“

m: m 0.5 5155! mama‘s on m
WHOM MUM or- A mm

Thom for H1: My .61 M8.
Mariam 314115 coma:
Robert Wdham Eshenour

1.953; '-

THFSiS

Ill!III/II/llllllIll/Illlllll/I/IH/Will/ll}Hill/Ill”!!!

o 7296 ’

A —~—‘—-—-<——— -._L‘._._LA77 ‘. _

This is to certify that the

thesis entitled

The Effects of Fish Predation on the

Bottom Fauna. of s Small Pond

presented by

Robert V. Eshenour

has been accepted towards fulfillment
of the requirements for

M. S. Zoolog

degree in

(361181 1 G 3:04

Major professor

Date Jul: 21+. 1953'

 

 

__4- __,

 

 

 

 

 

 

 

 

 

 

 

 

 

.J x \ t . ’—
r . '1 ,‘
. ‘i. ‘ 1.?
‘ I - I I ‘
L \
)V1€SI_J RETURNING MATERIALS:
Place in book drop to _ " '
LIBRARJES remove this checkout from _
“ your Y‘ECOY‘d. FINES W1”
be charged if book is ' ‘
returned after the date '
stamped below. ,
Y -‘ 1. ."~ 03 A .
\ I ("f )2. I ‘
‘2 _t‘ :
I‘IVK‘LEp .
74'1J.
, . \ _ ‘ . .{I _
ll - \ ‘ ’ ‘
1“ ‘11-‘13” 1" I "I n“- 1’ *‘. I , «3’7. 1 LL. 11’. -7"911113321175111"_
_—. 7... .1: \ _ I +1 ('1 ’ -M w . T, . J- -. .l I - l’."l(’—1_“_‘n.. f 3": $_'11“t:;7. 31:1: I '

THE EFFECTS OF FISH PREDATION ON THE BOTTOM FAUNA
OF A SMALL POND

By

ROBERT WILLIAM ESHFNOUR

A THESIS

Submitted to the School of Graduate Studies of Michigan

State College of Agriculture and Applied Science

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Zoology

1953
‘\\\.\“ ‘V.’ ,
\' , , \
.‘ . \ \

\ _- "

\7 M.

\ x

g ‘\‘\

:7 £:

L. N.‘

. W

, I f
‘32?" 1*

f /

,/
./
.(r A
/ f
.F . .. p
a a,

f /

r, .

r
,,,,,,

ACKNOWLEDGEMENTS

Special thanks are due to Dr. R. C. Ball, Department of
Fisheries and Wildlife, Michigan State College, who suggested
this problem and then gave much of his time in counseling and
guiding the writer. The assistance of Dr. D. W. Hayne, Depart-
ment of Zoology, Michigan State College, in the statistical
treatment of the data was appreciated. Edward Grassl, Edward
Bacon, Henry Hatt, and personnel at the Wolf Lake Hatchery
rendered valuable assistance in carrying out the field work
connected with the experiment.

The investigation was made possible by financial support
received through the Agricultural Experiment Station of
Michigan State College.

1) l.rlllll l
. u

I.

II.
III.

IV.

VI.
VII.
VIII.
IX.

TABLE OF CONTENTS

INTRODUCTION .

DESCRIPTION OF PONDS .

METHODS AND EQUIPMENT
Preparing the Fonds
Stocking Pond 5
Bottom Sampling

Field procedure .
Laboratory examination

Stomach Sampling . .
Recovery of Fish Population in Pond 5.

DISCUSSION OF DREDGE SAMPLING DATA
History . . .
Definition of Terms .
Analysis of the Tables . . .
Effects of Predation . . . . .
DISCUSSION OF STOMACH SAMPLING DATA

Feeding Habits . . . . . . . .
Forage Ratios . . . . . . . . . . . .

YIELD OF FISH AND BOTTOM FAUNA IN POND s
STATISTICAL ANALYSIS .

SUMMARY

LITERATURE CITED .

INTRODUCTION

Productivity is a term used to indicate the capacity of a
body of water to produce a crop of organisms. A study of prom
duction in a lake or stream requires a knowledge of three
fundamental concepts (Clarke, 1946): the amount of organisms
existing in an area at the time of observation; the amount of
organisms removed from an area per unit time by man, or in
other ways; the amount of organisms formed within an area per
unit time. The cycle between the death or removal of one crop
and the production of the next is complicated, and the inter-
dependency of the factors entering into this cycle is not in
all cases well understood. Many attempts have been made to
single out certain 'indeces' which might indicate, in general,
the capacity of an aquatic ecosystem to produce organisms.

The number and volume of bottom fauna organisms have been
used by many workers as an index of the productive capacity of
lakes and streams. Deevey and Bishop (1942) state that "In
evaluating the potential ability of a lake to produce fish,
probably no single standard is so important as an estimate of
the amount of bottom fauna,” emphasizing the importance of
measuring quantitatively the bottom organisms. Ball (1949)
used a comparison of production of benthic organisms in ferti-
lized and unfertilized ponds in studying the effects of fertie
lization on productivity and believed that an evaluation of
the standing crops of organisms will serve as a measure of the

relative productivity of fertilized and unfertilized ponds.

8

In his survey of the Horokiwi Stream, Allen (1951) found
that it was essential to undertake not only a quantitative
study of the bottom fauna, but also to determine the effects
of the selective feeding habits of the fish it supported on
its specific composition.

In this investigation a population of bluegills (Lepomis
macrochiggg) and pumpkinseed sunfish (Lepomis gibbosus), species
known to be dependent on the benthic organisms for most of
their food, was stocked in a small pond to determine the effects
of predation on the various invertebrate groups present. A
pond similar to the one stocked was kept without fish and used
as a control. Bottom samples were taken at the same rate from
each pond, and comparisons were made of the standing crOps of
organisne in the ponds. Stomach samples were taken periodically
and forage ratios determined as an additional check on the feeds
ing habits of the fish.

The population of fish added to the experimental pond was
of known weight, so the fish could be removed at the conclusion
of the experiment, the increase in weight noted, and conversion
factors for food materials to fTsA flesh calculated.

An analysis of variance was made to determine whether the
standing crops of food organisms in the ponds differed signi—
ficantly before and after fish were introduced, and also to see

if a significant difference existed between sampling stations.

DESCRIPTION OF PONDS

Ponds 4 and 5, on which the study was conducted, are located
at Wolf Lake State Fish Hatchery in Van Buren County about ten
miles west of Kalamazoo. They were chosen because of their
similar morphological, physical and biolOgical characteristics.

Both ponds are circular in outline, have a surface area of
one acre and a maximum depth of six feet at the outlet. The
average depth is approximately three feet.

The bottoms of the ponds were composed of three types
of material. In the shallower water and around the edges sand
was predominant. In the deeper water the bottom was primarily
muck-or a mixture of mud and bentonite, a material of clay con»
sistency with which the ponds had been treated to prevent water
1089 through the basin.

The source of water, a large spring, had a total hardness
of 160 parts per million. The temperature of the water as it
left the spring varied within a few degrees of 55° Fahrenheit.
Temperatures taken by means of a recording thermometer in Pond
4 varied from a high of 78° on July 23 to a low of 58° on
September 4. The mean temperature for the ten-week period was
69°. Although no thermal recordings were taken in Pond 5 it
is assumed that they closely paralleled those of Pond 4.

Turbidity was checked periodically with a Secchi disc,
and although Pond 5 was slightly more turbid than Pond 4 for
the first part of the summer, the disc was visible at the

greatest depth in each pond at all times.

4
ghgga sp. began to grow over the bottoms of both ponds
shortly after they were filled with water. By the fourth week
a solid mat of this alga blanketed three—quarters of each
basin, and little change was observed in the ghag§_until it
began to die at the end of the summer. A bed of Potamogeton

pectinatus, which covered an area of about 200 square feet,

 

established itself on the north side of Pond 5. Higher aquatic

vegetation was entirely absent in Pond 4.

_——._‘--

 

 

in ea

the

COD‘L

18
we;
1).".

a1".

 

METHODS AND EQUIPMENT
Preparing the Ponds

Both ponds were drained and allowed to remain dry for
several days in order to kill as much of the pond fauna as
possible. This was done so that the experiment could be
started with nearly the same standing crop of benthic organisms
in each pond.

Prior to draining, Pond 4 contained rainbow trout and
the bottom was covered by a thick growth of thga sp. Pond 5
contained a population of suckers and there was but little
vegetation left on the bottom, probably due to the feeding of
these fish upon the plants.

Pond 5 was drained June 5 and allowed to remain empty for
18 days. During all but the last four days of this period the
weather was dry and extremely hot, and the little vegetation
that remained at the time of draining was completely dried a:
and the bottom baked hard. In spite of this, larvae of many of
the bottom fauna groups were present in well-advanced instars
the first week of sampling, indicating survival of at least some
of the individuals.

Pond 4 could not be drained until June 19 and it was
allowed to remain empty for a period of only four days. Dur-

ing the four days that this pond was down the weather was cool

1 I

my: 1 ~ a \rA. :—
d m 095 a .99 #2. A: {a
n-” ed a .«I \d .5. :u a .
an Av w; 61v nu \Ph mg C

«w
in
0 ..

