Effects of a threadfin shad introduction upon black crappie and smallmouth buffalo pupulations in Roosevelt Lake
Item Type text; Thesis-Reproduction (electronic)
Authors Beers, Gary Delman, 1942-
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 25/09/2021 03:32:11
Link to Item http://hdl.handle.net/10150/551797 EFFECTS OF A THREADFIN SHAD INTRODUCTION UPON BLACK CRAPPIE AND
SMALLMOUTH BUFFALO POPULATIONS IN ROOSEVELT LAKE
by
Gary Delman Beers
A Thesis Submitted to the Faculty of
FISHERIES MANAGEMENT
In Partial Fulfillment of the Requirements For the Degree of
MASTER OF SCIENCE
In the Graduate College
THE UNIVERSITY OF ARIZONA
1 9 6 5 STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the Dean of the Graduate College when in his judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.
APPROVAL BY THE THESIS DIRECTOR
This thesis has been approved on the date shown below: ACKNOWLEDGMENTS
Sincere gratitude is expressed to Dr. William J. McConnell,
Head of the Arizona Cooperative Fisheries Research Unit, for his timely guidance throughout the study and for his suggestions in the preparations of the manuscript. Helpful advice on the manuscript was also received from Drs. Donald A. Thomson and William B. Heed, University of Arizona.
Special thanks are due Calvin Doner, Marley Conrad, and Leonard
Sexton for valuable help in the field. The cooperation of the Arizona
Game and Fish Department and the Arizona Cooperative Wildlife Research
Unit are gratefully acknowledged.
iii TABLE OF CONTENTS
Page
LIST OF T A B L E S ...... vi
LIST OF FIGURES...... vii
ABSTRACT...... viii
INTRODUCTION ...... 1
STUDY A R E A ...... 3
THREADFIN S H A D ...... 6 VO VOVO vO 0000^ Methods. Data Collection ...... Food Habit Analysis ...... Results...... Length-frequency Distributions Food Habits ......
BLACK CRAPPIE...... 13
Methods...... 13 Lengths and W e i g h t s ...... 13 Food Habit Analysis...... 13 Results...... 16 Food H a b i t s ...... 16 Condition Coefficients...... 16 Scale Analysis...... 19
SMALLMOUTH BU FFALO...... 21 Methods...... 21 Data Collection...... 21 Scale Preparation and Analysis...... 21 Body-scale Relationship ...... 22 Validity of the Scale Method...... 27
iv V
Page
SMAILMOUTH BUFFALO
Results...... 30 Length-frequency of Collections ...... 30 Past Growth R a t e s ...... 30 Condition Coefficients...... 30
DISCUSSION
Threadfin Shad— Black Grapple Relationship ...... 34 Availability of Small Threadfin Shad...... 35 Black Grapple Food H a b i t s ...... 35 Condition of Black Grapple...... 36 Food Habits of Threadfin Shad...... 41 Smallmouth Buffalo--Threadfin ShadRelationship...... 44 Growth Rates, 1952-64 ...... 44
CONCLUSIONS...... 49
APPENDIX A ...... 50
APPENDIX B ...... 51
LITERATURE CITED ...... 52 LIST OF TABLES
Table Page
1. Relative abundance of organisms in the diet of sixty threadfin shad longer than 12 cm (TL) from Roosevelt Lake, Spring of 1965...... 12
2. Percentage frequency of occurrence of each organism in the diet of 47 black crappie from Roosevelt Lake, Spring of 1965...... 17
3. Relation of "K" to length among 1006 black crappie according to 10 mm standard length frequency intervals, 1959-65 captures...... 20
4. Average calculated standard lengths in mm attained at successive ages by the V, VI, VII, and VIII year old smallmouth buffalo from Roosevelt Lake, 1960-65 captures. . 32
5. Relation of "K" to length among 955 smallmouth buffalo according to 20 mm standard length frequency intervals, 1960-65 captures...... 33
6. Average Kg%, values for various black crappie populations in other regions of the United Sta t e s...... 37
7. Comparison of average weights for each 10 mm standard length interval of black crappie from Roosevelt Lake in Arizona and Lake George in Florida...... 39
8. Comparison of the first two year's growth of the smallmouth buffalo spawn from 1952-53, 1957-58, and 1960 ...... 47
9. Average growth rates of V and VI year old smallmouth buffalo prior and subsequent to 1958...... 48
vi LIST OF FIGURES
Figure Page
1. Elevations of Roosevelt Lake, January, 1950 - December, 1964 ...... 5
2. Relative length frequencies of threadfin shad from Roosevelt Lake, 1964-65. Collected with gill nets ..... 10
3. Relative length frequencies of threadfin shad from Roosevelt Lake, October, 1961 and October, 1962. Collected by electric shocker...... 11
4. Relationship of standard length to the combined length of five terminal vertebrae of ten threadfin s h a d ...... 15
5. Length frequency of 96 threadfin shad found in the stomachs of 47 black crappie during the Spring of 1965 . . . 18
6. Photographs of Smallmouth buffalo scales (x4.5)...... 23
7. Scale radius-body length relationship established from measurements of 590 smallmouth buffalo ...... 28
8. Length frequencies of 955 smallmouth buffalo from Roosevelt Lake, 1960-65...... 31
vii ABSTRACT
An investigation of some effects of a recently developed
threadfin shad (Dorosoma petenense, Gunther) population upon the
existing black crappie (Pomoxis nigromaculatus, LeSueur) and small- mouth buffalo (Ictiobus bubalus. Rafinesque) populations, in Roosevelt
Lake, Arizona, was made in 1964 and 1965. Food habit analyses indicated
that small threadfin shad are the most important items in the diet of
the adult crappie. The black crappie was in excellent condition
(KgL = 4.06, based on 1006 fish) apparently as a result of a plentiful
supply of good forage size shad throughout the year. Scales of 480
smallmouth buffalo collected since 1959 reveal that the growth rate
has diminished since the shad introduction in 1958, apparently as a
result of food competition.
viii INTRODUCTION
During the past two decades, the threadfin shad (Dorosoma petenense Gunther) has attained widespread recognition as a valuable forage fish. Many food habit studies demonstrate that the shad is the principal food of predacious fish populations (spotted bass, Dendy,
1946; rainbow trout, Fetterolf, 1957; large-mouth bass, Melander,
1961a; striped bass, Stevens, 1957; and Goodson, 1964; carp, Melander,
1961b; white and black crappies, Finklestein, 1960; and catfish,
Melander, 1961b).
However, there is very little factual information about other
direct or indirect effects a shad introduction may have on the pre
existing fish populations. Fetterolf (1957) reported that the trout
in two Tennessee reservoirs were heavier and grew more rapidly as a
result of foraging upon shad. Recently, Melander (1961a) presents
similar results for the largemouth bass in Lake Mead. More recently,
McConnell and Gerdes (1963) described less favorable effects of a shad
introduction in a small lake in Arizona. They found that, due to
rapid growth of the shad progeny after one short spawning period in
the late spring, yearling centrarchids had difficulty finding suitable
fish forage during the succeeding fall, winter, and spring.
Since 1955, threadfin shad have been introduced into several
large; warm-water reservoirs on the drainages of the Colorado, Salt,
and Gila Rivers in Arizona. The shad are presently thriving in these 2 waters. Some effects of this introduction on a game fish and a rough
fish population in a large reservoir in Central Arizona is the topic
of this research.
