|
i
i
:
:
i
|
i
º
§:ºº
º
º
#.
i.
|
.
:
º
• * * * * * *
. . ; ) ºr ; ; ;
" ; , .
* , , , , , ,
tº tº
tº
twº-
A tº
º
sº
* * * * *
wº
* .
º
:
* * * *
"He kº ºr
: , ;
* . . . ;
* * * * * *
§ , , ..."
:: * * * * * , ,
tº a
º
º:
ſº º º
* * * * : .
|
º
* * * * ºr
º
º:
* jº
º' tº
º,” º º º & ºn “tº º
5 * *
º **
g :: * *
º * * * *
º **** * * * * * * * * * * º
A º
* : * :
º
ºw. " " ' "
º
*...º.o.º.w.
***** * *****
º * ,
* : * *
* . Mº $º
lºw is ºf
* * * * : * * *
º, tº sº it.
tº $. is . . . . . * * * * * * *
it * º:
w ; : : -º -
- ; : ºº º
tº r tº
sº the
º #:
* Hºº & tº sº.
t - ſº
Miº ºr y
tº ºsºti :
: º sº gº
tº º
ºt
\º
- . A " " ºf
*** * * * * * *
* tº tº
* * * : *
A
º
º tº
º: "º
** * * *
* * * * : *:
a sº
* * º:
* , º sº *.*.
º:
º t ".
* * * * * * * * * * * * * * * * *
* … º ºx at-º-º-º:

1. 3 1 7
P R O P E R T Y O F
6.
#
º ºgº
#
*ś §§§
.*
; ; N T W A V E R T T $
(ſ/ſ/lſº
-*.
-ºr



A COMPARISON OF THE DIET AND GROWTH OF BROWN
TROUT (SALMO TRUTTA) FROM THE SOUTH BRANCH
AND THE MAIN STREAM, AU SABLE RIVER,
MICHIGAN
By
Thomas E. Stauffer
A thesis submitted in partial fulfillment
of the requirements for the degree of
Master of Science
in Fisheries
School of Natural Resources
The University of Michigan
April 1977
Committee members:
Dr. James T. McFadden, Chairman
Dr. Frank F. Hooper
Dr. William C. Latta
ACKNOWLEDGMENTS
This research was supported by the George Mason gift to the
Michigan Department of Natural Resources. I wish to express my
appreciation to Drs. James T. McFadden and W. C. Latta for
encouragement and suggestions in the course of the project and for
critical review of the manuscript. I would like to thank Dr. Frank F.
Hooper for serving on my committee. I am especially grateful to
Gaylord R. Alexander for willingly providing comments and informa-
tion about fish populations of the Au Sable system, and for reviewing
the manuscript. Also I wish to thank William J. Buc for the
Au Sable trout population estimates. James R. Ryckman was a
great help with the statistical analysis of the data. Jack D. Rodgers Jr.,
James W. Strogen Jr., and Jere L. White assisted in the summer
collection of trout. Thanks also go to Margaret S. McClure for typing
the manuscript and to Alan D. Sutton for drafting the figures.
ii
TABLE OF CONTENTS
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . © C C C C
LIST OF TABLES . . . . . . tº C C C C C & © © C C C C C C C,
LIST OF FIGURES . . . . . . . . . . . . . . . . . tº e C Q @
INTRODUCTION . . . . . . & e º O e C C C C C e º e o Gº Gº º
METHODS tº dº º O & © e º © º o Q © 6 e s e o e e o e o e e e Q
RESULTS . . . . . . . . . . . . . . • * e o e s e e e o e o e
Stream-side Observations . . . . . . . . . © e º O O.
Stomach Sample Analysis . . . . . . . . © C Q @ º e
Age Structure . . . . . . . . . Q @ e º 'º e º e O © tº C
STATISTICAL ANALYSIS © e Q & D C e º e © C & © C C C & Gº
Stomach Sample Analysis © O O © e Q O O Q & O & © C
Length-weight Relationship . . . . . . . . . . . . ©
Seasonal Growth . . . . . . . . . . . . . . . . . © e
Page
ii
iv.
Vi
TROUT POPULATION SIZE AND g
FOOD PRODUCTIVITY . . . . . . • - - - - - - - - - © C C C
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . .
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES CITED . . . . . . . . . . . © C Q & Q Q @ e º
20
25
25
32
34
37
45
47
49
iii
LIST OF TABLES
Table
10.
11.
Mean volume of stonnach contents of brown trout
3. 0-5.9 inches total length from the South Branch
Au Sable River, May–September, 1976 . . . . . & O ©
Mean volume of stomach contents of brown trout
7.0-9.9 inches total length from the South Branch
Au Sable River, May–September, 1976 . . . . . tº C (9
Mean volume of stomach contents of brown trout
3. 0-5. 9 inches total length from the Main Stream
Au Sable River, May–September, 1976 . . . . . . . .
Mean volume of Stormach contents of brown trout
7.0-9.9 inches total length from the Main Stream
Au Sable River, May–September, 1976 . . . . . . . .
The most common families of aquatic organisms
in Stonach samples and their common names . . . .
Monthly averages of length and weight for brown
trout from the South Branch, Au Sable River, 1976. .
Monthly averages of length and weight for brown trout
from the Main Stream, Au Sable River, 1976 . . . . .
Results of the analysis of variance of stomach volume
data for brown trout from the South Branch and Main
Stream, Au Sable River, May to September, 1976 . .
Covariance test of the regressions of loge stomach
volume on loge total length for brown trout from the
South Branch and Main Stream, Au Sable River . . . .
Covariance test of the regressions of loge weight on
length for brown trout from the South Branch and
Main Stream, Au Sable River . . . . . . . . . . . .