 

[illfITIIITIJIltI Isl

$ i 1VN “ i
r; .1 3.. l
u m." TI... .1"

6

and rainy with the result that the ghggg sp. did not dry out
completely and the bottom remained soft. This condition was
reflected in the first week of sampling by a much higher ini-
tial level of abundance of bottom organisms than in Pond 5,
but by the second week the abundance level in Pond 5 was the
higher of the two. It is improbable that this rapid increase
was due entirely to the recovery of a decimated population of
organises, but more likely was the result of a combination of

factors as explained in a following section.
Stocking Pond 5

In order to establish the growth trend of the benthic
organisms after the ponds were refilled, no fish were stocked
until bottom sampling had been in prOgress for four weeks. It
was felt that a longer period of sampling the ponds while they
were devoid of fish would have been advantageous, but the time
limitations placed upon the investigation prevented this.

At the end of the four-week sampling period 160 pounds
of bluegills and pumpkinseed sunfish were added to Pond 5, and
sampling was continued, using Pond 4 as a control. The numbers

and weights of the fish stocked are shown in Table l.

7

Table 1. Size, numbers, and weights of fish stocked in Pond 5

 

 

 

Total length Number Weight

(inches (pounds)

Bluegills 5.0 - 4.4 357 31.0

4.5 - 6.4 428 57.0

6.5 - 8.5 175 59.5

Total 960 127.5

Pumpkinseeds 2.0 - 5.9 800, 27.5

4.0 a 6.0 _§Z 5.0

Total 857 38.5
Total for

both species 1797 160.0

 

Wilkins (m.s.) made a similar study on these ponds a year
earlier but used a different technique. Bluegills and pumpkin-
seed sunfish were used in his investigation also, but only 124
pounds were stocked. In the present investigation 160 pounds
Of fish were stocked in an attempt to induce a more complete
utilization of those benthic forms important in the diet of
the fish.

Wilkins' procedure differed in a second respect. Instead
of keeping one pond without fish as a control for the entire
experiment, Pond 4 was stocked and Pond 5 was used as a control
for the first phase of the investigation. For the second phase
Pond 4 was drained, the fish weighed, and 124 pounds of fish
constituting a second population whose numbers and species were
in prOportion to the original population, was stocked in Pond 5.

Pond 4 was immediately refilled and used as a control. Upon

8
transferring the fish population from Pond 4 to Pond 5, pro-
duction of organisms in Pond 4, released from predation, showed
a sharp increase as contrasted to a decreasing abundance of
invertebrates in Pond 5, subjected to predation by the same
population of fish that had originally existed in Pond 4. No
attempt was made to compare production of organisms in the

ponds prior to stocking.
Bottom Sampling

Field procedure

Beginning June 50 and continuing for a period of ten weeks,
twenty Ekman dredge samples per week were taken from each pond.
The Ekman dredge was used in preference to the Peterson dredge
because with it more samples could be taken in the allotted
time, resulting in a more complete coverage of the bottom. It
was felt that use of the Ekman was Justified for two reasons:
first, most of the organisms which are characteristically found
living in the bottom soils to any great depth are not available
to the fish as a source of food, and for practical reasons need
not be considered in this investigation; secondly, there was
very little material (Stones, sticks, etc.) present which might
impede the efficient Operation of the dredge in either pond.
It was found that the Ekman dredge sampled the Chg g quite
successfully even at the peak of its growth.

Stratified-random samples were taken in each pond by
laying out transects which radiated from the deepest part of

each pond to points equally spaced on the adjacent shoreline.

9

Sampling stations were then established at regular intervals
along these transects. This method gave as nearly complete
coverage as possible to the various types of bottom present
and at the same time insured sampling in all depths of water.

Ten samples a day were taken on the same two consecutive
days each week from each pond. This allowed one full week to
elapse between sampling at any particular station.

All samples were taken from a rowboat. As each sample
was taken it was raised to the surface and the entire dredge
scooped into a twelve~quart pail. The material in the dredge
was then washed into the pail and the contents dumped into a
20-mesh screen where the greater part of the bottom material
was washed away. The concentrated samples were then placed

in two-quart jars and taken to the laboratory for sorting.

Laboratory examination

The organisms were removed from the samples while still
alive and preserved in a solution of formalin. As time per-
mitted the preserved organisms were separated into taxonomic
groups, counted, and measured volumetrically. Organisms
which were too small to measure accurately were accumulated
for the entire week and the total volume of the groups to
which they belonged was determined. From this an average
volume for individual organisms was calculated. The volume
of a particular group in a sample was then calculated
by multiplying the number of individuals in the group by

the volume which had been determined for one individual

10
of that group. By using this method organisms were included
which Otherwise could not have been measured volumetrically.

In sorting individuals of the family Chironomidae (=Tendi-
pedidae) it was found that they naturally fell into three well»
defined size classes. In order to facilitate the conversion to
volume as explained above, these size classes were treated
separately in volumetric determinations, then recombined for
the final analysis of the data.

Weights of organisms have been employed in determining the
amount of bottom organisms present in a lake. In a fish food
study of Third Sister Lake in Michigan, Ball (1948) derived a
conversion factor of 0.98 for changing preserved volume in cubic
centimeters to live weight in grams. Because the discrepancy
is so slight, it was felt that for the purposes of this experiu
ment one cubic centimeter of preserved volume could be considered

equal to one gram live weight.
St omach Sa mp1 ing

Stomach samples were taken as an additional check on the
groups of organisms making up the diet of the fish in Pond 5.
Most of the fish removed for stomach samples were caught by
hook and line; the remainder were taken in wire traps. As the
fish were captured their stomachs were removed, slit open, and
these preserved in separate bottles of 5 percent formalin. The

species, weight, and length of each fish was recorded. The

11
contents of each stomach were examined in the laboratory with

the aid of a binocular microscOps.
Recovery of Fish Population in Pond 5

At the conclusion of the ten-week sampling period Pond 5
was again drained and the fish pOpulation removed for weighing.
Thus, the total weight gained by the fish during the six weeks
that they were in Pond 5 could be calculated and food convera
sion ratios determined.

Pond 5 is so constructed that the fish were forced to
collect at the deepest point, near the outlet, as the pond was
being drained. From here it was possible to seine them out
for weighing. -A thorough search of the emptied basin after
seining revealed only three small bluegills, indicating a prac~

tically complete recovery of the fish population.

12

DISCUSSION OF DBEDGE SAMPLING DATA
History

Numerous researchers in this country and others have inves~
tigated the role that bottom fauna plays in the food cycle and
its value as an index of the capacity of lakes or streams to
support fish pOpulations. .

Among the earlier workers in the field, Eggleton (1931,
1935, 1957) has contributed much to our knowledge of the distri»
bution, composition, and various ecological relationships of
the benthos. Even before Eggleton, however, Scott et a1 (1928)
made an Objective quantitative study of the bottom fauna in
certain Indiana lakes, correlating the occurrence of benthic
groups with various physical andlimnological features present.
More recent investigations have been made along these lines by
Deevey (1941), Lyman (1943), Ball (1949), Allen (1951), and a
number of others, to determine the reciprocal relationships of
the bottom fauna and the fish dependent upon it as a source of
food. In many of these later investigations the standing crOp
-ofbottom organisms has been used to give some idea of what a
body of water might produce as an end product in terms of fish.

Important in a study of this kind is the trophic-dynamic
aspect, defined by Lindeman (1942) as the point of view emphasiz~
ing the relationship of trophic or "energy-availing" processes

within the community unit to the process of succession. These

13
trophic relationships include all the biotic and abiotic factors
which enter into the food cycle relationships. In his discussion
of the relationship Lindeman points out the importance of con~
sidering the influence which the abiotic or non-living environ«
ment exerts on the biotic communities, and suggests that these
living and non~living communities are inseparable in an analysis

of food cycle relationships.

14

DISCUSSION OF DBEDGE SAMPLING DATA
Definition of Terms

The definitions of terms used in this manuscript closely
follow those suggested by Clarke (1946).

The volume or number of organisms existing in an area at
the time of observation will be referred to as the standing crop.
To determine production Clarke suggests that a knowledge of the
production rate, or the amount of organisms formed within the
area per unit time also be considered. This is a logical inclu~
sion in the concept of production, for without a knowledge of
the rate of turnover a measurement of the standing crop means

little in terms of production.
Analysis of the Tables

To measure directly the rate of production of benthic
organisms of Ponds 4 and 5 would have been a study in itself,
and time limits made it impossible. By introducing a fish
population the rate of production in terms of increase at the
higher trophic level of primary carnivore production was meas«
ured. This served our purpose in determining the degree to
i which the organisms were capable of supporting this primary
carnivore pOpulation, and at the same time reflected the rate

of production of the benthos.