The main objectives are: (1) to determine the role of the shad
in the diet and its effect on the condition of the black crappie
(Pomoxis nigromaculatus LeSueur); (2) to compare the growth rates of
the smallmouth buffalo (Ictiobus bubalus Rafinesque) before and after
the shad introduction in 1958; (3) to determine the relative abundance
of the small shad during the winter and spring; (4) to determine the
predacious and competitive aspects of the food habits of the large
shad during the spring.
For increased clarity in reading, particular methods will be
reported in conjunction with the results rather than in a general
methods section. STUDY AREA
Roosevelt Lake, the largest of the six lakes that constitute the Salt River Project water storage system, is located at the confluence of Tonto Creek and Salt River about thirty miles northwest of Globe, Arizona (Gila County). It lies at an elevation of 1900 feet in an area typical of the Lower Sonoran Life Zone. The reservoir has a maximum area of 17,335 surface acres, a maximum depth of 236 feet, and a maximum volume of 1,398,430 acre-feet.
Due to the large seasonal variations in the discharge of the
Salt River there have been large annual fluctuations in the lake level
(Figure 1). During the past fifteen years (1950-1964) the average annual fluctuation has been forty-eight feet. This lake, the first
Reclamation Project in the West, has not been completely full since
1941. Recently, the lake has experienced several drought periods when the deepest portions did not exceed one hundred feet (1947, 1951, 1956, and 1957). Cooperrider and Sykes (1938) explain the forces behind large seasonal fluctuations in the discharge of the Salt River.
Summers at Roosevelt Lake are usually hot and dry. Daily maximum air temperatures are about 100° F., with occasional 1- or 2- week periods with maximums about 105° F. The winters are mild, with only occasional freezing temperatures. A maximum water temperature of nearly 88° F. usually occurs in July and August. Minimum water
3 temperatures are near 56° F. during December and January, A surface temperature of 65° F. is usually attained by March. LAKE ELEVATION IN FEET iue . lvtos f osvl Lk, aur, 90 Dcme, 1964. - December, 1950 January, Lake, Roosevelt of Elevations 1.Figure 1950 'l9 5 f r l952 1 l952 1953 1963 11954111964 r 1962' f ' 1961 1955 1 5 1I960' 1135719561959 r 1958' 'l9 1950 YEAR in THREADFIN SHAD
METHODS
Data collection
Two sets of gill nets were utilized to collect shad for length- frequency analysis. The July sample was taken with bar mesh sizes of
3/8, 7/16, 1/2, 9/16, and 5/8 inch; whereas the remainder of the samples was obtained using bar mesh sizes of 5/16, 7/16, 9/16, 5/8,
11/16, 3/4, 1, and 1-1/16 inch after the first net was torn beyond repair by a-motorbbat in the Fall of 1964.
Variations in the field procedure were held to a minimum, and all sampling was done in the Windy Point area. Nets were set from the surface in ten to forty feet of water, and the fishing effort was constant (twelve hours, usually overnight) for all samples. Total length measurements in centimeters were made on a conventional fish measuring board.
Food habit analysis
The preserved stomach contents of shad larger than twelve centimeters were examined on three occasions for instances of predation, and a rough appraisal made on the relative abundance of each food organism. The foregut (esophagus and gizzard) contents from five shad were pooled, and six representative samples inspected under
6 7 magnification (X50). Four such poolings were examined from each collection date, and the organisms ranked as abundant, common, occasional or rare. The fish were placed in formalin as soon as possible after collection, and were dissected within two weeks.
Relative abundance estimates were based on the number of
organisms seen in the six samples from each pooling. The rating
system was as follows: abundant, over 50; common, 20-50; occasional,
10-20; and rare, less than 10 organisms. The estimates from each of
the four poolings for each date were combined by using the most
frequent ratings as the average for that organism. 8
RESULTS
Length-frequency distributions
The threadfin shad population was sampled once in 1964 and five times in 1965. The length-frequency distributions are given in
Figure 2 for each collection. Although the area of each of the mesh sizes in the first set was constant (100 square feet), the areas in the second set were unequal; the four smaller mesh sizes each had an area of 312 square feet, while the four larger mesh sizes each had an area of 275 square feet. Since the fish caught by all eight mesh sizes on one date would be treated as one sample, a pro rata adjustment was made for the catch from the larger meshes. The number of fish in each length group that were caught in the four larger meshes were expanded by a factor of 1.1. This factor is the ratio of the two different mesh areas, and is based on the assumption that an increase in net area will result in a proportional increase in the number of fish caught. This adjustment was made on the raw data before the histograms for the 1965 collections were constructed.
Additional threadfin shad data were procured from the Arizona
Game and Fish Department who collected the samples with electric shocking gear during the October of 1961 and of 1962 (Figure 3). Shad
larger than 9.0 cm (TL) are more frequent in the 1961 and 1962
collections than in the 1965 collections. This difference is due to
the inherent bias of the electro-fishing method for larger fish. 9
Food habits
Table 1 presents the relative abundance of each item found in the diet of the 60 large shad examined. Most of the stomachs contained a sizeable portion of debris which was more abundant than any other item. Figure 2. Relative length frequencies of threadfin shad threadfin of frequencies length Relative 2.Figure RELATIVE FREQUENCY 80—1 TNAD LENGTH STANDARD I ' I ‘ I ' I ' I ' I 8 0 2 14 12 10 8 6 N CM IN gill nets.gill from Roosevelt Lake, 1964-65. Collected with Collected 1964-65. Lake, fromRoosevelt
UY 94 N =270 1964 JULY A 16 N= 43 1965 JAN E 16 N= 233 1965 FEB PI 95 =lll N 1965 APRIL UE 95 =390 N 1965 JUNE A 16 N=4I7 4 = N 1965 MAY
Figure 3 Figure
NUMBER 20 = § ? 50— 0 — 0 — — . Relative length frequencies of threadfin shad threadfin of frequencies length Relative .
TNAD EGH IN LENGTH STANDARD coe, 92 Cletdb eeti shocker. electric by Collected 1962. October, rmRoeetLk, coe, 1961and October,Lake, from Roosevelt V T I 1 | 1 I 11 10 CM
12
14
16 1
I
C 16 N83 - 1961 OCT OCT 1962 N= 111 N= 1962 OCT
12
Table 1
RELATIVE ABUNDANCE OF ORGANISMS IN THE DIET OF SIXTY THREADFIN SHAD LONGER THAN 12 cm (TL) FROM ROOSEVELT LAKE, SPRING OF 1965
Month of Collection Organism February April May
Non-filamentous green algae AA A
Rotifers A A A
Cladocerans 0 0 0
Copepods 0 0 0
Ostracods - RR
Insect larvae - - R
Organic debris A A A
A = abundant, C = common, 0 = occasional, and R = rare BLACK GRAPPLE
METHODS
Lengths and Weights
The black crappie were weighed on a direct-reading spring scale calibrated in 0.1 and 1.0 pound intervals; these weights were later converted to grams. Total length measurements in centimeters were made on a conventional fish-measuring board. Subsequently, these lengths were changed to standard lengths employing the following formula based on twenty-five crappie: TL = 1.30 SL.
Food habit analysis
The diet of the black crappie was examined once in the fall of
1964 and periodically during the spring of 1965. The preserved stomach contents were inspected under low magnification (X15), and the food organisms identified to family or genus. Stomachs were taken from
crappie collected by gill nets, except for the May samples which were
obtained by creel checks. Stomachs were taken from the crappie in the
angler's catch only if the fish were caught with artificial lures.