Covariance test of the regressions of loge length on
loge growing season for brown trout from the South
Branch and Main Stream, Au Sable River . . . . . . .
iv.
Page
10
11
12
21
22
28
30
33
35
Table
12.
13.
14.
Page
Number of fish per acre for the South Branch and
Main Stream Au Sable River, September 27-
November 9, 1976 . . . . . . . . . . . . . . . . O © O 38
Annual weight gain, mean stomach volume, esti-
mated daily ration and ratio of daily ration to
stormach volume for brown trout from the South
Branch and Main Stream, Au Sable River . . . . . . . 40
Estimated volume of food eaten per acre during the
1976 growing season (May–October) for the South
Branch and Main Stream, Au Sable River . . . . . . . 42
LIST OF FIGURES
Figure
Page
Location of sampling sites on the South Branch and
Main Stream, Au Sable River, Crawford County,
Michigan . . . . . © C Q O © [º º O O C & Gº O O © O & C 4
Percent composition of the diet of brown trout from
the South Branch Au Sable River, May–September,
1976 © ſº Q © © O º © © © © © © . o © O tº & Q º © º Q O º 15
Percent composition of the diet of brown trout from
the Main Stream Au Sable River, May–September,
1976 . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Mean stomach volume versus total length for brown
trout from the South Branch and Main Stream,
Au Sable River, May–September, 1976 . . . . . . . . 19
Seasonal growth of brown trout from the South
Branch Au Sable River, based on summer and
fall sampling, 1976 . . . . . . . . . . . . . . . © e C 23
Seasonal growth of brown trout from the Main
Stream Au Sable River, based on summer and
fall sampling, 1976 . . . . . . . . . . . . . . Q Q & Q 24
Vi
INTRODUCTION
The Au Sable River System in Crawford County, Michigan, has
received much recognition for its excellent recreational value. Trout
fishing has undoubtedly contributed much to the Au Sable's fame,
especially with regard to the management philosophies developed
through great public interest and scientific research.
A section of the Au Sable system which has received much
attention is the Main Stream from Burton's Landing downstream to
Wakeley Bridge. This 8.9-mile stretch has been designated as a
"special regulation" or "quality fishing" area, with fly-fishing only,
a 12-inch size limit for brown trout, an 8-inch size limit for brook trout,
and a three-trout creel limit. Another section with fly-fishing only
regulations is the Mason Tract on the South Branch, however here the
minimum size for brown trout is only 10 inches, and the creel limit is
five trout. This popular section was originally set aside as a "wilder-
| |
Ile SS
area, and offers few access sites to its 9.7 miles of river.
Although both the Main Stream and South Branch are excellent trout
fishing rivers, the total number and size composition of their trout
populations differ. Both rivers have excellent water quality and diverse
insect populations. Coopes (1974) gave a thorough description of the
2
two rivers with respect to habitat (bottom type and trout cover), water
quality, and human impact and use.
The present study seeks to aid both anglers and biologists who
have an interest in the rivers. Knowing the major types of food actually
eaten by the trout will aid fishermen in selecting appropriate imitations.
These results, coupled with knowledge of the number and sizes of brown
trout in each river, will enable the serious angler to plan his trips more
effectively. These same population and diet data will also help biologists
understand the Main Stream and the South Branch trout populations.
The specific objections of the study were to identify the kinds of
organisms eaten seasonally by different sizes of trout, to determine the
mean volume of food eaten per trout and to compare food, growth and
numbers of South Branch trout with Main Stream trout.
METHODS
Monthly samples of trout were taken from May through September
1976. These months represent the major portion of the growing season
for wild trout (Cooper 1953; Alexander and Gowing 1976). The stations
sampled on the Main Stream were Burton's Landing, Louie's Landing,
Keystone Landing, Wa Wa Sum, Stephan's Bridge, Pine Road, Thunder-
bird Club, and Wakeley Bridge; stations on the South Branch were
Chase Bridge, Marlbar, castle, Downey's, Dogtown, and Smith Bridge
(Figure 1). The gear consisted of a 230-volt DC generator mounted on
a small boat. The copper-sheeted bottom of the boat served as the
negative electrode, and was used with two portable positive electrodes.
Each month, 25 brown trout from 3. 0-5.9 inches total length
and the same number from 7.0-9. 9 inches were taken from each river.
Immediately after collection, the trout were transferred to an ice chest
and then transported to the laboratory, where they were measured to
the nearest 0.1 inch (total length), weighed to the nearest 0.1 gram, and
scale-sampled. Stomachs were removed from larger fish and preserved
in 10% formalin, while the smaller fish were slit along the side and
preserved whole. A label containing pertinent data was inserted in each
stomach. After one week, the contents of each stomach and the
respective label were transferred to a vial containing 80% alcohol. Each
3
MAIN STREAM
Stephan's
Louie's wºº Bridge
Landing - Pine
Burton's sº Road
Landing 9
SOUTH BRANCH
Chase
Bridge
R2W.
+
Wakeley
Bridge
– T. 26 N.
Smith Bridge
Dogtown
+
Downey's
Castle 49
—T 25 N.
Marlbar
Figure 1. --Location of sampling sites on the South Branch
and Main Stream, Au Sable River, Crawford County, Michigan.
5
sample was then sorted to taxonomic order and family, and the number
of individuals was counted. The volume of each familial type was
determined by liquid displacement to the nearest 0.025 ml. Inpressions
of trout scales were made on cellulose acetate squares and were mag-
nified with a microprojector. Age (number of annuli) was determined
for each fish.
