
protozoea-one (Pi), P2, P3 and mysis-one (Mi) substages, respectively, or diatoms
alone or with Artemia, or diatoms and egg yolk from brown eggs (Experiment 2).

Treatment3

Mean survival
(% ± S.D.)

Mean meta. to PLi
(% ± S.D.)

Mean weight
(ium ± S.D.)

Diatoms/egg yolk Pi

87.8±7.1(A)b

48.8+17. 2(BC)

71. 8+1. 7(C)

Diatoms/egg yolk P2

93.6±63.6(A)

44.8+35. 1(BC)

74.3±2.9(C)

Diatoms/egg yolk P3

91.0±3.8(A)

74.8+1 1.7(AB)

78.3+1. 5(B)

Diatoms/egg yolk Mi

92. 2+5. 5(A)

89. 2+3. 8(A)

79.2+0. 3(B)

Egg yolk alone Pi

0% (B)

- (D)

— (D)

Diatoms alone

86.0+13. 6(A)

90.8+3. 8(A)

71.7±2.8(C)

DiatomsA4 rtemia P2

92.6+4. 1(A)

95. 2+3. 0(A)

109.9±0.8(A)

Diatoms/egg yolk
(brown shell) P3

92.8±4.0(A)

50.6±30.4(BC)

74.3+1. 7(C)

treatments include: diatoms composed of Chaetoceros gracilis and Skeletonema
costatum at 70,000 cells/ml each; egg yolk from hard-boiled chicken eggs (commercial
Grade A, white shell, except where otherwise indicated), screened to a 102 jum particle
size and fed at 16 mg/1 wet weight; Artemia nauplii fed at three animals/ml starting
at the P2 substage.

tetters in parenthesis represent rankings acording to Duncan’s multiple range test.
Numbers followed by the same letter(s) are not significantly different (P<0.05).

daily to note gross differences in development and mortality. Larval
and postlarval substages were identified as described by Cook and
Murphy (1971).

Data Collection and Analysis

The experiment was terminated when a minimum of 90 percent of
the live shrimp in any one treatment had metamorphosed to PLi. The
percent survival was determined from the total number of live shrimp
at the end of the experiment. The rate of metamorphosis was estimated
by dividing the number of postlarvae by the total number of live
shrimp times 100 at the termination of the experiment. Dry weights of
animals from the different treatments were measured at the conclusion
of experiments to estimate growth. This was done by dividing the
surviving larvae from each treatment (five cones) into three
approximately equal subsamples. The larvae were then rinsed with
deionized water and dehydrated for 48 hours in a drying oven at 60°C
before weighing.

Analysis of variance (ANOVA) and Duncan’s Multiple Range Test
(DNMR) were used to evaluate the significance of differences between
treatment means.

EGG YOLK AS SHRIMP LARVAL FEED

377

Table 3. Survival, rate of metamorphosis, and growth to the postlarvae-one (PLi)
substage of Penaeus aztecus larvae fed algae and egg yolk, at two particle sizes and
two feeding levels, or diatoms alone or with Artemia (Experiment 3).

Treatment3

Mean survival
(% ± S.D.)

Mean meta. to PLi
(% + S.D.)

Mean weight
(Aim + S.D.)

Diatoms/egg yolk

102 /im particle size

16.5 mg/1 wet weight

91.6+8.4(A)b

2.8+3. 9(C)

65. 2+7. 6(A)

Diatoms/egg yolk

102/zm particle size

8.25 mg/1 wet weight

90.0+6.7(A)

1.0±1.2(C)

62.9+4. 7(A)

Tetraselmis/ eg g yolk

102 /zm particle size

16.5 mg/1 wet weight

15.2±16.6(B)

0%(C)

33. 6+17. 7(B)

Tetraselmis /egg yolk

69 Aim particle size

16.5 mg/1 wet weight

1.0±0.7(C)

0%(C)

20.0±0.0(B)

Diatoms alone

98.8+3. 0(A)

37. 4+17. 7(B)

69.0+3. 5(A)

Diatoms/^ rtemia P2

87.5±7.9(A)

88.4±3.8(A)

80.6+7. 1(A)C

“Treatments include: diatoms composed of Chaetoceros gracilis and Skeletonema
costatum at 70,000 cells/ml each; egg yolk from hard-boiled chicken eggs (commercial
Grade A, white shell), fed at two different particle sizes and two different feeding levels
(mg/1), starting at the protozoea-2 (P2) substage; Tetraselmis chuii at 5000 cells/ml;
Artemia nauplii fed at three animals/ml starting at the P2 substage.

bLetters in parenthesis represent rankings according to Duncan’s multiple range test.
Numbers followed by the same letter(s) are not significantly different (P<0.05).
cUnusually low weight of larvae in diatom/Art P2 treatment was probably the result of
poor quality Artemia nauplii hatched from cysts that had been inadequately stored in
an open can for an extended period.

RESULTS

Adding Egg Yolk to the Algae Diet at Various Substages

As summarized in Table 1, survival was the same for P. setiferus
larvae fed diatoms and egg yolk at all substages. However, only those
animals fed egg yolk starting at the Pi, P3, or Mi substage survived
as well as those larvae fed diatoms only or diatoms and Artemia.
Survival of P. aztecus larvae was not significantly different when fed
diatoms with or without Artemia or egg yolk (Table 2).

Metamorphosis to PLi for P. setiferus (Table 1) fed diatoms and egg
yolk, starting at P3 or Mi, was the same as that of animals fed diatoms
only. Animals given egg yolk at stages earlier than P3 (Pi or P2) had
progressively lower metamorphic rates the earlier egg yolk was added.
Larvae fed diatoms only or diatoms plus Artemia had similar rates of
metamorphosis. For P. aztecus (Table 2), larvae fed diatoms and egg
yolk at the P3 or Mi substage metamorphosed at the same rate as those
fed diatoms only or diatoms and Artemia. Those started on egg yolk

378

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Table 4. Survival, rate of metamorphosis, and growth to the postlarvae-one (PLi)
substage of Penaeus vannamei larvae fed diatoms and egg yolk at three particle sizes
and two feeding levels, Tetraselmis chuii at two densities with or without egg yolk
and/or rotifers, rotifers alone, or diatoms alone or with Artemia (Experiment 4).

Treatment3

Mean survival
(%± S.D.)

Mean meta. to PLi
(% ± S.D.)

Mean weight
(/xm ± S.D.)

Diatoms/egg yolk

102 /xm, 4.0 mg/1

82.8±8.9(A)b

6.2±2.4(CD)

65.0±2.3(B)

Diatoms/egg yolk

102 /xm, 16.5 mg/1

89.6±4.2(A)

6.8+2. 8(C)

63.4+1. 1(B)

Diatoms/egg yolk

69 /xm, 16.5 mg/1

89.6+1. 8(A)

2.6+2. 7(CD)

61. 9+5. 8(B)

Diatoms/egg yolk

355 /xm, 16.5 mg/1

92.2±3.8(A)

2.2+1. 6(D)

60.9±6.9(B)

Tetraselmis

50,000 cells/ml

61. 4+15. 6(B)

0%(E)

37. 2+8. 2(C)

Tetraselmis /egg yolk
50,000 cells/ml

102 /xm, 16.5 mg/1

48. 3+13. 8(B)

0%(E)

28.3+4. 9(CD)

Tetraselmis/ egg yolk
10,000 cells/ml

102 /xm, 16.5 mg/1

17. 0+17. 5(C)

0%(E)

24.5±19.4(CD)

Tetraselmis/e gg yolk
10,000 cells/ml

102 /xm, 16.5 mg/1
plus rotifers P2

17.0±16.2(C)

0%(E)

34.0±4.7(CD)

Tetraselmis

10,000 cells/ml
plus rotifers P2

18.4+15. 0(C)

0%(E)

21. 1+7. 8(D)

Rotifers alone

0.3+0.5(D)

0%(E)

0%(E)

Diatoms alone

88.8+1. 9(A)

45. 6+5. 4(B)

61. 1+1. 0(B)

Diatoms /Artemia P2

87.8±9.7(A)

84.6+2. 1(A)

107.2+3. 1(A)

treatments include: diatoms composed of Chaetoceros gracilis and Skeletonema
costatum at 70,000 cells/ml each; egg yolk from hard-boiled chicken eggs (commercial
Grade A, white shell), screened to three particle sizes ( /xm ) and fed at two density levels
(mg/1 wet weight), starting at the protozoea-one (Pi) substage; Rotifers ( Brachionus
plicatilis) fed at ten rotifers/ml and added starting at the P2 substage or in the case of
‘Rotifers alone’, starting at the Pi substage; Artemia brine shrimp nauplii fed at three
animals/ml, starting at the P2 substage.

bLetters in parenthesis represent rankings according to Duncan’s multiple range test.
Numbers followed by the same letter(s) are not significantly different (P<0.05).

at the Pi or P2 substage metamorphosed at significantly slower rates
than all others except those given egg yolk at P3.