Since another forage fish, the red shiner (Notropis lutrensis
Baird and Girard), inhabits Roosevelt Lake, it was necessary to be able
to identify to genus the fish ingested by the black crappie. A
comparative study of the vertebral morphology was completed on the shad
13 14 and the shiner, but the resulting differences were so minute that it was necessary to seek out other differences. Further intergeneric
comparisons revealed that in the shiner the chromatophores on the
dorsal surface of the spinal cord were more numerous than in the shad
even after considerable digestion. When possible, an obvious enteric
difference was utilized to separate the genera; viz., the presence of
a gizzard structure in the shad.
The length of the shad ingested by the black crappie was
pertinent to the goals of this investigation; therefore, the relation
ship of standard length to the combined length of the five terminal
vertebrae preceding the hypural plate was established from ten shad
(Figure 4). 15
■ LENGTH OF 5 TERMINAL VERTEBRAE IN MM
Figure 4. Relationship of standard length to the combined length of five terminal vertebrae of ten threadfin shad. 16
RESULTS
Food habits
Findings from each stomach analysis were totaled for each collection and the per cent frequency of occurrence determined for each item (Table 2).
The combined length of the five terminal vertebrae of each ingested shad was measured to the nearest 0.5 mm. The length-frequency histogram for the 96 threadfin shad consumed by the 46 black crappie is shown in Figure 5. There was no relationship between predator length and prey length. All black crappie within the range of total length sampled (24 - 37 cm) apparently preferred the smallest shad available.
Condition coefficients
Condition coefficients, "K" factors, for 1006 black crappie were computed using the formula:
kSL = W x 105 13 where Kg^ is the condition index for the metric system, W is the weight in grams, and L is the standard length in millimeters. The presenta tion of the "K” factor with a subscript denoting the length used is in accordance with Lagler1s suggestion (1956). To minimize calculations the reciprocal method for "K" factor computations (Carlander, 1953) was used. 17
Table 2
PERCENTAGE FREQUENCY OF OCCURRENCE OF EACH ORGANISM IN THE DIET OF 47 BLACK CRAPPIE FROM ROOSEVELT LAKE, SPRING OF 1965
Tnt-al Length Range was 24-37 cm Month of Collection Number November February March April May X of Fish 1964(13) 1965 (3) (3) (12) (16)
Fish
T. shad 97 100 39 31 48 53
R. shiner 0 0 0 2 1 1
Chironomidae 3 0 61 67 38 41
Scales 0 0 0 0 13 5 iue5 Lnt feuny f 6 hedi hd on i thestomachs in found shad threadfin 96 of frequency Length 5.Figure UMBER f4 lc rpi uig h Srn f 1965.of Spring the during crappie of47black TOTAL TOTAL EGH N CM. IN LENGTH
18 19
This factor expresses, by virtue of its direct relationship to weight and indirect relationship to length, the relative plumpness of a fish. This indicator is directly dependent upon diet, and is often used to show the success of a fish population in a given environment.
Hile (1936) discusses in detail the concept and the significance of the "K" factor.
The computed "K" factors were averaged for 1-cm length groups and yearly collections (Table 3). The 1959, 1961, and 1962 data were furnished by the Arizona Game and Fish Department.
Scale analysis
Included in the initial plan of this investigation was an age and growth study on the crappie in Roosevelt Lake, but it was abandoned because of the confusing nature of the scale markings. The main
problem resided in the failure of the scales to record the changes in
the growth rate. Huish (1953) had similar difficulties aging crappie
in a Florida lake. Table 3
RELATION OF ”K" TO LENGTH AMONG 1006 BLACK CHAPPIE ACCORDING TO 10 MILLIMETER STANDARD LENGTH-FREQUENCY INTERVALS, 1959-1965 CAPTURES (Numbers of fish in parentheses!______10 millimeter Year of Collection standard length intervals______1959______1961______1962______1964______1965______Average
160 - 169 3.08 (1) 3.67 ( 4) 4.22 ( 3) 4.18 ( 2) 4.34 (2) 3.96 (12)
170 - 179 4.18 (13) 4.39 (27) 4.51 (1) 4.33 (41)
180 - 189 4.24 (18) 4.26 (39) 4.26 (10) 4.37 (7) 4.27 (74)
190 - 199 3.31 ( 2) 4.00 (59) 4.16 (37) 3.77 ( 2) 4.04(100)
200 - 209 3.47 ( 4) 4.07(132) 4.22 (36) 3.54 ( 2) 4.09 (6) 4.15(180)
210 - 219 3.85 ( 2) 4.02 (98) 4.11 (29) 4.05 ( 1) 4.04(130)
220 - 229 3.95 (10) 4.09 (40) 4.03 (16) 4.05 (66)
230 - 239 4.00 ( 9) 4.12 (12) 3.94 ( 9) 4.79 (1) 4.05 (31)
240 - 249 4.07 (17) 3.92 (28) 4.27 (14) 4.05 (59)
250 - 259 3.68 ( 2) 3.86 (49) 4.27 (33) 4.01 ( 2) 3.63 (3) 4.01 (88)
260 - 269 3.90 (48) 4.21 (51) 3.56 ( 2) 4.07 (2) 4.05(103)
270 - 279 3.79 (53) 3.95 (23) 4.27 ( 2) 4.28 (2) 3.86 (80)
280 - 289 3.84 (11) 3.93 (15) 3.66 ( 3) 4.03 (4) 3.89 (32)
290 - 298 3.31 ( 2) 3.59 ( 5) 4.28 ( 2) 3.65 (1) 3.68 (10)
Average 3.90 (47) 3.99 (566) 4.17 (337) 4.01 (28) 4.17 (28) 4.06 (1006) SMALLMOUTH BUFFALO
METHODS
Data collection
All data were taken from the smallmouth buffalo in the catch of commercial fishermen. Total length, weight, and scale samples were obtained from 417 specimens during 1964-1965; and in addition, like data for 538 smallmouth buffalo were provided by the Arizona Game and
Fish Department for 1960, 1961, and 1962. Weights were converted from pounds to grams, and total lengths changed to standard lengths in centimeters (TL = 1.2 SL based on sixty buffalofish). Scale samples were taken from the left side of the fish in an area midway between the lateral line and the anterior edge of the dorsal fin, and placed in marked envelopes.
Scale preparation and analysis
The scales were prepared for age and growth determination by soaking them in water for a sufficient time to dissolve the dried mucus and other extraneous material which was clinging to them. Next,
they were examined against a light source and selected for their
symmetry, clearness of markings, and average size in relation to the
other scales in the sample. Regenerate and lateral line scales were
rejected as unsuitable. Finally, the scales were dried and mounted
21 22 between glass slides. An Argus "300" slide projector cast the scale image on which radial measurements and age determinations were made.
The slide holder was replaced with a specially built holder that accommodated the scale mounts. The annuli of the projected scale image (X10) were identified and indicated on nomograph tags in the conventional manner (Carlander and Smith, 1944).
The year marks (annuli) on the scales of the smallmouth buffalo were unusually definite (Figure 6) and the ages were determined with ease. Structures which might be considered as false or accessory annuli were rare among the scales examined for this study.