Fall population estimates for several stations on the two rivers
were run by the Michigan Department of Natural Resources Fisheries
Division. At each station, trout were captured with an electroshocker,
fin-clipped, and returned to the river. Later, a second sampling with
the electrofishing gear yielded the number of marked and unmarked
trout. These data were then used for Peterson estimates of total
population size. The results were made available for this study, and
included the population size of brook, brown, and rainbow trout in each
inch group. Representative scale samples collected in the fall estimates
were used to determine the age of trout as described above.
RESULTS
Stream-side Observations
From observations made on the monthly sampling trips, it
appeared that the surface-feeding of trout on insects differed noticeably
between the South Branch and Main Stream. Although the monthly trips
occurred during the bright daylight hours, they were supplemented by
occasional evening visits to the two rivers. Without fail every month
on the Main Stream, daytime or evening, trout were surface-feeding to
some extent. Most of these fish appeared to be brook trout, with a few
smaller brown trout involved. These fish fed throughout the season on
the small black dance flies (Diptera: Empididae), but switched their
attention to larger insects during sizable emergences. These "hatches"
also stimulated the larger brown trout to surface-feed, especially toward
dusk. In May and June the fish fed intensively on emerging gray caddis
flies (Hydropsychidae, Brachycentridae). Small- to medium-sized
mayflies were eaten to a lesser extent. Much less surface-feeding on
dance flies occurred on the South Branch, and was confined to the evening.
Fairly good hatches of the mayflies Isonychia and Tricorythodes took
place, but the major event of the season, the Hexagenia (Michigan mayfly)
hatch, probably overshadowed all the others. The emergence of these
big insects prompted even the largest trout to surface-feed.
Ironically, I found no adult Hexagenia in the stomach samples,
although I did find some nymphs. The reasons for this were probably
two: (1) the sampling dates of June 15 and July 16 missed the peak of the
emergence; (2) adult Hexagenia eaten during the evening periods were
digested by the time we captured the fish on the following day. The latter
reason probably also explained why very few adults of other types of may-
flies were found in stomach samples. Nonetheless, I believe that the
South Branch trout benefitted greatly from Hexagenia, while Main Stream
fish did not have them as a food source. Although several nymphs were
found in the stomachs of trout from one station on the Main Stream, the
reports of knowledgeable anglers indicate that significant populations of
Hexagenia do not exist in the section of the river studied here.
Stomach Sample Analysis
Summaries of the data of brown trout diet for the Main Stream
and South Branch are presented in Tables 1 to 4. The most common
families of each order may be found in Table 5. For those who are
unfamiliar with insect taxonomy, this table also gives the common names
of the orders and families. Comparing the contribution of each taxon in
trout diet by percentages demonstrates, that in general; the types of food
eaten in both rivers were very similar. Also, it appears that shifts in
diet as the trout grow from small to large are of a similar nature in both
rivers. Comparisons of this sort may be more easily visualized in
Figures 2 and 3. (Notice that the label "others" in these figures was a
Table 1. --Mean volume Of stomach contents (ml) of brown trout 3.0-
5. 9 inches total length from the South Branch Au Sable River, May-
September, 1976 (sample size in parentheses).
Month Sea- Per-
Organism *. sonal cent
May June July Aug. Sept. mean compo-
(21) (25) (25) (25) (25) volume sition
Trichoptera 0. 175 0.023 0.025 0.012 0. 047 0.052 28. 2
Ephemeroptera 0.062 0.064 0.023 0.079 0.007 0.046 25. 0
Plecoptera 0.014 0.000 0.000 0.000 0.000 0.002 1. 3
Odonata 0.014 0.014 0.028 0.000 0.000 0.011 6. 0
Hemiptera 0.000 0.000 0.00 1 0. 004 0.000 0. 001 0. 6
Coleoptera 0.002 0.002 0.000 0.000 0.005 0.002 1. 0
Megaloptera 0.000 0.010 0.000 0.000 0.000 0.002 1. 1
Diptera 0.038 0.027 0.008 0.016 0.011 0.019 10.4
Mollusca 0.002 0.000 0.000 0.00 1 0. 003 0.001 0. 7
Fish 0.00 1 0. 000 0, 000 0.000 0.000 0.000 0. 1
Isopoda +
Amphipoda 0.000 0.000 0.00 1 0. 000 0.000 0.000 0. 1
Decapoda 0.000 0.000 0.000 0.000 0.000 0. 000 0. 0
Terrestrial 0.000 0.008 0. 007 0.00 1 0. 001 0.004 1. 9
Other 0.000 0.00 1 0. 000 0.000 0.000 0.000 0. 1
Unidentified 0.052 0.097 0.019 0.013 0.014 0.039 20.8
Annelida 0.019 0.000 0.008 0.000 0.000 0.005 2. 7
Monthly mean
volume 0.381 0. 246 0. 120 0. 126 0.088 0. 186 100. 0
Table 2. --Mean volume of stomach contents (ml) of brown trout 7.0-
9.9 inches total length from the South Branch Au Sable River, May-
September, 1976 (sample size in parentheses).