In terms of dry weights of shrimp at termination of both species
supplied diatoms and egg yolk , only P. aztecus larvae fed diatoms and
egg yolk (P3 or Mi) weighed significantly more than those given
diatoms only (Tables 1 and 2). Larvae in those treatments in which
diatoms were fed with Artemia nauplii starting at P2 (both species)
grew to significantly greater mean dry weights than all other groups.

EGG YOLK AS SHRIMP LARVAL FEED

379

Screening Egg Yolk to Various Particle Sizes

For P. aztecus larvae fed Tetraselmis chuii at 5000 cells per milliliter
and egg yolk starting at P2 (Table 3), animals fed egg yolk screened
to a 102 jum size survived better than those fed egg yolk screened to
a 69 /i m size. Metamorphosis and growth were not affected by particle
size.

For P. vannamei (Table 4), survival and mean dry weights at
termination were the same for all treatments fed diatoms and egg yolk
starting at Pi. Rate of metamorphosis was the same for those animals
fed diatoms and egg yolk, screened to a 102 /zm or 69 jum particle size,
but was slower for those fed diatoms and egg yolk, screened to 355 jum.

Adding Egg Yolk to the Algae Diet at Various Feeding Levels

For P. aztecus larvae fed diatoms and egg yolk, starting at the P2
substage at two different levels (Table 3), survival, growth, and
metamorphosis were the same for those animals fed egg yolk at 8.25
or 16.5 mg per liter. For P. vannamei larvae fed diatoms and egg yolk
starting at Pi (Table 4), animals fed egg yolk at the level of 4.0 mg
per liter survived, metamorphosed, and weighed the same as those
animals fed egg yolk at 16.5 mg per liter.

Adding Rotifers to the Algae and Egg Yolk Diet

Performance of P. vannamei larvae fed a combination of Tetraselmis
chuii at 10,000 cells per milliliter plus egg yolk starting at Pi plus
rotifers at 10 per milliliter starting at P2, was the same as that of larvae
fed T. chuii at 10,000 cells per milliliter plus egg yolk or rotifers
(Table 4). Survival of P. vannamei larvae fed rotifers only at 10 per
milliliter starting at the Pi substage was lower than any other
treatment group.

Adding Yolk from Brown Versus White Eggs to the Algae Diet, Adding
Egg Yolk to Two Levels of Algae, and Feeding Egg Yolk Only

For P. aztecus larvae fed diatoms and egg yolk from hard-boiled
brown eggs starting at the P3 substage (Table 2), survival and
metamorphosis to the PLi substage were the same as for larvae fed
cooked yolk from commercial grade, white eggs starting at the same
substage. Mean weights of shrimp at termination were greater for those
animals fed diatoms and commercial eggs versus those fed diatoms and
brown eggs.

As seen in Table 4, P. vannamei larvae fed 50,000 cells per milliliter
of T. chuii with or without egg yolk survived, metamorphosed, and
grew to the same mean weights. Those larvae given algae at 10,000
cells per milliliter plus egg yolk survived at a significantly lower rate
than those given the higher algae level but the mean weights were the
same for all three treatments.

380

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

For P. setiferus and P. aztecus larvae, animals could not survive or
metamorphose past the Pi substage when fed diets of egg yolk without
algae (Tables 1 and 2).

DISCUSSION

Survival of those penaeid species tested fed diets of diatoms and egg
yolk was the same or reduced as compared to those fed the standard
diet of diatoms and Artemia nauplii starting at the P2 substage.
However, larvae fed diatoms only also survived at the same mean rate
as animals fed diatoms and Artemia. If one compares only survival as
a means of evaluating the effectiveness of various diets, it would seem
unnecessary to add any animal food. This was also reported by Kuban
et al. (1984) and Wilkenfeld et al. (1984). All animals given egg yolk
along with diatoms at substages earlier than P3 metamorphosed at a
slower rate than those fed diatoms only, indicting an adverse effect of
egg yolk particles on the early larval substages.

For all experiments, the mean termination dry weights of larvae fed
diatoms plus egg yolk were less than those fed diatoms plus Artemia.
Of those species tested fed diatoms and egg yolk, only P. aztecus larvae
given diatoms plus egg yolk starting at the P3 or Mi substage had
significantly greater mean weights than those fed diatoms alone. These
results indicate that little or no benefit is derived from the egg yolk
supplement.

Observations of larvae in the early protozoeal substages (Pi, P2)
revealed problems in swimming caused by the fine, oily egg yolk
particles clinging to the setae of the animals’ appendages. Larvae fed
high levels of a nutritious algae regime (such as C. gracilis and S.
costatum) appeared to be strong enough to free themselves from the
sticky particles in the early protozeal substages. Further evidence of
this was obtained in a preliminary experiment with P. aztecus in
which larvae were fed a diet of 100,000 cells per milliliter Isochrysis
sp. and 40,000 cells per milliliter T. chuii up to the P2 substage. At
that point, the food regime was changed to one consisting of T. chuii
at 5000 cells per milliliter with egg yolk (102 /um and 6.0 mg per liter).
Survival of larvae as high as 83 percent was achieved — higher than that
of P. aztecus and P. vannamei treatments started on lower levels of T.
chuii with or without egg yolk (Tables 3 and 4). Animals in the two
latter-mentioned trials appeared sluggish during the early protozoeal
substages and could not free themselves from the burdensome yolk
particles.

Quinitio and Reyes (1983) obtained survival of P. monodon larvae
as high as 65.14 percent to M3 and PLi on a mixed diet of Tetraselmis
sp. (5000 cells per milliliter, egg yolk starting at the late Pi substage,
and rotifers ( Brachionus sp.) starting at the P2 substage. In a similar

EGG YOLK AS SHRIMP LARVAL FEED

381

treatment with the larvae of P. vannamei, a survival high of only 43
percent to M3 was obtained when larvae were fed 10,000 cells per
milliliter T. chuii plus egg yolk and rotifers, started at the Pi and P2
substages, respectively. Mean survival, growth, and metamorphosis of
P. vannamei larvae fed the algae/egg yolk/rotifer treatment were not
significantly different from that of P. vannamei fed T. chuii plus egg
yolk or T. chuii plus rotifers. Results for all groups were poor and
may not reflect the true value of rotifers as a food source for shrimp
larvae.

In conclusion, data reported in this paper strongly suggests that egg
yolk is not as effective as Artemia when added directly to the larval diet
of the three penaeid species tested. Results of trials performed in this
study, along with those reported for P. monodon (Quinitio and Reyes,
1983; P. Gabasa, 1983, personal communication, SEAFDEC, P.O. Box
256, Iloilo City, Philippines), indicated that various penaeid species
may have different nutritional requirements or preferences. Previous
results reported by Kuban et al. (1984) and Wilkenfeld et al. (1984)
support this. These results also suggest that growth, in terms of mean
dry weight, is a more sensitive indicator than survival or
metamorphosis or both, as a means of evaluating the effectiveness of
individual larval diets.

ACKNOWLEDGMENTS

This work is a result of a research program sponsored in part by a
United States-Israel Binational Agricultural Research and Development
Fund Grant, a grant from the Caesar Kleberg Foundation for Wildlife
Conservation to Texas A8cM University System, and by Texas A8cM
University’s Sea Grant College Program supported by the National
Oceanic and Atmospheric Administration, Office of Sea Grant, U.S.
Department of Commerce, under Grant no. 4-7-1 58-44 1054. The
authors wish to express their appreciation to Frank Kuban, Linos
Cotsapas, George Marsden, and Walter Bensousan for their generous
assistance in data collection and algae culture, and Ginny Mitchell for
help in preparing the final manuscript.

LITERATURE CITED

Cook, H. L., and M. A. Murphy. 1969. The culture of larval penaeid shrimp. Trans.
Amer. Fish. Soc. 98:751-754.

- — — . 1971. Early developmental stages of the brown shrimp Penaeus aztecus Ives,
reared in the labroatory. Fish. Bull. 69:223-239.

Chow, K. W. 1978. Microencapsulated egg diets for fish larvae. FAO/UNDP training
course in fish feed technology. Univ. Washington, Seattle, pp. 355-361.

Guillard, R. R. L. 1975. Culture of phytoplankton for feeding marine invertebrates. Pp.
26-60, in Culture of marine invertebrate animals, (W. L. Smith and M. H. Chanley,
eds.), Plenum Publishing Corporation, New York.

382

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Heinen, J. M. 1976. An introduction to methods of rearing penaeid shrimp larvae. Proc.
World Maricult. Soc. 7:333-344.

Hirata, H., Y. More, and M. Watanabe. 1975. Rearing of prawn larvae, P. japonicus,
fed soy cake particles and diatoms. Mar. Biol. 29:9-13.

Hudinaga, M., and J. Kittaka. 1966. Studies on food and growth of larval stages of a
prawn, Penaeus japonicus, with reference to the application of practical mass
culture. Inf. Bull. Plankt. Japan 13:83-94.