.mxsu macs .e econ ea

mcHHasmm amused an oeuooaaoo mesmw opuspbpno>cH .m bands

.o‘““ O .0 O 0
no. n
9... no."
RA in
a. e S. a:
so.oa «won
oa. om.
mo.” no“
no. an."
an. an
on.aa on.-
3.3 «ta
no. on.
do. a
mm. 9..
so. a
an. on.
na.a “ma
mg. mm.
mm. om
up. no.”
no; «2
mm. am.
oh. “a
ms. nu.
mo. on
sac ow
m~.n 3n
:3 one
an.» mam
om.~ no.3
“3.0” “mag
no. mo.
mm. mm
¢~.~ om.~
no.na sand
mu. on.
on. ma
am.n no.m
mz.nm was-
oo.n
oo.wo~
m.oam
wan.oa
om.
oom
odcoos

3. R.
mu. an
an.» 8;“
DJ w?
“m. no.

. 3
”he “he
3. a“
9% 8....
RA ok
me. o...
mud” O 0 mm. 0 0
nu. as.
E. on
9. on.
8. n
no. on.
mo. m
and om...
m~.H on”
so; ode
omnm men
on. o a.
Hm.~ nnm
ma. an.
94 N:
no. ro.
ow. HA
.34 .~
om. n «no

36

8. R

a. sum

3mm

3

on

2 u a

3.003

.elfioo 2 IBch 3. «o 2.083 on. Ede-0.3 no Each 0
£38:— nn lounge fie no «Semen I.- eleuaeuuo no we! 4

no. no.”
we.” as
no.2 odd
RA mom
no. 8.
m... an
o1 on.
no. u
no.“ on.»
we.“ nan:
no. 8.
8. a
no. S.
9. as
8. S.
3. d
a. O
s. a
80 “do
no. n
no. no.
8. n
64 8A
2.. mm
mm. 2..»
a .m a...
a. a...
and is
.0‘“‘ .0...
no. n
3. on.
X; mom
no. 8.
:1 n.
we. on.”
~a.m «on

on...

8...“...

Tim

min

2

9.

u . a.

8.09:.

:0 I O .0 0
no. n
80 “do
on an
3.: an.nu
$4 on“
i” O O O O 0
me. on
no. on.
3.. v .
SA. an .
no. 3.
.8. a
U0. “do
um. nu
no. no.
re. a
”N. “no
5. N
no. 3.
no. n
ON. 0:.
«o. m
$4 and
ow. no
0:. 0‘.
on. .— m2
.3. om.”
.— 5. a R:
.08“ O O 0 0 0
mm. mg.
m a. a 5:
no. 8.
so. a
we. ca.”
an. n man
S. n
on. ma
m. 9:
can."
an
on
w - n
once:

on. on.
on. on
and 3. 3
E. a as
no. no.
3. 9
no. 9.
B. s
2.. a oar"
mmum. mm. 0
no. 3 .
«c. a
no. no.
mo. 3 .
no. no.
me. on
80 “do
om. «N
9. on.
.3. a
2.. no. a
m... ow
no.4 cod.
3d new
”We “a.
S n3
o. 8.

9. 3
mm. on.
eo.n nun
sub“ 0 O O O 0
mo. m
K. “be
at. n 8:

mu.~

om. E

3.35

.35

on

9.

a u n

8.00:

<m<m<m<m<¢n<r<m<n<m<n<n<c<n<n<n<n<n<n<m<n

era-SE
.3502
8.2.33.8
ear—noon;
:33.» «m
888:8
load?
cocaine...
3.99vo I...
:39 «eat
9—3338
ISA—onus
9—390. «5
Seaman-m
I..." «one :30
o «no-o
3.3828
838023 «on .0
01::— one“...

' flag—HO.» «£0

.3 .5 «on Iowan-who uo 23>

3.3 Coo. bar-«go no Each #30...

8. 2.
mm. E
2. mm 3.?
B. a «we
no. no.
a” .. m... ..
8. a
a... a 8....
ma. a on”
8. 3.
8. a
S. om.
so. a
g“ .0 O 0 O
8. a
o1 om.
on. a
9A oc.o
«a.
80 “a.
3. 3
3. mm.
MM” 0 O mo 0 O
Re “No
on. a an:
8. 2.
en. 9.
“we no H
on. w {b

8. a

~49

SS

2

3

m .. a

3.3x

.8 .6. «on gage «o .9.
35:30.3 no hen-Ba use...
7: .3. :22: yo no.2

.33.: no .or

voice 833:3

.0 r 1- l. ‘ 1
IAIdOL. OM80... 0.x...

 

 

 

.oxmo mac: .m oooo so

wcaaaemm mmompo >9 douomaaoo madam cooppopum>cH .m manMe

ova“ .0...

no. w
mm. S.
m. 2
n.no on.um
3;. now
no. no.
u. ow
no. ou.
mo. u
no.mu no...
3.: nuou
ou. om.
uo. u
on. on.
mm.u mou
8"“ .0...
no. n
nn. no.u
n~.u nuu
mm. mm.
nn.u nu
nu. .
nn. ow
no.2 on.m
3... nmn
no... .3
nn.~u oouu
on. ou.u
an... on:
O
nu. nu
m. a.
.n um:
an. n.
$.u ozu
ou.: oo.o
ou.? on:
oo.~
3.1.:
uo.usu
zoom
on
ooa
ounces

.OCkU O 0...
no. n
ou. nu.
ou. m
Pic.— 2.3
n.. on
.0‘h0 000..
ou. m
no. 8.
no. n
nn.~ o:.n
un.u uuu

u. om.

n. no
AN. ONMO
m
um.” ma.
mo.u uo
ou. nu.
nu. nu
uu.n on.u
on...“ on...
au.m nm.u
on... Ru
ou. on.
a... w.
no. no.
«a. un
no. no.
an. un
nn. on.
R6 R»

up.“

no.5u

o.umu

ouou

ou

o:

ou . o

.80.:

8.0%“ O O...
uo. u
no. 8.
.0. a
Si 8. :
u~.u Nnu
Oghv o o o o o
no. 5
g“ 0.0 o.
.8. .—
nfln an.“
NH.N nwu
ou. om.
“WHO. 00%...
m. R.
z. o:

no. 8.
R. a
so. On.
8. o
a; n3
3; n.
u~.u n~.u
2... s.
”no no.
n~.~ mnu
g“ 00.. O
3. a
no. no.
go. .u
mm. om.
5n.» nan

mm.“

omwmu

a.nmu

«has

a

on

u o s

In...

.8930: inov or ..

.353» as £52.53 :0 no 280.3; 2:. 2.03.8.3 s... 0.53.? ¢
3.35:: a: 0:3:ch 3:. Me 28th can nrouroruc uo gong: <

.30
uo.
um.
um.
nn.nu
mm..."
.0180
80
.0150
no.
o~.n
nn.n

no.
nu.

ou.
mu.
80
um.
08.:
no.
nu.
Mum
aw.u
nu.
nu.u
no.
how
no.
nu.
R
on.»

u

on.
on
on.mu
auu

M0000
u
nu.u
sun
80
nu

OOE“ .00 O.
uo. u
5. on.
«u. uu
odou oanm
mm; gnu
no. no.
«c. N
no. no.
mo. m
Sum Sum
mo.n own
S. OH.
8. u
.2“‘ O. o o 0
mo. . m
3H. 8.
mu. mu
Nu. nu.
nu. nu
.2“0 0 O O .0
mo. m
.g““ 0....
we. n
ma. u omen
no.“ mum
Bo Odo
"so: .0 0‘00 0.
no. cu.
Ran w...—
so. cu.
nhu no.
8; ,3;
nw.~u moon

on.n

.nw.n

flofl

noun

cu

3

.. . u
as...

no.
no.
nn.ou
nm.u
000.3
uo.
no.u
3.
cu.
uo.
08.3
uo.

no.

n
on.nu
9..—

A
na.u
0c
8.
A

m
u
no.
m

a

m .
S.
OH.—

O
nu
ou.

ou.
nu
no.~
u.»

do.“

..onu
.onu

an

and

unannoua
coconoouom
ouooooomuuo
Id“: 0: CE
conuoauuu
cocoouaouo
.couuuanr
ouoaovoupe
ou.aauaum
£30003“.
£33.53
8390. «I

c «nun-ho:
lauuaooauooo
ou.-co
333905 :00
occu-ouoaouuuoo
.aoyn..oeux

<n<n<n<n<n<n<n<n<n<¢n<n<n<n<n<m<¢n<m<n<¢2

.Qgguflv

.3 .a. uo.. '30 no 2—0»
“.03 IIunIIuo «o '53 a?
.3 .3 .33 .3093 «o .0:
'3‘: no uo.—I:— «IE

7‘ .3. uo.—al- uo as
.1.- uo .08

8:8 88838

19

Certain dominant organisms determine the general complexion
of the invertebrate fauna in any body of water. An inspection
of Tables 2 and 3 reveals a few groups of organisms in each pond
that exhibit this dominance. However, high numbers of a parti~
cular group of organisms per square foot do not necessarily
mean correspondingly high volumes of the same group.

For instance, chironomids were numerically dominant in
both ponds, comprising 23.5 percent of the total numbers of
organisms taken from Pond 4 and 47.8 percent of the total from
Pond 5, but volumetrically the same group made up only 3.3
percent of the total in Pond 5. On the other hand the oling
chastes, though small in number, made up a considerably larger
volume than any other one group, comprising 10.1 percent of
the total number and 64.1 percent of the total volume in Pond
4, and 7.2 percent by number and 63.3 percent by volume in Pond
5.

It is generally considered that measurement of the volume
of the various faunistic groups constitutes a more valid repre~
sentation of their relative occurrence than does measurement of
their numbers. On the assumption that Ball's comparison of
preserved volume and live weight being nearly equal is correct,
preserved volume can be used in approximating the mass of bottom
organisms present. This is not meant to imply that numerical
measurement is not important, but simply that it alone cannot be
used as a measurement of standing crop and consequently is of
little use, in itself, in a study of the trophic-dynamic aspects

of production.