Body-scale relationship
Before past growth rates could be estimated by the nomograph method, it was necessary to establish a definite relationship between the body length and the scale size. Since this relationship should be founded on a wide range of data, and since the smallest buffalo fish collected during the study was 47 cm (TL), it was essential to procure some small fish elsewhere. Data from twelve specimens (25-40 cm, TL) of this species were obtained from Mr. James Stevenson (Fish Farm
Experimental Station, Stuttgart, Arkansas), and were only used in the body-scale relationship. Scale measurement values were clustered in
1-cm groups, and the average standard length was calculated for each group. The scatter of these points when plotted is best fitted with a
linear regression line (Figure 7) represented by the formula:
L = 11 + 1.4 S 23
Figure 6
PHOTOGRAPHS OF SMALLMOUTH BUFFALO SCALES (X 4.5)
Date of TL SL WT KSL Fig. 6 Capture mm mm grms factor Age
A March 1960 625 520 3810 2.694 VII
B June 1960 642 535 3765 2.621 VII
C May 1961 622 518 3765 2.708 VIII
D June 1964 522 436 2087 2.518 III
E June 1964 522 436 2268 2.736 VI
F June 1964 514 427 1950 2.504 VI Figure 6a Figure- 6B. NO 4S ro Figure 6C Figure 6D Ul \
Figure 6E Figure 6 f < IX) 27 where L is the standard length, 11 is the Y intercept, 1.4 is the
slope of the line, and S is the scale radius. Since the body-scale
relationship is linear, the direct-proportion nomograph method
described by Carlander and Smith (1944) was employed to estimate the
previous growth rates.
It should be noted that the Y intercept of 11 is not
necessarily the length of the fish at the time of scale formation,
but is a correction factor based on the available data. Proportional
growth between scale and body occurs only after the scale has formed.
Validity of the scale method
The validity of the scale method of determining age and
calculating growth rates of fishes has been proved for a variety of
species. Classic papers in this realm are: Taylor, 1916; Greaser,
1926; Van Oosten, 1926; Hile, 1936 and 1941; and Cooper, 1951. Of
prime concern to this study are the investigations that have
established this validity for the family Catostomidae: Stewart (1926)
on white sucker (Catostomus commersonnii); Frey and Pedracine (1938)
on bigmouth buffalo (Ictiobus cyprinellus), smallmouth buffalo
(I. bubalus), and black buffalo (I. niger); Spoor (1938) on white
sucker; Eschmeyer, Stroud, and Jones (1944) on bigmouth, smallmouth,
and black buffalo fish, river carpsucker (Carpiodes carpio). quillback
(C. cyprinus), and golden redhorse (Moxostoma erythrurum); Schoffman
(1944) on smallmouth buffalo; Martin (1963) on smallmouth buffalo; and
others. iue7 Saerdu-oy eghrltosi salse from established relationship length radius-body Scale 7. Figure - - - - - 70 -7 -10
STANDARD LENGTH IN CM i i i r i i i i i j j 1 1 2 2 3 35 30 25 20 15 10 5 ______esrmns f50salot ufl. rces enclose Brackets buffalo. smallmouth 590 of measurements aa rmAkna specimens. from Arkansas data ANTERIOR SCALE RADIUS RADIUS SCALE ANTERIOR | ______i ______| ______xO I CM IN ) (xIO 1 ______| ______L
28 29
In the present study, two criteria were investigated to determine the validity of annuli as year marks for this ictiobine species in Roosevelt Lake. First, periodic inspections of the scales revealed that an annulus was laid down once a year during May. This evert was observed during 1964 and 1965. Second, several comparisons of the growth marks on the opercular bones with those on the corresponding scales were made, and a high agreement was disclosed. 30
RESULTS
Length-frequency of collections
Length-frequency histograms (Figure 8) indicate that there has been a gradual decline in the size of the buffalo fish since 1960.
Because the commercial fishermen have used the same mesh size
(3-1/2 inch, bar measure) during the past six years, this decline is real and not due to increased catch of small fish due to decreased mesh size.
Past growth rates
Average past growth rates for age groups from each year were arranged to facilitate interyear comparisons (Table 4). Age groups that contained less than eight fish were not included in these results.
Condition coefficients
"K" factors for the smallmouth buffalo were computed in the same manner as described for the black crappie. The computed factors were averaged for 2-cm groups and yearly collections (Table 5). If an obvious error in length or weight measurement was realized through an unrealistic value for "K", the fish was excluded from all portions of the study. iue8 Lnt feunis f 5 salot ufl from buffalo smallmouth 955 of frequencies Length 8.Figure NUMBER 100^' 100 100 100 100 - 0 8 4 0- 4 - 0 6 20- 0- 0 — 0- 0- 0- - — - -
osvl ae 1960-65. Lake, Roosevelt 04 44 8 02 4 68 06 6 6 68 66 64 62 60 5658 54 5052 48 46 44 42 40 TNAD EGH N CM IN LENGTH STANDARD ---- -
90 N = 3I8 1960 1961 92 =73 N 1962 94 N |33 s 1964 95 =284 N 1965 s 147 Ns
Table 4
AVERAGE CALCULATED STANDARD LENGTHS IN MM. ATTAINED AT SUCCESSIVE AGES BY THE V, VI, VII, AND VIII YEAR OLD SMALLMOUTH BUFFALO FROM ROOSEVELT LAKE, 1960-65 CAPTURES ______Years of growth since 1958 (1959-64) are enclosed in boxes______Last Number Complete of Average calculated standard length in mm. Age Year of Fish at each annulus Group Growth Aged 1 2 3 4 5 6 7 8
V 1959 29 185 316 391 432 472 1963 22 163 278 348 383 413 1964 25 165 290 361 405 431
VI 1959 71 170 300 381 429 458 483 1960 14 170 290 361 413 440 463 1961 9 148 274 330 364 393 420 1963 14 150 1I" 270 340 383 413 437 1964 74 165 288 361 401 430 450
1959 53 190 338 430 488 526 557 581 1960 22 160 285 366 415 443 | 468 487 1961 28 160 275 340 376 407 432 453 1963 8 160 282 | 345 375 401 422 446 1964 43 155 275 348 383 414 436 450
VIII 1959 27 158 285 360 395 432 458 482 498 1960 21 141 266 345 376 410 435 460 474 1961 17 145 240 310 331 373 | 401 425 442 1964 11 150 265 330 358 391 418 442 450
u> M Table 5
RELATION OF "K" TO LENGTH AMONG 955 SMALLMOUTH BUFFALO ACCORDING TO 20 MM STANDARD LENGTH FREQUENCY INTERVALS, 1960r65 CAPTURES Numbers of Fish in Parentheses
SL Year (cm)______1960______1961______1962______1964______1965______Average
38.0- 39.9 2.49 ( 1) 2.49 ( 1) 40.0- 41.9 2.61 ( 4) 2.42 ( 1) 2.75 (22) 2.49 (18) 2.47 (45) 42.0- 43.9 2.54 ( 9) 2.66 ( 7) 2.40 ( 3) 2.41 (32) 2.40 (59) 2.42 (HO) 44.0- 45.9 2.40 (44) 2.50 (26) 2.37 (20) 2.35 (44) 2.34 (148) 2.24 (282) 46.0- 47.9 2.36 (56) 2.46 (51) 2.38 (22) 2.23 (16) 2.32 (31) 2.37 (176) 48.0- 49.9 2.35 (56) 2.40 (32) 2.33 (16) 2.41 (14) 2.31 (12) 2.36 (130) 50.0- 51.9 2.43 (45) 2.40 (16) 2.27 ( 7) 1.90 (-3) 2.42 ( 6) 2.42 (77) 52.0- 53.9 2.41 (40) 2.38 ( 6) 2.30 ( 4) 2.24 ( 2) 2.40 (52) 54.0- 55.9 2.38 (28) 2.33 ( 7) 1.91 ( 1) 2.30 ( 5) 2.36 (41) 56.0- 57.9 2.41 (13) 1.48 ( 1) 2.35 (14) 58.0- 59.9 2.79 ( 4) 2.60 ( 1) 2.53 ( 5) 60.0- 61.9 2.42 ( 4) 2.35 ( 1) 2.45 ( 5) 62.0- 63.9 2.95 ( 1) 2.95 ( 1) 64.0- 65.9 2.81 ( 6) 2.26 ( 2) 2.75 ( 1) 2.65 ( 9) 66.0- 67.9 2.63 ( 8) 2.63 ( 8)
2.42 (318) 2.45 (147) 2.35 (73) 2.41 (133) 2.36 (284) 2.35 (955) DISCUSSION
THREADFIN SHAD -- BLACK CRAPPIE RELATIONSHIP
Availability of small shad
Length frequency distribution data for 1965 (Figure 2) indicate that at least 50 per cent of the shad in each collection were under 8.0 cm (TL). Since shad have a rapid growth rate during their first summer of life (Kimsey, 1958; McConnell and Gerdes, 1964), it is certain that most of these fish were spawned in the late summer or early fall of 1964. Data collected in July of 1964 (Figure 2) and in the falls of 1961 and 1962 (Figure 3) show a similar abundance of small shad. These length distribution data clearly indicate that small shad are always available as a plentiful forage source in
Roosevelt Lake. This situation is probably the outcome of continual spawning by the threadfin shad population from early to late summer.
