nitlSNl”,NVIN0SHJLIWSZS3 I a V« 9 11 ""u B R AR I ES^SMITHSONIAN^INSTITUTIOI

_ m

CO — CO Zz CO

VARIES SMITHSONIAN INSTITUTION NOIlfUIlSNI NVINOSHXIWS S3I WV«ai z ^ to 2 co

<

z 4C o /£/

^ ■’#>/ <o 'vWlM. CO AftW'jShK co 1=1

VW O \%y.v¥o/ X :X '7|w O 06^70;/ X

# 1^^ | i*- X/- |

fUliSNI_NVIN0SHilWS^S3 I a va a n\l B RAR I ES^SMITHSONIAN JNSTITUTlOf ia^\ w ^ w /?3s£$8\ m /^rSiox w S

O ' __ "NjyAbtix^ o

J «gn

VARIES SMITHSONIAN INSTITUTION N0IXnXIXSNl~NVIN0SHXIWS~'S3 I &VM a I

? r ^ _ 2 r- z

^ m g ^35^ m X k> N^5?S?

<o — co __

nilXSNI NVINOSHXIINS $3 I d Vfcl 8 11 LIBRARIES SMITHSONIAN INSTITUTIO

Z ^ CO Z ,,. CO 2

s > r-r s > ^usty 2

1AR I ES ^SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHilWS^SI I H Va 8 I

so ^ _ _ _ CO zz . CO

yj

\ 52

a - 4gfeX 06

llto <

O x^v_d ^ o X 2

fUlXSNI^NVINOSHXMS^Sl I d VH 9 I 1 L I B R AR I ES^ SMITHSONIAN^INSTITUTIO £ * Z _ r* z <~

m

CO X CO ' ± CO

VARIES SMITHSONIAN INSTITUTION MOIXflXIXSNI NVINOSHXmS S3IHVH9I to _ _ ^ z « co z co

S X*vnfo>v < ^ S 5

z

t/W M*- <*> =

^*#r x

> vss' s ‘'w > - s '<&£< ;£y >

X w •••*' z « z

fUliSNI_NVINOSHilHS S3 1 avaa 11 LIBRARIES SMITHSONIAN INSTITUTIOI

* \ S — = w

lllSNl"JNVIN0SHimS*S3l ava 9 II^LI B R AR I ES^SMITHSONIANJNSTITUTION

rn '' z rn Xl^OSVA^ ^ rp

AR I ESWSMITHSONIAN~INSTITUTIONWNOUniUSNrNVINOSHi!WS S3 I 9VU 9 H

<2 _ _ _ z r <$ - § „ _ ^ £

2 ^ S S y^g*o$s

wm ' § s 7MmQ(mwm* w’m ■

> s '^Ng$r > v' S ^

illSN|ZNVIN0SHJLIWS</>S3 I M VH 9 n\l B RAR I ES SMITHSONIAN INSTITUTION

CO _ _ — CO „ 9

o ' n >ivJVAS*^ Q H O

ar i es^2 Smithsonian”* institution^ noixoxiisni^nvinoshxiiais S3i ava an

_ z r; z _ n >v/ s _ _

m

_ co ~ co - —

UIXSNI NVINOSHIIWS S31 WVbiail LIBRARIES SMITHSONIAN INSTITUTION 2 co z ,-.v z:

< /flSSI&X 1 , . ,•= S /5^>v ~ '■'

1 (|e J3J -

i ^ 2 > _

AR I ES^SMITHSONIAN^ INSTITUTION NQIXnXIX$NI_MV!NQSHXI!N$^S3 I B¥ H 8 I

^5x “ s , *% | ^

■ft *

q < c/ <£ »

m VoMK'y

Q Vgft. D.c^ «_ q

IJLIJLSNl'JNVIN0SHimSZS3 I a Va 9 M^LI B RAR I ES^. SMITHSONIANJNSTITUTlOf'

£0^ m V 2 /Z@S£S&s. 5> °

vZ-,^y rn m ^ w

AR I ESt”sMITHSONIAN~INSTITUTION</>NOIiniliSNI NVIN0SH1IINS S3 I ava 9 I t/> _ _ _ z r « z _ _ _ |2

z /pm ^

*p> «g »•' \^v 5, ' jg >

nillSN I Z NVINOSHJLHNS</> S 3 ! a Va 9 ! 1*L, B R AR I ES^SMITHSONIAN JNSTITUTIOt

^ 5 9|Br I § fl&yylJ § W*&

lume XXXII, No. 1

March, 1980

PUBLISHED QUARTERLY BY

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

GENERAL INFORMATION

MEMBERSHIP. Any person engaged in scientific work or interested in the promotion of science is eligible for membership in The Texas Academy of Science. Dues for annual members are $15.00; student members, $7.00; sustaining members, at least $25.00 in ad¬ dition to annual dues; life members, at least $300.00 in one payment; patrons, at least $500.00 in one payment; corporation members, $250.00 annually; corporation life members $2000.00 in one payment. Annual subscription rate is $45.00. Dues should be sent to the Secretary- Treasurer. Subscription payments should be sent to the Managing Editor.

TEXAS JOURNAL OF SCIENCE

Editor: G. ROLAND VELA, PhD.

Managing Editor: MICHAEL J. CARLO, PhD.

The Journal is a quarterly publication of The Texas Academy of Science and is sent to all members and subscribers. Single copies may be purchased from the Managing Editor.

Manuscripts submitted for publication in the Journal should be sent to the Manuscript Editor, P.O. Box 1 3066, North Texas State University, Denton, Texas 76203.

The Texas Journal of Science (USPS 616740) is published quarterly by the Talley Press, San Angelo, TX, U.S.A. (2nd Class Postage paid at Post Office, San Angelo, TX 76901). Please send 3579 and returned copies to the Editor (P.O. Box 10979, ASU, San Angelo, TX 76901.)

Volume XXXII, No. 1

March, 1980

CONTENTS

Instructions to Authors . . . . 2

Note from the Editor . . . 5

Algebraic Structure of Polars. By Ali R. Amir-Moez and Mohammed Goodarzi ...... 9

Tabosa- Delaware Basin as an Aulacogen. By D. H. Shurbet and S. E. Cebull . . 17

Woody Vegetation of Upland Plant Communities in the Southern Edwards Plateau.

By O. W. Auken, A. L. Ford, A. Stein, and A. G. Stein . 23

The Upper Incisors of the Giant Horse, Asinus giganteus. By Walter W. Dalquest . 37

A Cytological and Histochemical Analysis of the Ovarian Follicle Cells of the South

Texas Squid ( Loligo pealei). By Samuel A. Ramirez and Manuel Guajardo . 43

A Survey of Selected Plants for the Presence of Eukaryotic Protein Biosynthesis

Inhibitors . By Robyn Reynolds and James D. Irvin . . . 55

Reconnaissance Observations of Some Factors Influencing the Turbidity Structure of

a Restricted Estuary: Corpus Christi Bay, Texas. By Gerald L. Shideler . 59

Heavy-Mineral Variability in Fluvial Sediments of the Lower Rio Grande, Southwestern

Texas. By Gerald L. Shideler and Romeo M. Flores . . 73

NOTES SECTION

2-Alkyl-3-(2-Pyridyl)-Cinchoninic Acids. By Eldon H. Sund, Robert E. Cashon,

and Rodney L. Taylor . . 93

Central Texas Breeding of the American Woodcock, Philohela minor. By Doyle T.

Mosier and Robert F. Martin . . . . . 94

INSTRUCTIONS TO AUTHORS

Papers intended for publication in The Texas Journal of Science are to be sub¬ mitted to Dr. Roland Vela, Editor, P. O. Box 13066, North Texas State University, Denton, Texas 76203.

The manuscript submitted is not to have been published elsewhere. Triplicate typewritten copies (the original and 2 reproduced copies) MUST be submitted. Typing of both text and references should be DOUBLE-SPACED with 2-3 cm margins on STANDARD 814 X 11 typing paper. The title of the article 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 numbered. Include a brief abstract at the beginning of the text (abstracting services pick this up directly) followed by an introduction (understandable by any scientist) and then whatever paragraph headings are desired. The usual editorial customs, as exemplified in the most recent issues of the Journal , are to be followed as closely as possible.

In the text, cite all references by author and date in a 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; and 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: Jones, A. P., and R. J. Wilson, 1971— Effects of chlorinated hydrocarbons./. Comp. Phys., 37:116. (Only the 1st page number of the article is to be used.) For a PAPER PRESENTED at a symposium, etc., use the form: Jones, A. P., 1971— Effects of chlorinated hydrocarbons. WMO Sym¬ posium on Organic Chemistry, New York,N.Y. For a PRINTED PAPER use: Jones, A. P., 1971— Effects of chlorinated hydrocarbons. Univ. of Tex., Dallas, or Jones,

A. P., 1971— Effects of chlorinated hydrocarbons. Univ. of Tex. Paper No. 14,46 pp. A MASTERS OR Ph.D THESIS should appear as: Jones, A. P., 1971— Effects of chlorinated hydrocarbons. M.S. Thesis, Tex. A&M Univ., College Station. For a BOOK, NO EDITORS, use: Jones, A. R, 1971— Effects of Chlorinated Hydrocarbons. Academic Press, New York, N.Y., pp. 13-39. For a CHAPTER IN A BOOK WITH EDITORS: Jones, A. P., 197 1 —Structure of chlorinated hydrocarbons. A. P. Jones,

B. R. Smith, Jr., and T. S. Gibbs (Eds.), Effects of Chlorinated Hydrocarbons. Academic Press, New York, N.Y., pp. 13-39. For a BOOK WITH EDITORS: Jones, A. P., 197 1— . Effects of Chlorinated Hydrocarbons. J. Doe, (Ed.), Academic Press, New York, N.Y., pp 3-12. For an IN PRESS PERIODICAL reference, use: Jones, A. P., 1971— Effects of chlorinated hydrocarbons. J. of Org. Chem. , In Press. For an IN PRESS BOOK reference, use: Jones, A. P., 1971— Effects of Chlorinated Hydrocarbons . Academic Press, New York, N.Y. In Press. References MUST include article title and page numbers.

References such as unpublished data or personal communications need not be

listed in the LITERATURE CITED section. However, within the text they should be presented as: (Jones, C., unpubl. data) or (Jones, C., pers. comm.).

All tables are to be typed with a carbon ribbon, free of error, without hand¬ written notations, and be prepared for photographic reproduction. Tables should be placed on separate sheets with a marginal notation on the manuscript to indicate preferred locations. Tables should have a text reference, i.e., Table 2 shows ... or (see Table 2).

Figures are to be original inked drawings or glossy photographs NO LARGER than 6V2 X 4 Vi inches and mounted on standard 8V2 X 1 1 paper. Legends for figures are to be typed separately and lettering within the figure kept to a minimum.

All photographs, line drawings, and tables are to be provided with self- explanatory titles or legends. Each illustration should be marked on the back with the name of the principle author, the figure number, and the title of the article to which it refers.

Galley proof of each article will be submitted to the author. This proof must be carefully corrected and returned within 3 days to the Managing Editor’s Office (Dr. Mike Carlo, Managing Editor, P. O. Box 10979— ASU Station, San Angelo, Texas 76901). Page proof will not be submitted. Page charge ($35/page) and reprint costs MUST accompany the return of the corrected galley of the manu¬ script (Check or Purchase Voucher). A delay in the printing of the manuscript will occur if payment is not submitted with the return of the galley.

Reprint price list and page charge information will accompany galley proofs. Reprints are delivered approximately 6 to 8 weeks after articles appear.

NOTICE: IF YOUR ADDRESS OR TELEPHONE NUMBER CHANGES, NOTIFY US IMMEDIATELY SO WE CAN SEND YOUR GALLEY PROOF TO YOU WITHOUT LOSS OR DELAY.

NOTE FROM THE EDITOR:

The Texas Journal of Science, in its effort to provide objective and impartial review of all papers considered for publication, has pioneered the method of anonymity for both author and reviewer. This has worked wonderfully well and is now established as the standard operating procedure of the Journal. In maintaining the principle of double anonymity, the Editor has sought to protect the identity of those reviewers who choose to remain unidentified and will honor this commitment. As a result of this ‘secrecy’ there is a strong, and very reasonable, concern regarding the reviewers and the review process. The explanation that follows should answer many questions; if not, please contact me at your convenience.

Papers are treated as follows:

1. Manuscript (ms.) received at Journal Office

2. Receipt of ms. acknowledged

3. Ms. sent to reviewers

4. Ms. accepted or returned to author

5. Ms. resubmitted by author

6. Returned to same or new reviewers

7. Ms. accepted or rejected

The lists that follow show the addresses of reviewers used in 1978 and 1979. It is hoped that they reveal something of the professional stature and quality of the reviewers responsible for the contents of the Texas Journal of Science.

G. Roland Vela, Ph.D.

Manuscript Editor

REVIEWERS 1978

Texas: 76 Reviews

7239 Bridle Path

San Antonio, TX 78240 1

840 Mulberry St.

San Antonio, TX 78212 1

2319 Fowler

Denton, TX 76201 1

P.O. Box 13048

Denton, TX 76201 1

M.D. Anderson Hospital & Tumor Inst. Houston, TX 77030 1

National Marine Fisheries Service Galveston, TX 77550 1

Other: 49 Reviews

Sea-Arama Marine World

Galveston, TX 77552 1

Texas Archaeological Salvage Project Austin, TX 78758 1

Abilene Christian University

Abilene, TX 79601 1

Angelo State University

San Angelo, TX 76901 1

Baylor University

Waco, TX 76703 2

Mary Hardin Baylor

Belton, TX 76513 1

6

THE TEXAS JOURNAL OF SCIENCE

NTSU

Denton, TX 76203 7

Rice University

Houston, TX 77001 2

SMU

Dallas, TX 75275 1

Southwest Texas State University

San Marcos, TX 78666 2

Stephen F. Austin State University Nacogdoches, TX 75961 1

Texas A&M University

College Station, TX 77843 7

Texas A&M Marine Lab

Galveston, TX 77550 1

Texas A&M Research & Extension Walde, TX 78801 1

Texas Christian University

Ft. Worth, TX 76129 2

Texas Southern University

Houston, TX 77004 1

Texas Tech University

Lubbock, TX 79409 4

TWU

Denton, TX 76204 2

University of Houston

Houston, TX 77004 6

UTA

Arlington, TX 76019 5

University of Texas

Austin, TX 78712 13

University of Texas Marine Lab

Port Aransas, TX 78373 1

University of Texas Medical School Houston, TX 77030 1

University of Texas

El Paso, TX 79968 1

University of Texas

San Antonio, TX 78285 2

West Texas State University

Canyon, TX 79016 3

Argonne National Lab

Argonne, IL 60439 1

Bureau of Sports, Fisheries & Wildlife Fayetteville, AR 72701 2

Consulting Biologist

LaFayette, LA 79598 1

Eason Oil Co.

Oklahoma City, OK 73118 111

Fish & Wildlife

Alberto, Canada 1

Institute of Food & Agric. Science Gainesville, FL 32611 1

National Marine Fisheries Service

Miami, FL 33149 1

U.S. Nat’l Museum of Nat’l History Washington, D.C. 20013 1

Arizona State

Tempe, AZ 85281 2'

Brigham Young University

Provo, UT 84602 1

Duke University

Durham, NC 27702 1

Michigan State University

E. Lansing, MI 48824 1

NYU

New York, NY 10003 1

Ohio State University

Columbus, OH 43210 1

Oklahoma State University

Stillwater, OK 74074 2

Oregon State University

Corvallis, OR 97331 1

Purdue University

LaFayette, IN 47907 1

Southwest Oklahoma State Univ.

Weatherford, OK 73096 1

Tulane University

New Orleans, LA 70118 1

University of British Columbia Vancouver, B.C. Canada V6T 1W5 1

reviewers

University of California

Berkeley, CA 94720 2

University of California

Davis, CA 95616 1

University of Charleston

Charleston, SC 29401 1

University of Georgia

Athens, GA 30602 1

University of Louisville Water Resources Lab.

Louisville, KY 40208 1

University of Michigan

Ann Arbor, MI 48109 1

University of Natal

Piefermaritzhug, South Africa 1

University of North Carolina

Chapel Hill, NC 27514 1

The University of Oklahoma

Norman, OK 73069 2

University of Rhode Island

Kingston, R I 02881 1

Univ. of Science & Arts of Oklahoma Chickasha, OK 73018 1

University of Southern California

Los Angeles, CA 90007 1

University of South Florida

Tampa, FL 33620 1

West Virginia University

Morgantown, WV 26506 1

REVIEWERS 1979

Texas: 49 Reviews		Other: 23 Reviews
7223 Lavendale Circle Dallas, TX 75230	1	Texas A&I University Kingsville, TX 78363	1
Shuler Museum of Paleontology SMU, Dallas, TX 75275	1	Texas A&M University College Station, TX 77843	3
Southwest Foundation for Res. & Ed. San Antonio, TX 78228	1	Texas Christian University Ft. Worth, TX 76129	3
Univ. of Texas Health Science Center San Antonio, TX 78284	1	Texas Tech University Lubbock, TX 79409	1
Texas Parks & Wildlife Department Austin, TX 78701	1	TWU Denton, TX 76204	1
Univ. of Texas Marine Science Inst. Galveston, TX 77550	1	University of Dallas Irving, TX 75061	1
North Texas State University Denton, TX 76203	5	University of Texas Arlington, TX 76019	2
Pan American University Edinburg, TX 785 39	1	University of Texas Austin, TX 78712	8
SMU Dallas, TX 75275	3	University of Texas Dallas, TX 75221	1
Southwest Texas State University San Marcos, TX 78666	3	University of Texas El Paso, TX 79968	4
Stephen F. Austin State University Nacogdoches, TX 75962	3	West Texas State University Canyon, TX 79016	2

8

THE TEXAS JOURNAL OF SCIENCE

University of Texas Medical School

Houston, TX 77030 1

Clemson University

Clemson, SC 29631 1

Lamar University

Beaumont, TX 77701 1

Montclair State College

Upper Montclair, NJ 07043 1

Northwestern University

Evanston, IL 60201 1

Oklahoma State University

Stillwater, OK 74074 1

State University of New York

Albany, NY 12222 1

S.W. Oklahoma State University Weatherford, OK 73096 1

University of Arkansas

Fayetteville, AR 72701 1

University of Connecticut

Storrs, CT 06268 2

University of Georgia

Athens, GA 30602 1

University of Idaho

Moscow, ID 83843 1

University of Minnesota

Minneapolis, MN 55455 1

University of North Carolina

Chapel Hill, NC 27514 1

University of Southern California

Los Angeles, CA 90007 1

University of Utah

Salt Lake City, UT 84112 1

Georgia Inst, of Technology

Atlanta, GA 30332 1

Indiana University Medical School Indianapolis, IN 46202 1

West Virginia University

Morgantown, WV 26506 1

Patuxent Wildlife Research Center

U.S. Fish & Wildlife Service

Laurel, MD 20811 1

U.S. Geological Survey

Denver, CO 80225 2

U.S. Fish Wildlife Service

Tulane University Museum Nat’l History Belle Chase, LA 70037

1

ALGEBRAIC STRUCTURE OF POLARS

by ALI R. AMIR-MOEZ

Department of Mathematics Texas Tech University Lubbock 79409

and MOHAMMED GOODARZI

Departmen t o f Math ematics University of Teheran Teheran, Iran

Reviewed by: Dr. E. D. McCune, Dept, of Math. & Stat., Stephen F. Austin State University, Nacogdoches 75962

ABSTRACT

The idea of pole and polar with respect to a conic is generalized to the polars of a point with respect to a polynomial hypersurface in a Euclidean k-dimensional space. Then mappings which transform these polars to each other are studied.

INTRODUCTION

In a Euclidean plane the concept of the polar of a point with respect to a conic is the study of a function whose domain is the set of points and its range is the set of lines in the plane. To obtain the polar of a point with respect to a conic, one employs ideas such as the harmonic mean of 2 real numbers, symmetric functions of roots of polynomials, and Taylor series. Thus one simplifies tedious substitutions and algebraic simplifications by applying these ideas.

In this article we start with simple cases and then we give some generalizations. Finally we study an algebraic structure of the polars.

A SPECIAL CASE

Consider the conic

P(x,y) = ax2 + 2bxy + cy2 + 2px + 2qy + d = 0 (1)

Accepted for publication: January 16, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

10

THE TEXAS JOURNAL OF SCIENCE

and the line through (x0, y0), i.e.,

x = x0 + tC

y = y0 + tm.

The points of intersection of the line and the conic are obtained from

ax? + 2bx0y0 + cy % + 2px0 + 2qy0 + d

+ 2[(ax0 + by0) C + (bx0 +cy0)m]t

+ [aC2 + 2b Cm + cm2] t2 = 0

which is obtained by substituting Eq. (2) in Eq. (1).

We observe that Eq. (3) is of the form

(2)

(3)

1 / 32P

2! Idxo8 + 23x09y0

32P „ 32P

8m + I t2 = 0,

(4)

3pk dpk

where — ^ means — ^ (x0,y0), k=l,2.

To explain this fact and exploit it for generalizations we consider the Taylor Expansion of P(x, y) about (x0 , y0) which is

p(x,y) = p(x0, y0) + +a^(y-yo)

2!

32P

3xf(X-X°>2 + 2 3x03y0

32P

(x-x0)(y-y0) + g^r(y-y0)2

= 0,

and we write Eq. (2) as

x - x0 “ tC

y - y0 = tm.

This substitution will explain Eq. (4).

POLARS

11

POLAR OF A POINT WITH RESPECT TO A CONIC

Consider the point (x0, y0) and the conic (1). Let a line through M inter¬ sect the conic in 2 points A and B (Fig. 1). It is clear that A and B correspond to the roots tj and t2 of Eq. (3). Let H correspond to the hormonic mean of tt and t2 , i.e., the value of t which satisfies

Then the locus of H as the line changes is called the polar of M with respect to the conic. One observes that

t _ ~2t ! t2 ti + t2

Since t2 and t2 are roots of Eq. (4) substituting for the sum and product of roots, we obtain

t =

2P(x0 , y0) ap n ap

r — Z + - — m

ax0 ay0

Figure 1.