Month Sea- Per-
Organism sonal Cent
May June July Aug. Sept. mean compo-
(22) (25) (25) (25) (25) volume sition
Trichoptera 0. 281 0. 073. 0.059 0.065 0. 133 0. 118 16. 9
Ephemeroptera 0.034 0.029 0.094 0.059 0.001 0.044 6. 2
Plecoptera 0.018 0.02 1 0. 004 0.000 0.024 0. 013 1. 9
Odonata 0.173 0.038 0.072 0.032 0.007 0.062 8. 8
Hermiptera 0.000 0.006 0. 0 1 1 0. 006 0.086 0.022 3. 2
Coleoptera 0.000 0.028 0.003 0.005 0.005 0.008 1.2
Megaloptera 0. 016 0.000 0.000 0.000 0.010 0.005 0. 7
Diptera 0. 007 0.033 0.00 1 0. 0.53 0.018 0. 023 3. 3
Mollusca 0.038 0.036 0.018 0.044 0. 069 0.041 5. 9
Fish 0.000 0.012 0.031 0. 144 0.054 0.049 7. 1
Isopoda +
Amphipoda 0.000 0.000 0.000 0.000 0.000 0.000 0. 0
Decapoda 0.061 0. 000 0. 137 0.004 0.024 0.045 6.4
Terrestrial 0.009 0.062 0.027 0.044 0. 104 0.050 7. 2
Other 0. 198 0.008 0. 016 0.006 0.01.1 0.053 7. 6
Unidentified 0.116 0.240 0.285 0.056 0.112 0.168 23.3
Annelida 0.009 0.000 0.000 0.000 0.000 0.002 0.2
Monthly mean
volume 0.959 0.586 0. 758 0. 518 0. 702 10. 698 99. 9
10
Table 3. --Mean volume of stomach contents (ml) of brown trout 3.0-
5. 9 inches total length from the Main Stream Au Sable River, May-
September, 1976 (sample size in parentheses).
Month Sea- Per -
Organism - Sonal cent
May June July Aug. Sept. mean compo-
(25) (25) (29) (25) (25) volume sition
Trichoptera 0.042 0.047 0.016 0.003 0.019 0.025 20.0
, Ephemeroptera 0.024 0.035 0.042 0.006 0.011 0.024 19.4
Plecoptera 0.000 0.000 0.000 0.000 0.000 0.000 0. 0
Odonata 0.000 0.000 0.000 0.000 0.000 0.000 0. 0
Hermiptera 0.000 0.000 0.000 0.000 0.001 0.000 0. 2
Coleoptera 0.000 0.000 0.000 0.00 1 0. 000 0.000 0. 2
Megaloptera 0.004 0.000 0.000 0.000 0.000 0.001 0. 6
Diptera 0.010 0.014 0.01 1 0. 00 1 0. 000 0.007 5. 9
Mollusca 0.002 0.000 0.015 0.005 0.000 0.005 3. 7
Fish 0.022 0.000 0.000 0.00 1 0. 000 0. 004 3. 6
Isopoda +
Amphipoda 0. 069 0. 0 1 1 0. 0.25 0, 000 0.054 0.03.2 25. 3
Decapoda 0.000 0.000 0.002 0.000 0.000 0.000 0.3
Terrestrial 0.004 0.002 0.002 0.00 1 0. 000 0.002 1.4
Other 0.004 0.006 0.004 0.000 0.000 0.003 2. 3
Unidentified 0.025 0.025 0.023 0.016 0.011 0.020 16. 1
Annelida 0.000 0.006 0.000 0.000 0.000 0. 001 0. 9
Monthly mean
volume 0. 206 0. 146 0. 140 0.034 0.096 0. 125 99.9
11
Table 4. --Mean volume of stomach contents (m1) of brown trout 7.0-
9.9 inches total length from the Main Stream Au Sable River, May-
September, 1976 (sample size in parentheses).
- Month Sea- Per-
Organism - sonal Cent
May June July Aug. Sept. mean compo-
(25) (25) (26) (25) (25) volume sition
Trichoptera 0.319 0.097 0.171 0.079 0. 135 0. 160 20. 9
Ephemeroptera 0.023 0.210 0.043 0.000 0.007 0.057 7.4
Plecoptera 0.000 0.002 0.000 0.000 0.000 0. 000 0. 1
Odonata 0.020 0.018 0.000 0.000 0.000 0.008 1.0
Hemiptera 0.000 0.000 0.000 0.000 0.019 0.004 0. 5
Coleoptera 0.000 0.007 0.004 0.00 1 0. 000 0.002 0.3
Megaloptera 0.024 0. 0 10 0.000 0.000 0.000 0.007 0. 9
Diptera 0.010 0.038 0. 173 0.002 0.000 0. 047 5. 9
Mollusca 0.050 0.032 0. 047 0.051 0. 113 0. 0.59 7. 6
Fish 0. 186 0.038 0. 188 0, 246 0. 149 0.177 23. 1
Isopoda +
Amphipoda 0.055 0.061 0. 0 16 0.003 0.005 0. 028 3. 6
Decapoda 0.056 0.096 0.017 0.005 0.000 0.035 4. 5
Terrestrial 0.002 0. 167 0.004 0.008 0.030 0.042 5.4
Other 0.085 0.054 0.043 0.023 0.026 0.046 6. 0
Unidentified 0. 158 0. 122 0. 109 0.050 0.044 0.097 12. 6
Annelida 0.000 0.008 0.000 0.000 0.000 0.002 0.2
Monthly mean
volume 0. 988 1. 040 0.815 0. 468 0. 528 0. 768 100. 0
12
Table 5. --The most common families of aquatic organisms in stomach
samples and their common names.