A fall spawn in the forage fish population is very important to a warm water fishery. Kimsey (1957) stresses the need for a
satisfactory bass-of-the-year/forage fish ratio in fall when the bass
turn to a fish diet. McConnell and Gerdes (1964) relate an instance where, in a small lake, the failure of shad to spawn in fall resulted
in the lack of adequate forage for the yearling centrarchids during the
subsequent winter and spring. This condition was observed for three
34 35 consecutive years. Menu (1965) observed that the rapid growth of the shad-of-the-year in some Texas ponds limits their forage value later in the year.
Black crappie food habits
Only four kinds of food items were found in the stomachs of
47 adult crappie taken from Roosevelt Lake during November (1964) and
the spring of 1965 (Table 2). Threadfin shad were the prime forage in
the diet of the crappie, and its importance is attested to by the
occurrence of at least one shad in each stomach. Chironomid larvae
and pupae were rare among the fall foods, but common among the spring
foods. This shift in feeding habits could be the consequence of either
of two conditions: the immature insects are more abundant in the
spring; or, as Reid (1950) found in a Florida crappie population,
crappie eat more insects and less fish during the breeding season
(February-April). The plains red shiner was an uncommon item in the
diet which is not unexpected since it is present in limited quantities
in the lake. Some of the fish sampled in May had ingested several,
large, crappie scales. This was probably the result of "nipping11 or
incidental intake, not of lepidophagy.
The frequency of fish exceeded that of all other food items in
the diet of the crappie. This difference would have been much greater
if the foodstuffs were compared on a volumetric basis, because then
the shad would probably exceed 90 per cent of the diet. Like findings
are reported for the black crappie in Florida (Reid, 1950; Huish, 1958),
in Iowa (Kutkuhn, 1955), and in Arkansas (Finklestein, 1960). 36
A very important feature of this predator-prey relationship is that the crappie, although large enough (24-37 cm, TL) to consume longer quarry, were feeding predominately upon 6-7 cm (TL) shad
(Figure 5). The scarcity of large shad and the abundance of small shad in the lake are the main reasons underlying this size preference.
The relation of ingested shad lengths (Figure 5) to netted shad lengths from the spring of 1965 (Figure 2) is the critical consideration. Since the mode of each shad collection from 1965
(Figure 2) agrees with the mode in Figure 5, this indicates that the crappie were feeding upon the more available forage source, small shad.
The absence of large shad (longer than 9.0 cm, TL) in the diet attests to its scarcity as forage in the lake. Thus, without the progeny from the past fall spawn, the large crappie in Roosevelt Lake during the winter and spring months would have to rely on a less plentiful food source, the larger shad, and insects as the main staples of their diet.
Condition of the black crappie
Apparently as a result of the food and habitat available, the black crappie in Roosevelt Lake are in excellent condition. Carlander
(1944) lists the standards used for judging the condition of the crappie: poor, under 1.80; average, 2.10-2.60; and excellent, over
3.30. Data from collections during the last seven years givenan average Kg^ value of 4.06 for 1006 crappie. Since this value is higher
than those reported for some of the better populations of this species
(Table 6), it follows that the Roosevelt crappie are relatively heavier. 37
Table 6
AVERAGE Ksl VALUES FOR VARIOUS BLACK CRAPPIE POPULATIONS IN OTHER REGIONS OF THE UNITED STATES
No. Ave. SL Location Fish K Ranee Source
California (Clear Lake) 50 3.44 156-314 Murphy (1951)
Iowa 56 3.39 86-312 Carlander (1949)
Tennessee (Cherokee Lake) 85 3.38 — — Stroud (1949)
Florida (Lake George) 232 3.28 160-290 Huish (1953)
Iowa (Clear Lake) 169 3.25 79-252 Erickson (1952)
Indiana (East Lake) 33 3.09 70-189 Lewis (1950)
Minnesota 1917 3.08 70-380 Carlander (1944)
Tennessee (Norris Res.) 643 3.02 55-317 Stroud (1948) 38
Comparisons of overall Kg%, averages are usually liable to criticism; therefore, the average weight for each 1-cm length group
(SL) of the black crappie in Roosevelt Lake were compared with Lake
George, Florida. This Florida lake was used because, like Roosevelt,
it provides an abundant food supply (Huish, 1958) and a long growing
season (Huish, 1953) for the centrarchids. In addition, the plankton-
shad-piscivorous fish food chain is well-developed in both waters.
This collection (Table 7) demonstrates that for most length sizes the
Arizona population are 20 per cent heavier than their Florida counter
parts.
Since condition is directly related to diet, it is probable
that the excellent condition of the crappie in Roosevelt Lake is a
consequence of the threadfin shad introduction in the late 1950s.
In personal conversation Jack Hemphill, former fisheries
supervisor in Arizona, stated that during the early and middle 1950's
the crappie population in the reservoir consisted of few individuals
longer than 25 cm (TL); and, in addition, these fish were "knife-
bellied." In 1957 when the lake level was low, a rough fish removal
program was conducted by the Arizona Game and Fish Department. Marley
Conrad, commercial fisherman at Roosevelt Lake, participated in this
program and recalled that hundreds of stunted, "water-thin" crappie
all less than 25 cm (TL) were collected during the seining phases.