12

THE TEXAS JOURNAL OF SCIENCE

Thus a set of parametric equations for the polar is:

x - x0 = -£

y - y0 = -m

Here the parameters are elements of the ordered pair (£, m), the set of direction numbers of the line. Thus one may let (£, m) vary in such a way that

3p dp

t — £ + - — m ox0 oy0

2P(x0,yo) 3p o, »P

a — £ + r — m

dx0 3y0

-1

Therefore the parameters may be eliminated, and we get

a~(x-x0) + ^ (y-yo) = -2P(x0,yo).

Consequently the polar of M is a straight line. A more elementary treatment is found in Elements of Linear Spaces (Amir-Moez and Fass, 1962).

One observes that when (x0, y0) approaches a point on the curve the polar tends to the tangent line at (x0, y0).

NOTATIONS

In order to generalize the ideas of Eqs. (1) and (2) we would like to make use of simpler notations. Let V be a Euclidean space of dimension k. Vectors will be denoted by Greek letters, for example, £ e V means J^-(x1 , . . ., xk).Thusa poly¬ nomial of degree n in k variables xx , . . ., xk can be denoted by pn(|) or simply P(£)

whenever there is no confusion. When we write

3Pk

dx}

we mean

3pk

3x^

(hi,

..,hk).

DEFINITIONS

Let (tx , . . ., tn} be a set of non-zero real numbers. Then we give these definitions:

POLARS

13

(i) The 1st harmonic mean of this set, u = Uj , satisfies

U1 j=l tj

(ii) The 2nd harmonic mean of this set, u2 satisfies

u2 t 1 ^2 t 1 t3 tn_i tn

= 2

ji < h

where jl5 j2 = 1,2,.. .,n.

In general the m-th harmonic mean of the set, i.e. um, satisfies

J l < • • • <Jm Ji

t* • . . . • t;

, m = 1

'rn

n,

where jj,j2 = 1,2, . . .,n.

INTERSECTION OF A LINE AND A POLYNOMIAL HYPERSURFACE

A set of parametric equations of a straight line in a Euclidean k -dimensional space can be written in a vector form:

£ = r + t«,

where . . ., x^), f**(h1? . . ., h^) which is a fixed vector, and 5^(d1? . . .,dk)

which is a direction vector. The points of intersection of this line and the poly¬ nomial hypersurface P(£) = 0 is obtained very much the same way as in Eq. (1). For simplicity, again we make use of Taylor Expansion of P(£) and use £ - f = t5. Thus we get a polynomial equation of degree n in t, i.e.,

Antn + An_jtn 1 + . . . + A0 = 0,

(5)

where

3hj

. . +

4

0)

(P), j = 0, 1, . . n.

The symbolic power is a well-known notation. For example.

14

THE TEXAS JOURNAL OF SCIENCE

,(2)

ah,

3h,

(P) = df

a2p

3h?

+ 2di d:

a2p

ahidhj

a2p

ah?

THE SET OF POLARS

Let f ^(hj , . . h^) represent a fixed point M in the Euclidean k-dimensional

space V. Let the line £ = £ + t5 through this point intersect the hypersurface P(£) = 0 in n points M i , . . Mn. These points correspond to tls . . tn, the roots of Eq. (5). Let N correspond to the vector £ and um, the m-th harmonic mean of h , . . tn. Then the locus of N is called the m-th polar of M with respect to P(£) = 0. One makes use of symmetric function of the roots of Eq. (5) and for the m-th harmonic mean one obtains

urn A0

We only consider cases for which Am ^ 0, m = 1,. ..,n. Special cases should be discussed and incorporated in.

Thus the m-th polar of with respect to P(£) = 0, in vector form, will be

? = f + (6)

Here 5, the direction vector, is a parameter. One may see that Eq. (6) is equivalent to a set of k equations

xj = hj + (-l)m T^-dj, j = 1 , . . k.

rim

Since 5 is a direction vector one may choose it to vary such that

i.

Therefore, eliminating the parameter in Eq. (6), we obtain

3 3

(x'-hl)ah7 + -+ (Xk-hk)ah^

(m)

(P) + (-l)m + 1 (m!)Pft) = 0

This is a hypersurface with the equation Q(£) = 0, where Q(£) is a polynomial of degree m.

POLARS

15

ALGEBRAIC STRUCTURE OF POLARS

Let the set of polars f with respect to P(£) = 0 be S = {Si , . . ., Sn}, where, for example, Sm is the m-th polar. As was pointed out in Eq. (6) the m-th polar Sm was obtained by the use of

um

(-ir

n

(7)

the m-th harmonic mean of the roots of Eq. (5), where we had taken the case Am ¥= 0, m = 0, 1 , . . ., n. We now consider the set T = (ux , . . ., un). Indeed, there is a one-to-one correspondence between S and T by um**Sm. We can define

one can

a mapping on T, i.e., up->uq by multiplying up by (-l)p q Thus

Ap

define a mapping on the Euclidean k-dimensional space V such that the p-th polar would be transformed to the q-th polar. We shall call this mapping Apq. One observes that

ApqAqr Apr.

The set of mappings has all properties of a group except closure.

QUESTIONS

Since the Taylor Series of P(£) has been very useful, one might wish to inves¬ tigate generalizations of pole and polar with respect to an analytic function of k real variables.

A very interesting question is: “How can one complete the set of mappings in Eq. (7) in order to have a group?”

If the point (fq , . . ., hk)<*f approaches P(J) = 0, then the set of polars will become tangent to the hypersurface. The study of this case should be interesting.

One can maneuver around the cases in which some Am is 0. This is left to the reader.

The field of real numbers may be replaced by other fields. The investigation of this is also left to the reader.

LITERATURE CITED

Amir-Moez, A. R., A. L. Fass, 1962 -Elements of Linear Spaces. Pergamon Press, Oxford.

TABOSA-DELAWARE BASIN AS AN AULACOGEN

by D. H. SHURBET and S. E. CEBULL

Department of Geosciences Texas Tech University Lubbock 70409

ABSTRACT

The Tabosa-Delaware basin region is located near the rifted boundary of a postulated late Precambrian-early Paleozoic supercontinent, and cross sections of the region show structure and tectonic timing similar to that of the Southern Oklahoma Aulacogen. Hence, the Tabosa-Delaware basin succession may be demonstrative of stages of aulacogen develop¬ ment. We suggest that these basins represent the failed arm of a 3-prong fracture pattern, that the 2 formerly active arms also may have left their signature, and that the Tabosa-Delaware basin evolution is tied to that of the Gulf of Mexico.

INTRODUCTION

Older, as well as some recent, geologic literature (for example, Wilhelm and Ewing, 1972) treat the Gulf of Mexico as a feature whose origin and essential development is of Mesozoic age. Inherently, this view failed to suggest any rela¬ tionship between the origin of the Gulf and the proximal oil-bearing Paleozoic basins, such as the Tabosa-Delaware succession of basins in West Texas (Fig. 1). However, if the Gulf of Mexico (or at least the Proto-Gulf) has its origin in the late Precambrian-early Paleozoic breakup of a supercontinent (Pangaea I), as we and others have suggested (for example, Valentine and Moores, 1972; Keller and Cebull, 1973; Shurbet and Cebull, 1975), the Tabosa-Delaware basin is a Gulf-marginal feature located adjacent to the Paleozoic plate boundary. The position of the basin with respect to the boundary is similar to that of an aulacogen , and that it might be an aulacogen is hinted by Fig. 1 of Hoffman, et al., (1974) and indicated by Walper (1977). Such a proposal is supported, at least circum¬ stantially, by comparison of independently developed evolutionary cross sections of the Tabosa-Delaware basin region presented by Horak (1975) with those of the Southern Oklahoma Aulacogen (Anadarko-Oklahoma basin region) as out¬ lined by Hoffman, et al , (1974), which utilizes data by Ham (1969). This com¬ parison, shown in Fig. 2, illustrates the general similarity of structure and timing of tectonic events in the 2 regions. Clearly, more study of the early development of the Tabosa-Delaware region is required before its possible aulacogenic affinity is proclaimed with assurance. For example, the early graben stage of aulacogen

Accepted for publication: June 14, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

18

THE TEXAS JOURNAL OF SCIENCE

of the Tabosa-Delaware Basin in relation to the Marathon-Ouachita-Appalachian orogenic belt and the early Paleozoic continental margin (dotted line; King, 1975, Thomas, 1977). The Southern Oklahoma and Reelfoot Aulacogens, which developed along this margin, also are shown.

LATE CAMBRIAN - EARLY DEVONIAN

TABOSA BASIN

LATE DEVONIAN

LATE DEVONIAN - MISSISSIPPI

DELAWARE BASIN

EARLY PENNSYLVANIAN

\* \

DELAWARE BASIN

Figure 2. Series of schematic cross sections across the Southern Oklahoma Aulacogen (left; from Hoffman, et al. , 1974, utilizing data by Ham, 1969) and the Tabosa- Delaware basin region (right; Horak, 1975). Depiction of the early “graben stage” (late Proterozoic-Middle Cambrian) of aulacogen development is omitted from the succession of cross sections of Hoffman and others; Horak ’s cross sections are shortened slightly.

development (not illustrated in Fig. 2), which in the southern Oklahoma Aulacogen is characterized by extrusive and intrusive igneous activity and fault control of the sedimentary succession, is not documented in the Tabosa-Delaware region. Nonetheless, the similarities illustrated in Fig. 2 are impressive.

TABOSA-DELAWARE BASIN AS AN AULACOGEN

19

I

Figure 3. Schematic view of (A) possible 3-pronged supercontinent breakup pattern in region of the Tabosa and Delaware basins (late Precambrian - early Paleozoic) and (B) Paleozoic margin configuration and crustal-type distribution after breakup (approximately early-middle Paleozoic) but prior to “compressional phase” (late Paleozoic).

20

THE TEXAS JOURNAL OF SCIENCE

Our concept of probable plate-boundary configurations at the time of Pangaea I breakup (Cebull, et al, 1974; Shurbet and Cebull, 1975) is depicted in Fig. 3. It differs from that of Walper (1977) in a way that necessitates oceanic crust in the area southeast of the Marathon Mountains, unless continent-continent col¬ lision occurred as the hypothetical Proto-Gulf was closed. In the event of late Paleozoic -early Mesozoic continent-continent collision, present crustal structure southeast of the Marathon Mountains could be essentially continental. However, recent studies by Pinkerton (1978) suggest that crustal structure in this region is similar to the “filled -ocean” type that characterizes much of the region south of Ouachita system in Texas. Based on Rayleigh Wave dispersion, he derived amodel that shows a crustal thickness of 29.6 km. This thickness comprises 19.6 km of Paleozoic through Cenozoic sedimentary (or metasedimentary) rocks that rest on 10 km of “basaltic” material.

If supercontinent breakup in the Tab osa -Delaware basin region began by mantle upwelling and the consequent development of a 3-arm fracture pattern (Fig. 3A), the position of the active arms may be suggested by the general trend of the Marathon Mountains and the offset between the Marathon Mountains and the buried Ouachita system to the southeast (Fig. 3B). The latter has been interpreted as a possible transform offset (Cebull, et al, 1974, 1976; King, 1975; Thomas, 1976, 1977). In the 3-arm scheme, the Tab osa -Delaware basin represents only the failed arm, an arm that projects into the continental craton from the ancient ocean-continent boundary. Our assessment of the approximate former position of the boundary (for example, Cebull, et al., 1974) is supported generally by more recent studies of a somewhat different type (Thomas, 1977). The orientation of the failed arm may be similar to that of a failed arm proposed by Garrison and Ramirez-Ramirez (1978) for the region of the Llano Uplift of Texas.

If the Tabosa-Delaware Basin is a product ofaulacogen development, a premise yet to be proven, it is tempting to suggest that the other 2 arms of the tripartite are responsible, at least indirectly, for the ultimate location and orientation of the Marfa basin, which fronts the Marathon Mountains, and the Val Verde basin, which approximately parallels the Marathon-Ouachita offset. In any case, tectonic speculations concerning this interesting and economically important region must take fully into account present distributions of apparent crustal types. We believe our suggestions here, utilizing a “classical” but no doubt greatly simplified 3 -arm breakup pattern, satisfies this distribution. Furthermore, our view inextricably ties the development of the Tabosa and Delaware basins, as well as some other Paleozoic basins, to that of the Gulf (or Proto-Gulf) of Mexico.

LITERATURE CITED

Cebull, S. E., G. R. Keller, D. H. Shurbet, and L. R. Russell, 1974-Transform faults as ex¬ planation for offsets in southern Appalachian-Ouachita tectonic belt (abst.). Geol. Soc.

Am. Abst. with Prog., 6:341.

TABOSA-DELAWARE BASIN AS AN AULACOGEN

21

- , - , - , and - , 1976-Possible role of transform faults in

the development of apparent offsets in the Ouachita-southern Applachian tectonic belt. J. Geol., 84:107.

Garrison, J. R., Jr., and Ramirez-Ramirez , 1978-The Llano Uplift, Central Texas. Evidence for a Precambrian triple spreading system (abst.). Geol Soc. Am. Abst. with Prog., 10:106.

Ham, W. E., 1969 -Regional geology of the Arbuckle Mountains, Oklahoma. Oklahoma Geol. Survey Guidebook 17, 52 p.

Hoffman, P., J. F. Dewey, and K. Burke, 1974-Aulacogens and their genetic relation to geosynclines, with a Proterozoic example from Great Slave Lake, Canada. In R. H. Dott and R. H. Shaver (Eds.), Modem and Ancient Geosynclinal Sedimentation. S.E.P.M. Spec. Pub. No. 19, p. 38-55.

Horak, R. L., 1975 -Tectonic relationship of the Permian Basin to the Basin and Range Province. Exploration from the Mountains to the Basin. El Paso Geol. Soc. Guidebook, p. 61-94.

Keller, G. R., and S. E. Cebull, 1973-Plate tectonics and the Ouachita system in Texas, Oklahoma, and Arkansas. Geol. Soc. Am. Bull., 84:1659.

King, P. B., 1975 -Ancient southern margin of North America. Geol., 3:732.

Pinkerton, R. P., 197 8 -Rayleigh wave model of crustal structure of northeastern Mexico. M.S. Thesis, Texas Tech Univ., 52 p.

Shurbet, D. H., and S. E. Cebull, 1975 -The age of the crust beneath the Gulf of Mexico. Tectono physics, 28 :T25 .

Thomas, W. A., 197 6 -Evolution of Ouachita-Appalachian continental margin. J. Geol., 84:323.

- - , 1977, Evolution of Appalachian-Ouachita salients and recesses from reentrants

and promontories in the continental margin. Am. J. Sci., 277:1233.

Valentine, J. W., and E. M. Moores, 1972-Global tectonics and the fossil record. J. Geol., 80:167.

Walper, J. L., 1977 -Paleozoic tectonics of the southern margin of North America. Trans. Gulf Coast Assn. , Geol. Socs. Mtg., Austin, p. 230-241.

Wilhelm, O., and Ewing, M., 1972, Geology and history of the gulf of Mexico. Geol Soc. Am. Bull, 83:575.

WOODY VEGETATION OF UPLAND PLANT COMMUNITIES IN THE SOUTHERN EDWARDS PLATEAU

by O. W. AUKEN, A. L. FORD, A. STEIN1 , and A. G. STEIN1

Division of Allied Health and Life Sciences The University of Texas at San Antonio San Antonio 78285

Reviewed by: Dr. K. L. Carvell, Coll, of Agriculture & Forestry, W.V. Univ., Morgantown 26506 ABSTRACT

The woody vegetation of the cedar brakes of the southeastern Edwards Plateau, Texas was examined using the point-centered-quarter method to determine major community relationships. Density, dominance, frequency, and importance values of trees and shrubs were determined. Two geologically different areas were studied including outcroppings of both the Edwards and Glen Rose Limestone Formations. Similarity indices based on major community parameters were calculated and suggest the 2 upland areas are very much alike in regard to community structure and composition. In all, 24 woody species were identified; 29% were found exclusively on the Edwards Formation, 42% were common to both areas, and 29% were exclusively on the Glen Rose Limestone. The dominants on both formations, based on average importance values, were Juniperus ashei (52%), Quercus fusiformis (15%), and Diospyros texana (11%). The number of species, total density, as well as total dominance were not statistically different on the 2 formations.

INTRODUCTION

The Edwards Plateau area of west-central Texas covers about 1 X 107 ha of rough, well-drained land. The southern and eastern boundaries are marked by an area of faulting known as the Balconies Escarpment. On the north, the Edwards Plateau blends gradually into both the Rolling Plains and the High Plains and on the west, into the Trans-Pecos region (Gould, 1969).

A rainfall gradient exists across the Plateau ranging from approximately 38 cm/yr in the west to about 84 cm/yr in the east. Thornthwaite (1948) classified the western half as semiarid and the eastern half as dry subhumid; however, he considered the entire area mesothermal. Mean annual temperature for the entire

Present Address: Department of Education, The University of Texas at Austin 78712. Accepted for publication: June 22, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

24

THE TEXAS JOURNAL OF SCIENCE

Edwards Plateau is approximately 20 C (Arbingast, et al, 1976). Soils are usually shallow and throughout most of the region are underlain by limestone or caliche (Gould, 1969).

The cedar brakes region of the Edwards Plateau has been referenced by many early travelers in Texas, however, ecologically it has been studied very little. Roemer (1849), while traveling from New Braunfels to San Antonio through the blackland prairies along the eastern edge of the Edwards Plateau, wrote of the great fertility of the prairies as well as the cedar covered hillsides. DeCordova (1858) also wrote about beautiful prairies occurring between Austin and San Antonio. To the west of these prairies DeCordova (1858) described a chain of hills covered with a dense growth of mountain cedar and liveoak.

More recently, several investigators have studied the flora of the Edwards Hill Country. Although their concern was primarily with the geology of the region, Hill and Vaughan (1898) referred to the oaks, juniper, agarita, and yucca that occurred on the limestone slopes. Bray (1904a,b; 1906) discussed many of the timber species both in the bottomlands and in the upland areas. He (Bray 1904a, b; 1906) described dense thickets of both shin oak and Texas red oak in the upland sites, and made special mention of cedar breaks which are thick growths of juniper on the crumbly limestone hillsides. Metz (1934) discussed some of the early studies of Texas vegetation and included keys for species in Bexar County. A correlation between geologic strata and vegetational dominants was described by Tharp (1939). He (Tharp 1939) noted Texas red oak on the Walnut Springs Limestone; open stands of cedar on the Glen Rose; live oak and cedar on the Edwards; pecan, bur oak, elm, hackberry, and live oak along stream courses; and mesquite on many high stream terraces. Gould (1969) reported typically a grass understory with the overstory made up of live oak, shinnery oaks, junipers, and mesquites. Correll and Johnston (1970) refer to dense growths of Juniperus ashei, scrub oaks, and mesquite occurring on areas of the Edwards Plateau.

Van Auken, et al, (1979) studied the woody plant communities of the Buda Formation in the southeastern part of the Edwards and the streamside vegetation occurring along the intermittent streams in this same area. Van Auken, et al, (1979) found dominants in the upland areas to be Mexican juniper, live oak, and Texas persimmon. Dominants in the creekbottoms included the above 3 species as well as cedar elm and sycamore. Buechner (1944) studied the vegetation of the midsection of the Edwards Plateau (Kerr Co.) which included some of the thick cedar areas or the cedar brakes.

While all of the above works are very useful in describing the flora of the Edwards Plateau, only the work of Buechner (1944) and Van Auken, et al, (1979) attempt a quantitative ecological description. The purpose of this paper is to further quantitatively describe the upland woody plant communities occurring in the Texas Hill Country or the Cedar Brakes Region of the Edwards Plateau. Phytosociological comparisons are made of the major woody plant species existing on 2 geologically distinct strata that occur in the southern part of the Edwards Plateau.

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

25

MATERIALS AND METHODS

This study was conducted in the extreme southeastern portion of the Edwards Plateau very near the Balconies Escarpment. The woody vegetation of 18 forest stands in northwest Bexar County, northeast Medina County, and southeast Bandera County, Texas was studied. Fig. 1 shows the locations of the stands sampled with the inset showing their location in relation to the Edwards Plateau and the State of Texas. After a general visual survey, 12 sites located on the Edwards Formation and 6 on the Glen Rose Formation were selected.

Figure 1. Location of study sites in Bexar, Bandera, and Medina Counties, Texas. Stands 1-12 are on the Edwards Limestone and Stands 13-18 are on the Glen Rose Limestone. Inset shows the State of Texas, the Edwards Plateau (stipled) and Bexar County (black).

Various geological maps as well as site inspection were used to determine if the stands were located on the Edwards Formation or on the Glen Rose For¬ mation (Sellards, et al., 1932; Sellards and Baker, 1934; Barnes, 1974).

The Glen Rose typically consists of thin to medium bedded hard continuous limestone and dolomite strata alternating with marl or marly limestone. Hillsides in this formation weather into a terraced or staircase topography. The Edwards consists of layers of hard white rudistid limestone, usually crystalline, medium to massive bedded and containing considerable amounts of calcareous shell de¬ tritus (Sellards, et al, 1932).

All sites were mature stands without visible signs of fire or cutting, however, some of the stands were recently grazed by cattle and all stands had a prehistory

26

THE TEXAS JOURNAL OF SCIENCE

of grazing. Stands numbered 1-12 were on the Edwards Formation and Stands 13-18 were on the Glen Rose Formation (Fig. 1).