Jones, D. A., A. Kanazawa, and S. Abdel Rahman. 1979a. Studies on the presentation
of artificial diets for rearing the larvae of Penaeus japonicus Bate. Aquaculture 17:33-
43.

Jones, D. A., A. Kanazawa, and K. Ono. 1979b. Studies on the nutritional requirements
of the larval stages of Penaeus japonicus using microencapsulated diets. Mar. Biol.
54:261-267.

Kuban, F. D., J. S. Wilkenfeld, and A. L. Lawrence. 1983. Survival and growth of
Penaeus setiferus L. and Penaeus aztecus I. larvae fed Artemia beginning at the
protozoea-two substage versus mysis-one substage. J. World Maricult. Soc. 14:38-48.

- . 1985. Survival, metamorphosis and growth of larvae from four penaeid species

fed six food combinations. Aquaculture (accepted for publication).

Ling, S. W. 1962. Studies on the rearing of larvae and juveniles and culturing of adult
of Macrobrachium rosenbergii de Man. IPFC Current Affairs Bull. 35:78-87.

Loon, Ti Teow. 1976. The effects of different types of feed and feeding level on the
survival of Penaeus monodon larvae from zoea to mysis stage. SEAFDEC Training
Rept. 28 pp.

Mock, C. R. 1971. Larval culture of penaeid shrimp at the Galveston Biological
Laboratory. Contrib. Nat. Marine Fisheries Serv., Galveston, Texas, 344:33-40.

Mock, C. R., D. B. Revera, and C. T. Fontaine. 1980. The larval culture of Penaeus
stylirostris using modifications of the Galveston Laboratory technique. Proc. World
Maricult. Soc. 11:102-117.

Platon, R. R. 1978. Design, operation and economics of a small-scale hatchery for the
larval rearing of sugpo, Penaeus monodon Fabricus. SEAFDEC Aquaculture Ext.
Man. 1:1-30.

Quinitio, E. , D. del la Pena, and F. Pascual. 1983. The use of substitute feeds in larval
rearing of Penaeus monodon. Proc. First Internat. Biennial Conf. Warm Water
Aquaculture-Crustacea (in press).

Quinitio, E., and I. Reyes. 1983. The effect of different feed combinations using chicken
egg yolk in Penaeus monodon larval rearing. Proc. First Internat. Biennial Conf.
Warm Water Aquaculture-Crustacea (in press).

Reader’s Digest Association Limited. 1976. The cookery year. 25 Berkeley Square,
London, 3rd ed., 439 pp.

Sorgeloos, P. 1980. The use of brine shrimp Artemia in aquaculture. The brine shrimp
Artemia. Vol. 3. Ecology. Culturing, use of aquaculture. Universa Press, Wettern,
Belgium.

Sorgeloos, P. , and G. Persoone. 1975. Technological improvements for the cultivation
of invertebrate as food for fish and crustaceans. II Hatching and culturing of the
brine shrimp Artemia salina L. Aquaculture 6:303-317.

Wilkenfeld, J. S., A. L. Lawrence, and F. D. Kuban. 1983. Rearing penaeid shrimp
larvae in a small-scale system for experimental purposes. Proc. First Internat.
Biennial Conf. Warm Water Aquaculture-Crustacea (in press).

- . 1984. Survival, metamorphosis and growth of penaeid shrimp larvae reared on

a variety of algal ( Skeletonema costatum and Chaetocerus gracilis, or Isochrysis sp.
and Tetraselmis chuii) and animal foods (the brine shrimp Artemia or the nematode
Panagrellus redivivus). J. World Maricult. Soc. 15: 39-49.

SEA URCHINS FROM THE BRAZOS SANTIAGO
PASS JETTY, SOUTH PADRE ISLAND, TEXAS

by RICHARD R. FAIRCHILD

Department of Biology
Mt . Vernon Nazarene College
Mt. Vernon, Ohio 43050

and L. O. SORENSEN

Pan American University — Brownsville
Brownsville, Texas 78520

ABSTRACT

Data on occurrence and distribution of sea urchins (Echinoidea) are presented for the
Brazos Santiago Pass jetty on South Padre Island, Texas. Echinometra lucunter is
recorded from the Texas coast. Key words : sea urchins, South Padre Island, Texas.

The northwestern coast of the Gulf of Mexico in Texas has been
altered by the construction of a pair of jetties each at Sabine Pass,
Galveston, Freeport, Port Aransas, and Port Isabel (Hedgpeth 1953).
Sea urchins (Echinoidea) are among the inhabitants of these jetties.

Arbacia punctulata (Lamarck), the purple sea urchin (order
Centrechinoida, family Arbaciidae) has been collected from Woods
Hole, Maine, to the coast of Brazil (Agassiz 1872; Mortensen 1907). A
stable population of A. punctulata on the Brazos Santiago jetties was
reported by Hedgpeth (1953), Hildebrand (1955), and Breur (1962).

Arbacia punctulata has a relatively thick test which measures
approximately 20mm in height and 35 mm in diameter. It is
subconical with a flattened actional surface. The analsystem normally
is composed of four large periproctal plates surrounded by five genital
and five ocular plates (Jackson 1912). The test is covered by spines that
are moderately stout and frequently longer than the diameter of the
test (Harvey 1956). The color of this sea urchin, when alive, varies
from almost black to straw-color. Dry denuded tests usually are gray,
with a pinkish tint near the apex (Agassiz 1872).

A second species of sea urchin was reported from the Brazos
Santiago jetty by Whitten et al. (1950). They termed it a short spined,
green species and presumed it to be Lytechinus variegatus. This species
subsequently was confirmed as a jetty inhabitant by Breuer (1962).

Lytechinus variegatus (Leske), commonly termed the green sea
urchin, belongs to the order Centrechinoida and the family Echinidae

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

384

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

(Moore et al. 1963). This species is typically a shallow-water inhabitant
and is found on sandy bottoms. It often covers its spines with sand,
shell fragments, and organic debris. The test is nearly circular and
covered by short stiff spines. Living individuals are almost invariably
brownish green (Zeiller 1974).

Echinometra lucunter (Linnaeus) is a large sea urchin that is
common in tropical waters of the western Atlantic. This species
belongs to the order Centrechinoida, family Echinometridae (Zeiller
1974). E. lucunter is known for its ability to burrow into limestone
rocks in which it may take up permanent residence. This widely
distributed echinometrid ranges from the mouth of La Plata in the
south to Cape Hatteras in the north (Mortensen 1907). It has been
reported from the Blanquilla coral reef, which lies offshore from
Laguna Tamiahua, about 60 miles from Tampico in the state of
Veracruz, Mexico (Moore 1958). Fotheringham (1980) mentioned the
occurrence of Echinometra lucunter on the South Padre Island jetty.
He reported it and Arbacia sp. to be approximately the same size and
color, and to be almost equally abundant on South Texas jetties, but
gave no reference to his source of information.

METHODS

Our observations were based on collections made during the past
eight years from the north jetty of Brazos Santiago Pass. In addition,
urchins were collected throughout the lower Laguna Madre.

Tests of Arbacia punctulata, Lytechinus variegatus , and
Echinometra lucunter were prepared by submerging specimens in a 10
percent chlorox solution for several days. We measured test diameter
for a sample of 123 Arbacia punctulata and 98 Echinometra lucunter
from the north jetty of Brazos Santiago Pass. Tests of all three species
were placed on exhibit at the Pan American University Marine Biology
Laboratory, South Padre Island, Texas.

RESULTS AND DISCUSSION

Lytechinus variegatus, although previously reported as occurring on
the Brazos Santiago jetty, has been absent from there for at least eight
years. However, it is common in the shallow- water grass beds of
Laguna Madre adjacent to the mud flats. It is more frequently
associated with the seagrass, Thalassia, than with Syringodium.

Arbacia punctulata remains one of the dominant animals on the
Brazos Santiago jetty. Collectors have harvested large numbers of this
species during the past three years. To date, however, there seems to
have been no detrimental effect on the population.

SEA URCHINS FROM BRAZOS SANTIAGO PASS JETTY

385

Echinometra lucunter occurs in about the same abundance as
Arbacia punctulata on the Brazos Santiago jetty. It also has been
collected extensively during the past three years without any apparent
effect on the population. The two species occur in similar locations,
and many times an area will have nearly equal numbers of both
species.

Our identification of specimens as Echinometra lucunter has been
verified by David Pawson of the National Museum of Natural History,
Washington, D.C. Specimens of E. lucunter from the Texas coast are
unusual in being nearly circular; more typically E. lucunter is slightly
longer than broad, so that the skeleton or test is an elongate oval. This
is believed to be the first substantiated report of Echinometra lucunter
on jetties of the Texas coast.

Fotheringham (1980) reported species of Arbacia and Echinometra to
be approximately the same size and the same color; our collections
from the jetty did not confirm his statement. In a random sample of
A. punctulata, the mean diameter of 123 tests was 35.8 mm whereas
in a random sample of 98 E. lucunter, the mean diameter of tests was
67.3 mm. All A. punctulata were dark purplish-black, whereas E .
lucunter ranged in color from a rosy pink to the dark purplish-black
of A. punctulata.