It is obvious from the 1959-65 collections (Table 3) that a
sizeable number of the crappie are attaining a greater size
(25-38 cm, TL) than in pre-1958 times. The average "K" factors show 39 Table 7
COMPARISON OF AVERAGE WEIGHTS FOR EACH 10 MM STANDARD LENGTH INTERVAL OF BLACK CRAPPIE FROM ROOSEVELT LAKE AND LAKE GEORGE
Lake George Roosevelt Lake 100 x (B) - (A) SL cm (A) Wt. (B) Wt. (A)
16.0 - .9 144 (37) gms 178 ( 12) gms 23.6 %
17.0 - .9 1165 (34) 232 ( 41) 44016
18.0 - .9 197 (33) 271 ( 74) 37.5
19.0 - .9 247 (35) 299 (100) 21.0
20.0 - .9 272 (20) 357 (180) 31.2
21.0 - .9 317 (22) 402 (130) 26.8
22.0 - .9 381 (21) 461 ( 66) 20.1
23.0 - .9 476 ( 8) 525 ( 31) 10.2
24.0 - .9 549 (13) 596 ( 59) 8.6
25.0 - .9 589 ( 2) 665 ( 88) 12.9
26.0 - .9 758 ( 4) 754 (103) - 0.5
27.0 - .9 869 ( 1) 803 ( 80) - 7.6
28.0 - .9 947 ( 1) 900 ( 32) - 5.0
29.0 - .9 1014 ( 1) 944 ( 10) - 7.0
(232) (1006) Ave. =20.7
The last column lists the per cent heavier the Arizona crappie were than the Florida crappie. Numbers of fish in parentheses. 40 that along with a greater length these fish are heavier than before the introduction.
Melander (1961a) found similar increases in the weight of the largemouth bass in Lake Mead after threadfin shad was introduced as forage.
In summary, the threadfin shad are providing excellent forage for the black crappie. In comparison with some of the better crappie populations in other regions of the country, it is evident that the
Roosevelt population is definitely in better condition. On the basis of reports concerning the size and condition of this species prior to
1958, it is concluded that the introduction of the shad was responsible
for the subsequent improvement in condition of the crappie population.
The occurrence of shad as the major item in the diet of this
centrarchid is certain evidence that a satisfactory predator-prey
relationship exists between these two populations. A fall-spawn
provides excellent forage size shad for the large and, probably, the
small crappie during the critical winter and spring months. FOOD HABITS OF THE SHAD
The most numerous items in the diet of the large shad during the spring were algae, rotifers, and debris (Table 1). The bulk of
the items were Entromostraca (cladocerans, copepods, and ostracods).
Although volumetric determinations were not made during the analyses,
McConnell and Gerdes (1963) list the relative volume factors for most
of the items found in the stomachs of shad. They are: non-filamentous
algae, 8; rotifers, 2000; copepoda, 24000; ostracod, 36000; and
cladocera 36000. It is easy to see that the algae would have to be
3000-5000 times, and the rotifers 12-18 times as numerous as the
entromostraca to equal the latter's volume. The smaller items were
not this abundant and observations found that most of the bulk of the
stomach contents were debris and crustacea. Only one insect larva was
located among the stomach contents of the 60 large shad.
These cursory results are similar to the findings of more
detailed analyses of the diet of the threadfin shad in Arizona waters
by Kimsey (1958), Haskell (1958), and Gerdes and McConnell (1963).
Nothing was found among the stomach contents of the spring
shad samples that would indicate predation upon the early life history
stages of other fishes. Of the several studies concerning the food
habits of the shad, only one has reported definite predation:
Kimsey (1958) found that shad in the Salton Sea had a strong preference
for the eggs and the fry of the gulf croaker, Bairdiella icistius.
41 42
It is common knowledge that the diet of the black crappie and the smallmouth buffalo at some stage of their life history, like that of the shad, consists mostly of Entromostraca and immature insects.
Young crappie forage upon these items until they are large enough to assume a piscivorous role in the trophic system (Reid, 1950;
Erickson, 1951; Kutkuhn, 1955; Huish, 1958; Neal, 1961; and McConnell and Gerdes, 1964). On the other hand, buffalo fish feed upon microplankton (protozoa, algae, and rotifers) and Entromostraca as juveniles, and mainly upon the latter food and chironomidae as adults
(Forbes and Richardson, 1920; Bailey and Harrison, 1945; Moen, 1951;
Scidmore and Woods, 1960; and Johnson, 1963).
The extent of possible interspecific competition for food among these fishes was not included in this study; however, a few
statements can be made.
The severest competition would exist between the shad and the buffalo fish, since throughout life they occupy a similar food niche,
limnetic plankton feeders. A study of the food habits and feeding
habits of both species would be necessary before the aspects of
competition could be defined. The results of the buffalo fish growth
study (see next section in Discussion) indicate that probably the shad
are contending with the buffalo fish for food.
The shad-crappie food competition situation would not be
expected to be as severe as the foregoing, because crappie consume
significant amounts of entromostrans only during their yearling stage.
Beyond this stage they display a definite preference for insects and
fish (Kutkuhn, 1955). It must be kept in mind that the fact that two fish utilize the same food organisms at the same time does not constitute by definition a competitive situation. Many other factors, such as forage ratios, must be examined before any conclusion can be issued.
Several competent workers have pointed out that severe competition among fishes for food seldom exists for any length of time (Lagler,
1944; Frost, 1945; Hartley, 1948; Starrett, 1950; Nikolskii, 1953;
Larkin, 1956; and Kawanabe, 1959). SMALLMOUTH BUFFALO--THREADFIN SHAD RELATIONSHIP
Growth rates, 1952-64
The interyear age-group comparisons of growth rates (Table 7) fail to unveil any definite relationships, but do disclose a common trend. In general, the fish whose last complete growing season was in
1959 or 1960 did experience the best growth history and as a result attained a longer size at death. This trend is, also, shown in a simpler fashion by the length-frequency distributions of the yearly collections (Figure 8). This change in the length distributions is not the result of increased commercial fishing pressure that would
remove the larger fish and harvest most fish at a smaller size and earlier age. These two facts probably indicate that the conditions in
Roosevelt Lake during the years 1952-60 were more favorable than during the years 1957-64 for growth of the buffalo fish.
The fluctuating water level of the reservoir (Figure 1) and the
introduction of the threadfin shad were considered as possible causes
for the decline in lake conditions for buffalo fish growth. A rising water level is known to trigger breeding behavior (Walker and Frank,
1952; Canfield, 1922; Swingle, 1957), and a falling water level
severely limits reproduction (Shields, 1957) and food supply
(Eschmeyer, 1949) of the smallmouth buffalo. The shad-buffalo fish
relationship has not been commented upon by previous workers.
44 45
Logically, an increase in the water level over a period of several years would result in better growth conditions for the buffalo fish, by providing additional living area and food. Figure 1 shows that the water levels rose during the 1957-59, and dropped during the
1952-56 and 1960-64 year blocks. Therefore, if a comparison was to be made of the first two years growth of the spawn from 1952-53, 1957-58, and 1960; the middle group should display the best relative growth, if water levels are a serious factor in influencing growth rates. This comparison is shown in Table 8 and it is concluded that the fish who experienced rising water levels throughout their first two years of life had the poorest growth rate. Based on these limited findings it can only be concluded that the water level has no direct effect on the growth of the smallmouth buffalo in Roosevelt Lake.
There is a good probability that the threadfin shad introduction and subsequent population irruption has had an important effect on the growth rates of the buffalo fish since 1958. Six and seven year old fish in 1960 whose last complete year of growth was 1959 were used to represent pre-1958 growth rates. This was done by subtracting the increment of growth gained in 1959 which theoretically would result in the sizes of the five and six year old fish in 1958. Post-1958 growth rates were determined from two sets of five year old fish (1963 and
1964) and one set of six year old fish (1964). The average growth rates were calculated (Table 9) and comparison shows that the post-1958
group is definitely smaller at each annulus. 46
Although the decline in the growth rate of the buffalo fish and the introduction of the threadfin shad are synchronous further investigation would be necessary to establish a certain correlation.