The point-centered-quarter method (Cottam and Curtis, 1956) was used to collect quantitative phytosociological data. Transect starting points were ap¬ proximately 50 m inside the stands. A chain was laid out along a predetermined line through the stand and points were sampled 10 m apart along the chain. At each point the transect line was divided into 4 equal quarters by placing a rod across the chain perpendicular to it. For each quarter, the plant closest to the center point with a circumference of at least 3.0 cm at a point between 0 and 15 cm above ground level was selected. Since many of the woody plants in the Edwards are shrubby and have multiple stems, making measurements at or near ground level alleviates most problems relating to a choice of stems or to plants being excluded when measurements are made at breast height. The plant was then identified and the circumference and point-to-plant distance were recorded. Twenty-five points were sampled for each transect giving a total of 100 measure¬ ments/transect. In this manner, 1800 plants were measured and identified during the course of this study. This data allowed the calculation of total density and total dominance as well as the density, relative density, average dominance, dominance, relative dominance, frequency, relative frequency, and importance for each species. The above values were calculated for each stand and these values were averaged for each species and a standard deviation was determined (Steel and Torrie, 1960). Only the average values for importance, density, average dominance, and frequency are reported in this paper. Analyses of variance and Student’s t~ tests were also derived from Steel and Torrie (1960).

Plant identification and nomenclature follow from Correll and Johnston (1970). Species curves and density stability curves (Cox, 1972) were constructed for each transect but are not reported here. Size class curves (Spring, et al, 1974) were also plotted for the 3 dominant species.

RESULTS

Twenty -four woody species were encountered during this study. Of this total number, 7 species or 29% were found exclusively on the Edwards Formation, including Ulmus crassifolia, Prosopis glandulosa, Dasylirion texanum, Yucca sp., Bumelia celastrina , Celtis lindheimeri , and Condalia hookeri . Another 7 species were encountered only on the Glen Rose Formation including Sophora secundi- flora , Rhus lanceolata, Celtis sp., Celtis reticulata , Ungnadia speciosa, Bumelia lanuginosa , and Prunus serotina. The remaining 10 species or 42% were found to be common to both formations. This group of common species included Juniperus ashei , Quercus fusiformis , and Diospyros t ex ana, the 3 dominants.

The occurrence of each species expressed as a % of the total number of stands on each formation may be found in Table 1 . Only J. ashei was encountered in all stands in both areas. Q. fusiformis and D. t ex ana were present in all Edwards

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

27

stands and in all but one of the Glen Rose stands. Other widely distributed species on the Edwards Limestone were Berberis trifoliata, Rhus virens, Eysen- hardtia texana, and Ulmus crassifolia. On the Glen Rose,/?, virens, B. trifoliata, Quercus texana and S. secundiflora were widely distributed. The remaining species were less frequently encountered with 7 species occurring in only 1 transect.

TABLE 1

Summary of Presence Data for each Species Encountered Expressed as a % of the Total Number of Transects on each Geological Formation

Species	Edwards3	Glen Rose13
Juniperus ashei	100	100
Quercus fusiformis	100	83
Diospyros texana	100	83
Berberis trifoliata	75	50
Rhus virens	67	67
Eysenhardtia texana	42	33
Ulmus crassifolia	42	_ c
Quercus texana	33	50
Bumelia celastrina	25	...
Yucca sp.	25	—
P tele a trifoliata	17	17
Prosopis glandulosa	17	—
Celtis lindheimeri	17	—
Dasylirion texanum	17	...
Cercis canadensis	8	33
Acacia gr egg ii	8	17
Condalia hookeri	8	...
Sophora secundiflora	-	50
Bumelia lanuginosa		17
Celtis reticulata	—	17
Celtis sp.	—	17
Prunus serotina	—	17
Rhus lanceolata	~	17
Ungnadia speciosa	--	17

a Total of 12 transects. bTotal of 6 transects. cNot encountered in this area.

On the Edwards Formation 3 species appeared as dominants and accounted for 76.4% of the average importance value. These species were/, ashei (48.0 ± 8.7), Q. fusiformis (16.9 ± 5.9), and/). texana (11 .5 ± 5.1) (Table 2). All other Edwards species had importance values less than 10%. The same 3 species were dominants on the Glen Rose Formation, accounting for 78.4% of the average importance value. From Table 2, the importance values of these species were:/, ashei (56.3 + 14.0), Q. fusiformis (12.1 ± 10.1), and/), texana (10.0 ± 10.2). The remaining 14 Glen Rose species had importance values less than 10%.

Average of relative density + relative dominance + relative frequency. Expressed as %. Plants/ha.

Cm2 /plant.

Two Glen Rose species were unidentified. They had a total importance value of 0.4 ± 0.9.

28

THE TEXAS JOURNAL OF SCIENCE

TOTALS

Prunus serotina

Bumelia lanuginos

Ungnadia speciosa

Celtis reticulata

Celtis sp.

Rhus lanceolata

Co

©

“a

a*

o

2

n

o

s

a

3

Condalia hookeri

Cercis canadensis

Celtis lindheimeri

A cacia greggii

Bumelia celastrina

Ptelea trifoliata

Yucca sp.

Dasylirion texanui

Eysenhardtia texa

Prosopis glanduloi

Ulmus crassifolia

Quercus texana Berberis trifoliata

Rhus virens

Diospyros texana

Quercus fusiformi

Juniperus ashei

m

n

<0

o

1”

• a

a

©

2

3

s

54

69

(_4

| .

4s

M

.©

©

*-*

i—

to

p

'j

i~*

OS

oo

o

b

to

Vl

b

©

b

in

In

©

Vl

so

©

Qa

©

!

1

!

i

!

i

14-

1+

14-

1+

14-

1+

14-

1+

14-

1+

83 HI

to

©

h*

©

to

t-

to

Vl

OJ

w

-4

VI

oo

a 3

b

bs

b

sO

©

b

Vl

b

©

W

In

so

b

W *Q

o

2.

Vl

83

o

©

.©

©

o

4*

©

H-*

©

to

Vl

44*

©

to

as

£ 3

sO

to

b

bs

©

to

b

bs

bo

Vl

©

b

cb ri

3 »

sO

1+

14-

j

14-

!

14-

!

14-

1

i

14-

j

!

14-

p,

\Q

>-*

i-s

l-s

©

i—

9

©

i—

to

.©

©

w

os

4s

©

4*

u

in

Vi

b

sO

b

©

b

in

bs

b

to

i—*

©

Q

to

w

Os

w

i— '

b-s

w

to

Vl

-4

•4

©

to

Os

■-*

4*

00

Vl

Os

Vl

©

44*

©

Os

to

h-

o

©

Os

a

W 1— 1

00

Vl

Os

j

i

j

;

1

i

14-

1+

14-

1+

14-

1+

14-

1+

14-

w

14-

to

14-

to

in¬

to

i— *

1—*

i— *

i— i

4s

to

co

VI

•4

w

to

-4

**

oo

^ u

Os

4*

Vl

as

w

-4

Vl

-4

as

to

W

-4

to

3

m

to

vs

0 ^O*

i— »

h- *

1— *

to

t— *

1— »

00

Vl

OS

to

Vl

to

Os

Vi

Vl

Os

00

4*

f~*

oo

00

W

-4

00

©

00

4*

os

3

00

4*

1+

14-

in¬

14-

j

14-

;

in-

j

14-

i

1+

j

14-

pa

w

i— *

o

u>

b-L

Vl

*-*

©

50

i— *

H-

h- *

*— *

4*

ns

4*

w

41

i=»

to

w

os

05

to

05

ce

vi

to

VI

vi

Vl

sO

-4

Vl

SO

©

41*

as

00

44.

©

as

SO

w

Vl

OJ

to

W

Vl

SO

Vl

to

cn

Vl

©

to

4*

w

•4

SO

00

-4

)-*

Vl

I-*

Vl

Os

M

O*

to

bo

b

bs

w

b

©

b

©

b

bs

as

I

j

1

•

1

i

14-

\+

14-

1

!

1

£ «

i— »

-4

cn

i— *

oo

-4

VI

©

to

sO

i-»

4^

OJ

*— *

O

©

50 TO

00

to

4

©

sO

00

to

.©

-4

p

4^

©

Qs

Vl

W

P

©

-4

sO

b

bo

b

©

to

b

to

b

w

VI

D

as

w

o

w

-4

i—

Vl

1— *

to

oo

Os

q3

p

o

Os

©

oo

©

SO

©

N-

©

oo

00

►—

-4

as

bs

©

bs

w

b

to

Vl

SO

b

oo

b

to

bo

Vl

to

b

§ 2

!

1+

14-

1+

j

14-

j

14-

1

14-

!

14-

S

j

14-

1+

14-

1+

3 3

*-r] O

_*

-4

u»

o

00

i— *

sO

i— *

co

to

-4

©

50

Vl

to

*-*

-4

P

©

i—

©

;4

Vl

©

w

©

Vl

re

b

Vl

b

'so

b

©

b

to

*-*

to

b

In

p

©

O

©

b

©

b

©

b

©

b

©

b

©

to

w

b

oo

M

to

>—

*-*

*—

>-*

to

w

OJ

Os

as

to

©

-4

4*

Os

a s

to

i

j

I

1

j

14-

1+

14-

i+

14-

1+

14-

1+

14-

o

©

.©

©

Q. ^

©

b

©

b

©

b

to

b

« 3

43

to

1-1

w

Vl

Os

sO

©

w

N-*

Vl

to

p

©

9

p

©

c

ft)

sg

b

©

b

©

b

©

b

bo

to

►—

*-*

!-*

to

K-

4s

to

OJ

i-*

to

so

Os

Vl

►-*

©

to

CD ^

©

1+

14-

i

14-

j

14-

1

14-

!

1

14-

1

]

14-

4*

©

p

©

P

©

P

©

P

©

^ i

O

©

b

©

b

to

©

b

©

b

©

b

u>

to

b

.0)

to

Vl

OJ

w

Os

to

w

-4

W

©

as

SO

Comparison of the Importance, Density, Average Dominance, and Frequency of all Species Found on Both the Edwards Formation and the Glen Rose Formation

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

29

On the Edwards Limestone,/, ashei had the highest density with an average value of 1606 ± 1282 plants/ha. The other 2 dominants had much lower den¬ sities— Q. fusiformis (370 ± 247) and/). texana (310 ± 273). Other Edwards species with densities exceeding 100 plants/ha were R. virens ( 251 ± 322) and B. trifoliata (126 ± 174). The pattern on the Glen Rose Formation was somewhat similar. /. ashei again had the highest density with a value of 1556 ± 1086 plants/ha. D. texana ranked 3rd in importance but here ranked 2nd in density with 418 ± 544 plants/ha. Q. fusiformis ranked 2nd in importance but ranked 3rd in density with 224 ± 120, plants/ha. Two other species on the Glen Rose had densities greater than 100 plants/ha. They were S. secundiflora (214 ± 349), a species not encountered on the Edwards, and R. virens (160 ± 88). The total density on the Edwards was 2856 plants/ha and it was 2841 plants/ha on the Glen Rose. When total density values were compared using a Student’s f-test, they were not found to be statistically different at the 95% level.

Based on average dominance values from Table 2, the largest species found on the Edwards Limestone was Q. texana with an average dominance of 995.0 ± 1746.2 cm 2 /plant. The next largest species was Q. fusiformis (355.6 ± 503.3) followed by/, ashei (156.6 ± 101.5)./). texana , while ranking 3rd in importance, ranked 10th in average dominance with a value of 21.1 ± 15.4 cm2 /plant. On the Glen Rose Limestone Q. texana was again the largest species (658.2 ± 730.2 cm2/ plant). However, the next largest species was/, ashei (366.4 ± 305.5) followed by Q. fusiformis ( 87.2 ± 70.7)./). texana ranked 7th with 21.5 ± 23.2 cm2 /plant. Q. texana and Q. fusiformis were considerably larger on the Edwards than on the Glen Rose while /. ashei was much smaller on the Edwards and D. texana was the same size on both formations; however, these differences were not statistically significant.

Species with the highest frequency in the Edwards stands were /. ashei (0.86 ± 0.12), D. texana (0.37 ± 0.15), and Q. fusiformis (0.34 ± 0.12). The pat¬ tern was identical on the Glen Rose with only the numbers changing slightly (Table 2). In both areas, all other species had frequency values of less than 0.21. The mean number of species on the Edwards Formation was 7.0 ± 2.2 and 7.2 ± 2.0 on the Glen Rose (Table 3). Total dominance in m2/ha was 36.59 ± 29.06 for the Edwards vs. 48.26 ± 21.34 for the Glen Rose. None of the above differ¬ ences were statistically significant.

A size class distribution was prepared for each of the 3 dominant species. Table 4 shows the % of the total number of trees of each species in the 5 cm di¬ ameter size classes for /. ashei, Q. fusiformis, and D. texana. A large number of seedlings (1-5 cm diameter) of each species are present on both formations. It is apparent that these 3 species are successful, reproducing members of both com¬ munities.

Coefficients of similarity (Greig-Smith, 1964) were also calculated for the 2 communities based on data from Table 2 (see Table 5). The coefficient for density

30

THE TEXAS JOURNAL OF SCIENCE

TABLE 3

Summary of the Number of Species, Total Density, and Total Dominance for each Transect. Average Values for each Geological Formation are Shown along with 1 Standard Deviation (s.d.).

Formation	Transect Number	Number of Species	Total Densitya	Total Dominance^
Edwards	1	5	3812	34.05
	2	9	812	8.77
	3	10	670	24.89
	4	9	1104	12.98
	5	9	2066	20.59
	6	5	495	5.55
	7	5	5021	41.71
	8	6	5316	107.66
	9	7	3611	30.84
	10	8	3708	29.27
	11	8	4626	51.25
	12	3	3036	71.49
		7.0 ± 2.2C	2856 ± 1767	36.59 ±29.06
Glen Rose	13	4	819	51.34
	14	7	4411	18.15
	15	10	3575	79.89
	16	8	2901	32.90
	17	8	2387	59.41
	18	6	2955	47.86
		7.2 ± 2.0C	2841 ± 1210	48.26 ±21.34

a Plants/ha. bM2/ha. cx ± 1 s.d.

was highest (0.859) followed by importance (0.828) and frequency (0.821) with the coefficient for average dominance being the lower (0.630).

DISCUSSION

Bray (1904a) described the Edwards Plateau as a common meeting ground for species from the Atlantic forest belt, the southern Rocky Mountains, and the northern Mexican Highlands. Blair (1 950) also considered the Edwards as a special area and treated it as a separate biotic province containing a number of endemic species. The eastern species that occur in the Edwards are primarily limited to the rich bottomlands of the rivers dissecting the plateau. The western and south¬ western species occur mainly in the dry upland areas.

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

31

TABLE 4

Size Class Distribution for Juniperus ashei, Quercus fusiformis, and Diospyros texana in 5 cm (Diameter) Size Classes Expressed as a % of the Total Number of Trees of each Species for both Formations

Size Class	Juniperus ashei Edwards Glen Rose	Quercus fusiformis Edwards Glen Rose	Diospyros texana Edwards Glen Rose
1-5	50.3	36.5	43.3	23.1	77.5	83.7
6-10	21.2	24.1	18.7	23.1	18.1	10.5
11-15	9.5	12.7	12.0	29.2	1.4	2.3
16-20	7.8	4.1	11.3	12.3	2.2	-
21-25	4.5	8.6	6.0	7.7	0.7	3.5
26-30	3.0	4.1	2.0	1.5	-	-
31-35	0.8	3.2	3.3	1.5	~	-
36-40	1.6	1.6	-	-	-	-
41-45	0.5	1.9	1.3	-	-	-
46-50	0.5	0.6	1.3	-	-	-
51-55	0.2	0.6	-	-	-	-
56-60	-	0.3	-	-	-	-
61-65	-	0.3	-	-	-	-
66-70	-	0.3	-	-	-	-
70+	-	1.0	0.7	1.5	--	-
TOTAL	99.9	99.9	99.9	99.9	99.9	100.0

TABLE 5

Coefficients of Similarity Calculated for the Edwards and Glen Rose Communities

Coefficient of

Parameter Similarity

Density 0.859

Importance 0.828

Frequency 0.821

Average Dominance 0.630

The vegetation and the plant associations of the central Texas Hill Country have been only meagerly described. They were divided by Hill and Vaughan (1 898) into 3 simple topographical elements: 1) the flat-topped summits of the plateau, 2) the breaks and slopes of its borders and canyons, and 3) the streamways or rivers and their tributaries. Buechner (1944) partitioned the central part of the

32

THE TEXAS JOURNAL OF SCIENCE

Edwards Plateau into 5 separate areas. He (Buechner 1944) considered the flat- topped summits to be divisible into liveoak-shinoak divides and blackjack divides. Also, Buechner (1944) considered the erosional areas and the cedar breaks as separate. The present study deals with the breaks, slopes or erosional areas as they occur in the extreme southeastern part of the Edwards Plateau.

According to Bray (1904b) the hill and bluff timber of the Edwards includes cedar, live oak, cedar elm, hackberry, mountain oak, and shin oak as well as other species. Several species on the Edwards are limited in distribution to central or south -central Texas and northern Mexico. D. texana , J. ashei, C. lindheimeri, and U. crassifolia are examples (Brockman, 1968; Correll and Johnston, 1970). All of the above mentioned species, with the exception of shin oak, were en¬ countered during the course of this investigation.

Similarities may be noted between the vegetation of the upland Edwards Plateau and the piny on -juniper pigmy forests and sub -montane shrub associations of west Texas and western North America. Growth patterns are similar in that closed canopies are rarely observed and trees seldom exceed 9 m in height in both areas. Species types overlap in both areas as well. Both junipers and oaks are common on the Edwards Plateau. In the pigmy forests of the west, junipers are common and oaks are important in some areas (Woodin and Lindsey, 1954). Rainfall patterns overlap being 38-76 cm/yr on the Edwards Plateau (Gould, 1969), 25-38 cm/yr in the pinyon -juniper forests of Utah and northern Arizona (Woodbury, 1947), and 33-43 cm/yr in the same type forests in western Texas, New Mexico, and Colorado (Woodin and Lindsey, 1954). Additional overlap is indicated when one considers the higher precipitation levels required to support similar plant communities at lower altitudes and higher prolonged temperatures.

The current study indicates that no real differences exist between the plant communities of the Edwards Limestone and those of the Glen Rose Formation in the southeastern portion of the Edwards Plateau. When comparing density, dominance, frequency, and importance values of plants on both formations, no statistically significant differences are noted. Although several species were en¬ countered exclusively on one formation or the other, there was insufficient data to say that these same species are limited to the formation on which they were found. It is important to note that the 5 most important species on the Edwards Limestone, i.e.,/. ashei, Q. fusiformis, D. texana, R. virens, and Q. texana , are also the 5 most important species on the Glen Rose Limestone (Table 2). This fact serves to point out the vegetational similarity of the 2 geological areas. In the area of the Edwards Plateau examined, this report did not indicate the kinds of differences as noted by Tharp (1939) between the plant communities of the 2 geological formations.

In a previous study of the plant communities of the Buda formation, another limestone formation occurring in the southern Edwards Plateau, we found J. ashei,

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

33

Q. fusiformis and D. texana as dominants accounting for 80% of the total impor¬ tance of all species present (Van Auken, et al., 1979). In the present study they accounted for 76-78% of the total importance. Also, an average of 8.4 ± 2.7 species were found/stand, which is not significantly different from the number found on the Edwards or the Glen Rose. Total density values were 3605 ± 1448 on the Buda compared with 2856 ± 1767 on the Edwards and 2841 ± 1210 on the Glen Rose which are not significantly different.

Comparisons of the density, average dominance, dominance and frequency for the 3 most important species (J. ashei, Q. fusiformis, and D. texana) on the Edwards, Glen Rose and the Buda show no significant differences. The above data suggest that the soils derived from these limestones are very similar and that the plant communities are very similar because of the above. Data taken from the various county soil surveys (Taylor, et al, 1966; Hensell, et al, 1977; and Dittmar, et al, 1977) indicate that the soils from the above stands are in either the Tarrant-Brackett Association (shallow soils underlayed by limestone) or the Crawford -Bexar Association (moderately deep, stoney soils also underlaid by limestone). Although 2 of the stands were on the Crawford -Bexar soils, it should be noted that these soils were very shallow soils much like those of the Tarrant- Brackett Association and also very calcareous and slightly basic.

Buechner (1944) studied the cedar brakes in Kerr Co., Texas which is 15-20 mi deeper into the Edwards Plateau than the present study. Kerr Co. also includes a considerable section of the non -eroded portion of the Plateau and is at a higher altitude. It is difficult to make direct comparisons to Buechner’s (1944) work because of the different methods used. Buechner (1944) does state, however, that cedar comprised 80% or more of the arborescent vegetation.

Coefficients of similarity calculated for the Edwards and Glen Rose stands are also quite high (Table 5). Again, this suggests the close relationship of the upland plant communities in the southern Edwards Plateau region. If differences did exist in the plant communities occurring on these geological formations in the past, they have been obliterated possibly due to differential cutting, clearing, grazing, browsing, or fire.

LITERATURE CITED

Arbingast, S.A., L. G. Kennamer, R. H. Ryan, J. R. Buchanan, W. L. Hezlep, L. T. Ellis, T. G. Jordan, C. T. Granger, and C. P. Zlatkovich, 197 6 -Atlas of Texas. Bureau of Bus¬ iness Research, Univ. of Tex., Austin.

Barnes, V. E., 197 4 -Geological Atlas of Texas, San Antonio Sheet. Bureau of Economic Geology, Univ. of Tex., Austin.

Blair, W. F., 1950 -The biotic provinces of Texas. Tex. J. Sci., 2:93.

Bray, W. L., 19Q4a-Forest resources of Texas. USD A Bureau of Forestry Bull No. 47.

34

THE TEXAS JOURNAL OF SCIENCE

- , 1904b -The timber of the Edwards Plateau of Texas; its relation to climate, water

supply, and soil. USDA Bureau of Forestry Bull. No. 49.

- , 1906 -Distribution and adaptation of the vegetation of Texas. Univ. of Tex.

Bull. No. 82.

Brockman, C. F., 1968 -Trees of North America. Golden Press, Racine, WI.

Buechner, H. K., 1944-The range vegetation of Kerr County, Texas, in relation to livestock and white-tailed deer. Am. Midland Nat., 31:697.