Occurrence
Order and Family COrmn. On Nanne South Main
Branch Stream
Trichoptera Caddis flies
>
:::
Brachycentridae
Helicopsychidae >{<
>{<
:::
Goeridae
Glossosomatidae ><
>
k
>{<
Hydropsychidae
>{<
J
Lepidostomatidae
Leptoceridae - >k >{<
Philopotamidae
Ephemeroptera Mayflies 1
Ephemerellidae Hendrickson, Sulfur, Blue-
>{<
~Le
winged olive
Ephemeridae Brown drake and Michigan
mayfly ><
Baetiscidae >}<
Baetidae Blue-winged olive, Quill
Gordon, Sulfur - >{< ><
Heptageniidae Cahill, Quill Gordon, Sulfur,
March brown, Ginger quill,
Grey fox
Siphlonuridae Isonychia, Lead-wing coachman >|<
Tricorythidae Tricorythodes, Tiny white-winged
black >{< ><
(continued, next page)
13
Table 5. --continued
Order and Family
Common Name
Occurrence
South Main
Branch Stream
Plecoptera
Perlidae
Chlorope rlidae
Perlodidae
Pteronarcidae
Odonata
Gomphidae
Hemiptera
Corixidae
Belostomatidae
Coleoptera
Hydrophilidae
Dytiscidae
Neuroptera
Corydalidae
Diptera
Chironomidae
Simuliidae
Rhagionidae
Empididae
Mollusca
Physidae
Limnaeidae
Stoneflies
Dragonflies and Damselflies
Darners
True bugs
Water boatmen
Giant water bugs
Beetles
Fishflies
Two-winged flies
Midges
Black flies
Snipe flies
Dance flies
Molluscs
(continued, next page)
::
>{<
><
>
k
><
><
>{<
>{<
>k
J
K
><
>{<
•)
14
Table 5. --concluded
Order and Family
Common Name
Occurrence
South Main
Branch Stream
Perciformes
Cottidae
Percidae
Isopoda
Asellidae
Amphipoda
Gammaridae
Decapoda
Sculpins or Muddlers
Darters
Aquatic sow bugs
Scuds
Crayfish
>k
*
>{<
><
Common names of families of mayflies were taken from Caucci and
Nastasi (1975).
-*
15
Trichoptera
. 28%
Ephemeroptera
25%
Diptera NS
10%
Unidentified
Insects
21%
Trout 30-5.9 Inches Total Length
23%
Unidentified Trichoptera
Insects 17%
Odonata Other
9% 11%
Terrestrial
7%
Fish |
7%
Trout 7.0-99 Inches Total Length
Other 5%
Terrestrial 2%
Annelida 3%
Odonata 6%
Diptera 3%
Ephemeroptera
6%
Hemiptera 3%
Plecoptera 2%
Decapoda 6%
Mollusca 6%
Figure 2. --Percent composition of the diet of brown trout
from the South Branch Au Sable River, May–September, 1976.










16
Diptera *
6% O
Trichoptera
20%
Isopoda and Other 6%
Amphipoda
25%
Unidentified Fish 4 %
Insects
- 16 % Mollusca 4%
Trout 30-5.9 Inches Total Length
Diptera Ephemeroptera 7%
6%
Trichoptera
21%
ther
9%
Uniolentified
Insects
Terrestrial 5%
Decapoda 4 %
Isopoda and
Amphipoda 4%
Trout 7.0-99 Inches Total Length
Figure 3. --Percent composition of the diet of brown trout
from the Main Stream Au Sable River, May–September, 1976.

















17
pooled category consisting of the "other" category in Tables 1 to 4 plus
all taxa with percentages less than 2%. )
Comparing the diet of small fish of the South Branch to the Main
Stream, the percentages of Trichoptera, Ephemeroptera, and Diptera
were very similar. The major difference appeared to be Isopoda and
Amphipoda, which on the Main Stream comprised 25% of the diet while
being virtually absent from the South Branch. Odonata provided a fair
amount of food in the South Branch and not in the Main Stream; fish were
eaten in the Main Stream and not in the South Branch.
Small organisms were phased out of the diet of the larger fish and
were replaced by more substantial items. Specifically, Ephemeroptera
lost importance as a major source of food in both rivers; Isopoda and
Amphipoda declined drastically in the Main Stream. Trichoptera did not
lose importance in the Main Stream but were reduced moderately in the
South Branch. Replacing the small food items in the Main Stream were
Mollusca, Decapoda, and especially fish. In the South Branch, Odonata
and terrestrials increased in importance; also fish, Mollusca, and
Decapoda were major food items. Overall, the types of organisms
present in the diet of Au Sable brown trout were very similar to the
findings of Allen (1951), Lorz (1974), and Alexander and Gowing (1976).
Tables 1 to 4 show the average stomach volumes monthly and
over the entire growing season for each component taxon. The average
total volumes per fish for the entire 5-month period for the small and
large fish in the South Branch were 0.186 ml and 0.698 ml, respectively;
18
Main Stream small fish averaged 0.125 ml and large fish averaged
0. 768 ml. As previously discussed, the sampling scheme randomly
Selected trout in the intervals 3.0–5. 9 and 7.0-9. 9 inches. Due to
this method, a difference in mean length of trout in the categories arose
between the two rivers. South Branch and Main Stream srmal1 fish aver-
aged 4.41 and 4.56 inches, respectively; South Branch large fish aver-
aged 8.24 inches while Main Stream large fish averaged 8.81 inches. In
order to compare stomach volumes between the two rivers, correction
was made for this difference in length, because stomach capacity
increases with an increase in fish length.
TO determine this length-volume relationship, I calculated the
mean stomach volume for each inch group of trout for each river,
plotted the corresponding points, and fit the curves by eye (Figure 4).
These curves adequately represented the mean stomach volumes for
trout of all sizes between 3. 0-9. 9 inches, so a suitable correction factor
was employed on this basis. I arbitrarily elected the South Branch to
undergo correction using the following procedure. The mean length of
small Main Stream fish was 4. 56 inches, so the mean stomach volume
corresponding to this length on the curve for the South Branch fish was
equivalent to 0. 180 ml. Likewise, the mean length of large Main Stream
fish WalS 8. 81 inches, so the mean stomach volume for this length on the
South Branch curve was 0.885 ml. In effect, the corrected stomach
volumes were those volumes which would have been observed had the
mean lengths of South Branch and Main Stream fish been equal.