LITERATURE CITED

Agassiz, A. 1872-74. Revision of the Echini, Part 1-4. Mem. Mus. Comp. Zool., Harvard
Coll. 7:263-265.

Brener, J. P. 1962. An ecological survey of the lower Laguna Madre of Texas, 1953-1959.
Publ. Inst. Mar. Sci. 8:153-183.

Fotheringham, N. 1980. Beachcomber’s guide to Gulf Coast marine life. Gulf Publishing
Co., Houston, Texas.

Harvey, E. B. 1956. The American Arbacia and other sea urchins. Princeton Univ. Press,
Princeton, New Jersey.

Hedgpeth, J. W. 1953. An introduction to the zoogeography of the north western Gulf
of Mexico with reference to the invertebrate fauna. Publ. Inst. Mar. Sci. 3:187-191.

Hildebrand, H. H. 1955. A study of the fauna of the pink shrimp ( Penaeus duorarum
Burkenroad) grounds in the Gulf of Campeche. Publ. Inst. Mar. Sci. 4:169-232.

Jackson, R. T. 1912. Phylogeny of the Echini, with a revision of Palaeozoic species.
Mem. Boston Soc. Nat. Hist. 7:113-195.

Moore, O. M. 1958. Notes on the Blanquilla Reef, the most northerly coral formation
in the western Gulf of Mexico. Publ. Inst. Mar. Sci. 5:151-155.

Moore, H. B., T. Jutare, J. C. Bauer, and J. A. Jones. 1963. The biology of Lytechinus
variegatus. Bull. mar. Sci. Gulf Caribbean 13:23-53.

Mortensen, T. 1907. Echinoidea. The Danish Ingolf-Expedition 4(2):185-194.

Whitten, H. L., H. F. Rosene, and J. W. Hedgpeth. 1950. The invertebrate fauna of
Texas coast jetties; a preliminary survey. Publ. Inst. Mar. Sci. 1:53-87.

Zeiller, W. 1974. Tropical marine invertebrates of southern Florida and the Bahama
islands. John Wiley and Sons, New York.

OCCURRENCE OF MUSSELS IN THE ORBITS
OF A BLUE CRAB

by GEORGE N. GREENE, DAVID C. McADEN, and
WILLIAM B. BAKER, JR.

Houston Lighting & Power Company
P.O. Box 1700
Houston, Texas 77001

ABSTRACT

A blue crab ( Callinectes sapidus) was found with mussels ( Brachydontes recurvus)
attached in the orbits. Key words: blue crab, Callinectes sapidus, Brachydontes recurvus,
epibiosis, Galveston Bay.

A male crab ( Callinectes sapidus) with three mussels ( Brachydontes
recurvus) in the orbits was obtained on 26 October 1984 in a trawl near
the discharge structure of Houston Lighting 8c Power Company’s P. H.
Robinson Generating Station in Galveston Bay, Texas.

One mussel was found in the right orbit of the crab and two in the
left (Fig. 1). The mussel in the right orbit was 10 mm wide at its
widest visible part and protruded 7 mm from the orhit. It did not
occupy the entire orbit, the lateral 3 mm of which was open. It was
removed for identification. The large mussel in the left orbit had a
visible width of 8 mm and protruded 5 mm from the orbit. The
smaller specimen lay ventrolateral to the large one, was 5 mm wide,
and did not protrude.

The orbits showed no apparent distortion in size or shape when
compared to those of other crabs of similar size. However, the eyestalks
were pushed medially by the mussels, preventing retraction. The crab
appeared in good condition otherwise and no injuries or other
epibionts were visible to the naked eye. The carapace measured 160
mm wide and the total weight of the specimen was 263 grams.

This is the first time the authors have seen a mussel living attached
to a crab in their many years of experience on the Galveston Bay
system. No report of a mussel living attached to a crustacean was
found in the literature.

The interesting question is how the mussel larvae evaded the crab’s
defenses and were able to attach in the orbits and remain to grow to
their present size. The specimens have been retained in the permanent
collection at the Cedar Bayou Marine Laboratory of Houston Lighting
8c Power Company.

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

388

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Figure 1. Blue crab with mussels in orbits.

AN INSTANCE OF A LARGEMOUTH BASS,
MICROPTERUS SALMOIDES , FEEDING ON A WATER
SNAKE, NERODIA ERYTHROGASTER TRANSVERSA

by DENNIS PARMLEY and CHARLES MULFORD

Texas Parks and Wildlife Department
6200 Hatchery Road
Fort Worth, Texas 76114

Numerous studies have been conducted concerning the food habits
of adult largemouth bass ( Micropterus salmoides) and the most
frequently reported prey items are fish and crayfish (Carlander 1977).
Small mammals also are eaten occasionally (Snow 1971).

After reviewing current literature, it appears that reptiles rarely are
eaten by this fish. However, Mullen and Applegate (1970) reported
three worm snakes and a skink in the stomachs of largemouth bass
collected from an Arkansas impoundment that was in the process of
being filled; these terrestrial reptiles probably were washed into the
water as the reservoir was flooded. To our knowledge, no specific
record of an aquatic snake being eaten by M. salmoides has been
reported previously.

A largemouth bass weighing 350 grams and measuring 316 mm in
length was obtained on 4 September 1984 by one of us (CM) while
electrofishing in a small (250 surface acres) northcentral Texas
impoundment, Marine Creek Lake, located in Tarrant County. The
stomach of the fish contained a blotched water snake ( Nerodia
erythrogaster transversa) that measured 457 mm long. No other food
items were found in the stomach. The snake was well preserved and
mostly intact, with only small patches of skin missing, thus allowing
correct identification.

LITERATURE CITED

Carlander, K. D. 1977. Handbook of freshwater fishery biology. Iowa State Univ. Press,
Ames, Iowa, vol. 2, 423 pp.

Mullen, J. W., and R. L Applegate. 1970. Food habits of five centrarchids during filling
of Beaver Reservoir. Tech. Paper Bur. Sport Fish. Wildlife 50:1-16.

Snow, H. E. 1971. Harvest and feeding habits of largemouth bass in Murphy Flowage,
Wisconsin. Wisconsin Dept. Nat. Res. Tech. Bull. 50:1-24.

The Texas Journal of Science, Vol. XXXVII, No. 4, December 1985

i 'P

INDEX TO VOLUME XXXVII (1985),
THE TEXAS JOURNAL OF SCIENCE

Robert R. Hollander and Richard W. Manning

The Museum and Department of Biological Sciences,

Texas Tech University, Lubbock, Texas 79409

This index has separate subject and author sections. The subject index
has been modified somewhat from that of previous volumes of The Texas
Journal of Science. Key (or other important) words or phrases are
followed by an abbreviated title and initial page number of each article
in which they appeared. Scientific names of organisms were indexed only
to genus, followed by the initial page number of each article in which
that generic name was mentioned. Scientific names selected by authors
as key words, however, were treated as all other key words indexed.
Specific geographic areas or localities used by authors in titles or as key
words were entered as index headings with the exception of Texas,
because the majority of articles in the Journal dealt with that state. All
states (or countries) other than Texas appear as separate headings in the
index. Vernacular names of biological species and chemical compounds
were indexed only if used by authors in titles or as key words.

The author index includes the names of all authors, each followed by
the initial page number of the appropriate article. Also included is a list
of the names of colleagues who kindly served as reviewers of articles
submitted for this volume of the Journal.

SUBJECT INDEX

Ablabesmyia: 175
Abornia : 155
Abra: 269
Abstracts:

Aix: 215
Albunea : 269
Alligator. 215
Alpheus : 269
Amaranthus : 155
Amblema : 49
Ambrosia : 155
Ameiva: 33
Ammodramus : 215
Amnicola: 175
Ampelisca : 269
Anachis: 269
Anadara : 269
Anas: 215
Anchoa: 341
Ancinus : 269
Anodonta: 49
Aphylla: 175

Southwestern Developmental Biology
Conference, 103

Acacia : 155

Acanthohaustorius : 269
Acris: 215
Acroneuria : 175
Actinob della: 175
Aedicira: 269
Agkistrodon: 215
Aglaophamus: 269
Agraylea: 175
Air pollution:

atmospheric carbon monoxide, El Paso-
CD. Juarez, 235

392

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Apoprionospio : 269
Apseudes : 269
Arbacia : 383
Arcidens : 49
Aricidea: 269
Armandia: 269
Artemia : 371
Artiodactyla:

record, Symbos, Kaufman County, Texas,
311

Asellus : 175
Aspectrotanypus : 175
Aspergillus : 245
Astragulus : 155
Asychis : 269
Atherogenic diets:

fecal, intestinal bacteria, swine, 329
Aulodrilus: 175
Ay thy a : 215
Bacillus: 329
Bacteroides : 329
Baetis: 175
Baetisca : 175
Barrier islands:

vertebrate use, nontidal wetlands,
Galveston Island, Texas, 215
Batracob della: 175
Behavior:

comparative, sympatric centipedes, 253
Benthic:

damage assessment, oil spills,

macroinfauna, Texas continental shelf,
269

Betula: 175
Bifidobacterium: 329
Bivalvia:

range extension, Arcidens confragosus, 49
Blue crab:

mussels in the orbits, 387
bobwhite quail:

immune response, 133
Bootherium: 31 1
Bos: 311
Bouteloua: 155
Bovidae:

record, Symbos, Kaufman County, Texas,
311

Brachionus: 371
Brachydontes:

in orbits of blue crab, 387
Brassica: 201
Bratislavia: 175
Bufo : 215

Burmah Agate:

damage assessment, oil spills, 269
Cabira: 269
Caddo Lake:

Urnatella (Entoprocta), Texas and
Louisiana, 141
Caecum: 269
Caenis: 175
Calcitonin:

thyroid-parathyroid response, white-tailed
deer, 163

Calcium chloride: 163
Callianassa: 269
Callinectes:

mussels in the orbits, 387

explosives effects, marine organisms, 341
Callipepla:

foods, New Mexico, 155

immune response, 133
Callopistes: 33
Cambarellus: 175
Cambarus: 175
Camellia: 359
Capillary columns:

recent developments in gas

chromatographic trace analysis, new
concentration techniques, 259
Carassius: 75
Carbohydrates: 329
Carunculina: 49
Cassia: 63

Catoptrophorus: 215
Cellulose: 359
Cellulolytic activity:

factors influencing, soil fungus, 245
Centipedes:

comparative behavior, external color
patterns, sympatric taxa, 253
Ceratonereis: 269
Ceryle: 215
Chaetoceros: 371
Chaetogaster: 175
Chaoborus: 175
Chasmocarcinus: 269
Chelydra: 215
Chenopodium: 155
Cheumatopsyche: 175
Chilopoda:

comparative behavior, external color
patterns, sympatric taxa, 253
Chimarra: 175
Chione: 269
Chironomus: 175

INDEX TO VOL. XXXVII

393

Chlorides: 5
Cholestral: 329
Chone : 269
Chrysemys : 215
Chrysops : 175
Cistothorus : 215
Cladotanytarus : 175
Clinotanypus : 175
Clostridium : 329
Clymenella : 269
Cnemidophorus : 33
Coelotanypus : 175
Coleus : 201
Colinus : 133
Coloration, aposematic:

comparative, sympatric centipedes, 253
Commelina : 155
Compatability-incompatability:

vegetative plant tissue, graft development,
201

Concentration techniques:
recent developments in gas

chromatographic trace analysis, 259
Conchapelopia : 175
Continental shelf:

damage assessment, oil spills, benthic
macroinfauna, Texas, 269
Corbicula: 49, 175
Corbula : 269
Coryneura: 175
Cossura : 269
Crangonyx : 175
Crassostrea: 341
Cricotopus: 175
Crocodilurus : 33
Croton: 63, 155
Cryptochironomus : 175
Cucurbit a: 155
Cw/ex: 175
Culicoides : 175
Cyclaspis: 269
Cyrtonaias: 49
Cyst wall:

formation, Posthodiplostomum
metacercariae, 143
Damage assessment:
oil spills, benthic macroinfauna, Texas
continental shelf, 269
Deer:

thyroid-parathyroid response, white-tailed,
163

Demography:

Texas kangaroo rat, 5 1

Dm?: 175
Dicrodon: 33
Dicrotendipes : 175
Didelphis: 215
Didymella : 213
Difference equations:

modeling in social science, 147
Dionda :

reproduction, Fessenden Spring, Texas,

321

Diopatra : 269
Diplodonta : 269
Dipodomys : 51

Dissolved solids accumulation:

estimation in reservoirs, 25
Dithyrea: 155
Diversity:

pollution effects, Village Creek, Texas, 175
Donax: 269
Dracena: 33
Dubiraphia: 175
Echinometra : 383
EDTA: 163
Egg yolk:

use as feed, penaeid shrimp larvae, 371
Egret ta: 215
Einfeldia: 175
Entoprocta:

Urnatella from Caddo Lake, Texas and
Louisiana, 141
Ephemerella: 175
Epibiosis:

mussels in orbits of blue crab, 387
Eriogonum: 155
Eudorella: 269
Eukieferiella: 175
Euphorbia : 155
Explosives effects:

marine organisms, geophysical
exploration, 341
Fecal bacteria:

swine, atherogenic diets, 329
Feed:

use of egg yolk, rearing penaeid shrimp
larvae, 371
Feeding ecology:

scaled quail, New Mexico, 155
Fine structure of secondary walls:

sclereids, Rauwolfia serpentina , 359
Food habits:

scaled quail, New Mexico, 155
largemouth bass feeding on water snake,
389

394

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Foredune complexity:

North Padre Island, Texas, 63
Fulica : 215
Fungi, marine:

association, Spartina, Harbor Island,
Texas, 213
Fungus, soil:

factors influencing cellulolytic activity, 245
Fusobacterium : 329
Gallinago: 215
Gallinula: 215
Gallus : 133

Galveston Bay, Texas:

mussels in orbits of blue crab, 387
Galveston Island, Texas:

vertebrate use, nontidal wetlands, 215
Gas chromatography:

recent developments, trace analysis, new
concentration techniques, 259
Gastrophryne : 215
Gaura : 155

Geophysical exploration:

explosives effects, marine organisms, 341
Gidleya : 3 1 1
Glycera : 269
Glyptotendipes : 175
Graft development:
vegetative plant tissue, compatability-
incompatability, 201
Grubeulepis : 269
Gryllus : 253
Goeldichironomus : 175
Goniada: 269
Gutierrezia : 155
Gyptis : 269
Haematopota: 175
Haemonias: 175
Haploscoloplos : 269
Harbor Island, Texas:

marine fungi associated with Spartina, 213
Harnischia : 175
Hebetancylus : 175
Helianthus: 155
Helius : 175
Helobdella : 175
Heteromyidae:

variation, Texas kangaroo rat, 5 1
Heterospio : 269
Heterotheca : 155
Hemidactylus : 33
Hemipholis : 269
Hexagenia: 175
Hexapanopeus: 269

Hexatoma: 175
Hoffmanseggia : 155
Humicola: 245
Hyala : 269
Hydra : 175
Hydroporus : 175
Hydropsyche\ 175
//y/tf: 215
Hypercalcemia:

thyroid-parpthyroid response, white-tailed
deer, 163
Hypocalcemia:

thyroid-parathyroid response, white-tailed
deer, 163
Ilyodrilus: 175
Immunoelectrophoresis:

immune response, bobwhite quail, 133
Immunology:

immune response, bobwhite quail, 133
Intestinal bacteria:

swine, atherogenic diets, 329
Intrapopulational variation:
caudal osteology, Cnemidophorus tigris,
33

Ipomoea: 63
Isochrysis : 371
Isolda: 269
Isotomurus : 175
Ixtoc:

damage assessment, oil spills, 269
Joint sets:

Precambrian-Paleozoic rock succession,
Llano Uplift, 123
Juncus : 215
Kalliapseudes : 269
Kentropyx: 33
Kiefferulus : 175
Kincaidiana : 175
Kinosternon: 215
Labrundina : 175
Laccophilus : 175
Lacerta: 33
Laevapex : 175
Lampsilis: 49
Largemouth bass:
temperature tolerance, 75
feeding on water snake, 389
Larsia : 175
Lepidasthenia: 269
Lepomis : 143
Leptophlebia : 175
Leptosphaeria: 213
Libellula: 175

INDEX TO VOL. XXXVII

395

Limnodrilus : 175
Lindra : 213
Linopherus : 269
Linum : 155
Lipids: 329
Listriella : 269
Lithospermum : 155
Litocorsa : 269
Llano Estacado:

records, spotted skunk and long-tailed
weasel, 355
Llano Uplift:

joint sets, Precamprian-Paleozoic rock
succession, 123
Long-tailed weasel: 355
Louisiana:

Urnatella from Caddo Lake, 141
Lovenella : 269
Lucina : 269
Ludwigia: 321
Lulworthia : 213
Lumbriculus : 175
Lumbrineris : 269
Lycopersicon : 201
Lytechinus : 383
Macoma : 269
Macrobenthos:

pollution effects, Village Creek, Texas, 175
Macrobrachium: 371
Macroinfauna:

damage assessment, oil spills, Texas
continental shelf, 269
Macro mia: 175
Magelona : 269
Marine organisms:

explosives effects, geophysical exploration,
341

Marphysa : 269
Mathematics:

monodiffric laplace and fourier
transforms, 85

difference equation modeling, social
science, 147
Mediomastus : 269
Melospiza : 215
Menetus: 175
Mentzelia : 155
Metacercariae:

cyst wall formation, Posthodiplostomum ,
143
Mexico:

atmospheric carbon monoxide, El Paso-
CD. Juarez, 235

Micropholis : 269
Micropogonias : 341
Micropterus : 75, 389
Minuspio : 269
Models:

estimation, dissolved solids accumulation
in reservoirs, 25

elements, difference equation, social
science, 147
Mollusca:

range extension, Arcidens confragosus, 49
Monarda : 155
Monoculodes : 269
Monoxide:

atmospheric, El Paso-CD. Juarez, 235
Musk ox:

record, Symbos, Kaufman County, Texas,
311

Mussel: 387
M us tela :

records. Llano Estacado, Texas, 355
Myocaster: 215
Mysidopsis : 269
Vaw: 175
Nanocladius: 175
Nassarius : 269
Na tarsia: 175
Natica: 269
Nectopsyche : 175
Nephtys: 269
Nereis : 269
Nerodia: 215, 389
New Mexico:

foods, scaled quail, 155
Ninoe : 269

Nongeographic variation:

Texas kangaroo rat, 51
Notomastus: 269
Nucula : 269
Nuculana : 269
Nycticorax : 215
Nyctiophylax: 175
Odocoileus: 163, 215
Oecetis: 175
Oenothera : 63
Ogy rides: 269
Oil spill:

damage assessment programs, benthic
macroinfauna, Texas continental shelf,
269

Onuphis: 269
Ophidonais: 175
Opistocysta: 175

396

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Opuntia : 155
Orethemis : 175
Orthocladius : 175
Oryzomys : 215
Osteology:

intrapopulational variation,
Cnemidophorus tigris, 33
Owenia : 269
Oxyurostylis : 269
Pachydiplax : 175
Padre Island, Texas:
variation in vegetation density and
foredune complexity, 63
Pagurus : 269
Paleanotus: 269
Palemonetes : 175
Paleozoology:

record, Symbos, Kaufman County, Texas,
311

Palpomyia : 175
Panaeus : 371
Panicum : 63, 155
Parachironamus : 175
Paracladopelma : 175
Paraonides : 269
Paraonis : 269
Paraprionospio : 269
Parasite:

vesicle contribution to cyst wall formation,
Posthodiplostmum metacercariae, 143
marine fungi associated with Spartina,
Harbor Island, Texas, 213
Paratanytarus : 175
Parathyroid harmone:
thyroid-parathyroid response, white-tailed
deer, 163
Paspalum : 155
Penaeus : 341, 371
Pentaneura : 175
Petricola : 269
Peptostreptococcus : 329
Phalacrocorax: 215
Phase olion: 269
Phaseolus : 155
Photis : 269
Phyllodoce : 269
175

Physicochemical:

pollution effects, Village Creek, Texas, 175
AVarw: 175
Pimep hales : 75
Pinnixa : 269
P//2W5: 227

Pisum: 359
Plant ago: 155
Plathemis: 175
Platyischnopus : 269
Plegadis : 215
Pleospora: 213
Podilymbus : 215
Pogonias : 341
Pollution:

effects. Village Creek, Texas, 175
Polyodontes: 269
Polypedilum : 175
Porphyrula: 215
Portulaca: 155
Posthodiplostomum : 143
Potamothrix : 175
Potamyia : 175
Potthastia : 175
Predation:

largemouth bass on water snake, 389
Primacord: 341
Prionospio : 269
Pristina : 175
Procambarus : 175
Procladius: 175
Procyon : 215
Progomphus : 175
Prosimulium: 175
Prosopis : 155
Protankyra: 269
Protohaustorius : 269
Psectrocladius: 175
Psectrotanypus : 175
Pseudeurythoe : 269
Pseudo chironomus: 175
Pseudorthocladius: 175
Quercus : 155, 175
215

Rana: 215
Range extension:

Arcidens confragosus, 49
Rattus: 215
Rauwolfia : 359
Renilla : 269
Reproduction:

Texas kangaroo rat, 51

Dionda episcopa, Fessenden Spring,
Texas, 321
Reptilia:

variation, caudal osteology,
Cnemidophorus tigris, 33

largemouth bass feeding on water snake,
389

INDEX TO VOL. XXXVII

397

Reservoirs:

estimation of dissolved solids
accumulation, 25
Rheocricotopus : 175
Rheotanytarsus : 175
Rhynchelmis : 175
Rio Grande fluvial:

suspended sediment variability in surface
waters, 5
Robackia: 175
Rodentia:

variation, Texas kangaroo rat, 51
Rorippa : 321
Rumex : 155
Salix: 175
Scaled quail:

foods, New Mexico, 155

immune response, bobwhite quail, 133
Sciaenops : 341
Scirpus : 215
Sclereids:

fine structure, secondary walls, Rauwolfia
serpentina , 359
Sclereid walls:

fine structure, Rauwolfia serpentina , 359
Scolelepis : 269
Scolopendra :

comparative behavior, external color
patterns, sympatric forms, 253
Sea urchins:

from Brazos Santiago Pass, South Padre
Island, Texas, 383
Secondary walls:

fine structure, sclereids, Rauwolfia
serpentina, 359
Sediment:

correlation, water quality, streams of East
Texas, 227
Sediment variability:

surface waters, lower Rio Grande fluvial
system, 5
Sedum : 201
Seismic exploration:

explosives effects, marine organisms, 341
Septoria : 213
Sesbania: 215
Sesuvium : 63
Set aria: 155
Shrimp larvae:

egg yolk as feed, rearing, 371
Sialis: 175
Sigambra: 269
Sigmodon: 215, 355

Siltelianthis : 63
Siltelianthus: 63
Skeletonema: 371
Skrjabingylus: 355
Slavina: 175
Smicridea: 175
Social science:

difference equation modeling, 147
Solanum: 155, 201
Somatochlora : 175
South Padre Island, Texas:

sea urchins, Brazos Santiago Pass, 383
Spartina:

variation, vegetation density, foredune
complexity, North Padre Island, Texas,
63

marine fungi associated, Harbor Island,
Texas, 213

vertebrate use, nontidal wetlands,
Galveston Island, Texas, 215
Spectrum analysis:

triple-channel, pulse wavetrain, comparing
methods, 189
Speocarcinus : 269
Spilogale:

records, Llano Estacado, Texas, 355
Spiophanes : 269
Spirogyra: 321
Sphaerium: 175
Sphaerulina : 213
Sporobolus: 155
Sporotrichum: 245
Spotted skunk: 355
Squilla: 269
Stenelmis: 175
Stenelus: 175
Stenochironomus : 175
Stenonema: 175
Stephensoniana : 175
Sthenelais: 269
Stictochironomus : 175
Streams:

pollution effects, Village Creek, Texas, 175

correlation, suspended sediment and water
quality, Texas, 227
Streptococcus : 329
Structural geology:

joint sets, Precambrian-Paleozoic rock
succession, Llano Uplift, 123
Sylvilagus: 215
Symbiocladius: 175
Symbos :

record, Kaufman County, Texas, 311

398

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Synchelidium : 269
Syringodium: 383
Tamarix : 215
Tanypus: 175
Tany tarsus: 175
Taxodium : 175
Teiidae:

variation, caudal osteology,
Cnemidophorus tigris , 33
7eiwj: 33
Tellina : 269

Temperature tolerance:

largemouth bass, 75
Tenax-GC: 259
Terebellides: 269
Terebra: 269
Tetraselmis : 371
Texas kangaroo rat: 51
Thalassia : 383
Thamnophis: 215
77wryx: 269
Thermal stability:

reproduction, Dionda episcopa, Fessenden
Spring, Texas, 321
Thermoactinomyces : 245
Thermomonospora : 245
Thienemanniella : 175
Thyone : 269
Tipula : 175
Trace analysis:

recent developments, gas chromatography,
new concentraction techniques, 259
Transforms:

laplace and fourier, 85
Tribelos : 175
Trichoderma : 245
Trichophoxus: 269
Tricladida: 175
Tricorythodes: 175
Triplasis : 155
Twfo/ejc: 175

Tupinambis : 33
Typha: 215
Uniola : 63
Unionidae:

range extension, Arcidens confragosus, 49
Upogebia: 269
Urnatella :

Caddo Lake, Texas and Louisiana, 141
Variation:

intrapopulational, caudal osteology,
Cnemidophorus tigris, 33
nongeographic, Texas kangaroo rat, 5 1
vegetation density and foredune

complexity. North Padre Island, Texas,
63

Vegetation density:

North Padre Island, Texas, 63
Vegetative plant tissues:

compatability-incompatability, graft
development, 201
Veillonella: 329
Verbena : 155
Verbesina: 155
Virgularia: 269
Vitrinella: 269
Water quality:

correlation, sediments, streams, East
Texas, 227

Wavetrain, triple-channel:

spectrum analysis, comparing methods,
189

Wetlands:

vertebrate use, Galveston Island, Texas,
215

White-tailed deer: 163, 215
Wildlife:

use, nontidal wetlands, Galveston Island,
Texas, 215
Xenanthura: 269
Xironodrilus : 175