Correll, D. S., and M. C. Johnston, 1910-Manual of the Vascular Plants of Texas. Texas Re¬ search Foundation, Renner, TX.

Cottam, G., and J. T. Curtis, 1956— The use of distance measures in phytosociological sampling. Ecol., 37:451.

Cox, G. W., 1911-Laboratory Manual of General Ecology . W . C. Brown and Co., Dubuque, IA.

DeCordova, J., 1858 -Texas: Her Resources and Her Public Men. J. B. Lippincott and Com¬ pany, Philadelphia, PA.

Dittmar, G. W., M. L. Deike, and D. L. Richmond, 1911 -Soil Survey of Medina County, Texas. USDA Soil Conservation Service.

Gould, F. W., 1969 -Texas plants-A checklist and ecological summary. Texas Agr. Exp. Sta. Bull. M P-5 85.

Greig -Smith, P., 1964 -Quantitative Plant Ecology, 2nd Ed. Butterworth, London.

Hensell, J. L., G. W. Dittmar, and F. Taylor, 1911-Soil Survey of Bandera County, Texas. USDA Soil Conservation Service.

Hill, R. T., and T. W. Vaughan, 1898 -Geology of the Edwards Plateau and Rio Grande Plain adjacent to Austin and San Antonio, Texas, with reference to the occurrence of underground waters. U. S. Geolog. Survey Ann. Report, 18:193.

Metz, M. C., 1934 -A flora of Bexar County, Texas. Ph.D. Dissertation, Catholic Univ. of Am., Washington, DC.

Roemer, F., 1849 -Texas with Particular Reference to German Immigration and the Physical Appearance of the Country. Standard Printing Co., San Antonio, TX (Original, published in Germany, Trans, by Oswald Mueller. 2nd pub. 1935).

Sellards, E. H., W. S. Adkins, and F. B. Plummer, 1932- The Geology of Texas. Vol. I: Strat¬ igraphy. Bureau of Economic Geology (Bull. 3232), Univ. of Tex., Austin.

- -, and R. L. Baker, 1934-The Geology of Texas. Vol. 2: Structural and Economic

Geology. Bureau of Economic Geology (Bull. 3401), Univ. of Tex., Austin.

Spring, P. E., M. L. Brewer, J. R. Brown, and M. E. Fanning, 1974-Population ecology of loblolly pin e Pinus taeda in an old field community. Oikos, 25:1.

Steel, R. G. D., and J. H. Torrie, 1960 -Principals and Procedures of Statistics. McGraw-Hill, New York, NY.

Taylor, F. B., R. B. Hailey, and D. L. Richmond, 1966-Soil Survey of Bexar County, Texas. USDA Soil Conservation Service.

WOODY VEGETATION-SOUTHERN EDWARDS PLATEAU

35

Tharp, B. C., 1939 -The Vegetation of Texas. The Anson Jones Press, Houston, TX.

Thorn thwaite, C. W., 1948 -An approach toward a rational classification of climate. Geogr. Rev., 38:55.

Van Auken, O. W., A. L. Ford, and A. Stein, 1979 -A comparison of some woody upland and riparian plant communities of the southern Edwards Plateau. Southw. Nat., 24:165.

Woodbury, A. M., 1947 -Distribution of pigmy conifers in Utah and northeastern Arizona. Ecol., 28:113.

Woodin, H. E., and A. A. Lindsey, 1954-Juniper-Pinyon east of the Continental Divide, as analyzed by the line-strip method. Ecol., 35:473.

THE UPPER INCISORS OF THE GIANT HORSE, ASINUS GIGANTEUS

by WALTER W.DALQUEST

Department of Biology Midwestern State University Wichita Falls 76308

Reviewed by: Dr. W. S. Strain, Prof. Emeritus, Dept, of Geol, Univ. of Texas, El Paso 79902 INTRODUCTION

In 1901 J. W. Gidley described a new species of horse, Asinus giganteus, based on a single tooth of relatively enormous size found in southwest Texas (for use of Asinus rather than Equus for most American Pleistocene horses, see Dalquest, 1979). The tooth (Collection of American Museum of Natural History) had been referred to Asinus crenidens by Cope (1 899). When sectioned (by Gidley) 35 mm below the occlusal surface the tooth revealed small, intricately folded lakes, a short, broad protocone, and small pli caballin.The large size and distinctive features of the enamel pattern convinced subsequent writers (e.g. Savage, 1951) that the species was valid, even though based on a single tooth from an indefinite type locality. Some confusion with Asinus pacificus (Leidy), another large extinct horse, existed, but Lundelius (1972) noted that A. pacificus had teeth that were smaller than those of the holotype of A. giganteus , and had long, slender rather than short, broad protocones.

Although A. giganteus has been known for more than 75 yr, very few specimens have been referred to the species. These include an upper tooth from the Holloman local fauna of Oklahoma (Hay and Cook, 1930; Dalquest, 1977) and 2 upper teeth from the Gilliland local fauna of Knox County, Texas (Hibbard and Dalquest, 1966). Both of these faunas are of earliest Pleistocene age. A lower tooth from an early Pleistocene deposit in Meade County, Kansas, was questionably referred to A. giganteus by Hibbard and Dalquest (1966). I am aware of no other fossils referred to the giant horse.

DISCUSSION

Of interest then is the discovery of horse premaxillaries of enormous size from the Seymour Formation of Knox County, Texas. The specimen was found on the Bruce Burnett Ranch, 100 m west of State Highway 267 and a km or less southwest of the bed of Pearlette ash at the type locality of the Seymour Formation

Accepted for publication: February 19, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

38

THE TEXAS JOURNAL OF SCIENCE

(Hibbard and Dalquest, 1966: 5, plate 1). The fossil was embedded in a half¬ meter thick layer of brownish, very hard, caliche. The incisors were perpendicular to the bedding plane, and pressure from the growth of amesquite tree root beneath the caliche bed caused the caliche to crack along the anterior faces of the incisors, and “pop up.” Part °f the enamel of the front surfaces of some of the teeth adhered to the counterpart and were destroyed by weathering. Only the bone of the terminal 75 mm of the muzzle is preserved, and this is so decayed that it would scarcely be recognizable were it not for its association with the incisor teeth. The caliche bed was excavated for 4 m around the site, but nothing else of interest was recovered.

The specimen (MWSU 11785) includes the incisor teeth held in place by the caliche matrix. Teeth other than left 1 3 are in their natural positions. Left 1 3 is displaced downward and outward 2 or 3 mm. A bit of the enamel of the anterior face of left 13, most of the anterior faces of both I2’s, and the anterior face of left 17, are lost. The horse was quite young when it died, with the Ii’s just beginning to wear. Infundibulata were large and prominent but the hard matrix has not been cleaned from the pits.

In the measurements of individual teeth that follow, the first is the greatest dimension, transverse to the longitudinal axis of the skull in 17 but almost parallel to it in 13. The second measurement is taken at right angles to the first. Measure¬ ments were made at the occlusal surface.

Breadth across Ii’s is 1 19.3 mm but, allowing for the displacement of left 13, the true distance was approximately 116 mm. 17 measures 23.3 x 13.2 mm; 12, 23.0 x about 12.8 mm; 12, 26.0 x 12.4 mm.

The medial edges of 17 ’s are flattened and the 2 teeth are appressed (Fig. 1A). I2’s are broadly oval. I5’s are almost unworn and the large, oval anterior cusp, smaller oval middle cusp, and tiny, rounded posterior cusp, are apparent. These resemble the unworn upper IJ’s of a modern domestic horse.

The shape of the incisors may, to a degree, be a function of age (Olsen, 1964). Striking is the enormous breadth of the incisor row, and the shape of the row formed by the teeth. The complete upper incisor row is rarely preserved in Pleistocene fossil horses, and the lower incisor row only slightly more often (the breadth of the lower incisor row is usually slightly less than that of the upper row). Exami¬ nation of numerous modern and fossil horse skulls revealed none with the breadth of the incisor row as great as that of the fossil. Horse incisors are strongly wedge- shaped. Maximum breadth of incisors and incisor rows is attained at an early stage of tooth wear, and thereafter wear shortens the transverse breadth of the incisor row. The fossil probably represents maximum breadth for this individual, and at a greater age the incisor row would be somewhat shorter.

In the table that follows, the stage of wear of specimens in the Midwestern State University Collection is approximately that of the fossil, and when more than 1 specimen was available only the maximum breadth obtained is cited.

As expected, only A. pacificus approaches A. giganteus in size, and the incisor row of even this very large species is exceeded by that of A. giganteus by 24 mm

UPPER INCISORS OF ASINUS GIGANTEUS

39

Figure 1. A. Upper incisor row of Asinus giganteus Gidley. B. Of a large male African lowland zebra, Equus burchellii (Gray). Maximum breadth across tooth row of zebra is 67.5 mm; specimens to scale.

TABLE 1

Breadths of Upper Incisor Rows of Some Large Recent and Pleistocene Horses

Species	Incisor Breadth	Where Found and Authority
Asinus giganteus (Gidley)	116.0	Texas. MWSU Coll.
Asinus paci ficus (Leidy)	92.0	Aguascalientes, Mexico. MWSU Coll.
Asinus pacificus (Leidy)	*90.0	Mexico, Mexico (Hibbard, 1936).
Asinus scotti (Gidley)	*78.0	Texas (Johnson, 1937).
Asinus niobrarensis (Hay)	78.0	Nebraska (holotype, Hay, 1913).
Equus simplicidens Cope	*76.0	Idaho (Gazin, 1936).
Equus caballus Linnaeus	76.0	Texas. MWSU Coll.
Asinus lambei (Hay)	74.0	Yukon Territory (Harington and Clulow, 1973).
Asinus excelsus (Leidy)	73.8	Aguascalientes, Mexico. MWSU Coll.
Asinus calobatus (Troxell)	73.6	Aguascalientes, Mexico. MWSU Coll.
Equus burchelli (Gray)	67.9	Mozambique, Africa. MWSU Coll.

*Breadth of muzzle at posterior alveolar borders of I3’s.

(almost 1 in). Thus the largest known Pleistocene horse other than A giganteus possessed an incisor row breadth only 79% as great.

40

THE TEXAS JOURNAL OF SCIENCE

In every Pleistocene or Recent horse or zebra skull examined, the upper incisors form a smoothly rounded arc. In the Seymour horse the upper incisors form a trapezoid (Fig. 1 A). The 4 anterior incisors make an almost straight line, with the I5’s turned sharply backward from the I2’s.

CONCLUSIONS

Because the specimen from the Seymour formation is so very large and comes from a deposit where teeth referred to A. giganteus are known to occur, it is referred to that species. The specimen suggests that the head of A. giganteus might have appeared, in life, quite different from the head of a modern horse. If the muzzle was stout, as the breadth of the incisor row suggests, the head might have been short and bulldog-like in appearance.

LITERATURE CITED

Cope, E. D., 1899 -Vertebrate remains from the Port Kennedy bone deposit./. Acad. Nat. Sci., Philadelphia, PA, 11:193.

Dalquest, W. W., 1977-Mammals of the Holloman local fauna, Pleistocene of Oklahoma. Southwest Nat. , 22:255.

- , 1978-Phylogeny of American horses at Blancan and Pleistocene age. Annal.

Zool. Fennica, 15:191.

Gazin, C. L., 1936 -A study of the fossil horse remains from the upper Pliocene of Idaho. Proc. U. S. Nat. Mus., 83:281.

Gidley, J. W., 1901-Tooth characters and revision of the North American species of the genus Equus. Bull. Amer. Mus. Nat. Hist., 14:91.

Harington, C. R., and F. V. Clulow, 1973— Pleistocene mammals from Gold Run Creek, Yukon Territory. Canadian J. Earth Sci., 10:697.

Hay, O. P., 1913-Notes on some fossil horses with descriptions of four new species. Proc. U. S. Nat. Mus., 44:569.

- , and H. J. Cook, 1930-Fossil vertebrates collected near, or in association with,

human artifacts at localities near Colorado, Texas; Frederick, Oklahoma; and Folsom, New Mexico. Proc. Colorado Mus. Nat. Hist. , 9:4.

Hibbard, C. W., 1955 -Pleistocene vertebrates from the Upper Becerra (Becerra Superior) Formation, Valley of Tequixquiac, Mexico, with notes on other Pleistocene forms. Contr. Mus. Paleo., Univ. Michigan, 12:47.

- , and W. W. Dalquest- 1966-Fossils from the Seymour Formation of Knox and

Baylor counties, Texas, and their bearing on the late Kansas climate of that region. Contr. Mus. Paleo., Univ. Michigan, 21:1.

Johnson, C. S., 1937-Notes on the craniometry of Equus scotti. J. Paleo., 11:459.

Lundelius, E. L., 1972-Fossil vertebrates from the late Pleistocene Ingleside fauna, San Patricio County, Texas. Bureau Econ. Geol., Univ. Texas, Rept. Invest., 11 A.

UPPER INCISORS OF ASINUS GIG ANTE US

41

Olsen, S. J., 1964-Mammal remains from archaeological sites, Part 1, southeastern and south¬ western United States. Papers of the Peabody Museum of Archaeology and Ethnology, Harvard University, Vol. LVI, No. 1, Fig. 2, p. 7.

Savage, D. E., 1951 -Late Cenozoic vertebrates from the San Francisco Bay region. Univ. California Pubis. Geol. Set, 28:215.

■

A CYTOLOGICAL AND HISTOCHEMICAL ANALYSIS OF THE OVARIAN FOLLICLE CELLS OF THE SOUTH TEXAS SQUID (LOLIGO PEALEI)1

by SAMUEL A. RAMIREZ and MANUEL GUAJARDO

Division of Allied Health and Life Sciences,

University of Texas at San Antonio,

San Antonio 78285

ABSTRACT

The oocyte and follicle cell complex of the squid, Loligo pealei, from the South Texas Gulf of Mexico was studied. Since the processes of oogenesis, vitellogenesis and choriono- genesis are highly interrelated and coordinated, these are described as a unit. Six stages in oocyte development and maturation are proposed. The role of the follicle cells (epithelium- syncytium) is studied by cytological and histochemical techniques. This study indicates that follicle cells (epithelium-syncytium) undergo a high degree of cytodifferentiation which is coordinated with oocyte development and have a secretory function that contributes to the maturation of the oocyte. Oocyte development may be directly dependent on the activity of the follicular epithelium.

INTRODUCTION

Ovarian follicle cells have different functions during oogenesis in different organisms (Arnold and Williams -Arnold, 1977; Bloom and Fawcett, 1975; Hoar, 1965, 1969; Sadlier, 1973). Although the specific function of the follicle cells in many organisms is not fully established, circumstantial evidence suggests some possible functions such as yolk granule production (Arnold and Williams -Arnold, 1977; Bottke, 1974; Nelsen, 1953), coat formation (Anderson, 1974; Cowden, 1968; Nelsen, 1953), and transport of ions and molecules synthesized in the fol¬ licle cells into the oocyte (Anderson, 1974; Arnold and Williams- Arnold, 1977; Fujii, 1960; Nelsen, 1953; Raven, 1961, 1967; Selman and Wallace, 1972). The follicle cells of the squid , Loligo pealei , show a high degree of coordinated dif¬ ferentiation with the oocyte (Arnold and Williams-Arnold , 1976;Cowden, 1968; Ramirez and Guajardo, 1977; Selman and Arnold, 1978; Selman and Wallace, 1972). Studies have shown that these differentiated follicle cells become secretory cells (Anderson, 1974; Bottke, 1974; Raven, 1961, 1967; Selman and Wallace, 1972) although their products have not been fully analyzed.

Contribution No. 78-14 from Center for Applied Research and Technology, University of Texas at San Antonio, San Antonio, Texas 78285

Accepted for publication: August 7, 1978.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

44

THE TEXAS JOURNAL OF SCIENCE

Cowden (1968) and Selman and Arnold (1978) staged the maturing oocytes of the squid (Loligo brevis and L. pealei) with light and electron microscopic tech¬ niques and have described the ultrastructure of the follicle cells and oocytes. Five to ten stages have been described according to size and structure of the oocyte and follicle cells. These stages will be used as a basis for proposing 6 stages in this report. Cytological and histochemical observations are reported which support earlier studies (Arnold and Williams- Arnold, 1976, 1977; Cowden, 1968; Ramirez and Guajardo, 1977; Selman and Arnold, 1978; Selman and Wallace, 1972), and pro¬ vide additional insight on the function of the follicle cells during the oocyte de¬ velopment of the South Texas squid, Lo ligo pealei.

METHODS AND MATERIALS

Adult female squid were collected in the Gulf of Mexico, approximately 25 mi from Port Aransas, Texas over a period of 18 mo. Specimens were collected by a 10.5 m otter trawl aboard the University of Texas R/V Longhorn. The ovaries of approximately 30 squid were immediately removed and fixed in Bouin’s solu¬ tion (Galigher and Kozloff, 1971) or calcium-formalin (1% calcium in 10% form¬ alin) (Roozemond, 1967). The tissues were routinely processed and embedded in paraffin. The ovaries were sectioned transversely with 6-8 serial sections/slide.

Tissues were stained with iron-hematoxylin -eosin (Lillie and Fullmer, 1976; Pearse, 1975) for general morphology, Feulgen reaction (Humason, 1962; Lillie, 1977) for DNA, periodic acid-Schiff (PAS) (Pearse, 1975) for polysaccharides, Azure B (Flax and Himes, 1952; Swift, 1966) for RNA and DNA, and fast green (Lillie and Fullmer, 1976) at pH 3.0 for basophilic (mucopolysaccharides) mat¬ erial. Calcium-formalin fixed material was frozen and serially sectioned on a cryostat. Cryostat sections (4-6 cross sections/slide) were stained for lipids with Sudan Black B, III, IV and oil red 0 stains (Lillie and Fullmer, 1976), and mounted with glycerogel (glycerol gelatin) (Lillie and Fullmer, 1976; Pearse, 1975).

Material was examined with a Zeizz research microscope and photographed with Type 55 P/N Polaroid film. All measurements were made with an occular micrometer.

RESULTS

The squid ovary is unpaired and supported by a median mesentery to the dorsal wall at the apex of the coelom. Developing oocytes are clustered around vesicular tissue and a wide range of developmental stages can be found within each cluster (Fig. 1). Each oocyte is surrounded by a single layer of follicular epithelial cells that becomes a syncytium in the mature stages. Six stages of oogenesis can be recognized based on the structure of the follicle cells and oocyte changes (Figs. 1, 2, 3, 4, 5; Table 1). Stage I oocytes are less than 50 pm in diameter and are sur¬ rounded by a single squamous-shaped follicle cell. These immature oocytes have

THE SOUTH TEXAS SQUID

45

Figure 1. Section through ovary shows 4 stages of oocyte development. A large vessel (large arrow) can be seen in the middle of the section. (Hematoxylin-eosin stain, scale 100 /im),

a large germinal vesicle that occupies approximately 40% of the oocyte. As these oocytes begin to grow, the diameter will increase to 50-100 jum, while the single follicle cell proliferates to form a contigious layer of squamous-shaped follicle cells (Fig. 2). The maturing oocyte (Stage II) increases in diameter to 1 00-200 jum while the follicular cells continue to proliferate mitotically and become cuboidal in shape (Fig. 1). The follicle cells continue their active mitotic activity during Stage III forming follicular folds that penetrate into the growing oocyte (Figs. 1 , 2, 4). These folds eventually occupy approximately 80% of the 200-800 /im in diameter oocyte. Vitellogenesis marks the beginning of Stage IV (Fig. 3). Yolk granules begin to form as the cuboidal follicle cells become low columnar cells, stop dividing and form a follicular syncytium (Fig. 3). With the accumulation of yolk granules and oocyte growth (800 qm to 1.5 mm in diameter), the follicular syncytium is pushed out (Stages V, VI) until the chorion is formed during Stage VI (Figs. 4, 5). After the chorion is formed (Stage VI), the follicular syncytium begins to slough off, leaving the oocyte enclosed only by its chorion (Fig. 5).

46

THE TEXAS JOURNAL OF SCIENCE

Figure 2. Higher magnification of Stage I oocyte with a large germinal vesicle (GV) and several follicle cells beginning to form a complete follicular epithelium around the young oocyte. Note the different Feulgen reaction of the follicle cell nuclei at different stages of oocyte development as indicated by staining intensity. (Feulgen reaction, scale 50 /tm).

Cytochemical reactions of the follicular epithelium (syncytium) and oocyte are given in Tables 2 and 3. The first changes seen in the follicle -ovarian complex are in the nuclei of the follicle cells. During the transition of the follicle epithelium from a squamous-shaped single cell to follicular syncytium, the nuclei change from a dense (heterochromatic) Feulgen positive reaction (Stages I, II, III; Fig. 2) to a moderate Feulgen positive reaction in Stage IV (Fig. 3) to a dispersed (eu- chromatic) weak Feulgen reaction in Stage V (Fig. 3) as the follicular epithelium becomes a syncytium. The Azure B (DNA) reaction shows a similar pattern as the Feulgen reaction except that the DNA reaction peak occurs later (Stage III). Within the nuclei, the nucleoli are also undergoing structural and cytochemical changes. The number of nucleoli increases from 1 in Stage I to several in Stage V as the Azure B (RNA) affinity also increases (Table 2).

THE SOUTH TEXAS SQUID

47

Figure 3. Feulgen reaction varies between Stage IV and V follicle cell nuclei, Stage IV nuclei dense while Stage V nuclei is dispersed (euchromatic). (Feulgen reaction, scale 50 jJm),

The cytochemical reactions of the follicle cells’ cytoplasm are not easily de¬ tected during Stages I and II due to the squamous-like nature of the follicle cells. The Azure B (RNA) reaction is weak in Stage III and its intensity increases through Stage V (Table 2). Sudanophilic reaction (lipids, steroids) parallels the Azure B reaction indicating an active synthetic period. The sudanophilic material later is observed increasing in the ooplasm. The polysaccharide (PAS) reaction is not seen in the follicular cytoplasm until chorionogenesis starts in the oocyte (Stages IV, V; Fig. 4). Basophilic material (mucopolysaccharides) is first detected in Stage II and remains at a moderate level through Stage V when the follicular syn¬ cytium begins to slough off.