19
1.2T
0.
8
*
0.O
46
0.2 H.
Total Length (Inches)
Figure 4. --Mean stomach volume versus total length for
brown trout from the South Branch (solid line) and Main Stream
(broken line), Au Sable River, May–September, 1976.

20
Comparison of these corrected stomach volumes showed that
South Branch trout ate more than Main Stream trout. For small fish the
comparison was 0.180 ml to 0.125 ml, which yielded a ratio of 1.44:1
for South Branch: Main stream. The South Branch: Main Stream ratio
for large fish was less striking--0. 885 ml to 0.768 ml, or 1. 15:1.
However, the absolute difference between the volumes for the two rivers
is greater for large fish (difference = 0. 117 ml) than for small fish
(difference = 0.055 ml).
Age Structure
The mean weight and length for fish in each age group as
determined by scale readings are shown in Tables 6 and 7. Plotting the
average total length against age yielded growth histories for South Branch
and Main Stream trout (Figures 5 and 6). Knowledge concerning the
periodicity of wild trout growth (Cooper 1953 and 1961) permitted the
plotting of these curves by eye, thus depicting the most probable pattern
of growth for trout in the two rivers. Also, smoothing of the curves was
necessary because of the bias in the summer sampling scheme. Sampling
was designed to obtain fish in certain length intervals; stratifying these
fish by age introduced a bias to the fish at the lower and upper size
boundaries. The data from fish sampled in the fall population work were
not biased in this way, and were further strengthened by the large sample
sizes. Therefore in drawing the smoothed curves, more emphasis was
given to the fall data. These smoothed curves showed the inherent dif-
ference between the growth in length of South Branch and Main Stream
trout. At first the difference was slight, but gradually became greater.
21
Table 6. --Monthly averages of length and weight for brown trout from
the South Branch, Au Sable River, 1976.
Summer - Late fall (November)
Age Month Num- Mean Mean Age Num- Mean
ber total weight ber total
Of length (grams) Of length
fish (inches) fish (inches)
O May © º © & © tº gº © O 115 4. 1
June © º º • Q @ O e e I 82 7. 6
July 19 3. 1 5. 1 II 29 10. 9
Aug. 24 3.6 7.8 III 47 13. 2
Sept. 23 3.9 11.2 IV 11 17. 0
V 1 26. 0
Mean © Q @ 3. 6 8. 2
I May 21 5. 0 21.4
- June 27 5.4 26. 6
July 17 6. 6 49. 6
Aug. 23 7. 8 75. 2
Sept. 23 7. 9 81. 5
Mean tº C & 6. 5 50s 6
II May 23 8. 9 116. 9
June 23 9. 2 107. 1
July 13 9.4 119.6
Aug. 2 8. 6 91.4
Sept. 3 9. 0. 115. 2
Mean e Q C 9. 1 113.0
22
Table 7. --Monthly averages of length and weight for brown trout from
the Main Stream, Au Sable River, 1976.
Summer : Late fall (Sept. - Oct. )
Age Month Nunn - Mean Mean Age Num- Mean
ber total weight ber total
Of length (grams) Of length
fish (inches) fish (inches) -
0 May e C e O © O © e O O 79 3. 5
June 2 3. 8 10. 2 I 74 6. "
July 10 3. 1 5. 0 II 63 9. 5
Aug. 16 3.7 8.7 III 46 11. 5
Sept. 21 3. 9 10.4 IV 9 13. 0
Mean O 3, 7 8, 7 V 1 16. 3
I May 25 4. 9 19. 5
June 24 5. 3 26. 0
July 21 5. 6 30. 3
Aug. 11 5. 8 33. 4
Sept. 10 6, 7 5.1. 8
Mean © O C. 5. 5 28, 9
II IMay 24 8. 6 106, 8
June 22 9. 1 121. 3
July 22 8. 8 110.3
Aug. 20 8. 5 98. 5
Sept. 15 8. 8 107. 0
Mean © O © 8. 8 109. 1
III May 1 9.6 165. 1
June 2 9. 4 133. 5
July 2 9. 6 144. 0
Aug. 2 9. 4 141. 2
Sept. 4 9.5 121.4
Mean e i e a 9. 5 135. 3
23
11
68O2
º-Fº–Fº
|
4
->
| | | | | L | | | _l
Jan. Jan. Jan. Jan. Jan. Jan.
O | | ||| IV
Age Group and Month
Figure 5. --Seasonal growth of brown trout from the South
Branch Au Sable River, based on summer and fall sampling, 1976.

24
14 T-
11
68.O2
-º-F-º
|
4
º-
| | | | | | | l L _l
Jan. Jan. Jan. Jan. Jan. Jan.
O | || ||| |V
Age Group and Month
Figure 6. --Seasonal growth of brown trout from the Main
Stream Au Sable River, based on summer and fall sampling, 1976.

STATISTICAL ANALYSIS
The two sources of brown trout observations --the summer sam-
pling and fall population runs--provided information to statistically
compare various aspects o: diet and growth for the two rivers. Although
most monthly summer samples consisted of 25 each of small (3.0–5.9
inches) and large (7.0-9.9 inches) trout from each river, the first
South Branch sample had only 21 small fish. Therefore to make all
sample sizes equal, a characteristic which simplified the application of
some statistical techniques, the appropriate numbers of samples were
randomly deleted from each category so that the total number of samples
for the summer data was 21 samples X 2 sizes X 2 rivers X 5 months =
420 samples. Due to the different statistical treatment of the fall popula-
tion samples, there was no need for equal sample sizes. Thus none of
the 557 fall brown trout samples were eliminated.