AUTHOR INDEX

Aguirre, M., Jr., 235
Alnes, J. R„ 123
Applegate, H. G., 235
Baca, E. J., 245
Baker, W. B„ Jr., 387
Barclay, C. M., 175
Berry, R. L., 341
Best, T. L., 155
Blum, M., 63

Brown, R. D., 163
Buckner, J. E., Jr., 341
Cebull, S. E., 123
Chang, M., 227
Chao, C. C., 163
Curry, E. W., 147
Cusak, T. M., 141
Daneshi, T., 85
Deeter, C. R., 85

Fairchild, R. R., 383
Fuze, D. M„ 371
Garza, J. M., 143
Ghaoui, L., 259
Gonzalez, H. F., 133
Granillo, A. B., 227
Greene, G. N., 387
Guest, W. C, 75
Barrel, R. 175

INDEX TO VOL. XXXVII

399

Hendricks, F. S., 33

McCullough, D. A., 355

Rashin, E. B., 227

Hollander, R. R., 355

McCullough, J. D„ 141

Schram, A. C., 133

Jones, J. K., Jr., 51, 355

McDonald, J. N., 311

Shideler, G. L., 5

Jones, J. R., 63

Meade, T. G„ 143

Smartt, R. A., 155

Koehn, R. D., 213

Meador, C. D., 133

Sorensen, L. O., 383

Kruse, O. E., 189

Mia, A. J., 359

Walling, D„ 147

Landry, A. M., Jr., 341

Moore, R., 201

Ward, G. H„ Jr., 25

Lawrence, A. L., 371

Mueller, A. J., 215

Wayne, L. M„ 321

Lewbel, G. S., 269

Mulford, C, 389

Webster, W. D., 51

Lewis, D. H., 329

Neck, R. W., 253

Weisner, S., 259

Linton, T. L., 341

Nonnast, J. H., 189

Whiteside, B. G., 321

Mather, C. M., 49

Ortega, J., 245

Wilkenfeld, J. S., 371

McAden, D. C, 387

Parmley, D., 389

Zlatkis, A., 259

SOUTHWESTERN DEVELOPMENTAL BIOLOGY CONFERENCE, ABSTRACT

AUTHOR INDEX

Abbott, M. K., 104

Huettermann, A., 113

Peeples, E., 118

Allen, R. D„ 106

Ivatt, R. J., 117

Percy, J., 106

Allison, J. E., 118

Janer, L., 106

Pierron, G., 1 14

Arnott, H. J., 106

Jeffery, W. R., 109

Sauer, H. W„ 113, 114

Aufderheide, K. J., 113

Kalthoff, K., 106

Saugstad, J. A., 1 17

Beckingham, K., 108

Kaufman, T. C., 104

Schelling, M. E., 118

Bell, P. B., Jr., 112

Kingsley, S., 110

Seto, F., 119

Bloch, D. P, 115

Klass, M. R., 100

Shipley, G. L., 113, 115, 118

Bradshaw, W. S., 110

Klein, W. H., 112

Silverman, H., 105

Buchanan, E., 116

Lamoreux, M. L., 117

Smith, J., 1 15

Campbell, M., 106

Lee, T., 112

Stanley, A. J., 118

D’Agostino, J. B., Ill

Leeper, L. L., 105

Steffens, W. L., 105

Dietz, T. H„ 105

Lopez, L., 110

Subtelny, S., 108

Duke, J., 106

Marrs, J., 115

Thomas, T., 115

Durica, D. S., 115

McArthur, B., 1 15

Tobin, S. L., 117

Field, J. B., Ill

McCarthy, B. J., 117

Tomlinson, C. R., 109

Franklin, L. E., 119

McGarth, S. A., 1 10

Uzman, J. A., 107

Frazier, M. L., Ill

Mifflin, R., Ill

Wacker, K., 106

Genovese, G., 104

Nader, W., 113, 114

White, H. U., 114

Gumbreck, L. G., 118

Nazimiec, M. E., 108

Wilt, F. H., 107

Harris, C., 110

Nelson, E., 107

Winkler, M., 107

Henney, H. R., Jr., 1 14

Nessler, C. L., 106

Yamanaka, M., 1 17

Henning, S. J., 104, 105

Nordstrom, J. L., 1 15

Yang, M., 112

Hershey, J., 107

Papaconstantinou, J., Ill

REVIEWERS

Adams, W.

Clark, W. J.

Golightly, C.

Amoss, M. S., Jr.

Conrad, G. S.

Guest C.

Bechler, D. L.

Coutant, C. C.

Hannan, H. H.

Bragg, L. H.

Dalquest, W. W.

Harper, D. E., Jr.

Branson, B. A.

Davis, C. A.

Harrel, R. C.

Burrus, C. S.

Dehn, P. E.

Harry, H. W.

Cain, J. R.

Dial, B. E.

Hart, J.

Capen, C. C.

Ditton, R. B.

Hubbs, C.

Chamberlin, G. W.

Engstrom, M. D.

Kemp, W. M.

Cichra, C. E.

Ethridge, R.

Kimber, C.

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

400

Koenig, K. J.

Nixon, E. S.

Sweet, M.

Kohler, E. M.

Owen, R. D.

Szaniszlo, P. J.

Lawhorn, B.

Rabalais, N. A.

Taber, W. A.

Longley, G.

Roberts, L. S.

Tsutsui, E. A.

Lundelius, E. L., Jr.

Sartin, A.

Tunnell, J. N., Jr.

Martyu, R. W.

Schmidly, D. J.

Tuttle, J. R.

Mauseth, J. D.

Schreure, B.

Ubelaker, J. E.

McCullough, J. D., Jr.

Scifres, C. J.

Van Auken, O. W.

Measel, J. W., Jr.

Scudday, J.

VanBlaricom, G. R.

Messer, J. K.

Shideler, G. L.

Ward, G. H.

Mitchell, R. W.

Siehr, D. J.

Weller, M. W.

Mueller, D. M. J.

Strawn, K.

Wicksten, M. K.

Neck, R. W.

Swank, W. G.

Yuen, T. S.

INSTRUCTIONS TO AUTHORS

Scholarly manuscripts in any field of science or technology, including science education,
will be considered for publication in The Texas Journal of Science. Prior to acceptance,
each manuscript will be reviewed by at least two knowledgeable critics, the appropriate
Associate Editor, and the Editor. Manuscripts intended for publication in the Journal
should be submitted to the Editor, J. Knox Jones, Jr. (The Museum, Box 4499, Texas Tech
University, Lubbock, Texas 79409), in accordance with the following instructions.

No manuscript submitted to the Journal is to have been published or submitted
elsewhere. Manuscripts must be double-spaced throughout (including tables, legends, and
cited literature) and submitted in triplicate on typed or clear machine-produced copies on
8.5 x 1 1-inch paper, with margins of approximately 1.5 inches.

The centered title of the article (usually 10 words or less) should be followed by the
name(s) of the author(s) and institutional or business address(es), including zip-code, both
also centered on the title page. Each manuscript should have a brief, concise Abstract,
terminating with up to five key words. The following text can be subdivided into sections
as appropriate (examples follow): introductory information is self evident and thus usually
needs no heading; materials and methods (acknowledgments frequently can be placed here
as well); results; discussion; summary or conclusions; literature cited. Major sections are
centered and capitalized; secondary divisions are italicized (underlined) left flush; tertiary
headings are italicized at the beginning of paragraphs.

Cite all references in text by author and date in chronological order — Jones (1971); Jones
(1971, 1975); (Jones, 1971); (Jones, 1971, 1975); (Jones, 1971; Smith, 1973; Davis, 1975);
Jones (1971), Smith (1973), Davis (1975); Smith and Davis (1985); (Smith and Davis, 1985).
If more than two authors, use Jones et al. (1976) or (Jones et al., 1976). Citations to
publications by the same author(s) in the same year should be designated alphabetically
(1979a, 1979b). Be sure all citations in text are included in the Literature Cited section and
vice versa. Hypothetical examples of proper citations are given below.

Davis, G. L. 1975. The mammals of the Mexican state of Yucatan. Unpublished Ph.D.
dissertation, Texas Tech Univ., Lubbock, 396 pp.

Jones, T. L. 1971. Vegetational patterns in the Guadelupe Mountains, Texas. Amer. J.
Bot., 76:266-278.

- . 1975. An introduction to the study of plants. John Wiley and Sons, New York,

xx+386 pp.

Jones, T. L., A. L. Bain, and E. C. Burns. 1976. Grasses of Texas. Pp. 205-265, in Native
grasses of North America (R. R. Dunn, ed.), Univ. Texas Studies, 205:xx+l-630.

Smith, J. D. 1973. Geographic variation in the Seminole bat, Lasiurus seminolus. J.
Mamm., 54:25-38.

Smith, J. D., and G. L. Davis. 1985. Bats of the Yucatan Peninsula. Occas. Papers Mus.,
Texas Tech Univ., 97:1-36.