The ooplasm’s cytochemical reactions parallel and/or interrelate with the changes seen in the follicle cells (epithelium-syncytium). Activity in the germ¬ inal vesicle was not analyzed due to the dispersed nature of the nucleoplasm

48

THE TEXAS JOURNAL OF SCIENCE

Figure 4. PAS stained section of ovary shows Stage V oocyte with chorion precursor droplets (arrow) beginning to coalesce to form the chorion. Chorion precursor droplets and yolk granules are PAS positive.. (PAS reaction, scale 50 /im).

(Davidson, 1976) however, the nucleoli were observed in the Azure B stained material. The number of nucleoli increases during oogenesis. The nucleoli migrate to the periphery of the germinal vesicle where they are transported across the envelope into the ooplasm and have a strong affinity for Azure B. The ooplasm in Stage I reacts weakly to Azure B (RNA) stain and increases to a strong reaction in Stage III (Table 3). In Stage IV the RNA reaction is marked by the increase in yolk granules and basophilic reaction. With the onset of vitellogenesis (Stage IV), the ooplasm becomes strongly basophilic, sudanophilic and gives a strong PAS reaction (Table 3; Fig. 4) which is maintained through Stage VI. During Stage V basophilic and PAS positive droplets begin to accumulate below the follicular syncytium which gives rise to the chorion in Stage VI (Table 3; Fig. 4, 5). When the follicular epithelium is completely retracted (Stage VI), the chorion is fully formed and demonstrates strong PAS and basophilic reactions.

THE SOUTH TEXAS SQUID

49

DISCUSSION

The highly coordinated development and cytodifferentiation of the squid oocyte and follicular epithelium is evident in its structural characteristics. The intimate relationship between the oocyte and follicular epithelium may indicate that the follicle cells are involved in some control and regulation of the oocyte development and the synthesis of yolk granules and other substances as in other animal systems (Anderson, 1974; Raven, 1961). Earlier cytochemical studies (Arnold and Williams- Arnold, 1976, 1977; Cowden, 1968; Ramirez and Guajardo, 1977; Selman and Arnold, 1978 ; Selman and Wallace, 1972) and the present study demonstrates that the metabolic patterns in the oocytes of Loligo pealei are similar to established patterns for oocytes with large amounts of yolk (Ander¬ son, 1974; Davidson, 1976; Hoar, 1965, 1969;Nelsen, 1953; Raven, 1961 ;Sad- lier, 1973). Selman and Wallace (1972) used tritiated leucine to show that the

Figure 5.

Portion of 2 Stage VI oocytes with the follicular epithelium (fe) beginning to slough off. The chorion (ch) is well formed. (Hematoxylin -eosin stain, scale 50 /im).

50

THE TEXAS JOURNAL OF SCIENCE

TABLE 1

Maturation Stages of Loligo pedlei Oogenesis from the South Texas Gulf of Mexico

Stage	Size	Histological Appearance
I	<50 jUm	Immature oocytes surrounded by a single squamous follicle cell.
	50-100 JJm	Growing oocyte with a large germinal vesicle (40% of oocyte) surrounded by several squamous follicle cells.
II	100-200 JJm	Active proliferation of follicle cells and change from squamous to cuboidal in shape.
III	200-800 jUm	Follicle cells continue to proliferate and folds of follicle epithelium penetrate the growing oocyte. Follicle cells become col¬ umnar and form a syncytium.
IV	800 JJm -1.5 mm	Vitellogenesis is evident and accumulation of yolk pushes the follicular layer out.
V	800 Jim -1.5 mm	Vitellogenesis continues and chorion be¬ gins to form. Follicular syncytium is pushed out.
VI	800 JJm -1.5 mm	Chorion formation is complete. Follicular syncytium is sloughed off.

follicle cells were the site of protein synthesis and that the material initially syn¬ thesized in the follicle cells was subsequently transferred to the oocyte to form the yolk granules. Oviductal eggs incubated in tritiated leucine showed no direct' incorporation of the tritiated leucine by the oocyte indicating a dependence on the synthetic activity of the follicle cells. Electron microscopic studies have shown that a large amount of microvillar extensions into the oocyte exist but no pinocy totic vesicles have been seen to indicate transport of synthesized molecules by this method (Bottke, 1974; Selman and Arnold, 1978). However, a definite pattern of metabolic activity can be seen cytochemically (Tables 2, 3) which sug¬ gests the transfer of material and close relationship of the follicular epithelium to the oocyte .

The cytochemical reactions of the follicular cell nuclei and cytoplasm and of the oocyte support the intimate relationship seen histologically. Initially, as growth is first seen in the oocyte -follicular syncytium complex, the nuclear ac¬ tivity appears to undergo marked changes as demonstrated in the nuclei’s Feulgen and Azure B (DNA) staining reaction (Table 2) during the active period of the follicular epithelium.

THE SOUTH TEXAS SQUID

51

TABLE 2

Cytochemical Reaction of the Ovarian Follicle Cells of Loligo pealei from the South Texas Gulf of Mexico

		Cytoplasm		Nucleus	Nucleolus
Oocyte Stage	Lipid Stain3	PAS Azure B	Basophilia	Feulgen	Azure B DNA	Azure B RNA
I	-	_b	-	+	+	_b
II	-	-	+	+++	++	+
III	+	+	++	++	+++	++
IV	++	+ ++	++	+	++	+++
V	+++	++ +++	++	+	+	+++
VI	_c	+c ++c	+c	+	+	+
a Sudan Black B, III, IV and Oil Red 0. bNot visible. cFollicular syncytium is being sloughed off. +++Strongly positive reaction ++Moderately positive reaction. +Weakly positive reaction. -Negative reaction
			TABLE 3
		Cytochemical Reaction of the Oocytes of Loligo pealei from the South Texas Gulf of Mexico
		Ooplasm			Chorion
Oocyte Stage	Lipid RNA Stain3 Azure B	Basophilia	PAS	Basophilia PAS
I	-	+	-	-	_b	_b
II	+	++	-	-	_b	_b
III	+	+++	+	-	_b	_b
IV	+	+	++	+d	„b	_b
V	++	-	+++	+++d	++c ++c
VI	+++	-	+++	+++d	+++	+++

a Sudan Black B, III, IV and Oil Red 0. bNot present. cPrecursor Droplets. dYolk Platelets.

+++Strongly positive reaction. ++Moderately positive reaction. +Weakly positive reaction.

-Negative reaction.

52

THE TEXAS JOURNAL OF SCIENCE

The synthesis of RNA, as demonstrated by Azure B staining, shows that some coordinated activity is occurring. In Stages II and III, the ooplasm shows great amounts of RNA. This has been reported to be the result of germinal vesicle ac¬ tivity during the lampbrush stage of oogenesis (Davidson, 1976; Raven, 1961, 1967). While this activity diminishes in the germinal vesicle, the RNA synthesis increases in the cytoplasm and in the nucleoli of the follicular epithelium. This suggests that ribosomal RNA as well as messenger RNA is being synthesized for normal metabolic activity of the follicle cells and possibly contributed to the developing oocyte as suggested by Raven (1967) and Davidson (1976).

The interaction between the oocyte and the follicle cells could also be steroidal and/or hormonal in nature. This activity has been observed in vertebrate and mammalian systems (Bloom and Fawcett, 1975;Hoar, 1965, 1969;Sadlier, 1973). Sudan and Oil Red O stains were used to analyze the follicular-oocyte complex for sudanophilic material which suggests the presence of steroid (lipid) material in the cells and in the oocyte. As with some of the other cytochemical reactions, sudanophilic material was seen in the follicle cells increasing through Stage V as the ooplasm and yolk granules also increase in their sudanophilic properties (Tables 2, 3).

The process of vitellogenesis in the squid is a poorly understood process and subject to conjecture as shown by Seim an and Arnold (1978) in their ultrastruc- tural studies and by the radioactive tracer studies by Selman and Wallace (1972). Yolk may be produced by the oocyte itself (autosynthetic), by cells other than the oocyte (heterosynthetic), or by a combination of these 2 processes (Ander¬ son, 1974). The present cytochemical studies together with previous studies (Arnold and Williams- Arnold, 1976, 1977; Cowden, 1968; Selman and Arnold, 1978; Selman and Wallace, 1972) support the idea of a heterosynthetic process in the squid. Results reported here (Tables 2, 3) show an increase of cytoplasmic basophilia (mucopolysaccharides) in the follicle cells prior to vitellogenesis (Stages I-III) while the ooplasm does not show any basophilic reaction. With the onset of yolk granule formation in Stage IV and subsequent stages, an increase in baso¬ philia is seen in the ooplasm (yolk granules) suggesting a transfer of material from the follicle cells to the oocyte. Likewise, in the formation of the chorion, baso¬ philic droplets are seen first forming between the oocyte and follicular epithelium that eventually coalesce to form the chorion (Fig. 4). A similar shift in activity is noted in the production of polysaccharides as demonstrated by the PAS reaction. PAS positive material is first seen in the follicle cell cytoplasm during Stage IV prior to vitellogenesis and chorionogenesis. As the PAS positive material increases in the follicle cells, an increase occurs in the yolk granules and chorion precursor indicating a flow of material from the follicle epithelium to the ooplasm.

The histological observations show that the follicular epithelium (syncytium) development is closely coordinated with the 6 stages of oocyte development of the Loligo pealei from the Gulf of Mexico. Cytochemical data suggests that prod¬ ucts from the follicular epithelium are transferred to the oocyte and may be con¬ tributing to the maturation of the squid oocyte as suggested by Anderson (1974)

THE SOUTH TEXAS SQUID

53

and Raven (1961). The nature of the products has not been fully characterized other than knowing that the product is basophilic, sudanophilic and PAS positive, but these findings suggest that oocyte development is under some control of the follicular epithelium as in other animal systems (Anderson, 1974; Bloom and Fawcett, 1975 ; Davidson, 1976; Hoar, 1969; Raven, 1961).

ACKNOWLEDGEMENTS

This paper was partially supported by the Bureau of Land Management, Contracts Nos. AA550-CT6-17 and AA550-CT7-1 1 .

LITERATURE CITED

Anderson, E., 1974-Comparative aspects of the ultrastructure of the female gamete. In G. H. Bourne, J. F. Danielli and K. W. Jeon (Eds.), Review of Cytology, Supplement 4. Acad¬ emic Press, New York, pp. 1—70.

Arnold, J. M., and L. D. Williams-Arnold, 1976-The egg cortex problem as seen through the squid eye. Amer. Zool. , 16:421.

- , and - , 1977-Cephalopoda: Decapoda. In A. D. Giese and J. S. Pearse

(Eds.), Reproduction of Marine Invertebrates, Vol.4. Academic Press, New York, pp. 243- 290.

Bloom, W., and D. W. Fawcett, 1975 -A Textbook of Histology . W. B. Saunders Co., PA, pp. 805-906.

Bottke, W., 1974-The fine structure of the ovarian follicle of Allotheuthis subulata Lam. (Mollusca, Cephalopoda). Cell Tissue Res., 150:463.

Cowden, R. R., 1968-Cytological and cytochemical studies of oocyte development and development of follicular epithelium in the squid, Loligo brevis. Acta Embryol. Morph. Exp., 10:160.

Davidson, E. H., 1976 -Gene Activity in Early Development. 2nd Ed. Academic Press, New York.

Flax, M. H., and M. H. Himes, 1952-Microspectrophotometric analysis of metachromatic staining of nucleic acids. Physiol. Zool., 25:291.

Fujii, T., 1960-Comparative biochemical studies on the egg yolk proteins of various animal species. Acta Embryol. Morphol. Exp., 3:260.

Galigher, A. E., and E. N. Kozloff, 1911-Essentials of Practical Microtechnique. Lea & Fe- biger, PA.

Hoar, W. S., 1965 -Comparative physiology: Hormones and reproduction in fishes. Ann Rev. Physiol, 27:51.

- , 1969-Reproduction. In W. S. Hoar and D. J. Randall (Eds.), Fish Physiology,

Vol. 3. Academic Press, New York, pp. 1-72.

Humason, G. L., 1962— Animal Tissue Techniques. W. H. Freeman and Co., San Francisco.

54

THE TEXAS JOURNAL OF SCIENCE

Lillie, R. D., 1911 -H. J. Conn’s Biological Stains. 9th Fd. The Williams & Wilkins Co., Balt¬ imore.

- — , and H. M. Fullmer, 191 6 -Histopathologic Technique and Practical Histochemistry ,

4th Ed. McGraw-Hill Book Co., New York.

Nelsen, O. E., 19 53 -Comparative Embryology of the Vertebrates. McGraw-Hill Book Co., Inc., New York.

Pearse, A. G. E., 197 5 -Histochemistry , Theoretical and Applied. 3rd Ed., Vol. 1. Churchill Livingstone, New York.

Ramirez, S. A., and M. Guajardo, 1977-Histological and cytochemical study of ovarian fol¬ licle cells of the squid, Loligo pealei. J. Cell Biol., 75:174.

Raven, C. P., 1961 -Oogenesis: The Storage of Developmental Information. Oxford, Per- gammon Press, New York.

- -, 1967-The distribution of special cytoplasmic differentiations of the egg during

early cleavage in Limnaea stagnalis. Develop. Biol., 16:407.

Roozemond, R. C., 1967 -Thin layer chromatographic study of lipid extractions from cryo¬ stat sections of rat hypothalmus by same fixatives. 7. Histochem. Cytochem., 15:526.

Sadlier, R. M. F. S., 1913-The Reproduction of Vertebrates. Academic Press, New York, pp. 1-35.

Selman, K., and J. M. Arnold, 1978-Anultrastructuraland cytochemical analysis of oogenesis in the squid, Loligo pealei. J. Morph., 152:381.

- , and R. A. Wallace, 1972-A role for the follicle cells during vitellogenesis in the

squid Loligo pealei. Biol. Bull., 143:477.

Swift, H., 1966-The quantitative cytochemistry of RNA. In G. L. Wied (Ed.), Introduction to Quantitative Cytochemistry . Academic Press, New York.

A SURVEY OF SELECTED PLANTS FOR THE PRESENCE OF EUKARYOTIC PROTEIN BIOSYNTHESIS INHIBITORS

by ROBYN REYNOLDS and JAMES D. IRVIN

Department of Chemistry

Southwest Texas State University

San Marcos 78666

ABSTRACT

A number of selected plant seeds and leaves were screened for the presence of inhibitors of eukaryotic protein biosynthesis. All the plant extracts tested contained significant amounts of inhibitory compounds, most of which were not inactivated by heat treatment. The plant seed from Aleurites fordii was found to contain the greatest inhibitory activity which was caused by a protein.

INTRODUCTION

In recent years a number of proteins from various plants have been shown to be potent inhibitors of eukaryotic protein synthesis but possess varied biological properties. One protein purified from Phytolacca americana (pokeweed) is a powerful antiviral agent (Irvin, 1975; Ussery, et al , 1977). The 2 proteins abrin (from Abrus precatorius) and ricin (from Ricinus communis ) are very potent toxins (Olsnes and Pihl, 1976). Another protein, alpha sarcin from Asperigillus giganteus , has been shown to be an anti-tumor agent (Olsen and Goerner, 1965). The site of action of all of these proteins has been shown to be upon the eukaryotic ribosome (Dallal and Irvin, 1978; Olsnes and Pihl, 1976; Schindler and Davies, 1977).

In this communication we report the results of a survey of selected plants for the presence of proteinaceous inhibitors of eukaryotic protein synthesis. The selection of the plants for this study was based upon previous reports of the pres¬ ence of toxins or lectins in the plant and the ease of obtaining suitable quantities of material for the purification of potential inhibitors.

MATERIALS

The ground seeds of Caragana arborescens, Cytsus scoparius, Euonymus europaeus, Laburnum alpinum, Robinia pseudoacacia, Sophora japonica, and

Accepted for publication: June 4, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

56

THE TEXAS JOURNAL OF SCIENCE

Ulex europaeus were purchased from P. L. Biochemicals, Milwaukee, WI. The leaves of Phoradendron serotinum were harvested locally from wild plants and those from Jatropha multi fida were obtained from ornamental plants. The fruit from Aleurites fordii were the kind gift of Dr. James M. Spiers; USD A, Poplarville, MS.

METHODS

All operations were performed at 0-4 C. Solution E consists of 10 mM tris (hydroxymethyl)aminomethane -HC1 , pH 7.5; 0.1 mM 2-mercaptoethanol; and 0.2 mM ethylenediamine tetraacetate.

Ten grams of ground or chopped seeds were allowed to soak for 15 min in Solution E containing 100 mM KC1 with the exception of R. pseudoacacia which were soaked in water. The mixture was homogenized 5 min in a Lourdes homo¬ gen izer followed by centrifugation for 15 min at 8,000 x g in an IEC refrigerated centrifuge. The supernatant was filtered through cheesecloth and was recentrifuged for 20 min at 27,000 xg to remove excess lipids. The supernatant obtained from the 2nd centrifugation was dialyzed against Solution E for 12-14 hr. The activity of this crude extract was determined by its ability to inhibit in vitro polyphenylalanine synthesis on Artemia salina ribosomes as previously described (Irvin, 1975).

Crude extracts were also obtained from the leaves and stems of P. serotinum and from the leaves of/, multifida. In this procedure, 50 g of leaves were homo¬ genized with a Waring blender in 100-200 ml of Solution E, 100 mM KC1, fol¬ lowed by filtration through cheesecloth with mild suction. The filtrate was then dialyzed against Solution E for 12-14 hr. The crude extracts from these plants were also tested for inhibitory activity. Protein concentrations were determined by the method of Kalb and Bernlohr (1977).

RESULTS AND DISCUSSION

The results presented in Table 1 demonstrate the presence of heat labile in¬ hibitors of protein synthesis in 3 of the 1 0 plant sources surveyed . The heat lability of these extracts suggests that the inhibitory factors are proteins and thus may be similar to abrin, alpha sarcin, the pokeweed antiviral protein, and ricin which also inhibit protein synthesis and are purified from plants.

The remaining 7 plant extracts also inhibit protein synthesis at fairly low doses but the failure of heat treatment to destroy the activity suggests that the active principles are not proteins or very heat stable ones.

Of particular interest is the extract from R. pseudoacacia which is very active and could only be extracted in the absence of salt in the media. The most potent inhibitory extract, that from the seeds of the tung fruit {A. fordii), has been chosen for further investigations and we have partially purified a basic protein from this source which absorbs to phosphocellulose ion exchange resin and thus appears to be similar in properties to the pokeweed antiviral protein (Irvin, 1975).

EUKARYOTIC PROTEIN BIOSYNTHESIS INHIBITORS

57

TABLE 1

Inhibition of Protein Synthesis by Plant Extracts

Source	IDSoa (fig Protein)	Heat Lability*5
Seeds
Aleurites fordii	0.017	+
Car ag ana arbor escens	1.450	-
Cytsus scoparius	0.220	-
Euonymus europaeus	2.100	-
Laburnum alpinum	5.350	-
Robinia pseudoacacia	0.050	-
Sophora japonica	13.200	-
Ulex europaeus	7.210	-
Leaves
Jatrophia multifidia	0.970	+
Phoradendron serotinum	1.650	+

aThe inhibitory does which produces 50% inhibition of protein synthesis. ^Heat lability is defined as sensitivity (+) to heating at 90° for 15 min.

ACKNOWLEDGEMENTS

The authors wish to thank Mrs. Roxie Smeal for her help in preparing the typescript. This work has been supported by Robert A. Welch Foundation Grant AI-605 and by Organized Research Funds from the State of Texas.

LITERATURE CITED

Dallal, J. A., and J. D. Irvin, 1978-Enzymatic inactivation of eukaryotic ribosomes by the pokeweed antiviral protein. FEBS Letters , 89:257.

Irvin, J. D., 1975 -Purification and partial characterization of the antiviral protein from Phytolacca americana which inhibits eukaryotic protein synthesis. Arch. Biochem. Biophys. 169:522.

Kalb, V. F., and R. W. Bernlohr, 1977 -A new spectrophotometric assay for protein in cell extracts. Anal. Biochem., 82:362.

Olsen, B. H., and G. L. Goerner, 1965 -Alpha sarcin, a new antitumor agent. I. Isolation, purification, chemical composition, and the identity of a new amino acid. Appl. Microbiol. , 13:314.

Olsnes, S., and A. Pihl, 1976-Abrin, ricin, and their associated agglutinins. In P. Cuatrecasas (Ed.), The Specificity and Action of Animal, Bacterial and Plant Toxins. Chapman and Hall, London, pp. 131-173.

Schindler, D. G., and J. E. Davies, 1977-Specific cleavage of ribosomal RNA by alpha sarcin. Nucl. Acid Res. , 4:1097.

Ussery, M. A., J. D. Irvin, and B. Hardesty, 1977-Inhibition of polio virus replication by a plant antiviral peptide. Ann. N. Y. Acad. Sci., 284:431.

RECONNAISSANCE OBSERVATIONS OF SOME FACTORS IN¬ FLUENCING THE TURBIDITY STRUCTURE OF A RESTRICTED ESTUARY: CORPUS CHRISTI BAY, TEXAS1

by GERALD L. SHIDELER

U.S. Geological Survey

P. O. Box 6732

Corpus Christ i 78411

ABSTRACT

Corpus Christi Bay is a shallow restricted estuary that is typical of the Texas Coastal Plain. On the basis of synoptic reconnaissance measurements of light transmissivity and sus¬ pended-sediment concentrations at 6 monitoring stations, a time sequence of turbidity structures was determined along the longitudinal trend of the Bay and its tidal inlet. Measure¬ ments were made on 6 observation dates extending over a 16-mo period. Longitudinal turbidity structures were highly variable in time and space. Structures ranged from avertically homogeneous water column, to a well-stratified column showing an increasing turbidity gradient with depth. Mean sediment concentrations also showed high variability.