.oxsu ouos .e ecom mu
mauaaemm omnmud no omxmp museum opnnompno>cu mo

madao> one Hogans >n acupumoaaoo pcmopom .H oudmum

 

I I 9:. TOTAL NUMBER
V/A °/. TOTAL VOLUME

 

 

 

 

 

 

 

 

k\\\\\\\\\\\\\\\\\\\\\\\\

 

 

 

93

70—

eo-r

50o

4o-r

30

zo-r
o

 

 

 

.oemo nuoa .m ocoa cu
mouaoEMm mmoepo up some» mddonm manpomppo>cH mo

oesno> new umbesc >9 acupumoqaoo unsound .m ensmum

 

E] % TOTAL NUMBER
7A °/. TOTAL VOLUME

 

 

 

 

 

L

\\\\\\\\\\\\\\\\\\\\\\,

 

 

 

 

34’

l 1 T I
o O O O O O
[s m In V [O N

l0-

 

 

84
In comparing Tables 2 and 3 two things become evident:
(1) those groups which are common to both ponds make up approxi~
mately the same percentage COmpositiOn Of volume and number in
each case; (2) in spite Of the proximity Of the ponds to one
another and their similarities, there were two groups which

were not common to both ponds, i.e., Chaoborus sp. and D011—

 

chOpodidae. It will be noticed that all of the dolichopodids
were taken very early in the sampling period and that they were
Of little importance in the final data. The Chaoborus sp. in
Pond 4 were likewise of little importance in the final analysis,
and the environment which the pond Offered did not appear tO be
favorable for their deveIOpment.

As seen in Figures 1 and 2, four groups Of Organisms made
up the bulk Of the total both by number and by volume. In Pond
4 the four dominant groups in order of their volumetric importe
ance are the Oligochaeta, Ephemsrida, Gastrcpoda, and midges;
in order Of their numerical. importance, they are nudges,
Ephemerida, Gastropoda, and Oligochaeta. In Pond 5 the groups
Oligochaeta and Gastropoda have exchanged places in order Of
volumetric importance, but the order Of numerical importance is
the same as in Pond 4.

In spite Of the relatively large volume of Oligochaetes,
snails, and mayfly larvae, certain members Of these groups were
not accessible to the fish and made up little or no part Of their
diet, as will be shown later.

During the early part of the sampling period tubificids

‘Were taken in large numbers. By the time the fish were placed

25
in Pond 5, most of the tubificids had disappeared, and from
this point on they occurred in steadily decreasing numbers so
that none were being taken at the conclusion of the experiment.
It was thought advisable to eliminate this group altOgether due
to the ephemeral nature Of their occurrence and the great diffi-
culty experienced in separating them from the ghgzg and detritus

with which they were closely associated.
Effects Of Predation

The effects Of fish predation On the bottom fauna of Pond
5 are shown graphically in Figures 3, 4, 5, and 6. The numbers
and volumes of all organisms per square foot in Ponds 4 and 5
may be compared by referring to Figures 3 and 4. In Figures 5
and 6 the same comparison is made of "fish food" organisms only.

Included as fish food organisms were those groups which
were determined by stomach sampling to be actually utilized in
the diet of the fish. As a result the Oligochaetes, Hexagenia
sp., and all snails larger than a size which occurred in the
stomach samples were eliminated from the group designated as
fish food organisms. This made possible a more valid comparison
between organisms subject to predation and those not subject to
predation. The discussion will be confined chiefly to the
organisms comprising the fish food group.

An examination of Figures 5 and 6 shows the same general
trend in paucity and abundance whether considered numerically
0r volumetrically. The numerical data, however, show more incon—
Sistency from week to week and do not reflect the actual

h-'

 

.mxdq MACE

.m mom e mocom cu wouaaamm owdosd no nopoeanoo

poom opmzdm poo msmucmMHO Hum we mnmnaoz

 

..n oosmus

 

scum
_

hmoooq
_

>431

 

 

    

v
ouoooozhz_ /

Ia...-

m OZOduIII
V ozom.|||

Tom
.lOO.
.IOD.

ICON

Iona

loom
dz

 

 

 

mumCHo n. <0H55om on www oamwuumsm own mucosa noon
ooHHmouoo ow oamowo mmsownnm no endow o moo m.

sown rows.

 

 

PmDoad

_ _

_ _

 

>43...

 

 

 

m 020d III
v 020m .llll

omega-01.5,:
In;

Ion.

Too._
:0 n.
Too.~
Iona
Toon
Ton.»
Tood
Tone
T006.
To no
Iooo
oo

 

 

 

30
difference in fOOd—available levels between the ponds as
accurately as the volumetric data.

The first week of sampling revealed a substantially higher
volume of organisms in Pond 5 than in Pond 4 as shown in
Figure 6, but by the second week the volume of organisms in
Pond 4 had approximately doubled that in Pond 5. This marked
change was due primarily to an increase of chironomids (from

.04 cc. to .39 cc. per square foot) and Centrgptgium sp. (from

 

a trace to .13 cc. per square foot), and could have been due to
a condition which made these forms unavailable to the dredge,
sampling inaccuracies, natural growth, or more likely a combi«
nation Of these factors. Between the third and fourth weeks

Of sampling Font S experienced a heavy emergence Of Centrop~
gglgg sp. and chironomids, indicating that larval forms of these
insects in late instars of their develOpment must have been
present in spite Of draining.

It is believed that the method Of sampling used precludes
the possibility that this difference in levels between the two
weeks was due to any great inaccuracy Of the sampling method
itself. In view of the late instar forms present during the
second and third weeks after filling the ponds, it can be assumed
that the larval forms were unavailable to the dredge due to their
activity in burrowing deeper into the bottom material to escape
drying. If this were the case, they presumably did not move to
the surface of the pond bottom until after the first week Of

sampling.

 

31
Between the third and fourth weeks of sampling in Pond 5
the general emergence of chironomids and mayflies of the genus

ggntrOptilum reduced the benthic level of abundance in this

 

pond below that of Pond 4. The week following this emergence
the fish were introduced into Pond 5, and from that time through
the end of the sampling period the abundance level in this pond
did not approach that of Pond 4, nor did it ever again reach its
previous high of the second and third weeks.

Pond 5, and to a lesser extent Pond 4, contained an
abundance of dragonfly larvae which frequented the shallower
areas around the periphery of each pond, but they were seldom
taken in dredge samples because of their avoidance reaction in
shallow water. Immediately upon being introduced into the pond,
the fish were observed to begin praying heavily upon these
larvae, almOst to the exclusion of the other benthic forms.
Three days after introducing the fish a trip around the edge
Of the pond revealed only five Of these organisms where hundreds
were present a few days earlier. It is questionable that any
serious predation on the other bottom inhabitants occurred until
the dragonfly larvae had been reduced to a low level.

The trends shown for both ponds were greatly influenced
by the abundance of individuals of the groups Chironomidae and

gentrOptilum sp., two Of the most important food groups in the

 

diet of the fish, as will be shown in a feeding habit analysis.
The percent composition that these groups made up of the total

fish food organisms varied in Pond 4 from a low of 21 percent

.oxmq Macs .m ncm e mused :u mounuamm owuonv no

Oopooauoo msmucsmuo noon smug mo muooaoz .m oudMua

 

dll I..nrr,r.!k 4 Vi.

 

.ommg mnog .m pom e mocom cu managemm omoono no

empoouuoo msmqumeo noon smug mo mesono>

 

‘..e r. 4. . In . H’s? suiéj.
... ..
. ... . . I

.m 93m;

 

kdmm

 

 

Pmnwad >435
_ _ _ - _ H _
DUO-40015.2.
\\:// 1m:
\ III
\ lal
\ I,
\\ z
\ I
\ I
\ I
/
/
I ~
rlllL
n OZOdIII
¢ DZOQIII.

IN.O

Ind

lcd

1.0.0

.lmd

T. 0.0

lad

IO._
00

 

 

 

36
the second week to a high of 68 percent the last week, while
in Pond 5 the extremes were 50 percent the first and last weeks
and 87 percent the second week.

The average volume per week Of fish food organisms for the
foura-week period before fish were stocked was .50 cc. for Pond
4 and .49 cc. for Pond 5, practically identical levels. For
the six-week period that fish were in Pond 5 the average volumes
per week Of these organisms had increased to .86 cc. per square
foot in Pond 4 and decreased to .45 cc. per square foot in Pond
5. It appears that predation by the fish was effective in
reducing the level of abundance in Pond 5, while the level of
abundance in Pond 4 increased markedly in the same period of
time.

From the time the fish were introduced into Pond 5 until
the end of the experiment the volumes and numbers of organisms
for both ponds show much the same trends but at different levels
of abundance. There are two major decreases in the volume Of
fish food organisms during this period, both due primarily to
emergence of mayflies and/or midges. The first reduction
Occurred early in the period, between the third and fourth
Weeks in Pond 5 and between the fourth and fifth weeks in Pond
4. The effects of these emergences, especially in Pond 5, may
be seen in Figures 3, 4, 5, and 6.

Immediately after the first emergence the organisms in
Pom 4 increased steadily in volume until the second major

emergence Of the mayfly and midge groups between the eighth

37

and ninth weeks. The organisms in Pond 5, however, did not show

a recovery until the sixth week, after which they increased at
a lower level of abundance than in Pond 4.

Between the eighth and tenth weeks a second major period

of emergence again reduced the level of organisms in Pond 4 as

The last week of sampling in Pond 4 shows a surprising
.increase in volume of food organisms following the emergences

(of the preceding week. This unusual condition resulted primarily

ffirom two samples which apparently were taken from areas having
EL high concentration of the large midge larvae of the genus

(Ihironomus. The resulting increase in volume and numbers is

Ireadily discernible in Figures 5 and 6.