An average "K" factor of 2.35 was found for the 955 buffalo fish collected since 1960. These fish are in average condition in comparison with other regions of the United States (see Carlander,
1953, pp. 77-79, 309-310). 47
Table 8
COMPARISON OF THE FIRST TWO YEAR'S GROWTH OF THE SMALUIOUTH BUFFALO SPAWN FROM 1952-53, 1957-58, and 1960
First Two Years1 Year Age at Year of Number of Growth Spawned Capture Capture Fish 1 2
1952 VIII 1959 27 158 285 1953 VIII 1960 21 141 266 1953 VII 1959 53 190 338
Ave. = 171 309
1957 VIII 1964 11 150 265 1957 VII 1963 8 160 282 1958 VII 1964 43 155 275 1958 VI 1963 14 150 270
Ave. = 154 273
1960 V 1964 25 165 290
Ave. = 165 290 48
Table 9
AVERAGE GROWTH RATES OF V AND VI YEAR OLD SMALLMOUTH BUFFALO PRIOR AND SUBSEQUENT TO 1958
Standard Length in mm at Each Annulus Age No. 1 2 3 4 5 6
(Last Year Growth) 1958
V (VI 1959) 71 170 300 381 429 458
VI (VII 1959) 53 190 338 430 488 526 557
Ave. 180 319 406 459 487 557
(First Year Growth) 1959
V (1963) 22 162 278 348 383 413
V (1964) 25 165 290 361 405 431
VI (1964) 74 165 288 361 401 430 450
Ave. 164 286 358 397 426 450 CONCLUSIONS
Since its introduction in 1958, the threadfin shad has apparently had important effects on the black crappie and smallmouth buffalo populations in the reservoir.
The effects on the black crappie population have been beneficial. The shad is the main food item in the adult crappie
diet, and is probably the prime cause of the excellent condition
(Kg^ = 4.06) of the crappie population since 1958. Fall spawns provide
an abundance of small forage fish during the succeeding winter and
spring.
The effects on the ictiobine population have been possibly
detrimental. The growth rate of the smallmouth buffalo fish has
diminished since 1958.
49 APPENDIX A
History of Theodore Roosevelt Dam
The construction of T. Roosevelt Dam, the first Reclamation
Project in the West, was started in the fall of 1906, and completed in the spring of 1911 at a total cost of ten million dollars. On
March 18, 1911, when President T. Roosevelt presided at its dedication, it was the largest man-made dam in the world. Built of large blocks
(10-50 cubic feet) of calcareous sandstone, the dam is 245 feet high,
165 feet thick and 210 feet wide at its base, and 16 feet thick and
700 feet wide at its crest. Today, this structure is the largest masonary dam in the world.
The Salt River Project water storage system consists of six dams and their respective lakes: Roosevelt (Roosevelt Lake), Horse- mesa (Apache Lake), Mormon Flat (Canyon Lake), and Stewart Mountain
(Saguaro Lake) on the Salt River; Bartlett (Bartlett Lake) and Horse shoe (Horseshoe Lake) on the Verde River. These dams and their lakes serve three multifarous purposes: irrigation, hydro-electric power, and recreation.
50 APPENDIX B
The common ictiobine species in Roosevelt Lake, 1964-65
Several authorities (Moore, 1957; Miller and Lowe, 1964) report that the bigmouth buffalo, Ictiobus cyprinellus, is the connnon species in the Salt River drainage. On the basis of recent observations
(1964-65) the abundance of this species in Roosevelt Lake is questioned.
The species that appears in the commercial catch has a subterminal mouth which identifies it as the smallmouth buffalo, 1^. bubalus, or the black buffalo, JL niger. Using the key given in Forbes and Richardson
(1920), the Roosevelt species was identified as 1^. bubalus.
Krumholz (1943) found differences in the Webberian ossicles of these
three closely related species that is of useful taxonomic value. The
ossicles from three specimens were removed, examined, and identified
as I. bubalus.
51 LITERATURE CITED
Bailey, R. M ., and H. M. Harrison. 1945. Fishes of Clear Lake, Iowa. Iowa State Coll. J. Sci., 20(1):57-77.
Canfield, H. L. 1922. Care and feeding of buffalo fish in ponds. U.S. Bur. Fish., Econ. Giro., No. 56, 3 pp.
Carlander, K . D. 1944. Notes on the coefficient of condition, K, of Minnesota fishes. Minn. Bur. Fish. Res. Invest. Rept. 41, revised, 40 pp. typed.
1949. Project No. 39. Yellow pikeperch management. Prog. Rept. Iowa Coop. Wildlife Res. Unit, Jan.-Mar.
1953. Handbook of freshwater fishery biology with the first supplement. Wm. C. Brown Co., Dubuque, Iowa. 429 pp.
Carlander, !. D., and L. L. Smith. 1944. Some uses of nomographs in fish growth studies. Copeia, 1944(3):157-162.
Cooper, E. 1951 Validation of the use of scales of brook trout, Salvelinus fontinalis, for age determination. Copeia, 1951(2):141-148.
Cooperrider, K. L., and G. G. Sykes. 1938. The relationship of stream flow to precipitation on the Salt River watershed above Roosevelt Dam. Arizona Univ. Agr. Expt. Sta. Bull. 76, 69 pp.
Greaser, C. W. 1926. The structure and growth of the scales of fishes in relation to the interpretation of their life history, with special reference to the sunfish Eupomotis gibbosus. Misc. Publ. Univ. Mich. Mus. Zool., 17:1-82.
Bendy, J. S. 1946. Food of several species of fish of Norris Reservoir, Tennessee. J. Tenn. Acad. Sci., 21(1):105-127.
52 53
Erickson, J. G. 1952. Age and growth of the black and white crappies, Pomoxis nigro-maculatus (LeSueur) and P. annularis (Rafinesque), in Clear Lake, Iowa. Iowa State Coll. J. Sci., 26(3): 491-505.
Eschmeyer, R. W. 1944. Harvesting the fish crop. Trans. N. Am. Wildl. Conf., 9 :202-211.
Eschmeyer, R. W . , R. H. Stroud, and A. M. Jones. 1944. Studies of the population on the shoal areas of a TVA main-stream reservoir. Rept. Reelfoot Lake Biol. Stat., 8:70-122.
Fetterolf, C. M . , Jr. 1957. Stocking as a management tool in Tennessee reservoirs. Proc. 10th Ann. Conf., S. E. Assoc. Game and Fish Comm., pp. 275-284.
Finklestein, S. L. 1960. A food study of four Centrarchidae of Lake Fort Smith, Crawford County, Arkansas. Progr. Rept., Appendix B, Federal Aid to Fisheries Project, Arkansas F-8-R-3 (Experimental Introduction of threadfin shad). 64 pp. Mimeo.
Forbes, S. A., and R. E. Richardson. 1920. The fishes of Illinois. Illinois Nat. Hist. Survey Bull., 8:1-358, Second Ed.
Frey, D. G., and H. Pedracine. 1938. Growth of the buffalo in Wisconsin lakes and streams. Trans. Wise. Acad. Sci., Arts, and Lett., 31:513-525.