Wind appeared to be the dominant forcing agent influencing turbidity in the bayhead sector, where it both generates waves that resuspend bottom sediment and regulates fluvial- sediment influx from the Nueces River. Turbidity in the baymouth sector appeared to be mainly influenced by tidal-forcing effects from Aransas Pass inlet. Neither the sediment- discharge characteristics of the Nueces River nor the mean water density of the Bay had any discernible influence on Bay turbidity.

INTRODUCTION

The Texas coast along the northwest Gulf of Mexico is characterized by a well- developed barrier island chain and an extensive backbarrier lagoonal-estuarine system. These coastal features were formed during the latter stages of the Holocene rise in sea level that commenced approximately 18,000 yr ago. The drowning of Pleistocene fluvial channels and subsequent barrier construction during the last few thousand years resulted in the development of the shallow “bar-built” type of restricted estuary (Schubel, 1971) along the Texas coast, of which Corpus Christi Bay is a representative example (Fig. 1).

An estuary’s circulation pattern is greatly influenced by its physical configuration and by the external driving forces of river flow, tidal flow, and wind stress. As

Approved for publication by the Director, U.S. Geological Survey Accepted for publication: May 15, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

60

THE TEXAS JOURNAL OF SCIENCE

Figure 1. Location map of the study area showing sites of monitoring stations and bay bathymetry.

noted by Stommel (1951), any of these external forces can dominantly regulate the circulation of a particular estuary, and consequently, its resulting patterns of sedimentation. The purpose of the present reconnaissance study was to establish a comparative time sequence of estuarine turbidity structures along the longitudinal trend of Corpus Christi Bay, and to attempt to relate these structures to ambient environmental conditions. This was done in an effort to gain some insight into sedimentary processes indigeneous to the shallow bar-built variety of coastal- plain estuary that is characteristic of the Texas Gulf Coast.

ENVIRONMENTAL SETTING

Corpus Christi Bay is a relatively shallow estuary, generally less than 5 m deep (Fig. 1). An exception is the Corpus Christi ship channel that is maintained for navigation by dredging to a depth of approximately 15 m. The Bay has a slight northwest-southeast elongation, and is separated from the Gulf of Mexico by the Mustang Island barrier. The Bay’s main tidal inlet (Aransas Pass) is near the city of Port Aransas. The main fluvial flow into the estuarine system is from the Nueces River which discharges directly into satellite Nueces Bay. In turn, shallow Nueces Bay (< 1 m deep) has water exchange with adjacent Corpus Christi Bay via a narrow causeway-connected inlet. Bottom sediment within Corpus Christi Bay is mainly mud in the interior, whereas muddy and shelly sand is concentrated in the marginal areas (Univ. of Texas, 1974). Observations during the present study indicate that the composition of the Bay’s suspended sediment is mainly inorganic silt and clay detritus, with a subordinate organic skeletal fraction dominated by diatoms.

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

61

Both meteorological forces and astronomical tides substantially influence bay circulation. Astronomical tides are both diurnal and semi-diurnal; the tidal range in the adjacent open Gulf during fair weather is generally less than 0.3 m (Marmer, 1954), and decreases bay ward. Local prevailing winds are onshore from the south¬ east, and are most consistant during the summer. During the winter, stronger northerly winds frequently are associated with the passing of polar cold fronts southward into the Gulf of Mexico. Visual observations indicate that the response of the Bay’s circulation system and associated turbidity patterns to changing wind conditions is rapid (less than a few hours), mainly because of the Bay’s shallowness. The Bay is susceptible to both “norther” storms during the winter, as well as to tropical storms and hurricanes during the summer and fall seasons.

METHODS

Reconnaissance field work consisted of obtaining a time sequence of water- column measurements at 6 monitoring stations (2-7) along the longitudinal trend of Corpus Christi Bay and its associated tidal inlet (Fig. 1). Field sampling was conducted on 6 dates that represent all seasons, over a total observational period of 16 mo. The sampling dates were: October 20, 1975; January 19, 1976; May 11, 1976; June 7, 1976; August 9, 1976; February 14, 1977. At each moni¬ toring station, vertical transmissivity and temperature profiles were obtained respectively to determine turbidity structure and thermostructure ; profile measure¬ ments were made by means of a light-beam transmissorneter (2 5 -cm optical path) and attached temperature sensor. Surface and near-bottom water samples also were collected by meansof a 3-liter Van Dorn bottle for laboratory analyses. In addition to bay samples, a surface-water sample was obtained near the mouth of the Nueces River (Station 1) for approximating the rate of fluvial-sediment influx during the sampling date.

Surface and near-bottom water samples from each station were analyzed in the laboratory for salinity using an induction salinometer. Water densities then were determined from salinity and temperature, and expressed as sigma-T values [cq = (density - 1) x 1000] . The samples also were analyzed for suspended- sediment concentrations in terms of total mass (mg/fi). Mass determinations were determined gravimetrically by filtration on prewashed 0.45 qm Millipore filters. Mean values and standard deviations of both bay-water density and sediment concentrations then were determined on the basis of the 12 station measurements (one anomalous measurement was deleted from the May suite). Because of the Bay’s shallowness, the mean and standard deviation values of bay -water density and sediment concentrations were considered to be representative of the water column along the monitored transect (Table 1).

Comparative vertical-transmissivity cross-sections were constructed to illustrate the turbidity structure along the longitudinal transect of the Bay during the 6 sampling dates. In addition, statistically significant differences in mean values of

62

THE TEXAS JOURNAL OF SCIENCE

TABLE 1

Water-Column Characteristics along the Monitored Transect and Fluvial-Sediment Influx from the Nueces River

Observation Date	Mean Bay-Sediment Concentrations (mg/C)	Mean Bay-Water Density (of)	Fluvial-Sediment Influx (g/sec)
Mean	Standard Deviation	Mean	Standard Deviation
October 20, 1975	11.8	6.9	17.9	0.8	80
January 19, 1976	27.3	17.7	22.0	0.1	24
May 11, 1976	14.4	7.0	20.0	0.9	7,504
June 7, 1976	25.0	11.5	18.4	0.7	352
August 9, 1976	14.7	13.1	18.2	4.7	2,774
February 14, 1977	17.7	4.6	17.1	2.8	301

turbidity in terms of suspended-sediment concentrations (mg/C) among the 6 sampling dates were determined by means of a robust t-statistic test at the 95% confidence level (Table 2), using an SAS computer program (Barr, et al., 1976). The t-statistic tests also were used to determine significant differences in mean bay -water density because of its potential influence on suspended -particle set¬ tling velocities and overall turbidity (Table 2). Differences among the 6 sets of

TABLE 2

Results of t-Statistic Tests for Monthly Comparisons of the Bay’s Water Column in Terms of Mean Sediment Concentration and Density

Compared Months	t-Values for Total Sediment Mass (rag/C) Comparisons	t-Values for Water Density (at) Comparisons
October vs. January	2.81*	16.45*
October vs. May	0.89	5.52*
October vs. June	3.40*	1.37
October vs. August	0.68	0.17
October vs. February	2.44*	0.93
J anuary vs. May	2.24*	6.74*
January vs. June	0.37	15.81*
January vs. August	1.97	2.77*
January vs. February	1.82	5.85*
May vs. June	2.63*	4.49*
May vs. August	0.08	1.27
May vs. February	1.33	3.18*
J une vs. August	2.04	0.14
June vs. February	2.05	1.46
August vs. February	0.73	0.65

* Significant difference at the 0.05 level of confidence (degrees of freedom = 22, to.os = 2.07; for May comparisons, degrees of freedom = 21, to. os = 2.08)

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

63

field measurements along the monitored transect were interpreted in terms of variations in ambient environmental conditions during the sampling periods.

Local wind data were obtained from the U.S. Weather Service at Corpus Christi, and local tidal data at the Aransas Pass inlet were obtained from standard tide tables. Stream-discharge rates (m3/sec) from the Nueces River (Mathis gage) were obtained from U.S. Geological Survey’s water data reports (1975-1977), Austin District. Estimates of the relative rates of river-sediment influx (g/sec) during the sampling dates were based on the measurements of sediment concentration (mg/C) from the river-water samples and on the average daily stream-discharge rates (Table 1).

DISCUSSION

The observed longitudinal turbidity structures are the composite responses to several complexly interrelated environmental variables. Differences among the observed structures could reflect variations in any of the following: wave and tidal conditions, wind conditions, Nueces River discharge, and water density. Under natural field conditions, these environmental factors may represent a multivariate system. Therefore, the individual influence of each variable on the turbidity structure may not be totally resolvable. In addition, the actual response time of the bay turbidity structure to changing environmental conditions is un¬ known. Consequently, it was frequently impossible to distinguish residual effects generated prior to the monitoring periods. In spite of these limitations, some insight can be acquired regarding the dominant forcing agents by comparing the observed sequence of bay -turbidity structures formed under different field conditions.

October 20, 1975/May 11, 1976/February 14, 1977 Structures

An informative comparison can be made of the 3 turbidity structures comprising this sequence because they occurred under a similar set of ambient field conditions (Fig. 2). All 3 structures formed during the same tidal phase (waning ebb tide and accelerating flood tide), with a similar onshore wind direction (southeasterly) and under similar calm sea state conditions (SS 1 ). The absence of significant waves during this sequence is especially noteworthy because visual observations indicate that waves within the shallow bay are highly influential in establishing turbidity patterns through the resuspension of bottom sediments. Consequently, a distin¬ guishing aspect of these 3 structures is that they were generated largely through processes not associated with wave activity.

The 3 turbidity structures within the bay interior (Stations 2-5) are basically similar in that each exhibits some degree of turbidity stratification and an in¬ creasing turbidity gradient with increasing depth. This suggests that the water column was not vertically homogenized by wave activity. In terms of overall mean transect turbidity based on sediment concentrations (mg/C), the t-statistic tests indicate that the only significant difference among the 3 sampling dates is

64

THE TEXAS JOURNAL OF SCIENCE

Figure 2. Comparative sequence of transmissivity profiles illustrating bay turbidity struc¬ tures for the following observation dates: October 20, 1975; May 11, 1976; February 14, 1977. Contour interval is 10%T/0.25 m. Also illustrated are ambient tidal-current variations (shaded interval is sampling period) and daily wind vectors.

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

65

that February was more turbid than October (Tables 1,2); this was not associated with a corresponding significant difference in mean density (at) of the water column. Furthermore, significant water density differences do occur between May and both October and February which were not associated with significant turbidity differences. These observations indicate that water density had no discernible influence on bay turbidity.

In attempting to relate the 3 turbidity structures to external forces, the most prominent environmental variable during the sequence was the fluvial-sediment influx from the Nueces River (Table 1). Maximum sediment influx was in May during a high-water stage at an approximate rate of 7, 504 g/sec, whereas minimum influx was in October (80 g/sec). Although the river-sediment influx was nearly 2 orders of magnitude greater in May than in October, this was not manifested in a corresponding contrast at Bay Station 2, which is closest to the Nueces River mouth. The combined average concentration of surface and bottom waters at Station 2 was 9.5 mg /£ in October, compared with only 13.0 mg /£ in May. Further¬ more, in terms of transmissivity, higher overall turbidity actually occurred at Station 2 in October when the river influx was minimal. In addition, overall mean transect turbidity based on sediment-concentration measurements (mg/£) was not significantly higher in May than in October. These relationships suggest that river-sediment influx, even during a high-water stage, was not highly in¬ fluential in contributing to the variations observed among the turbidity structures from the 3 sampling dates.

As a working hypothesis, the apparent absence of river influence on bay turbidity during these periods is attributed to the entrapment of fluvial sediments within the adjacent shallow Nueces Bay which functioned as an effective settling basin. Entrapment within Nueces Bay would have been facilitated by the absence of significant wave action to maintain sediments in suspension, and by set-up ef¬ fects and wind-drift currents toward the northwest generated by the southeasterly onshore winds; this would have inhibited the dispersal and mixing of relatively turbid Nueces Bay waters with the less turbid waters of Corpus Christi Bay. Northerly flow into Nueces Bay during a period of southeasterly onshore winds has been indicated by longitudinal turbidity bands on aerial photographs (Fig. 3), possibly reflecting bottom-sediment resuspension and transport by Langmuir circulation. Conversely, the outflow of turbid Nueces Bay waters into Corpus Christi Bay during a time of relatively strong northwesterly offshore winds also has been documented by aerial photography (Fig. 4). Consequently, it appears that Nueces Bay may function as a release valve for fluvial-sediment influx into Corpus Christi Bay — a valve that is regulated mainly by wind direction.

The greatest differences among the 3 turbidity structures are found in the vicinity of the tidal inlet (Stations 6, 7), ranging from a highly stratified water column in October to vertically homogeneous conditions in February. These inlet differences can be reasonably interpreted as the result of tidal forcing effects. As the inlet stations in all 3 structures were occupied during the same basic tidal

66

THE TEXAS JOURNAL OF SCIENCE

Figure 3. Oblique aerial photograph of Nueces Bay -Corpus Christi Bay Inlet taken on July 19, 1978. Longitudinal turbidity bands indicate northerly flow into Nueces Bay (top of photo) from Corpus Christi Bay under conditions of on¬ shore winds; resultant daily wind vector was from the southeast (135 ) with a speed of 10 knots.

phase (accelerating flood tide), the variations could reflect differences in a com¬ bination of the following: (1) tidal-current velocities, (2) duration of flooding, and (3) residual effects from previous tidal phases. Turbidity stratification at the inlet was best developed in October, a period characterized by the relatively highest tidal-current velocities, the longest duration of flooding, and the longest previous ebb phase. These tidal conditions also might account for the significantly lower mean turbidity in October relative to February, possibly reflecting the more efficient prior seaward flushing of relatively turbid ebb waters and greater sub¬ sequent exchange by cleaner oceanic flood waters. More effective tidal flooding during October may have been enhanced by stronger onshore winds (12.8 km/hr), as compared with the weaker February winds (6.7 km/hr). The stronger onshore winds in October also would have more effectively inhibited the dispersal of relatively turbid Nueces Bay waters into Corpus Christi Bay, thus further con¬ tributing to the lower overall transect turbidity during October relative to February.

January 19, 1976 Structure

The turbidity structure for January formed entirely during a waning ebb tide, with a relatively strong (30.1 km/hr) southeasterly onshore wind, and very choppy seas (Fig. 5). The sediment influx rate from the Nueces River was the lowest

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

67

Figure 4. Aerial photograph of Nueces Bay-Corpus Christi Bay Inlet taken at 12,500 ft on January 21, 1973. Turbid plume of Nueces Bay water moving into Corpus Christi Bay under conditions of strong offshore winds; resultant daily wind vector was from the northwest (290 ) with a speed of 22 knots.

among the 6 sampling dates (24 g/sec). The structure consists of a homogeneously turbid inner-bay sector (Stations 2-4), becoming somewhat stratified toward the tidal inlet. Compared with the October/May/February sequence, the main dif¬ ferences in ambient conditions during the January sampling period were much stronger southeasterly winds, substantial wave activity, and the absence of prior contiguous flood-tide effects. In addition, the mean water-column density (ot = 22.0) along the transect during January was significantly higher than during all other sampling periods. Because the strong onshore winds would tend to inhibit the dispersal of turbid Nueces Bay waters into Corpus Christi Bay, the greater inner-bay turbidity in January is attributed to a higher degree of bottom-sediment resuspension and vertical mixing by waves toward the head of the bay ; this bay sector is especially susceptible to intense wave action generated by strong south¬ easterly onshore winds because of maximum fetch. In terms of overall mean

68

THE TEXAS JOURNAL OF SCIENCE

figure 5 . Comparative sequence of transmissivity profiles illustrating bay turbidity struc¬ tures for the following observation dates: January 19, 1976; June 7, 1976; August 9, 1976. Contour interval is 10%T/0.25 m. Also illustrated are ambient tidal-current variations (shaded interval is sampling period) and daily wind vectors.

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

69

transect turbidity based on sediment concentrations (mg/C), January had the highest (27.3 mg/C) and most variable (std. dev. = 17.7) turbidity. January was significantly more turbid than both May and October. This difference is attributed both to a greater degree of bottom-sediment resuspension by waves resulting from the stronger southeasterly winds, and to the absence of contiguous prior replace¬ ment by less turbid flood -tide Gulf waters. The greater mean turbidity in January relative to May when fluvial-sediment influx was more than 2 orders of magnitude greater, once again supports the inference that river influx was not particularly influential on bay turbidity.

June 7, 1976 Structure

The June turbidity structure formed during a complete flood-tide phase and an accelerating ebb-tide phase (Fig. 5). Winds were from the northeast at 12.5 km/hr, and the seas were slightly choppy. Fluvial-sediment influx from the Nueces River was at a moderate rate (352 g/sec). Mean water-column density (crt = 18.4) along the transect was significantly lower than during both May and January. The June turbidity structure is characterized by a homogeneously turbid bay head sector (Station 2), with the rest of the Bay showing varying degrees of turbidity stratification; the degree of stratification increases toward the tidal inlet. Relative to the previously discussed observation dates, the most distinguishing environ¬ mental variable during June was a change in wind direction, with northeasterly winds oriented essentially normal to the Bay’s longitudinal trend. In addition, the tidal conditions during sampling were different.

Of importance are the more turbid conditions (41 mg/C) at the bayhead (Station 2), relative to May turbidity (12 mg/C) when the rate of river sediment influx was more than an order of magnitude greater. This is interpreted as being largely the combined result of both a higher degree of sediment resuspension by waves, and the more effective dispersal of turbid Nueces Bay waters into Corpus Christi Bay by set-up effects and wind-drift currents generated by the north¬ easterly winds. It appears that, in addition to offshore winds, alongshore winds parallel to the Gulf Coast which have strong northerly components also may be effective agents for flushing out the Nueces Bay settling basin. In terms of sediment concentrations (mg/C), overall mean transect turbidity was significantly higher in June than during both May and October. This is attributed to the more effective dispersal of turbid Nueces Bay water into Corpus Christi Bay, the greater observed wave activity, and possibly to variations in tidal conditions.

August 9, 1976 Structure

The August turbidity structure formed during a waning flood tide, with relatively weak (6.7 km/hr) southerly winds, and very calm sea state conditions (Fig. 5). The sediment influx from the Nueces River was relatively high (2,774 g/sec), being second only to the high-water stage influx during May. The mean water density (at = 18.2) along the transect during August was significantly lower than

70

THE TEXAS JOURNAL OF SCIENCE

during January. The distinguishing environmental variables during August were the southerly winds, high flu vial -sediment influx, and flood -tide conditions during the entire sampling period. The August structure is characterized by a high degree of stratification, apparently reflecting an absence of significant wave homogeni¬ zation. Similar to the May/June comparison, the relatively high fluvial-sediment influx during August is associated with a lower concentration at Station 2 (20 mg/fi) compared with the lower fluvial influx but higher Station 2 concentration (41 mg/fi) during June. Once again, this is attributed to the confinement of fluvial sediments within Nueces Bay by set-up and currents generated by the onshore southerly winds during June. The overall mean transect turbidity during August was not significantly different than that of any other sampling period. However, the tidal-inlet sector (Stations 6,7) had relatively cleaner waters than during any other period, with transmissivity values (%T/0.25 m) reaching a maximum of 76% at Station 7. These were the most transparent conditions observed during the study, probably reflecting the relatively long period of exchange by cleaner flood-tide waters prior to sampling.

SUMMARY AND CONCLUSIONS

Turbidity characteristics along the longitudinal trend of Corpus Christi Bay were highly variable in time and space. Transmissivity values (%T/0.25 m) of the water column along the monitored transect ranged from zero to a maximum of 76%. Longitudinal turbidity structures based on transmissivity ranged from a vertically homogeneous water column, to a well -stratified column which had an increasing turbidity gradient with depth; the structures changed with varying ambient conditions. Sediment concentrations among individual transect stations showed substantial spatial variability during a given sampling date, with standard deviations ranging from 4.6 mg/£ in February to 17.7 mg/C in January. Temporally, mean sediment concentrations along the entire transect ranged from 1 1 .8 mg/C in October, to 27.3 mg/C in January. Mean transect concentrations were signif¬ icantly lower in October than during January, June, and February; they also were significantly lower in May than during June and January.

Turbidity toward the bayhead sector appeared to be largely influenced by wind, whereas the rate of fluvial -sediment influx from the Nueces River had no discernible influence on bay turbidity. A working hypothesis is suggested whereby sediment influx from the Nueces River enters adjacent Nueces Bay, which appears to function as a shallow storage basin that entraps sediment at times when winds have southerly or southeasterly onshore components. However, during periods of westerly and northerly offshore or alongshore winds, set-up effects and wind- drift currents appear to flush and disperse relatively turbid Nueces Bay waters southward into the head of Corpus Christi Bay. In essence, the influx of fluvial sediments into the estuarine system could be regulated mainly by wind direction. Winds further influence turbidity structures by generating waves that resuspend

TURBIDITY STRUCTURE OF CORPUS CHRISTI BAY

71

bottom sediments, especially within the bayhead sector. Turbidity structures toward the baymouth sector appear to be influenced largely by tidal-forcing effects associated with Aransas Pass Inlet. Variations in mean density of the Bay’s water column during the observed periods ranged from a maximum value (at = 22.0) in January 1976, to a minimum value (at = 17.1) in February 1977; these density variations had no discernible systematic effect on mean bay turbidity. Winds and tides appeared to have been the dominant forcing agents influencing the observed bay -turbidity structures. These observations are of a reconnaissance nature, and more detailed future long-term monitoring of the Bay would be necessary to verify the relationships suggested by the present study.

ACKNOWLEDGEMENTS

The author extends his appreciation to F. Firek, C. Stelting, B. Willingham, and G. Harrison for field and laboratory assistance during the study.

LITERATURE CITED

Barr, A. J., J. H. Goodnight, J. P. Sail, and J. T. Helwig, 1976-/4 Users Guide to SAS-76. SAS Inst. Inc., Raleigh, N. C., 329 pp.