38

DISCUSSION OF STOMACH SAMPLING DATA
Feeding Habits

Analyses of the feeding habits of the bluegill have been
made by numerous investigators (Forbes, 1903; Muttkowski, 1918;
Leonard, 1940; Howell, 1941). More recently the importance of

(comparing the occurrence of food organisms in the stomachs with

 

inhese organisms as they are found in the fish's environment has 3
t>een recognized (Hess and Swartz, 1941; Allen, 1942; Ball, 1948; ’
IBaJl.and Tanner, 1951). These investigators have shown that
f'ish exhibit a selectivity in their feeding habits and that
iihe presence of an organism in a fish's environment does not
riecessarily mean that this organism will be used as a food by
the fish.

The numtpr of fish killed for stomach samples was kept
esmall so that the weight increase of the original population
vvould be altered as little as possible. At the conclusion of
tlie experiment the weights of all fish removed were added to
tlie weight of the population remaining for computation of cone
Version factors from organisms to fish flesh.

Although only 40 stomachs were taken during the last 4
Weeks of the experiment, the data exhibit a constancy (Figure
7') from which can be drawn certain conclusions.

Only 5 of the 40 fish taken for stomach samples were pump-
klinseeds; the remainder were bluegills. Other investigators

(Pearse, 1931; Ball, 1948; Patriarche and Ball, 1949) have

Table 4.

from Pond 5.

Number of fish ......... 40

Weight range (gm.) ..... 194135
Average weight (gm.)... 63.2
Length range (mm.) ..... 96.5~188.0
Average length (mm.)... 147.3

Total weight (pounds).. 5.57

39

Food of bluegills and pumpkinseed sunfish taken

 

Type of food

AQUATIC INSECTS
Diptera
Ceratopogonidae
Chironomidae
Chironomidae (adults)
Midge pupae
Ephemerida
Caenis
CentrOptilum
Centroptilum
Odonata
Anisoptera
Zygoptera
Zygoptera (adults)
Hemiptera
Corixidae
Corixidae (adults)
Coleoptera
Hydrophilidae (adults)
Trichoptera
MOLLUSCS
Gastropoda
Pelecypoda
ARTHOPODS
Cladocera
Hydracarina
ANNELIDS
Oligochaeta
TERRESTRIAL INSECTS
CHARA
OTHER VEGETATION

Percent
by
number

a:
-0 OFF H009

-9 mo» away

Percent Percent Percent
by by of total
volume‘ volume“ stomachs

(0‘20! ()3 0101

WHO) H0103 09 (no
N
NCO ONO O- NO
H

01
mow wpw H-GH

HN
U‘IQU'IU) 010') NN [MN 00‘) NNQ U'ILOO GIN-QC

omoo 00 mm mm mm mmm omo ommo

' H 00" OH
03

' 03 HP ‘10!
Q

2.5

-m om ON PG
om HH up mo

5%:é

030)

 

IWithout Chara
l""‘With Chara

40
found considerable variation in the feeding habits of the two
species even when they were taken from the same body of water.
The differences between species in this investigation was so
slight that it was not considered important enough to warrant
a separate analysis Of their feeding habits.

The data concerning the fish removed for the feeding habit
study are shown in Table 4. The results of this experiment
seem to confirm those of Wilkins (m.s.) and Ball (1948) who
found little variation in the food taken by different age
classes of bluegills which were older than young-of—the-year.
No young—of-the-year fish were stocked in this experiment.

For these reasons all size_classes and both species will be
considered as one group in the feeding habit analysis.

A summary of the fOOds found in the stomachs of fish taken
from Pond 5 is shown in Table 4. The percent that each food
group makes up of the total, both numerically and volumetrically,
as well as the number Of stomachs that a particular food group
was found in, expressed as a percent of the total stomachs,
may be seen in the table. Because ghggg constituted a large
percent of the total volume, the percent composition by volume
Of the invertebrate groups has been shown both with and without
Elissa-

The volumes of the various Organisms found in the stomachs
were not measured directly, but were calculated by determining
a conversion factor (average volume of an organism) from total

volumes and total numbers taken by the dredge. By multiplying

41
the number of a particular group found in the stomachs by the
conversion factor, a close estimate of the volume of organisms
was Obtained.

The volume of gaggg taken as food was determined by the
following method: as the stomachs were examined, the percent
that thgg made up of the total volumewas estimated. The
volume that the invertebrate organisms made up in the same
stomach was calculated as outlined atpve. With these two facts
known the total volume of food in a stomach was found by divide
ing the volume of invertebrates in the stomach by 100 less
the percent composition by volume that the 9h; g constituted
of the total. The volume of thgg in a stomach was then found
by multiplying the total volume in the stomach by the percent
composition of ghggg.

At the time stomach sampling was started Ch; 9 was the
dominant item in the diet of the fish, comprising from 50 to
100 percent of the total volume of food in all but one out of
the first ten stomachs examined.’ Stomach sampling apparently
started at a time when vegetation was at a peak in the diet of
the bluegills and pumpkinseeds, as evidenced by the steady
decline of ghggg in the stomachs of the fish (Figure 7) from
the sixth week of sampling to the end of the experiment.

Other investigations of the feeding habits of bluegills
have revealed this same phenomenon of changing over from a diet
of invertebrate organisms to one composed largely of vegetative
matter during the summer months. Ball (1948) found that blue—

gills in Third Sister Lake ingested plant foods at rates

.m doom a“ madmm acupon no mESHo> mg» mp“;

mnomEOPm swam QH dunno mo mESHo> ecu mo somwnmmsoo .5 onsmfim

 

 

)L

 

O.n

O.¢

0.0

Fudm
\oo

 

 

O.¢

@zjdszm no xww?
n.» O.n 9N
_ p

4234“. ZOPhOm Ill
<m<Io Ill

O.N
_

 

”Jon". aha H

ION

.lom

10¢

low

low

ION

low

low

JO)
Adkoh.

 

 

 

44
increasing from two percent of the total volume in May to a
high of 36 percent in July. This increase was in an inverse
ratio to the volume of invertebrates present in the lake,
leading Ball to conclude that the bluegills turned to plant
food as the supply Of animal food diminished. Howell, Swingle,
and Smith (1941) working on ponds in the South came to the same
conclusion, and suggested that the fish "preferred" the animal
fOOd.

At the time the fish were stocked in Pond 5 the standing
crOp of fish food organisms had just been reduced by an emer»
genes. A further reduction in the volume of food organisms
took place during the two weeks following this emergence,
presumably due to predation by the fish. This was the period
during which the fish utilized ghaga to the greatest degree as
indicated by stomach analyses, and it corresponds closely to
the period through which the invertebrate food level of the
pond was at a minimum. A suggested conclusion is that the fish
began eating more ghgga because the supply of invertebrate food
was reduced to a level which could not meet the dietary require-
ments of the fish. This interpretation lacks definite proof,
however, and it is quite possible that other factors, such as
nutritional elements which may be contained in the plant mate-
rial making it preferred over the invertebrate food, influenced
the fish to feed on ghagg. The last two weeks of the sampling
period actually showed a decrease in volume of the benthos,

but the consumption of Chara instead of increasing as might be

45
expected also showed a decrease (Table 7), indicating further
that a simple inverse relationship between consumption of ghggg
and volume of invertebrates present may not constitute the
entire explanation of the recorded feeding habits.

Cladocera were by far the most abundant invertebrates
eaten during the period of stomach sampling, making up 75.2
percent of the numerical total. This figure creates a false
impression of the importance of the group as a fish food, as
indicated by the low percent (3.1) it comprised of the total

volume even when Chara is eliminated from the calculations.

 

Centroptilum ranks next to the Cladocera in numerical
importance and volumetrically it is by far the most important
group of invertebrates. _It was utilized as a source of food
by 72.5 percent of the fish. The creeping mayfly of the genus

Caenis, having habits which are very similar to Centroptilum,

 

was not utilized to the extent that the latter was, comprising
only 1.4 percent of the organisms by number and 1.3 percent by
volume. This group occurred in only 30 percent of the stomachs
examined.

Two remaining groups were of major importance in the diet,
Gastropoda and Chironomidae in that order. The percent compo—
sition that these two groups made up of the total is not in
agreement with the findings of other workers, the chironomids
occurring as a low percent (9.4 percent by volume and 5.6
percent by number) and the gastropOds as a much higher percent

(15.1 percent by volume and 12.1 percent by number) of the

46
total. Ball and Tanner (1951) found that mollusks constituted
4.3 percent of the organisms by volume in the stomachs of pump»
kinseed sunfish and only a trace in bluegill stomachs, while
the numerical composition was about the same. Midges, howeven
made up about 25 percent for both species volumetrically, and
between 41 and 53 percent by number. Ball (1948) and Patriarche
(and.Ball (1949) also found large prOportions of midges and small
,prOportions of gastropods in the stomachs of bluegills which
they examined.

The reason for this departure from the apparently normal
feeding habits of the fish in Pond 5 is not evident. It is
;possible that the chironomids were not readily accessible as a
:food because of the screening effect of the ghggg which covered
the bottom .

The remainder of the organisms utilized in the diet may
'be considered insignificant as a source of food. The group
Oligochaeta accounted for 5.94 percent of the total volume,
although only two individuals of this group were taken as food.