Stomach Sample Analysis
A three-way analysis of variance was run for the mean stomach
volume and for the mean volume of each of six major organisms in the
diet. The three components of the test were river, month, and size;
the number of replicates in each cell was 21. This design tested the
hypothesis that the mean volume of food type was different for each of
the components included in the study. The components tested were as
follows: river, month, size, river X month, river X size, month X size,
25
26
and river X month X size. For example, in the test of total volume,
the hypothesis for the "river" main effect was as follows: the mean
stomach volumes of South Branch and Main Stream fish were signifi-
cantly different. For the river X size interaction, the hypothesis was
that the difference in mean volume between South Branch fish and Main
Stream fish was different for small fish than it was for large fish.
All possible interactions were tested. As in the previous section,
correction factors for comparing stomach volumes between the South
Branch and Main Stream were derived from Figure 4. This correction
was again arbitrarily applied to South Branch fish, and was calculated
as follows:
mean stomach volume at length equal to
Main Stream mean length
correction factor = -
mean Stonach volume at observed mean
length
This equation yielded the following factors for small and large trout:
0. 18
c. f. (for small fish) = — = 1.06
0. 17
0. 88
c. f. (for large fish) = — = 1.28
0. 69
In Order to make the analysis of variance tests meaningful, the correc-
tion factors were applied to the stomach volumes prior to running the
tests. This consisted simply of multiplying the South Branch small fish
stomach volumes by 1.06 and the South Branch large fish volumes by 1. 28.
27
The results of all seven of the analysis of variance tests appear
in Table 8. Total volume, the most important test, showed a significant
difference between months and between sizes, but not between rivers as
I had anticipated. Therefore, the difference in total stomach volume
between the South Branch and Main Stream as calculated in the previous
section could have been due to the variability between samples. The
significant size interaction simply reflected the greater capacity of large
fish's stomachs. The test of Trichoptera yielded results similar to the
total volume test. The amount of Dipters eaten was significantly higher
in the South Branch, probably because of the abundance of Rhagionidae
larvae. Like the tests of total volume and Trichoptera, the test of
Mollusca showed significant differences between months and between
sizes. However, the seasonality was reversed from the trend of the
first two tests; instead of higher volumes in the early season, more
mollusks were eaten in the late season. This probably indicated that as
the caddis and other preferred foods became scarce, more food of lower
preference, i.e., mollusks, were eaten. As expected, the test for
volume of fish eaten showed that large brown trout ate more fish than
did small trout. Also as expected, Main Stream fish ate significantly
more isopods and amphipods than did the South Branch fish. Monthly
means showed that more were eaten early in the season.
An overview of Table 8 shows that major differences occur
between months and between the two size categories. For the most part,
significant differences were not found between the South Branch and Main
f
Stream with respect to total volume or types of food eaten.
28
Table 8. --Results of the analysis of variance of stomach volume data for
brown trout from the South Branch and Main Stream, Au Sable River,
May to September, 1976 (degrees of freedom are in parentheses).
Component
River Month Size River X
(1, 400) (4,400) (1,400) Month
(4,400)
Test
River Month River X
X. Size X Size Month
(4,400) (4,400) × Size
(4,400)
Total
volume ><><>< >{< ><>|<
Trichop-
tera
volume
k
><
>{<
><
Ephemer-
optera
volume
Diptera
volume
>{<
Mollusca
volume >|<
>{<
>;
>{<
Fish'
volume
Isopoda and
Amphipoda
volume ><>|<>{< >|<>{<
k
•)
>{<
k
><
* Significant at the 0.05 level.
** Significant at the 0.01 level.
*** Significant at the 0.005 level.
29
The analysis of variance described above tested the hypothesis
that there was a difference between the South Branch and Main Stream
for stomach volume on the basis of one mean value computed from the
entire size range of trout. Because this test lumped the fish from each
river into one category, I employed an alternate test to determine any
difference between mean stomach volumes between the two rivers.
This method, the analysis of covariance, determines significant dif-
ferences between two regression lines on the basis of slope and adjusted
means (Steel and Torrie 1960). The regression lines which I used were
log-log relationships derived from Figure 4. By testing the regression
line for the South Branch against that of the Main Stream, I attempted to
establish a difference between the respective volumes of food eaten for
all inch groups of trout (between 3.0 and 9.9 inches).
For South Branch brown trout, the regression equation was as
follows:
loge (stomach volume) = -6. 6439 + 2. 7714 loge (length).
The coefficient of determination (r2) was 0.374. For Main Stream
brown trout, the equation was as follows:
loge (stomach volume) = -8. 1559 + 3.3678 loge (length),
r2 = 0.348.
Table 9 shows the covariance test of these regressions. The test for
equality of slopes was accepted at the 0.05 level of significance, while
the test for equal adjusted means was rejected. This indicated that the 2.
30
Table 9. --Covariance test of the regressions of loge stomach volume
on loge total length for brown trout from the South Branch and Main
Stream, Au Sable River.