Marmer, H. A., 1954-Tides and sea level in the Gulf of Mexico. Gulf of Mexico, its origin, waters and marine life. U. S. Department of the Interior, Fish and Wildlife Service Bul¬ letin No. 89, pp. 101-108.

Schubel, J. R., 1971 -The classification of estuaries. In Schubel, J. R. (Conv.), The Estuarine Environment-Estuaries and Estuarine Sedimentation. AGI Short Course Lecture Notes, pp. II— 1—8.

Stommel, J., 1951 -Recent developments in the study of tidal estuaries. Woods Hole Oce¬ anographic Institution, Technical Report Reference No. 51-33.

Univ. of Texas Bureau of Economic Geology, 1974-Environmental geologic atlas of the Texas coastal zone. Corpus Christi Sheet, 1:125,000.

■

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS OF THE LOWER RIO GRANDE, SOUTHWESTERN TEXAS1

by GERALD L. SHIDELER

U.S. Geological Survey Corpus Christi 78411

and ROMEO M. FLORES

U.S. Geological Survey Denver , CO 80225

ABSTRACT

The variability of heavy minerals in modern fluvial sediments of the Rio Grande between El Paso and Brownsville, Texas, was studied in an effort to evaluate relative effects of provenance and stream -transport processes on mineralogical composition. The heavy-mineral assemblage is characterized as a pyroxene, hornblende, and opaque mineral suite, containing minor quantities of epidote, tourmaline, garnet, and zircon. The 3 dominant minerals show the greatest variability, as well as significant downstream trends over a 1400-km transport distance. Stepforward regression analysis indicates a linear pyroxene trend attributed to hydraulic shape sorting during transport. A curvilinear hornblende trend is attributed to both shape sorting and to local hornblende-rich source rocks near El Paso, Texas. A curvilinear opaque-mineral trend indicates both hydraulic density sorting and the presence of local source rocks that are rich in opaque minerals within Big Bend National Park. Local variations in relative mineral abundance are attributed to a combination of local source-rock differences and hydraulic sorting effects. The heavy-mineral composition of the Rio Grande sediments does not appear to be greatly affected by tributary influx from either the Pecos River or the Rio Conchos.

INTRODUCTION

The Rio Grande is one of the major fluvial systems in the southwestern U.S. The river originates within the southern Rocky Mountains of south-central Colorado and flows southward through central New Mexico; it then flows south¬ eastward to the western Gulf of Mexico between southwestern Texas and north¬ eastern Mexico (Fig. 1 ). This study was concerned with evaluating the downstream variability of heavy minerals in modern fluvial sediments along the lower Rio Grande valley between El Paso, Texas and the Gulf of Mexico. Previous work on

Approved for publication by the Director, U.S. Geological Survey.

Accepted for publication: April 10, 1979.

The Texas Journal of Science, Vol. XXXII, No. 1, March, 1980.

74

THE TEXAS JOURNAL OF SCIENCE

105° 100°

Figure 1. Map of the lower Rio Grande valley study area showing locations of sample stations.

fluvial heavy minerals near the present study area includes studies of the middle Rio Grande in central New Mexico (Rittenhouse, 1943, 1944) and of the upper Pecos River in New Mexico (Sid well, 1941). Heavy-mineral studies also have been made of Rio Grande delta sediments and of adjacent Continental Shelf sediments originally derived from the Rio Grande (e.g. van Andel and Poole, 1960; Flores and Shideler, 1976; and Shideler and Flores, 1976).

The early work of Rittenhouse (1943) on heavy minerals of the Rio Grande stressed the complex interrelationships between source rock characteristics and transport processes that determine the heavy mineral distributions in fluvial sediments. Ever since the introduction of the hydraulic equivalence concept by Rubey (1933), it has been widely recognized that the hydraulic behavior of heavy minerals is jointly influenced by their physical properties (size, shape, density), availability, and the dynamics of the transporting medium. A discussion of the interrelationships and influence of size, shape, and density on hydraulic sorting has been presented by Briggs (1965). In his study, Briggs noted that deviations from expected theoretical relationships among certain minerals observed in some Tertiary sandstones could be explained on the basis of restricted size availability of the anomalous mineral groups. The objective of the present study was to evaluate the downstream variability of heavy minerals in the lower Rio Grande

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

75

over a distance of 1400 km, in an effort to gain insight into the relative importance of source rocks and stream transport processes in establishing mineralogical com¬ position. Stream processes of particular interest were hydraulic sorting and trib¬ utary dilution.

METHODS Field Techniques

Samples of modern fluvial sediments were obtained from 8 sample stations along the lower Rio Grande between the mouth of the river at the Gulf of Mexico and El Paso, Texas (Fig. 1). The sample stations are: Station 1— east of the city of Brownsville about 8 km above the river mouth; Station 5— north side of the City of Laredo; Station 7 --City of Del Rio; Station 9— Pecos River mouth, less than 2 km above its junction with the Rio Grande Station 10 — Boquillas Canyon area of Big Bend National Park; Station 11— City of Presidio below mouth of Rio Conchos; Station 12— City of Presidio above mouth of Rio Conchos ; Station 13— northwest side of the city of El Paso. The station numbering system is sequential, but excludes some stations that were occupied for purposes other than heavy mineral analysis.

At each station, 4 sediment samples were obtained several meters apart to evaluate “within -station” mineral variability. Vertical channel samples were acquired from the river bed and/or bank to determine average composition, using a cylindrical sampling tube (10-cm dia. x 12-cm depth). All field sampling was done within a 6-day period during a low -water stage of the river.

Analytical Techniques

In the laboratory, the field samples were dispersed and wet-sieved; the 63 jiim-125 jum (3 0 - 40) sand fractionswere then separated by sieving for heavy mineral analysis. This narrow size range was used so that any mineral variations resulting from grain-size effects would be constant among the sample stations, thus accentuating variations resulting from mineral shape and density factors. The very fine sand fractions were cleaned with a dilute hydrochloric acid solution (10%), followed by thorough washing. Heavy -mineral separations were then performed by the centrifuge-frozen bromoform method (e.g. Carver, 1971), using a liquid nitrogen freezing agent. Heavy -mineral separation efficiency was standardized by using the following constant conditions: (1) centrifugation at constant speed (1200 rpm) and duration (20 min), (2) use of a standard bromoform volume (15 ml), and (3) use of a standard sample size (1-2 gm). The separated heavy-mineral fractions were weighed, and mounted on glass slides in a Lakeside 70 medium (RI = 1 .54).

Petrographic analysis consisted of identification and point-counts of heavy mineral grains along random line traverses. A total of 200 translucent grains were identified and point-counted, and an additional 100 grains were point-counted to determine the percentage of opaque minerals. All point counting was done by the same operator.

76

THE TEXAS JOURNAL OF SCIENCE

The heavy-mineral percentages at the 8 sample stations were analyzed statis¬ tically to determine significant local and downstream mineral variations. Mineral percentages at the individual stations were plotted as a function of distance from the river mouth. Each mineral group was subjected to a stepforward regression analysis to determine any significant downstream trends. This analysis consisted of fitting polynomials in successive stages and testing for significance at each stage. The mineral percentages were then fitted with the optimum least-square regression equation at the 0.05 significance level. The regression analysis was conducted by using a U.S. Geological Survey STATPAC computer program (D0094). Significant local variations also were determined between individual sample stations and selected groups of stations (Table 2); this evaluation was done with a t-statistic test at the 95% confidence level, using aSAS1 computer program (Barr, et al , 1976). Significant downstream and local variations were then inter¬ preted in terms of known geological conditions within the lower Rio Grande valley.

REGIONAL PHYSIOGRAPHY AND GEOLOGY

The studied 1400-km sector of the lower Rio Grande extends from El Paso, Texas, downstream to the river mouth near Brownsville, Texas. The mean annual precipitation along the lower Rio Grande valley increases downstream from less than 8 in (20 cm) near El Paso, to about 25 in (64 cm) near Brownsville. Normal annual temperatures along the lower Rio Grande range from about 64 F (18 C) in some sectors upstream from the Pecos River junction, to a high of about 74 F (23 C) downstream from Laredo (Orton, 1969). Two major tributaries of the lower Rio Grande are the Rio Conchos and the Pecos River, which respectively join the Rio Grande near the towns of Presidio and Del Rio. The Pecos tributary originates in the Sangre de Cristo Mountains of northern New Mexico, and the Rio Conchos originates in the Sierra Madre Occidental of Chihuahua, Mexico.

The studied sector of the lower Rio Grande valley traverses 3 separate physio¬ graphic provinces that are progressively lower in relief downstream (e.g. U.S. Dept. Interior, 1970). The river segment from El Paso to the east side of Big Bend National Park (Stations 13-10) is within the Basin and Range Province, which is characterized by mountain ranges and intervening plains. Elevations above sea level are mostly within the 2,000-5,000 ft (610-1,524 m) range, but some peaks are higher. The segment from Big Bend to approximately Del Rio (Stations 10-7) is within the Great Plains Province, which is characterized by a hilly terrain and elevations within the 1 ,000-2,000 ft (305-610 m) range. From south of Del Rio to the Gulf of Mexico (Stations 5,1), the Rio Grande crosses the nearly flat Gulf Coastal Plain Province, which slopes to sea level.

A generalized geologic map of the lower Rio Grande valley and adjacent areas shows that different bedrock types are found within the 3 physiographic provinces (Fig. 2). Local bedrock from El Paso to the Big Bend area (Basin and Range

1 Any trade names are used for descriptive purposes only and do not constitute endorsement by the U.S. Geological Survey.

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

77

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).

Province) is a relatively heterogeneous assemblage of sedimentary, igneous, and metamorphic rocks that range in age from Quaternary to Precambrian. Abundant Tertiary intrusive and extrusive igneous rocks are found in this province, and are essentially absent from the provinces further downstream. As noted by Maxwell, et al , (1967), a large concentration of these igneous rocks, which are composed mainly of alkali basalt, riebeckite rhyolite, and granite, occurs within the Big Bend area. These workers also noted the partial or complete alteration of the mafic minerals from these rock types into limonite and brown opaque grains. Local Precambrian rocks between El Paso and Presidio consist of metasedimentary greenschist facies and amphibolite facies that are extensively veined by pegmatite (King and Flawn, 1953). Several areas of outcropping Paleozoic carbonate and clastic sedimentary rocks also occur near the Rio Grande, mainly on the northern side; the only nearby Paleozoic outcrop in Mexico is in a small uplift traversed by the Rio Conchos. The local bedrock is more homogenous near the Rio Grande between the eastern Big Bend area and the Del Rio area (Great Plains Province) than further upstream. The province is underlain mainly by Cretaceous strata

78

THE TEXAS JOURNAL OF SCIENCE

dominated by carbonate rocks (Figs. 3A, 3B). Downstream from the Cretaceous outcrop belt lies the Gulf Coastal Plain; this province is underlain mainly by heterogeneous Tertiary (Pliocene, Miocene, Eocene) and Quaternary clastic sediments that decrease in age gulfward. In general, the lower Rio Grande valley includes source rocks of a wide variety in terms of both age and lithology.

Figure 3. A. View toward Mexico of Rio Grande incised in Cretaceous strata of the Great Plains Province just above the junction of the Pecos River; B. Upstream view of Pecos River near its junction with the Rio Grande; C. Downstream view of Rio Grande from small spillway near Presidio (Station 11) that causes local turbulence; D. Local heavy-mineral concentration (dark band) along sand/gravel river bank at Station 11, reflecting local hydraulic sorting effects immediately downstream from spillway turbulence.

DISCUSSION OF RESULTS

Total Heavy-Mineral Percentages

The total heavy-mineral content in fluvial sediments along the lower Rio Grande varies greatly. The mean weight percentages of heavy minerals within the studied size fraction (63-125 gm) at the 8 sample stations range from 0.8-18.3%. Mean weight percentages at individual stations are as follows: Station 1 - 1.0%,

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

79

Station 5 - 2.6%, Station 7 - 15.2%, Station 9 - 0.8%, Station 10 - 5.1%, Station 11 - 18.3%, Station 12 -2.3%, Station 13 - 7.2%.

No well-defined downstream trend in total heavy -mineral percentages is apparent, and most of the variability is probably attributable to local hydraulic sorting effects. An illustrative example is Station 11 at Presidio, which has the highest heavy mineral content (Figs. 3C, 3D). At this station, turbulence resulting from a small man-made spillway has resulted in the local fractionation and con¬ centration of heavy minerals immediately downstream from it. Similar effects from natural turbulence probably have contributed to the variability at other sample stations. The lowest heavy -mineral content occurs at Station 9, suggesting that effluent from the Pecos River tributary is relatively deficient in heavy minerals compared to that of the Rio Grande.

The heavy -mineral assemblage of the lower Rio Grande is differentiated into 8 mineral groups: pyroxene, hornblende, epidote, tourmaline, garnet, zircon, opaque minerals, and miscellaneous minerals. The miscellaneous group includes minerals that occur in very minor quantities and consists of rutile, kyanite, staurolite, and micas.

The mean percentages of each mineral group at individual sample stations are given in Table 1 . Quantitatively, the opaque minerals represent the most abundant group; they consist mainly of limonite, ilmenite, hematite, and magnetite. The opaque -mineral content ranges from 37.0% at Station 1 to 66.0% at Stationll. The pyroxene group is second in abundance, and contents range from 13.8% at Station 9 to 27.9% at Station 1. The pyroxene group consists mostly of augite, but also contains a minor amount of hypersthene. The hornblendes are the next most abundant group, ranging from 5.8% at Station 10 to 27.5% at Station 1. The hornblendes consist of both brown and subordinate green varieties. On the average, the brown variety constitutes about two-thirds of the hornblende group. The remaining 5 mineral groups are much less abundant, and no individual group exceeds 8% at any of the sample stations. Of these minor groups, zircon is most abundant, and tourmaline is genetically significant. The tourmaline group consists of both brown and blue (indicolite) varieties; the latter variety is derived chiefly from pegmatites (Krynine, 1946). On the basis of the foregoing, the lower Rio Grande heavy-mineral assemblage is dominated by opaque minerals, pyroxene, and hornblende.

TABLE 1

Mean Percentages of Mineral Groups at Individual Sample Stations

Station	Pyroxene	Hornblende	Epidote	Tourmaline	Garnet	Zircon	Misc.	Opaque
1	27.9	27.5	0.5	2.5	1.1	0.9	1.8	37.0
5	22.2	12.1	3.9	2.0	2.4	5.1	1.7	50.0
7	16.0	6.1	2.5	1.2	1.6	6.3	0.9	64.5
9	13.8	24.2	3.3	3.7	2.3	7.2	1.8	43.0
10	22.1	5.8	1.7	0.8	1.2	5.1	1.3	61.0
11	19.4	5.9	1.7	0.7	1.0	3.6	1.2	66.0
12	16.8	9.8	1.9	1.0	1.6	4.0	2.2	62.2
13	13.9	19.0	2.2	2.0	1.6	5.6	1.3	53.7

80

THE TEXAS JOURNAL OF SCIENCE

In terms of the more abundant components, the lower Rio Grande suite is similar to the middle Rio Grande suite of central New Mexico, as described by Rittenhouse (1944). The more abundant heavy minerals of the middle Rio Grande suite are magnetite, ilmenite, pyroxene, epidote, and hornblende. The only significant disparity is epidote, which is subordinate to zircon in the lower Rio Grande suite.

Total Variability

The total percentage variability of each of the 8 identified heavy mineral groups is illustrated by longitudinal plots of mineral mean percentages at the 8 sample stations along the studied 1400-km length of the Rio Grande (Fig. 4). Station 9 is included in the plots to illustrate differences in heavy -mineral assem¬ blages between the Rio Grande channel and the mouth of the Pecos River tribu¬ tary. The total percentage variability illustrated by the longitudinal plots is the net result of combined downstream, local, and random variation effects. Greatest total percentage variability is shown by the opaque minerals, hornblende, and pyroxene groups, which are also the most abundant constituents. Local variations in abundances of opaque minerals and hornblende tend to be greatest in the vicinity of the Pecos River junction (Station 9). In comparison with nearby suites from the Rio Grande, the suite from the Pecos River is rich in hornblende, and deficient in opaque minerals and pyroxene. Variations in the relative abundance of the remaining 5 mineral groups are minor, partially reflecting their uncommon occurrence.

In general, only minor chemical weathering effects on ferromagnesian minerals were noted petrographically during this study. The relatively minor influence of both selective weathering and abrasion on minerals of the Rio Grande assemblage during transport also had been previously noted by Rittenhouse (1943). Therefore, it appears that the observed mineral variability is mainly attributable to source rock and hydraulic effects.

Downstream Trends

In an effort to differentiate significant systematic downstream trends in mineral composition, the mineral percentages of 33 individual samples obtained at the 8 stations were plotted as a function of distance from the river mouth. The data were then fitted with an optimum least-square regression equation by using stepforward regression analysis. The regression analysis indicates that only the 3 most abundant mineral groups (pyroxene, hornblende, opaque minerals) show significant downstream trends at the 0.05 probability level along the 1400-km segment of the Rio Grande from El Paso to Brownsville (Fig. 5).

The pyroxenes show a trend of increasing percentages downstream. This trend is best characterized by the linear regression equation (Y = 25 .07 - 1 .22X), which suggests that 31% of the total pyroxene variability can be accounted for by distance of transport. Because the most prolific source of pyroxene is probably crystalline

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENT

81

%

if) °

LU ,o

ft 0

DOWNSTREAM

Km

Miscellaneous

Zircon

Garnet

Tourmaline

Epidote

l

Brownsville

5 79 10 II 12

SAMPLE STATIONS

13

El Paso

Figure 4. Longitudinal graphs illustrating total variability of mineral mean percentages between El Paso and Brownsville; Station 9 is the Pecos River mouth.

bedrock of the upper Rio Grande, the systematic downstream increase in pyroxene is interpreted as a trend resulting largely from fluvial suspension-transport proc¬ esses. Hydraulic sorting by shape during suspension could result in the preferential transport and downstream concentration of the relatively bladed pyroxene grains.

82

THE TEXAS JOURNAL OF SCIENCE

-« — DOWNSTREAM

0 500

I ■ ■ . i _ I

k m

1^ 5 7 9 10 II 12 _ 13

STATION LOCATIONS

Figure 5. Least-square regression curves illustrating significant regional trends along the lower Rio Grande at the 0.05 probability level; Station 9 is the Pecos River mouth. A. Pyroxene trend; B. Hornblende trend; C. Opaque-mineral trend. The independent variable (X) values in the regression equations are in hundreds of mi.

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

83

The remaining 69% of unexplained pyroxene variability is attributed to a combi¬ nation of local hydraulic sorting and source-rock effects, sampling and analytical variability, and random error.

The downstream trend in the percentages of hornblende-group minerals is best characterized by a 5th -degree polynomial equation (Y = 25.37 - 3.50X + 0.00045X5), in which distance of transport accounts for 42% of the total horn¬ blende variability. The observed curvilinear trend shows a percentage minimum in the Big Bend area (Station 10,11 , 12), and both an upstream and a downstream increase in hornblende. Similar to the pyroxene trend, the downstream increase in hornblende below the Big Bend area is interpreted as mainly reflecting hydraulic shape sorting during suspension that resulted in a downstream concentration of bladed hornblende grains. The increase in hornblende upstream from the Big Bend toward El Paso (Station 13) is attributed to provenance; it could reflect the local influx of a hornblende-rich mineral assemblage derived from Tertiary volcanic source rocks near El Paso and farther north within the New Mexico drainage basin. The remaining 58% of unexplained hornblende variability is attributed to a combination of local hydraulic sorting and source-rock effects, sampling and analytical variability, and random error.

The downstream trend of the opaque-mineral group is best characterized by a 7th-degree polynomial regression equation (Y = 38. 28 + 4.20X- 0.000005 IX7), in which distance of transport accounts for 43% of the total opaque-mineral variability. The curvilinear downstream trend shows a percentage maximum near the Big Bend area, and is largely the inverse of the hornblende trend. The opaque- mineral trend, like the hornblende trend, is interpreted as the combined result of both provenance and suspension-transport processes. The high percentages within the Big Bend area (Station 10, 1 1 , 12) are attributed to the local abundance of opaque minerals derived from Tertiary igneous source rocks. However, the subsequent downstream reduction in opaque minerals is attributed mainly to hydraulic density sorting during transport. The relatively high -density opaque minerals would tend to be concentrated locally near the prolific Big Bend source area, and to decrease in abundance with transport downstream. The remaining 57% of unexplained opaque-mineral variability is attributed once again to a combination of local hydraulic sorting and source -rock effects, sampling and analytical variability, and random error.

The reason for the absence of significant downstream trends among the other 5 mineral groups is conjectural, but is probably related to their minor quantitative importance. At such low levels of occurrence (<8%), any downstream trends may be completely obscured by local and random variation.

The observed downstream trends of the pyroxene, hornblende, and opaque minerals are in agreement with similar trends exhibited by amphiboles and opaque minerals in the Godavari River of India (Naidu, 1964), and by hornblende and pyroxene in some short -headed streams of western New York (Flores, 1971). These studies attributed trends in these mineral groups to hydraulic sorting by shape and density factors. However, the observed hornblende and pyroxene

84

THE TEXAS JOURNAL OF SCIENCE

trends along the lower Rio Grande are in contrast to the situation along the lower Mississippi River. The same 2 heavy mineral groups are major components of the Mississippi River assemblage, but they exhibit no downstream trends over ap¬ proximately a 900 -km transport distance between Cairo and New Orleans (Russell, 1937; Davies and Moore, 1970). The reason for this contrast between the 2 fluvial systems is conjectural. However, some possibilities include: 1) dif¬ ferences in hydraulic characteristics of the 2 rivers, 2) differences in availability of the 2 mineral groups along the length of the drainage basins, or 3) a combi¬ nation of the 2 forgoing factors.

Local Variability

In an effort to gain additional insight into local sources of heavy -mineral variations, the mineral mean percentages at individual sample stations were selec¬ tively grouped and compared. Significant compositional differences were then established at the 95% confidence level by a t-statistic test (Table 2), and dif¬ ferences were interpreted in terms of known geologic conditions. Comparisons were made at 3 levels: 1) source-rock province comparison, 2) stream-sector comparisons, and 3) individual station comparisons.