‘Hexagenia sp. were common in the bottom samples but were coma

 

pletely absent in the stomachs. This was probably due to the
burrowing habits of the nymphs (Lyman, 1943) which make them
relatively safe from fish predation except at the time of
emergence.

All debris was eliminated from consideration, as suggested
by Leonard (1940) who assumed that the organisms occurred in
the debris in about the same proportion that they occurred in

the recognizable material.

47
Organisms smaller than Cladocera, or the smaller midges
observable without the aid of a binocular micros00pe, were not
considered in the feeding habit study. Because all fish stocked
were at least two inches in length it was believed that the
volume of these small organisms would not justify the addi-

tional work necessary to isolate them.

Forage Ratios

Measurement of the standing crop of bottom fauna in a body
of water has been prOposed as an indication of its ability to
:support a population of fish. This concept fails to recognize
the fact that fish show a certain selectivity in choosing the
(organisms which make up their diet and, moreover, that many
organisms present are unavailable as a source of food.

Muttkowski (1918), in a study of the fauna of Lake Mendota,
VNaS one of the first to recognize that the various food groups
Eire not present in the stomach contents of fish in the same
;pr0portion as in the environment of the fish. Surber (1930)
Idsed a comparison of the numericad. occurrence of faunistic
ggroups in the stomachs of the fish with the same arOups as they
occur in the environment in an effort to quantitatively deter-
rnine the degree of selectivity exercised by them in taking the
food available in their environments.

The term "forage ratio" was prOposed by Hess and Swartz
(1941) and was defined as "the ratio of the percentage which a
given kind of organism makes up of the total stomach contents

to the percentage which this same organism makes up of the total

48
population of food organisms in the fish's environment", and
they suggest that these percentages may be calculated from
numbers, volume, or weight. This was probably an outgrowth of
work done by Hess and Rainwater (1939) in which these ratios
were set up as a measure of preference for certain organisms.
Leonard (1942) applied this technique in his study of the winter
feeding habits of brook trout fingerlings.

Allen (1942) uses the term "availability factor" for this
same relationship and prOposes a method for comparing bottom
fauna populations as sources of available food. Because the
availability factor varies with the size, structure, and habits
of an animal, he concludes that "although the availability factor
appears to bear a definite relationship to the extent to which
a species is available,.... it is not in itself a direct measure
of availability."

The forage ratio as described by Davis (1938) was used in
this experiment to compare the relative occurrence of organisms
found in the stomachs with those taken by dredge sampling. Ball
(1948) considers volume measurement a better index to the benefit
a fish receives from food organisms, and number measurement a
better index of the effort used in procuring the food (8616C“
tivity).

All groups which did not occur in the stomachs or were not
present in large enough volume to be considered important in
the diet were eliminated from consideration. This excluded

terrestrial insects as well as some aquatic invertebrates. The

49
group Hemiptera was represented in the stomachs by only three
individuals, and these ocCJrred as an unmeasureable amount in
the bottom samples. Cladocerans were omitted as a group because
their large numbers indicated that they were more important as a
food than was actually the case when their rather small total
volume was considered.

In determining the forage ratios, the percent composition
that the 12 groups shown in Table 5 made up of the total number
taken in stomachs and by dredge was recalculated on the basis
of these groups comprising 100 percent of the total.

In inspection of Table 5 reveals an extremely high forage
ratio for the group Gastropoda. This could be due partly to
error introduced in sorting the snails taken in dredge samples
into a size class utilized by the fish as food and a size class
too large to be selected as food. Even if an allowance were
made for bias in the arbitrary division of this group into size
classes, the ratio would still be high when compared to the
results of other workers. 9

Five groups, Anisoptera, Centroptilum, GastrOpoda, midge
pupae, and Trichoptera had forage ratios of greater than one,
indicating a selectivity on the part of the fish for these
organisms. The Trichoptera might well be excluded from this
grouping as it exhibited a forage ratio of just slightly over
one (1.01). The inclusion of Anisoptera and nudge pupae in
this classification might also be questioned on the grounds
0f low numerical occurrence of the former and low percentage

0f total volume shown by the latter.

50

A comparison of the creeping mayflies Caenis and Centrop-
tllgm shows a lower forage ratio for the first group than might
be expected. The apparent preference which the fish exhibited

for the Centroptilum may have been due to the more active

 

swimming habits of this mayfly which would present it to the
fish's view more often and for longer periods of time than

the more sluggish Caengg. A contributing factor to the higher

 

ratio shown by Centroptilum was probably the avoidance reaction

 

to the dredge which these organisms showed, with the result
that they were present in the fauna of the pond in greater
numbers than indicated in the bottom samples.

The low forage ratio of .29 shown for the group Chironomidae
appears to be in direct disagreement with the findings of Ball
(1948), Ball and Tanner (1949), and Patriarche and Ball (1949)
who observed forage ratios of 2 to 3, 3.3, and .45 to 4.7
respectively. It is quite possible that the low forage ratio
observed was the result of a situation in which the chironomids
were made unavailable by the heavy mat of ghggg, described
earlier, which covered the pond bottom.

The remaining groups of organisms were of minor importance
in contributing to the diet of the fish and were all represented
by forage ratios of less than one, indicating that either the
fish did not select these forms, or the habits of the groups

made them unavailable as a fish food.

 

51
YIELD OF FISH AVD ROTTOM FAUNA IN ”ONE 5

There is recognition of the need for knowledge of standing
crOps of fish and their yield to fishermen in the intelligent
management of our lakes and streams. This knowledge alone,
h0wever, is not enough. If these waters are to be nanaged in
a way which will produce more fish for more peovle, it will
be necessary to estimate the amount of fish a particular body
of water will produce as indicated by its fooi resrouces. This
phase of the experiment was designed to contribute to that
type of information.

The term "fish production", as used by Picker (1948),
indicates the amount of additional material which is formed
by the conversion of food at that level of the food cycle
which is occupied by fish, and “yield" as the amount of matter
passing from any one level to a particular organism at the next
level. At any level this yield is often termed the "crop".

In this investigation the production of fish in Pond 5 is
considered as the weight increase of the standing crOp of fish
from the time they were stocked until they were removed at the
conclusion of the experiment. The yield is defined as the
total weight of fish removed at the conclusion of the experiu
ment. I

Ricker (1948) outlines a method for determining fish
production and yield in a body of water. The first part of

his proposal parallels the plan that was followed in this

investigation,

i.e. a quantitative study of the standing

52

crop

of food organisms and anemalysis of the stomachs of the fish

present.

He suggests controlled feeding experiments as the

next step to determine the rate of digestion for the various

organisms being utilized as fish food and the efficiency of

conversion of this

Table 5.

more important invertebrates collected
in stomachs and dredge samples, and

forage ratios for these groups.

fond to flesh.

Percentage composition by number of the

 

% of total % of total Forage

in stomachs in dredge ratios
Anisoptera .83 .41 2.02
Caenis 2.21 6.01 .37
Centroptilum 39.47 15.73 2.51
CeratOpOgonidae 5.80 6.02 .96
Chironomidae 17.57 60.86 .29
Coleoptera 1.47 1.62 .91
Gastropoda 28.24 2.47 11.43
Hydraoarina .18 .34 .53
Midge pupae 2.48 2.09 1.19
Pisidium .09 .69 .13
Trichoptera 1.57 1.56 1.01
Zygoptera .09 2.20 .04

53

In this experiment, by calculating the difference in
benthic levels between Pond 4, used as a control, and Pond 5,
the weight of organisms utilized in producing the observed
increase in weight of fish was determined directly. With this
knowledge of the amount of food consumed and the increase in
weight of fish for the period, it was possible to work out
conversion factors without going through a complicated investi-
gation of digestive rates, efficiency of conversion of indi-
vidual organisms, and life cycles of the various insects.

The weight of fish which shOuld be stocked in Pond 5 to
produce an almost complete utilization of the bottom fauna
presented a problem. It appeared that the 160 pounds which
were stocked fell short of the weight that would have most
efficiently utilized the food resources of the pond.

Wilkins (m.s.) stocked 124 pounds of fish in the same
pond a year earlier and observed a rate of increase of 4.2
percent of the original weight per week for the period from
July 9 to September 9, as compared to an increase of 5.9 per~
cent of the original weight per week in the present experi—
ment. At the same time Wilkins found an average standing crop
of food organisms equal to 1.15 cc. per square foot, a figure
that was more than twice as high as the .45 cc. per square foot
found in the current investigations. If increasing the weight
of fish stocked from 124 to 160 pounds was responsible for so
great a reduction in the standing crOp, then the 10gical con-

clusion is that the carrying capacity of the pond is being

54
approached. However, the standing crop of fish food organisms
in Pond 4 was prOportionately low when compared to the stand-
ing crOp which Wilkins reported when the organisms were released
from predation, indicating that the low standing crOp in the
present investigation may not have been due entirely to an
increased intensity of predation.

There are three possible explanations for the rapid
increment of growth observed in this experiment: (1) a greater
proportion of smaller fish was stocked for this investigation
than was stocked by Wilkins. Growth studies indicate that
young fish gain weight more rapidly than older fish; (2) most
of the pumpkinseed sunfish in the smaller size class were in
poor condition when stocked, with the result that they possessed
a greater growth potential; (3) the abundance of thgg in Pond
5, utilized extensively in the diet of the fish, provided an
easily accessible source of food.