Frost, W. E. 1 1946. On the food relationships of fish in Windermere. Biol. Jaarb. Dodonaea, 13:216-231.
Gerdes, J. H., and ton. J. McConnell. 1963. Food habits and spawning of the threadfin shad in a small, desert impoundment. Ariz. Acad. Sci., 2(3):113-116.
Goodson, L. F., Jr. 1964. Diet of striped bass at Millerton Lake, California. Calif. Fish and Game, 50(4):307.
Hartley, P. H. T. 1948. Food and feeding relationships in a community of fresh water fishes. J. Anim. Ecol., 17:1-14. 54
Huish, M. T. 1953. Life history of the black crappie of Lake George, Florida. Trans. Am. Fisheries Soc., 83:176-193.
1957. Food habits of three Centrarchidae in Lake George, Florida. Proc. 11th Ann. Conf., S. E. Assoc. Game and Fish Comm., pp. 293-302.
Hile, R. 1936. Age and growth of the cisco, Leucichthys artedi (LeSueur), in the lakes of the northeastern highlands, Wisconsin. Bull. U.S. Bur. Fish., 44:265-428.
1941. Age and growth of the rockbass, Ambloplites rupestris (Rafinesque), in Nebish Lake, Wisconsin. Trans. Wise. Acad. Sci., Arts, and Lett., 33:189-337.
Johnson, R. P. 1963. Study on the life history and ecology of the bigmouth buffalo. J. Fish. Res. Bd. Canada, 20(6):1397-1429.
Kawanabe, H 1959. Food competition among fishes in some rivers of Kyoto Prefecture, Japan. Mem. Coll. Sci. Univ. Kyoto, Series B, 26(3):253-268.
Kimsey, J. 1957. Fisheries problems in impounded waters of California and the lower Colorado River. Trans. Am. Fisheries Soc., 87:319-332.
1958. Possible effects of introducing threadfin shad (Dorosoma petenense) into the Sacramento-San Joaquin delta. Inland Fisheries. Admin. Rept. No. 58-16, Calif. Dept. Fish and Game. 21 pp. Mimeo.
Krumholz, L . A. 1943. A comparative study of the Webberian ossicles in North American Ostariophysine fishes. Copeia, 1943 (1):33-40.
Kutkuhn, J. H. 1955. Food and feeding habits of some fishes in a dredged Iowa Lake. Iowa Acad. Sci., 62:576-588.
Lagler, K. 1944. Problems of competition and predation. Trans. N. Am. Wildl. Conf., 9:212-219.
1956. Freshwater fishery biology. Wm. C. Brown Co., Dubuque, Iowa. 421 pp. 55
Lewis, W. M. 1950. Fisheries investigations on two artificial lakes in southern Iowa. II. Fish populations. Iowa State Coll. J. Sci., 24(3):287-324.
Martin, R. E. 1963. Growth and movement of smallmouth buffalo, Ictiobus bubalus (Rafinesque), in Watts Bar Reservoir, Tennessee. Diss. Abst. 1963, 24(10):3030-3031.
Martin, R. E., S. I. Auerbach, and D. J. Nelson. 1964. Growth and movement of smallmouth buffalo in Watts Bar Reservoir, Tennessee. ORNL-3530, UC-48-Biology and Medicine, 100 pp.
McConnell, Wm. J., and J. H. Gerdes. 1964. Threadfin shad, Dorosoma petenense. as food of yearling centrarchids. Calif. Fish and Game, 50(3):170-175.
Melander, W. C. 1961a. Completion Report, Job R-l, Federal Aid to Fisheries Project, Arizona F-7-R-5 (Studies of the ecology, distribution, migration, and utilization of the threadfin shad, Dorosoma petenense. in Lake Mead).
1961b. Completion Report, Job R-l, Federal Aid to Fisheries Project, Arizona F-7-R-4 (Studies of the ecology, distribution, migration, and utilization of the threadfin shad, Dorosoma petenense, in Lakes Mead and Mohave).
Menn, C. T. 1965. Completion Report, Job E-6, Federal Aid to Fisheries Project, Texas F-6-R-12 (Experimental stocking of large- mouth bass and threadfin shad in ponds in south Texas).
Miller, R. R., and C. H. Lowe. 1964. An annoted check list of the fishes of Arizona. In The vertebrates of Arizona, by C. H. Lowe. Univ. Ariz. Press, pp. 133-152.
Moen, T. 1954. Food of the bigmouth buffalo in northwest Iowa lakes. Iowa Acad. Sci., 61:561-569.
Moore, G. A. 1957. Fishes. In Vertebrates of the United States, by W. F. Blair et al. McGraw-Hill Book Co. Inc., pp. 31-210. 56
Murphy, G. 1951. The fishery of Clear Lake, Lake County, California. Calif. Fish and Game, 37(4):439-484.
Neal, R. A. 1961. White and black crappies in Clear Lake, summer 1960. Iowa Acad. Sci., 68:247-253.
Nikolskii, 1953. On the law of the food relationships among freshwater fishes. Proc. Symp. Ichthyol., pp. 261-281.
Reid, G. K. Jr. 1950. Food of the black crappie, Pomoxis nigromaculatus (LeSueur), in Orange Lake, Florida. Trans. Am. Fisheries Soc., 79:145-154.
Schoffman, . J. 1944. Age and growth of the smallmouth buffalo in Reelfoot Lake, Tennessee. J. Tenn. Acad. Sci., 19(l):3-9.
Scidmore, W J., and D. E. Woods. 1960. Some observations on competition between several species of fish for summer food in four southern Minnesota lakes in 1955, 1956, and 1957. Minn. Fish and Game Invest., Fish Series, 2:13-24.
Shields, J. T. 1957. Experimental control of carp reproduction through water drawdowns in Fort Randall Reservoir, South Dakota. Trans. Am. Fisheries Soc., 87:23-33.
Spoor, W. A 1939. Age and growth of the sucker, Catostomus conmersonnii (Lacepede), in Muskellunge Lake, Vilas County, Wisconsin. Trans. Wise. Acad. Sci., Arts and Lett., 31:457-505.
Starrett, W . C. 1950. Food relationships of the minnows of the Des Moines River, Iowa. Ecology, 31:216-233.
Stewart, N. H. 1926. Development, growth, and food habits of the white sucker, Catostomus conmersonnii. (LeSueur). Bull. U.S. Bur. Fish. 42:147-184. 57
Stroud, R. H. 1948. Growth of the basses and black crappie in Norris Reservoir, Tennessee. J. Tenn. Acad. Sci., 23(1):31-99.
1949. Rate of growth and condition of game and panfish in Cherokee and Douglas Reservoirs, Tennessee, and Hiwassee Reservoir, North Carolina. J. Tenn. Acad. Sci., 24(1): 60-74.
Swingle, H. S. 1957. Revised procedures for commercial production of big- mouth buffalo in ponds in the southeast. Proc. 10th Ann. Conf., S.E. Game and Fish Comm., pp. 162-165.
Taylor, H. F. 1916. The structure and growth of the scales of the squeteaque and the pigfish as indicative of life history. Bull. U.S. Bur. Fish., 34:285-330.
Van Oosten, J. 1929. Life history of the lake herring (Leucichthys artedi. LeSueur) of Lake Heron as revealed by its scales, with a critique of the scale method. Bull. U.S. Bur. Fish., 44:265-428.
Walker, C. W . , and P. T. Frank. 1952. The propagation of buffalo. Progr. Fish Cult., 14(3): 129-130.