Source-rock province comparison : The lower Rio Grande sample stations were grouped into 2 provinces characterized by basically different bedrock materials. A downstream province extending from Brownsville to Del Rio (Stations 1+5+7) is essentially a sedimentary province. These sedimentary deposits range in age from Quaternary to Cretaceous, and are predominantly clastic sediments; the province is essentially devoid of crystalline igneous-metamorphic source rocks. In contrast, an upstream source -rock province extending from Boquillas Canyon in the Big Bend National Park to El Paso (Stations 10+11+12+13) contains a relatively large proportion of crystalline rocks, largely Tertiary igneous rocks. The collective mineral assemblages from sample stations within these 2 provinces were statistically compared to evaluate the influence of their contrasting local source rocks (Table 2).

The comparison indicates that significant differences between the 2 source- rock provinces are found in only 2 of the 8 mineral groups, namely, the opaque- mineral group and the tourmaline group. The mineral assemblage from the up¬ stream crystalline-rich province is significantly higher in opaque minerals and lower in tourmaline than the assemblage from the downstream province. Individual station comparisons indicate that these differences are not the result of tributary influx, as discussed in a subsequent section. The higher content of opaque minerals in samples from the upstream province is in agreement with the opaque -mineral downstream trend, suggesting provenance effects reflecting the local abundance of opaques derived from igneous source rocks within the Big Bend area, as well as reinforcing secondary downstream effects from subsequent hydraulic density sorting. The higher concentration of tourmaline in the downstream sedimentary province could be anticipated from the recycled nature of the sedimentary clastic

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

85

Results of t-:	TABLE 2 statistic Tests for Station Comparisons of Mean Mineral Percentages
STATION COMPARISON: 1+5+7 vs 10+11+12+13 (d.f. = 28,^05 = 2.04)
Significant Mineral Differences	Mean Percentages	t-value
Tourmaline Opaque	^1+5+7 = 2.40, ^10+11+12+13 = 1.55 xl+5+7 = 50.50, x10+ll+12+13 = 60.77	2.69 2.82

STATION COMPARISON: 1+5+7 vs 10+11 (d.f. = 20, 1.05 = 2.08)

Hornblende	xl+5+7 = 15.29^10+11 = 5.90		3.02
Tourmaline	xl+5+7 = 2.40, x10+ll = 1.29		3.49
Garnet	3 1+5+7 = 1.73, XJ_0+11 = 1.15		2.13
Opaque	xl+5+7 = 50.50, x10+ll = 63.00 STATION COMPARISON: 1+5+7 vs (d.f. = 18, t.05 = 2.10)	12+13	2.84
Pyroxene	X1+5+7 = 22.08, x12+13 = 15.37 STATION COMPARISON: 10+11 vs (d.f. = 16, 1.05 = 2.12)	12+13	2.73
Pyroxene	Xio+11 = 21.05^x12+13 = 15.37		3.00
Hornblende	x10+ll = 5.90, x12+13 = 14.41 STATION COMPARISON: 1 vs (d.f. = 6, 1.05 = 2.44)	7	4.22
Pyroxene	xl = 27.97, x7 = 16.05		4.96
Hornblende	xl = 27.55^7 = 6.15		11.58
Epidote	X1 = 0.57,32 = 2.57		3.24
Garnet	X1 = 1.12X7= 1.65		2.48
Zircon	X1 = 0.90, x7 = 6.30		6.95
Opaque	xl = 37.0037 = 64.00		5.97

STATION COMPARISON: 5 vs 7 (d.f. = 6, ^05 = 2.44)

Hornblende	35 = 12.1737 = 6.15	3.82
Opaque	x5 = 50.00, x7 = 64.50	2.94

STATION COMPARISON: 1 vs 5 (d.f. - 6, t.05 = 2.44)

Hornblende	3l = 27.5535= 12.17	10.24
Epidote	X1 = 0.5735 = 2.92	7.86
Garnet	3l = 1.12,35 = 2.42	2.70
Zircon	31 = 0.90, x£= 5.12	9.47
Opaque	xl = 37.00, x5 = 50.00	2.88

86

THE TEXAS JOURNAL OF SCIENCE

Table 2 (Continued)

Significant Mineral Differences	Mean Percentages	t-value
Pyroxene	STATION COMPARISON: 9 vs 10 (d.f. = 8, t.05 = 2.30) x9 = 13.87, x10 = 22.11	5.42
Hornblende	x9 = 24.25, x10= 5.85	3.84
Epidote	x9= 3.30, x10 = 1.78	2.64
Tourmaline	x9= 3.75, x10 = 0.83	4.64
Garnet	x9 = 2.35, x10= 1.23	2.31
Opaque	x9 = 43.00, x10= 61.00	2.85
Hornblende	STATION COMPARISON: 10 vs 11 (d.f. = 8, 1.05 = 2.30) No Significant Mineral Differences STATION COMPARISON : 1 1 vs 1 2 (d.f. = 6, t.05 = 2.44) xll = 5.97, x12 = 9.80	3.00
Hornblende	STATION COMPARISON: 12 vs 13 (d.f. = 6, f05 = 2.44) x12 = 9.80, x13 = 19.02	3.34
Garnet	STATION COMPARISON: 7 vs 10 (d.f. = 6, f 05 = 2.44) x7 = 1.65, x10= 1.12	3.13

deposits. This local source-rock effect might have been augmented by transport effects resulting froiji shape sorting, whereby the elongated tourmaline grains would tend to be concentrated downstream. These 2 conditions working in concert could explain the significant downstream increase in the relatively minor tourmaline group. The blue indicolite tourmaline variety found in fluvial sediments from the downstream sedimentary province was probably derived from pegmatitic source rocks between El Paso and Presidio within the upstream crystalline -rich province.

The absence of significant provincial differences among the quantitatively important pyroxene and hornblende groups is noteworthy. If provincial source- rock composition was the dominant factor in controlling their variability, signifi¬ cantly higher concentrations of both pyroxene and hornblende should be found in sediments from the upstream province where crystalline rocks are abundant. However, since thisisnot true, hydraulic sorting appears to have been the dominant factor controlling the well-defined downstream trends of pyroxene and hornblende. In essence, any local provincial variations in pyroxene and hornblende that might have been attributed to provenance appear to have been obscured by opposing

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

87

effects resulting from subsequent transport processes; the bladed pyroxene and hornblende grains were probably concentrated downstream by hydraulic shape sorting. The absence of detectable provincial differences among the other minor mineral groups (epidote, garnet, zircon, misc.) might be partially attributed to the low levels of their occurrence; however, their absence also supports the infer¬ ence that provincial variations in source rocks were of only minor importance in establishing the total heavy -mineral variability.

Stream-sector comparisons . A second method of evaluating local variability in mineral percentages along the Rio Grande is to compare mineral content in 3 geographic sectors of the river channel, arbitrarily defined as follows: 1. lower sector- -from the mouth of the Rio Grande at Brownsville to Del Rio (Stations 1+5+7); 2. middle sector-- the Big Bend area, from Boquillas Canyon to just below the mouth of the Rio Conchos tributary at Presidio (Stations 10+11); and 3. upper sector — from just above the mouth of the Rio Conchos tributary at Presidio to El Paso (Stations 12+13). Statistical comparisons of the collective mineral assemblages from the 3 stream sectors were conducted to determine significant differences. A comparison of the lower and middle sectors showed significant differences in the following 4 mineral groups: hornblende, tourmaline, garnet, and opaque minerals. Relative to the middle sector, the lower sector contains significantly more hornblende, tourmaline, and garnet, but significantly less opaque minerals. The relationships of both hornblende and opaque minerals in these sectors are in agreement with their respective downstream trends, which indicate hydraulic sorting according to shape and density during suspension transport. Shape sorting could also explain the tourmaline relationship, although the significance of the garnet relationship is enigmatic.

A comparison of heavy minerals in the middle and upper sectors of the Rio Grande showed significant differences in both pyroxene and hornblende. The middle sector has a higher pyroxene content and a lower hornblende content than the upper sector, thus agreeing with the downstream trends of the 2 minerals. The higher pyroxene content in the middle sector is attributed to downstream concentration by shape sorting, whereas the lower hornblende content might be attributed to anomalously high local concentrations of hornblende in waters in fluxing from near and above El Paso.

Individual-station comparisons . A third method of evaluating local mineral variability is to compare selected pairs of individual sample stations for significant differences. Within the sedimentary province of the lower sector of the Rio Grande (Stations 1+5+7), comparisons were made between each pair of sample stations. Sediment samples from the Del Rio Station (7) are significantly lower in hornblende and higher in opaques than those from both the Laredo (5) and Brownsville (1) Stations; the Del Rio samples are also significantly lower in pyroxene than are Brownsville samples. These relationships are consistent with the opaque mineral,

88

THE TEXAS JOURNAL OF SCIENCE

hornblende, and pyroxene downstream trends that indicate hydraulic sorting by shape and density; this interpretation is further supported by a significantly higher content of opaque minerals and a lower content of hornblende in samples from the Laredo Station relative to samples from the Brownsville Station. The Del Rio samples are also significantly higher in zircon, epidote, and garnet than are samples from the Brownsville Station; these differences could be caused by local hydraulic sorting effects, local source rock effects, or a combination of both. A comparison of sediment samples from the Laredo (5) and Brownsville (1) Stations also shows the same significant upstream increase in zircon, epidote, and garnet. Inasmuch as the Laredo and Brownsville Stations both occur in basically similar bedrock (Cenozoic clastic sediments), this consistent relationship suggests that the variability of the minor mineral groups also may be controlled dominantly by hydraulic sorting.

A comparison of sediment samples from the Pecos River Station (9) and from the Big Bend area Station (10) shows substantial differences in the 2 mineral assemblages. Sediments from the Pecos are significantly richer in hornblende, epidote, tourmaline, and garnet; whereas the Big Bend sediments are richer in pyroxenes and opaque minerals. These differences could reflect provenance and/or distance of transport. It should be noted that the regional trend reversals in opaques and hornblende occur downstream from the Pecos River mouth, and would be compatible with dilution effects from the influx of Pecos sediment. Therefore, in order to evaluate this possibility, a comparison was made of samples from Rio Grande Stations immediately below (Station 7) and above (Station 10) the mouth of the Pecos River. Any net downstream changes in the Rio Grande mineral assemblage resulting from Pecos dilution effects should be most pronounced between these 2 stations. However, this comparison shows that the only statistically significant difference is in garnet content, which is higher downstream from the Pecos River junction, in agreement with the station 9-10 comparison. The notable absence of significant differences among the major mineral groups (pyroxene, hornblende, opaque minerals) illustrates that net mineralogical effects of Pecos River sediment influx on the Rio Grande assemblage are minimal. In view of the substantial contrast in the opaque-hornblende-pyroxene contents of the 2 assem¬ blages, the absence of local net downstream changes in these dominant minerals suggests that the quantity of sediment influx from the Pecos was insufficient to modify the basic characteristics of the Rio Grande assemblage. This inference is supported by surface measurements of suspended sediment concentrations taken concurrently with the heavy -mineral sampling which showed only 20 mg/1 of Pecos influx at Station 9, compared to 2356 mg/1 of Rio Grande sedimen t measured upstream at Station 10. This great contrast in sediment load (2 orders of magnitude) could explain the inability of the Pecos influx to significantly modify the pro¬ portions of the more abundant mineral species within the relatively turbid Rio Grande. If any Pecos River dilution effects have contributed to the regional opaque and hornblende trends, they appear to be subordinate to the hydraulic sorting effects.

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

89

Within the middle and upper Rio Grande sectors, a comparison of sediment samples from the Big Bend Station (10) with samples from the Presidio Station below the mouth of the Rio Conchos (1 1) shows no significant differences, thus suggesting relatively uniform conditions throughout the Big Bend area. A com¬ parison of samples from the 2 Presidio Stations (11 and 12) is used to evaluate the influence of sediment influx from the Rio Conchos. The only observed dif¬ ference between samples from the 2 stations is that sediments from the station above the Rio Conchos have a significantly higher content of hornblende. This suggests possible local dilution of the Rio Grande mineral assemblage by a horn¬ blende-deficient Rio Conchos assemblage; however, the Rio Conchos does not appear to have any other significant effect. A comparison of sediment samples from the upper Presidio Station (12) with samples from the El Paso Station (13) indicates only a significantly higher content of hornblende at El Paso, which is consistent with the downstream hornblende trend.

CONCLUSIONS

The heavy -mineral assemblage of the Rio Grande between Brownsville and El Paso, Texas is dominated by opaque minerals, pyroxene, and hornblende. Minor components include epidote, tourmaline, garnet, and zircon. The 3 dominant mineral groups have the greatest total variability , including significant downstream trends over a 1400 -km transport distance. A downstream linear trend of increasing pyroxene accounts for 31% of the total pyroxene variability; it appears to reflect hydraulic sorting by shape during suspension transport. The downstream curvilinear trend of hornblende, which accounts for 42% to total hornblende variability, shows lowest percentages within the Big Bend area. A downstream increase in hornblende content below the Big Bend is attributed to hydraulic shape sorting, whereas an upstream increase is attributed to local hornblende-rich volcanic source rocks near El Paso and in New Mexico. The downstream curvilinear trend of opaque minerals accounts for 43% of the total opaque -mineral variability. The trend shows maximum contents within the Big Bend area, reflecting a local abundance of opaque minerals derived from igneous source rocks. A downstream reduction in opaque minerals is attributed to hydraulic density sorting during transport. The unexplained downstream variability of the 3 dominant mineral groups (pyroxene - 69%, hornblende - 58%, opaque minerals -57%) is attributed to a combination of local hydraulic sorting and source -rock effects, sampling and analytical variability, and random error. The minor mineral groups show relatively low variability and no significant downstream trends. In the establish¬ ment of downstream trends, the effects of hydraulic sorting during transport appear to dominate the effects of regional source rocks. Local mineral variability along the Rio Grande is partially attributed both to local source rock differences and to local hydraulic sorting effects. Dilution effects resulting from the influx of sediment to the Rio Grande from both the Pecos River and Rio Conchos ap¬ pear to be of only minor importance in contributing to the total variability of the Rio Grande heavy -mineral assemblage.

90

THE TEXAS JOURNAL OF SCIENCE

The observed downstream trends have relevance to stratigraphic studies of ancient fluvial sandstones, especially in relation to provenance determinations and lithostratigraphic correlations based on relative proportions of heavy minerals. The initial proportions of heavy minerals in the source area, as established by source rock composition, can be progressively modified down paleoslope because of the cumulative effects of hydraulic sorting by shape and density. Consequently, these effects must be considered when formulating inferences regarding source rock composition. In addition, caution must be exercised in establishing litho¬ stratigraphic correlations based on mineral percentages. It is apparent that proximal and distal sections of the same fluvial sandstone can exhibit substantially different proportions of heavy minerals, with the proximal sections most closely reflecting the true composition of the original source rocks. Lastly, this study illustrates the potentially “noisy”nature of fluvial heavy-mineral assemblages within the stratigraphic record, whereby much of the regional variability actually may result from nonsystematic and indeterminable causative factors.

ACKNOWLEDGEMENTS

The writers express their appreciation to A. T. Miesch, U.S. Geological Survey, for assistance in the application of computer programs used in this study. Appre¬ ciation is also expressed to D. K. Davies, Texas Tech University, for reviewing the manuscript.

LITERATURE CITED

American Association of Petroleum Geologists, 1973 -Geologic Highway Map of Texas, Map No. 7.

Barr, A. J., J. H. Goodnight, J. P. Sail, and J. T. Helwig, 1976-/J User’s Guide to SAS-76. SAS Inst. Inc., Raleigh, N. C., N. N., 329 pp.

Briggs, L. I., 1976-Heavy mineral correlations and provenances. J. Sed. Petrol , 35:939.

Carver, R. E., 1971 -Heavy mineral separation.//? R. E. Carver (Ed.), Procedures in Sediment¬ ary Petrology. John Wiley and Sons, Inc., New York, N.Y., 653 pp.

Davies, D. K., and W. R. Moore, 1970-Dispersal of Mississippi sediment in the Gulf of Mexico. J. Sed. Petrol, 40:339.

Flores, R. M., 1971 -Variations in heavy -mineral composition during transport of short¬ headed stream sands (abst.). AAPG-SEPM Ann. Mtg. Program, Houston, Texas, pp. 338.

- , and G. L. Shideler, 1976-Concentrating processes of heavy minerals on the outer

continental shelf off southern Texas, Gulf of Mexico (abst.). Geol. Soc. Am. Prog., 8:868.

King, P. B., and P. T. Flawn, 195 3 -Geology and mineral deposits of Pre-Cambrian rocks of the Van Horn area, Texas. Univ. of Texas Pub. 5301, Austin, Texas, 218 pp.

Krynine, P. D., 1946-Tourmaline group in sediments./. Geol. , 54:65.

HEAVY-MINERAL VARIABILITY IN FLUVIAL SEDIMENTS

91

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

Naidu, A. S., 1964— Lithologic and chemical facies changes in the recent deltaic sediments of the Godavari River, India. Deltas in Their Geologic Framework . Houston Geol. Soc., pp. 125-157.

Orton, R. B., 1969-Climates of the States-Texas. Climatography of the United States, U.S. Dept, of Commerce, 46 pp.

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

- , 1944-Sources of modern sands in the middle Rio Grande Valley, New Mexico.

J. Geol, 52:145.

Rubey, W. W., 193 3 -The size-distribution of heavy minerals within a water-laid sandstone. J. Sed. Petrol. ,3:3.

Russell, R. D., 1937 -Mineral composition of Mississippi River sands. Geol. Soc. Am. Bull., 48:1307.

Sanchez Mejorada, S. H., and Ernesto 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, 1976-Maps showing distribution of heavy minerals of the South Texas Outer Continental Shelf. U.S. Geol. Survey, Misc. Field Studies Map MF-84 1.

Sidwell, R. G., 1941 -Sediments of Pecos River, New Mexico./. Sed. Petrol., 11:80.

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

van Andel, Tj. H., and D. H. Poole, 1960-Sources of recent sediments in the northern Gulf of Mexico. J. Sed. Petrol. , 30:91 .

NOTES SECTION

2 -ALKYL -3-(2-PYRIDYL)-CINCHONINIC ACIDS. Eldon H. Sund, Robert £ Cashon, and Rodney L. Taylor, Department of Chemistry, Midwestern State University, Wichita Falls 76308.

Seven 2-alkyl-3-(2-pyridyl)-cinchoninic adds (2-alkyl-3-(2-pyridyl)-4-carboxyquino- lines) were prepared by the interaction of the requisite l-(2-pyridyl)-2-alkanone with isatin under Pfitzinger conditions (W. Pfitzinger, 1886,/. Prakt. Chem. , 33(2): 100) as modified by Henze and Carroll (H. R. Henze and D. W. Carroll, 1954,/. Amer. Chem. Soc., 76:4580). Table 1 lists yield, decomposition temperature, and elemental analyses for these 2-alkyl-3- (2-pyridyl)-cinchoninic acids.

TABLE 1

2-Alky 1-3 -(2-Pyridyl)-Cinchoninic Acids

R	% Yield	Decomposition Temperature °C±1%	C	Calculated H N	C	Analyses Found H	N
ch3	49	234	72.73	4.55	10.61	72.26	4.67	10.90
c2h3	43	228	73.38	5.04	10.07	73.14	5.29	9.94
«-c3h7	60	202	73.97	5.49	9.59	74.01	5.65	9.41
iso-C3H7	88	207	73.97	5.49	9.59	73.89	5.61	9.31
n- C4H9	69	190	74.48	5.93	9.15	74.18	5.78	9.01
iso-C4H9	44	176	74.48	5.93	9.15	74.38	5.99	8.95
ft-CsHi 1	83	112	75.00	6.25	8.75	74.70	6.24	8.47

Experimental

The l-(2-pyridyl)-2-alkanones were synthesized (T. L. Gore, H. N. Rogers, Jr., R. M. Schumacher, E. H. Sund and T. J. Weaver, 1971,/ Chem . Eng. Data , 16:491), while the re¬ mainder of the reactants were obtained commercially and used without further purification. Elemental analyses were performed by the Huffman Microanaly tical Laboratories, Wheatridge, CO. Melting points determined in either open or sealed capillaries resulted in a slow, indistinct decomposition over a wide temperature range which varied with rate of heating. Sharp de¬ composition points were recorded using a PTC melting point meter (Hot Bench). The ac¬ curacy of the meter is ± 1%. The following example illustrates the general procudure for the synthesis of the 2-alky l-3-(2-pyridyl)-cinchoninic acids.

2-Pen tyl-3-(2-Pyridyl)- Cinchoninic A cid

A mixture of 7.3 g (0.05 mole) isatin, 10.0 g (0.05 mole + 5% excess) of l-(2-pyridyl)- 2-heptanone, and 25 ml of a 34% KOH in a 50% ethanol-water solution and 40 ml of water was stirred on a steam bath for 72 hr. The solvent was removed by a water aspirator until a moist paste remained, which was dissolved in water, and the solution extracted with ether to

94

THE TEXAS JOURNAL OF SCIENCE

remove any unreacted ketone. Addition of concentrated hydrochloric acid to pH 8.0 produced a small amount of brownish gray, noncombustible matter, which was discarded. Further addition of concentrated HC1, to pH 5.5, resulted in the formation of a thick, yellow-tan precipitate which was removed by filtration. The 6.6 g (83%) of the crude 2-pentyl-3- (2-pyridyl)-cinchoninic acid, thus obtained, was treated with activated charcoal and recrys¬ tallized from 95% ethyl alcohol (dec. point 112 C).

We gratefully acknowledge financial support by the Robert A. Welch Foundation (Grant No. AO-413 ). -Reviewed by: Dr. John Fitch, Southwest Texas State University, Department of Chemistry, San Marcos 78666, and Dr. G.