The weight of fish that a given body of water can produce
per unit area varies greatly with the climatic conditions of
the region in which the water occurs, as well as between bodies
of water in similar climatic regions.

Viosca (1935) found a one-tenth acre pond in Louisiana
that contained fish at the rate of 860 pounds per acre, and
he estimates the total standing crop of South Louisiana spring-
fed creeks to be between 300 and 500 pounds per acre. The pro-
duction of fish in other southern lakes and streams has been

comparably high. Thompson and Bennett (1939) investigated

55
seven Illinois lakes, ranging in size from 1 to 12 acres, that
supported from 232 to 1,143 pounds of fish per acre; Swingle
and Smith (1939), working on small ponds in Alabama, reported
130 to 200 pounds per acre in unfertilized ponds and 578 pounds
per acre in a fertilized pond; Tarzwell (1941) reports a total
standing crOp of 219 pounds per acre in an Alabama pond less
than 2 acres in surface area; in studying production in ferti~
lized terrace—water ponds in Alabama, Rwingle and Smith (1940)
found production as high as 657 pounds per acre.

Lakes in more northern latitudes have been found to
characteristically produce smaller standing crops than those
in the South, although at least one worker, Juday (1938),
reports 357 pounds per acre (of which 124 pounds consisted of
game and panfish) in Lake Wingra, Wisconsin. The total weight
of fish reported for 6 lakes in Michigan by Esohmeyer (1938)
varied from 21 to 194 pounds per acre, the lakes with the
greater shoal areas producing the most fish. Other workers in
Michigan (Brown and Ball, 1943) and Wisconsin (O'Donnell, 1943)
have found standing crops of 86.6 to 186 pounds per acre on
four lakes.

It is interesting to compare the above to three Nova
Scotian lakes found by Smith (1938) to be supporting fish
populations of 17.0, 19.9, and 36.0 pounds per acre. These
lakes were from 45 to 55.8 acres and of the acid-bog type,

exhibiting little growth of higher aquatic vegetation.

56

From the preceding information it appears that the weight
of fish in Pond 5 approaches an average of those weights found
in temperate lakes. It is quite pOssible, however, that a
shallow, relatively rich pond with abundant vegetation, such
as Pond 5, could be expected to support a population closer
to the maximum rather than the average Observed for this
part of the country. That this may be true is indicated by
the high rate of production of fish in Pond 5, an increase of
36 percent of the original weight taking place in a six-week
period.

By the time Pond 5 was drained in September the original
160 pounds of fish stocked had increased to 217.1 pounds, a
net gain of 57.1 pounds in six weeks. It is reasonable to
expect that this rate of gain, if applied to the population
for the entire growing season for one year, would have resulted
in an increase of around 80 pounds or 50 percent of the ori-
ginal weight, allowing for the cooler weather before and after
the period during which the growth was measured.

The weight of food organisms required to produce 57.1
pounds of fish, as determined by the difference in benthic
levels between Ponds 4 and 5, was 64.3 pounds. The result-
ing conversion ratio of 1.1 for organisms to fish flesh
(Table 6) is extremely low, and obviously does not represent
all the food consumed by the fish. For this reason a second
conversion factor was determined which included ghagg as a

constituent of the matter used as food by the fish.

57
Because a better method was lacking, gaggg was regarded
as being eqdivalent in food value to an equal volume of the
benthos. The reasoning involved in calculating the volume of
thgg ingested follows: if the difference in the levels of
the standing crOps of the two ponds is the result of fish
predation in Pond 5, and 64 percent of the stomach contents
were composed of gaggg, then the 64.3 pounds of food which ’
we supposed the fish had consumed must be only 100-64 or 36
percent of the total food ingested. If .36 of the total food

consumed is equal to 64.3 pounds, then the total food cone

 

sumed must be 178.? pounds. The difference between the weight
of invertebrates consumed and the total weight of food ingested
is the weight of ghggg (114.4 pounds) eaten by the fish in
the six—week period.

The conversion rate based on total food consumed is 3.1,
a figure which compares favorably with those ordinarily associ-
ated with hatchery production, but lower than the ratio of
5 to 1 as used by several German workers and suggested by
Richardson (1921) for fish living primarily on animal food.
Moore (1941) found a specimen of Lepomis cyanellus which
showed a conversion rate of 1.9 for a six-week period of con-
trolled liver feeding. He considered this efficiency higher
than average. The figure 3.1 indicates a slightly more effi-
cient conversion than is actually the case due to the ingestion
of terrestrial insects and other organisms that did not occur

in the bottom samples. The effects that emergence might have

58

had on the level of food organisms in Pond 5 were balanced by

similar emergences in Pond 4.

Table 6. Calculation of conversion ratios.

 

Pounds per acre

 

 

 

With Ehara Without Chara
FOOd level in Pond 4 132.5 132.5
Foou level in Pond 6 68.2 68.2
Difference in levels 64.3 64.3

-Weight of ghaga consumed 114.4

Total food consumed 178.7 64.3
Weight of fish removed 217.1 217.1
Weight of fish stocked 160.0 160.0
Increase in weight 67.1 57.1

Conversion ratio lf—f%-“ %%f% = 1.1

Table 6 shows a summary of the yield of fish flesh in

Pond 5 and the weight of bottom organisms and Chara used to

produce the observed increase in weight.

The total standing

crop of organisms in Pond 5 during the period the fish were

present was 365 pounds per acre.

fish food organisms.

This figure includes non—

 

59
RTATISTICAL ANALYSIS

The dredge sampling data have been examined statistically
by subjecting them to an analysis of variance in order to deter-
mine where the variation in volumes of fish food organisms had
their origin. Data used in the analysis have been confined to
fish food organisms only.

Bartlett (1947) regards a correlation between the mean
and the variance as a condition which is undesirable in an
analysis of variance. Such a correlation was found to exist
in the data from Ponds 4 and 5. To correct this condition
Bartlett suggests a transformation of the data in order to
change the scale of the measurements and make the analysis
more valid. The transformation of all values to legarithms
was found by Ball and Hayne (1952) to produce a set of data
adapted to an analysis of variance and this technique was
followed here. Data for the first week were eliminated for
both ponds because in several samples the volume of fish food
organisms was zero, and no logarithmic transformations may be
made of a zero value. It was believed that this action was
justified and may have been beneficial by eliminating a period
when sampling inaccuracies introduced by the experimenter may
have been at their peak.

The analysis of variance (Table 7) shows highly signifio
cant differences among weeks and also among stations, implying

that some of the weekly changes in Ponds 4 and 5 (Figureéfi are

-~ _. T—{l'

 

60
Table 7. Analysis of variance of the legarithms

of volumes of food organisms in Ponds 4 and 5.

 

 

 

 

 

Pond 4
Source of variance D.F. S.S. M.S. "F" ratio
Total 179 27.1 0.15
Among weeks 8 7.8 0.98 10.88*’
Before and after fish 1 4.4 4.40 8.88*
Within periods 7 3.4 0.49
Among stations 19 5.4 0.28 3.11"
Error 152 13.9 0.09
Pond 5
Source of variance D.F. 8.3. M.S. "F" ratio
Total . 179 31.1 0.17
Among weeks 8 3.5 0.44 3.14"
Before and after fish 1 0.4 0.40 0.91
Within periOds 7 3.1 0.44
Among stations 19 6.1 0.32 2.29"
Error 152 21.5 0.14

 

statistically significant. The coefficient of variation amounts
to about 36 percent of the mean for each weekly determination
in Pond 4 and 48 percent of the mean in Pond 5. Although these

values are large, they are not unreasonable for data of this

81
type. Significant differences among stations implies that
certain consistent differences in volumes of fish food organisms
occur at scattered points on the floors of the two ponds. This
means that the bottom fauna are not distributed in a uniform
manner but are more concentrated in some areas than in others.
This condition was reflected in the occurrence of much greater
concentrations of certain invertebrate groups at some stations
than at others.

Differences among weeks may be examined to discover
whether the volume of the standing crOp was perceptibly changed
by the introduction of fish in Pond 5. In the analysis, a
comparison of levels of bottom fauna before and after the date
of introducing the fish into Pond 5 shows that no significant
change occurred in that pond, but that the increase in Pond 4
was significant at the 5 percent level. Instead of reducing
the level of bottom fauna in Pond 5 to any great degree, it
appears that the fish population was instrumental in holding
the level close to that which existed before fish were intro-
duced, while the organisms in Pond 4, not subject to predation,
showed a much greater variability which is reflected in the
marked increase in the standing crop of fooa organisms during
this period. Apparently the fish were introduced before the
standing crop of benthos had reached its peak, with the result
that the bottom organisms were growing at a rate great enough
to balance predation by the fish in Pond 5. In Pond 4, where

the effects of predation were not felt, the growth potential

 

wt a--' 1n}

of the benthos resulted in a

standing oron of organisms.

substantial increase in the

62

63

SUMMARY

It was the purpose of this investigation to determine
the relationships between the bottom fauna of a small pond and
an introduced population of fish dependent upon them for the
greater part of their food.

Two similar ponds were used in the investigation, one
containing a populating of bluegills and pumpkinseeds; the
other, without fish, was used as a control. Results indicated
that the fish which were stocked in Pond 5 utilized selected
groups of organisms in their diet to a degree that was effec»
tive in limiting the standing crOp of benthos in this pond to
a level which was only half that observed in Pond 4. Apparently
the standing crOp of invertebrates was well below the maximum
that might have been produced if fish had not been present, as
evidenced by a doubling of the standing crOp of organisms in
Pond 4.