Consecutively-paged journal volumes and other serials should be cited only by volume
number and pagination. Serials with more than one number and that are not consecutively
paged should be cited by number as well (Smiths. Misc. Coll., 37(3): 1-30).

Illustrations are acceptable only as original inked line drawings or photographic prints.
They normally should be no larger than 4.5 x 6.5 inches and mounted on 8.5 x 11 paper
or backing. Each figure should be marked on the back with the name of the author(s) and
figure number. If confusion might result as to arrangement of a figure, label “top.” All
legends for figures must be typed (double spaced) on a separate sheet(s) of paper. All
figures must be referred to in text — as “Figure 3” or “(Fig. 3).”

All tables are to be typed, double-spaced, and headed by the legend, on a single page(s)
for each table. All should be cited at the appropriate place in text as “Table 1” or “(Table
1).”

402

THE TEXAS JOURNAL OF SCIENCE— VOL. XXXVII, NO. 4, 1985

Some important specific points for authors: (1) do not break words at the right-hand
margin of text; (2) footnotes are to be avoided except as absolutely needed in tables; (3)
scales for illustrations should be on the figure, not in the legend, to avoid errors when
illustrations are reduced for publication; (4) be sure all lettering or other marks on
illustrations will be clearly evident after reduction of them to Journal page size; (5) the
editor should be notified immediately of any change in address of the responsible author,
whose telephone number also should appear on correspondence; (6) in order to make
papers more readable for the general scientific public, abbreviations are to be avoided in
text except for standard mathematical or chemical formulae (where an abbreviation might
be used many times to save space, write out the full term the first time used and give the
abbreviation, which can be used thereafter, in parentheses); (7) consult recent issues of the
Journal, beginning with 38(1), for all matters of style. v

The principal author will receive galley proofs along with edited typescript and a reprint
order form. Proofs must be corrected and returned to the editor within five days; failure
to return proof promptly will result in delay of publication. Reprint order forms should
be returned directly to Texas Tech Press (P.O. Box 4240, Lubbock 79409), not to the editor.

Charges of $35 per printed page (or part thereof), or partial payment, strongly are
encouraged by members of the Texas Academy of Sciences when grant or institutional
funds are available for that purpose. Some contribution, even if modest, is expected for
any paper that exceeds 10 printed pages. Nonmembers of the Academy are required to
cover all page costs except as rarely excepted by the Secretary-Treasurer. All authors are
provided with page-charge information when their manuscript is accepted for publication.

DATES OF DISTRIBUTION OF VOLUMES 34-36,
TEXAS JOURNAL OF SCIENCE

34(1)

34(2)

34(3-4)

27 July 1982
30 September 1982
21 December 1982

35(1)

35(2)

35(3)

35(4)

28 March 1983
24 June 1983
30 September 1983
9 February 1984

36(1)

36(2-3)

36(4)

13 April 1984
7 November 1984
17 June 1985

THE TEXAS ACADEMY OF SCIENCE, 1985-86
OFFICERS

President:

President-Elect:

Vice-President:

Immediate Past President:
Secretary-T reasurer:

Editor:

AAAS Council Representative:

William J. Clark, Texas A&M University
Billy J. Franklin, Lamar University
Lamar Johanson, Tarleton State University
Michael J. Carlo, Angelo State University
Fred S. Hendricks, Texas A&M University
J. Knox Jones, Jr., Texas Tech University
Ann Benham, University of Texas at Arlington

DIRECTORS

1983 D. Lane Hartsock, Austin
Katherine Mays, Bay City

1984 E. D. McCune, Stephen F. Austin State University
Jim Neal, U.S. Fish and Wildlife Service

1985 George B. McClung, San Angelo
Barbara Schreur, Texas A&I University

SECTIONAL CHAIRPERSONS

I — Mathematical Sciences: John Seaman, Jr., Baylor University

II — Physical Sciences: John Hubisy, College of the Mainland

III — Earth Sciences: James O. Jones, University of Texas at San Antonio

IV — Biological Sciences: Randy Moore, Baylor University

V — Social Sciences: B. Thomas Gray, Southwest Texas State University

VI— Environmental Sciences: Dean V. Ferguson, Austin

VII — Chemistry: Lynn Melton, University of Texas at Dallas

VIII — Science Education: Thomas R. Koballa, Jr. University of Texas at Austin

IX — Computer Sciences: H. R Haiduk, Amarillo College

X— Aquatic Sciences: Darrell S. Vodopich, Baylor University

COUNSELORS

Collegiate Academy: Shirley Handler, East Texas Baptist College

Helen Oujesky, University of Texas at San Antonio
Junior Academy: Ruth Spear, San Marcos

Peggy Carnahan, San Antonio

COVER PHOTO

Albert Zlatkis, the Academy’s Distinguished Texas Scientist, 1985

2nd CLASS POSTAGE
PAID AT LUBBOCK
TEXAS 79401

3024 I 85

Library Acquisitions
Smithsonian Institute
Washington DC 20560

A

ISSN 0040-4403

2 -XT g >' 5

TITUTION N0I101I1SNI_ NVINOSH1I WS^ S 3 I d Vd 8 I l^LI B R AR I E S^SMITHSONIAN
^ Z \ 00 ^ _ _ ^ 05

IdVdan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIWS

z r- _ _ 2 r-

X ” X^3vX O . “! O - .*V

, m

w - co = - c/>

TITUTION NOlinillSNI NVINOSHlItMS S3ldVd8IT LIBRARIES SMITHSONIAN
vV co 2: _ _ _ co z . co

. . s < s , < x< , s:

-- V\ - ✓ - * E >\W r,

lg ji) * mw 8 mx * ®

muM> i %M, ?

r ,
r<T.^mts%s-i

X
co

o lo

> 5 2 T > XgSTg

2 00 Z CO * Z -

lavaan libraries Smithsonian institution NoumusNi nvinoshiiwc

~ co x co —

O XV£IX — 0

Z _J 2 _j - z V

TITUTION NOlinillSNI NVIN0SH1I1AIS S3ldVd8IT LIBRARIES SMITHSONIAN

Z £ » Z i— Z

m xX ° ~ /'ooTXv O

m W ^ X^VAst^X m

- co £ (j) _

yvyan libraries Smithsonian institution NoumusNi nvinoshihms

X ^ ^ ^ Z CO z

— x. £ Xx-llc/X < A s <

2 ^ Z mNA ~h 'W%y/Mr

° wm x fmf'Wj o

£ 8 JW* £

Z , / '^ H-

5 | ^ >•* I

TITUTION NOlinillSN^NVINOSHlIWS^SB I UVB a nZLI B RAR I ES^SMITHSONIAN

— — \-i ,

O X>*os*iX

Z _

dVdan li brar i es Smithsonian institution NoumusNi nvinoshiiws

^ _ Z r-

^Oias^V m Z V m ~ N^ua^X m

TITUTION NOlinillSNI- NVINOSHIIIAIS^SB I a VB 8 l“l~LI B R AR I ES ^SMITHSONIAN

»V. co z _ _ . co z CO

^ < Xv**°^/X S , <

q ./ ^ « ' 2 *.

X O

fn

X
CO
O

? •

5 2 -.XT >' _ _ z

UTION ^NOlinillSNI NV1N0SH1IWS S3 I dVdan LI BRAR I ES^SMITHSONIAN IIS

/BBil LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IINS S
r- • ___ 2 r~ _ . 2 r- z

m 2 x m 2 rn ^ « °

m -■ x^uusjjx m

c r> tz w ± co

UTION NOlinillSNI NVIN0SH1IIAIS S31BVBQH LIBRARIES SMITHSONIAN ir
co 2 ^ _ . £ g ^ 2

^ ^ /x-'7 yV * z =5 i

X /,y^ o ££ X o X ^ ^ o

\

CO

- xg^ocX > ’•T' s >

2 CO 2 CO * 2 CO

/yen LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IIMS S

— CO 2 ^ _ CO — CO

w ^ „ < to ^ xS35£S

>A —1 /°) x4 « f »A . x *-; / Aa — J / ^

— /O.

cc

< IQI

q: ^

o 5 5 “

UTION N0lifUllSNl‘JNVIN0SHJ.IWS:2S3 I BVa 8 IT 'll B RAR I ES^SMITHSONiAN-!

2 <~ ,, Z t~ 2 r~

:vrX O ATstitX _ w O XtlvnroX. ~ x<oa*^rNv O

03
23
>

2 ^

m ■' ^ m N^SsSS

lyan LIBRARIES SMITHS0NIAN~~l NSTITUTION^ NOlinillSNI ~NVIN0SH1IWS S

2 CO CO 2 CO

y -H
c
H

O
2

fd8H LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S

r- z

m L ^F 7 £ m xtiiDCX « X^^X m ^ «

to y — w — to

ution niA-^-Asni NViNOSHiUMS S3iavaan libraries Smithsonian h>

to Z _ to z ...,. to 2

. s

V

3