Although certain organisms (Oligochaeta, Hexagenia, large
snails) made up a large part of the total volume of inverte-
brates collected by dredge, their burrowing habits or morpho»
logical characteristics prevented their extensive utilization
as a fish food. Mayflies, snails, and midges were the most
important foods as revealed by stomach analyses.

The fish in Pond 5 turned to ghggg as a main food item
during the middle of the summer. This alga was the most impor-

tant food during the first weeks of stomach sampling, but its

 

84
presence in the stomachs decreased steadily until the end of
the investigation, when the volume of ghgga being consumed
became negligible. It appears that §Q§;g_was used as a supple«
ment to the diet of invertebrates, but actual proof as to why
the fish turned to this plant food is lacking.

An average standing crOp of 585 pounds of bottom fauna
per acre resulted in an increase in the weight of fish stocked
from 180.0 to 217.1 pounds per acre in a six—week period.
Calculation of the weight of food consumed by comparison of
the standing crops of food organisms in Ponds 4 and 5 revealed
a conversion ratio for food ingested to weight of fish pro-
duced of 1.1 to l, or 3.1 to 1 when 9h; 1 was included in the
calculations. These rates represent a more efficient conver-
sion than is actually the case, due to ingestion by the fish
of terrestrial insects and other material which was not taken
by dredge sampling.

A statistical evaluation of the data indicated that fish
predation in Pond 5 was effective in holding the benthic popu-
lation at a constant level, while the benthic population in
Pond 4 showed a significant increase in volume during the same

period of time.

 

LITERATURE CITED

Allen, K. Radway

1942 Comparison of bottom faunas as sources of available
fish food. Trans. Am. Fish. Soc., 1941, 71:
275-283.

1951 The Horokiwi Stream. A study of a trout pOpulation.

New Zealand Marine Dept. Fish. Bull. No. 10.

Ball, Robert C.

1948 Relationship between available fish food, feeding
habits of fish and total fish proauction in a Michia
gan lake. Tech. Bull. 208, Michigan State College
Ag. Exp. Sta.

1949 Experimental use of fertilizer in the production
of fish—fOOd organisms and fish. Tech. Bull. 210,
Michigan State College Ag. Exp. Sta.

---------- and Howard B. Tanner
1951 The bi010gical effects of fertilizer on a warm«
water lake. Tech. Bull. 223, iflohigan State
College Ag. Exp. Sta.

.......... and Don W. Hayne

1952 Effects of the removal of the fish population on
the fish-food organisms of a lake. Ecol., 33(1):
41-48.

Bartlett, M. S.
1947 The use of transformations. Biometrics, 3:39-52.

Brown, C. J. D. and R. C. Ball
' 1943 A fish population study of Third Sister Lake.
Trans. Am. Fish. Soc., 1942, 72: 177-188.

Clarke, George L.
1948 Dynamics of production in a marine area. Ecol.
MonOg., 18: 321-335.

Davis, H. S.
1938 Instruction for conducting stream and lake surveys.
U.S. Bureau of Fisheries, Fish. Ciro. No. 28, pp.
1-55.

66

Deevey, Edward S.

1941

a...-- av"a---

Eggleton, Frank E.

1931

u-Luw“-‘i‘—‘-

-l—--—--—-B

Eschmeyer,
1938

Forbes, S.
1903

Hess, A. D.
1939

63—-‘--‘--

Limnological studies in Connecticut. VI. The
quantity and composition of the bottom fauna of
thirty-six Connecticut and New York lakes. Fool.
Mong., 11: 413~455.

and James 8. Bishop

Limnology. Sect. 2. A fisheries survey of impor«
tant Connecticut lakes. Bull. 83, Conn. Geol.

and Nat. Hist. Surv., pp. 89-121.

A

A limnolOgical study of the profundal bottom fauna

of certain freshwater lakes. Ecol. Mong. 1: 213“ ”

332.

A comparative studywof the benthic fauna of four
northern Michigan lakes. Papers of the Mich. Acad.
of 801., Arts, and Lett., Vol. XX, 1934.

Productivity of the profundal benthic zone in Lake
Michigan. Papers of the Mich. Acad. of Sci., Arts,
and Lett., Vol. XXII, 1938.

R. W.

The significance of fish population studies in lake
management. Trans. Third N. Am. Wildlife Conf.,
pp. 458a463.

A.
The food of fishes. Bull. Ill. State Lab. Nat.
Hist., 1(3): 19:70.

and J. H. Rainwater

A method for measuring the food preference of trout.

Copeia, 1939, No. 3, pp. 154-157.

and Albert Swartz

The forage ratio and its use in determining the food

grade of streams. Trans. Fifth N. A. Wildlife
Conf., pp. 182~184.

Howell, Henry H.

1941

“his.--““D

Bottom organisms in fertilized and unfertilized
fish ponds in Alabama. Trans. Am. Fish. Soc.,
1941, 71: 185-179.

, H. S. Swingle, and B. V. Smith

Bass and bream food in Alabama waters. Alabama
Cons., 1(4): 3.

mo“: 0.: "wt-mm: 4 .ir

67

Juday, Chauncey
1938 Fish records for Lake Wingra. Trans. Wise. Acad.
Sci., Arts, and Lett., Vol. XXXI, pp. 533—534.

Leonard, Justin w.

1940 Further observations on the feeding habits of the
Montana grayling (Thymalus montanus) and the_blue-
gill (Lepomis macrpchirugj in Ford Lake, Michigan.
Trans. Am. Fish. Soc., 1939, 89: 244-258.

-3 I.-.-_‘-¢J.I_-

1942 Some observations on the winter feeding habits of
brook trout fingerlings in relation to natural food
organisms present. Trans. Am. Fish. Soc., 1941,
71: 219-227.

Lindeman, Raymond L.
1942 The trOphic-dynamic aspect of egolOgy. Ecol. 23:
399-418. ., ;
Lyman, F. Earle
1943 A pre-impoundment bottom fauna study of Watts Bar
Reservoir area (Tennessee). Trans. Am. Fish. Soc.,
1942, 72: 52-82.

1943 Swimming and burrowing activities of mayfly nymphs

of the genus Hexa enia. Annals Ent. Soc. Am.,
June, 1943, 38(2): 250-258.

Moore, Walter G.
1941 Studies on the feeding habits of fishes. Ecol.,
Jan., 1941, Vol. 22, No. 1, pp. 91-98.

Muttkowski, Richard A.

1918 The fauna of Lake Mendota. A qualitative and
quantitative survey with special reference to the
insects. Trans. Wise. Acad. 801., Arts, and Lett.,
19(1): 374-482.

O'Donnell, D. John

1943 The fish population in three small lakes in northern
Wisconsin. Trans. Am. Fish. Soc., 1942, 72: 187-
198.

Patriarche, Mercer H. and Robert 0. Ball
1949 An analysis of the bottom fauna production in ferti-
lized and unfertilized ponds and its utilization by
young-of—the-year-fish. Tech. Bull. 207, Michigan
State College Ag. Exp. Sta.

68

Pearse, A. S.

1921 Distribution and food of the fishes of Green Lake,
Wisconsin, in summer. U. S. Bur. Fish. Bull. 37,
pp. 256-272.

Richardson, Robert E.

1921 The small bottom and shore fauna of the Middle and
Lower Illinois River and its connecting lakes,
Chillicothe to Grafton: its evaluation; its sources
of food supply; and its relation to the fishery.
Bull. Ill. Nat. Hist. Sur., Vol. 13, pp. 359-522.

Ricker, William E.
948 Production and utilization of fish populations.
Ecol. Ionog., Oct. 1948, 18: 373—391.

Scott, Will, Ralph D. Hile, and Herman Spieth
1928 A quantitative study of the bottom fauna of Lake
Wawasee (Turkey Lake). Publ. No. 77, Dept. of
Cons., State of Indiana.

Smith, M. W.
1938 A preliminary account of the fish pOpulations in
certain Nova Scotian lakes. Trans. Am. Fish. Soc.,
1937, 87: 178-183. '

Surber, E. W.
1930 A quantitative method of studying the food of small
fishes. Trans. Am. Fish. Soc., 1930, 80: 158-183.

Swingle, H. S. and E. V. Smith
1939 Fertilizers for increasing the natural food for fish
in ponds. Trans. Am. Fish. Soc., 1938, 88: 128-134.

---------- and E. V. Smith
1940 Fish production in terrace-water ponds in Alabama.
Trans. Am. Fish. Soc., 1939, 89: 101-105.

Tarzwell, Clarence M.
1941 The fish population of a small pond in northern
Alabama. Trans. Fifth N. Ant Wildlife Conf., pp.
245-251.

Thompson, D. H. and G. W. Bennett
1939 Fish management in small artificial lakes. Trans.
Fourth N. Am. Wilklife Conf., pp. 311-317.

69

Viosca, Percy, Jr.
1935 Statistics on the productivity of inland waters:
The master key to better fish culture. Trans. Am.
Fish. Soc., 1935, 85: 350-358.

Wilkins, Lloyd P.
1952 Relationships of a fish population to the inverte—
brate fauna in two small ponds. Thesis, Michigan
State College. Unpublished.

" mm m "vita“! ‘

 

 

 

 

'l

w»

HICHIGHN STQTE UNIV. LIBRRRIES

31293104807296