Degrees g
Source of ... ..., F P
freedom Cl Cl
Total 419 1533. 40
Overall regression 1 530. 07
Equal regressions 2 24. 91 12.45 5. 3 0.05
Equal slopes 1 5. 15 5. 15 2. 2 0. 05
Equal adjusted
Iſle allS 1 19.76 19. 76 8. 4 0.05
Error 4 16 978. 45 2. 35
31
South Branch fish ate significantly more food than did the Main Stream
fish of the same sizes. The three assumptions for the analysis of
covariance are as follows:
(1) the X's are fixed and measured without error,
(2) the regression of Y on X after removal of block and treatment
differences is linear and independent of treatments and blocks, and
(3) the residuals are normally and independently distributed with
Zero mean and a common variance. The assumption of normality is not
necessary for estimating components of the variance of Y, but does
affect the validity of the F-test (Steel and Torrie 1 960). In this case,
the residuals were not quite normally distributed. This stemmed from
transforming stomach volumes to logarithms -–fish with empty stomachs
Were arbitrarily assigned a volume of 0.001 ml, so the logarithms for
"empty" stomachs were quite low compared to those with measurable
quantities. As a result, residuals for "empty" stomachs were clustered
on the lower tail of the distribution, causing the mode of the residuals to
be somewhat greater than zero (about 0.5 log units). I thought that this
could have been responsible for the low coefficient of determination for
both regressions (r” = 0.374 and 0.348). Therefore, I ran regressions
of the same data excluding fish with empty stomachs. The resulting r2
values were not much better, leading me to believe that the inherently
high variability between stomach samples accounted for the low rº
values. Since trout from both rivers showed similar variablity, I felt
that the results of this covariance test were valid.
32
Length-weight Relationship
In the past few years, controversy arose concerning the condition
of brown trout in the "quality water" of the Main Stream. Fishermen
reported that trout caught in this area were "skinnier" than normal.
White et al. (1975) compared the growth and condition factor of brown
trout in the "quality water" with those in adjacent sections of the Main
Stream. They documented the abnormally thin condition of the trout in
and below the special regulation water, noting that the condition factor
decreased from age I to age IV, whereas the condition of fish upstream
increased for older trout. Because this aspect of the trout's biology
was relevant to my investigation, I compared the length-weight relation-
ship of South Branch and Main Stream trout sampled during the summer.
I employed the analysis of covariance technique similar to Cooper's
(1961) treatment, except that I used semi-log rather than log-log
regressions. The regression equation for the South Branch took the
form loge (weight) = 0. 31018 + 0.5064 (length), and for the Main Stream,
loge (weight) = 0. 52666 + 0.4744 (length). The coefficients of
determination were 0.978 and 0.977, respectively. The analysis of
covariance for these regressions appears in Table 10. The hypothesis
of equal slopes was rejected at the 0.05 level of significance, which
precluded the need for the test of adjusted means. This showed the
inherent difference between the condition of South Branch and Main
Stream fish. Although Main Stream trout were heavier for their length
in the lower sizes, South Branch trout were heavier for their length in :
the upper size ranges. The intersection of the two regression lines
33
Table 10. --Covariance test of the regressions of loge weight on length
for brown trout from the South Branch and Main Stream, Au Sable River.
Source - pºss Sum of Mean F P
freedom **** Square
Total 419 507. 36
Overall regression 1 495. 44 >
Equal regressions 2 0. 53 0. 27 9. 7 0.05
Equal slopes 1 0. 52 0. 52 19. 1 0.05
Error 4 16 11. 39 0.03
34
occurred around 6.0 inches of length. Because of the high coefficients
of determination, I applied the regression lines to fish longer than 9.9
inches (the limit of my data). This extrapolation showed the increased
difference between points on the two regression lines in the range of
legal-sized fish (over 12 inches). This evidence supports the findings
of White et al. (1975) for fish Over 6. 0 inches.
Seasonal Growth
An apparent difference in growth between South Branch and Main
Stream trout is shown in Figures 5 and 6. These curves were drawn on
the basis of both summer and fall samples of trout. Because of the
limitations of the Summer data as discussed above, the statistical
treatment of growth data was confined to the fall samples. The analysis
of covariance was used to test the hypothesis that South Branch and Main
Stream trout had equal total lengths at the end of successive growing
Sea. SO11S for all lengths of fish sampled (2.0-26. 0 inches), or in other
words, that South Branch and Main Stream fish grew at the same rate.
The regression equations used for this test were log-log transformations
of the fall data from Figures 5 and 6. The regression equation for the
South Branch was loge (length) = 1.3966 + 0.8716 loge (age + 1); the
equation for the Main Stream was loge (length) = 1. 2592 + 0.8863 loge
(age + 1). The r2 values were 0.896 and 0.837, respectively. The
results of the covariance test of these regressions appear in Table 11.
The test for equality of slopes was accepted at the 0.05 level of
35
Table 11. --Covariance test of the regressions of loge length on loge
growing season for brown trout from the South Branch and Main
Stream, Au Sable River.
Source bºº Sum of Mean F P
freedom **** Square
Total 556 150. 71
Overall regression 1 128. 41
Equal regressions 2 2. 24 1. 12 30. 9 0.05
Equal slopes 1 0.01 0.01 0. 2 0.05
Equal adjusted means 1 2. 23 2. 23 61. 6 0.05
Error 553 20. 07 0.04
36
significance; the test for equality of adjusted means was rejected.
This indicates that South Branch trout were significantly longer at
the end of successive growing seasons than were Main Stream trout.
TROUT POPULATION SIZE AND
FOOD PRODUCTIVITY
The preceding treatment has compared the diet, condition,
and growth of individual trout sampled from the South Branch and Main
Stream. All of these factors reflect the availability of food in the
streams, since condition and growth arise from diet. However, this
investigation to be complete needs to consider also the effects of
population size on the individual trout's diet.
The results of fall population estimates by inch group for the
South Branch and Main Stream for 1976 are given in Table 12. It was
readily apparent that the Main Stream had a much higher population
than the South Branch, especially for brown trout. However, I
previously showed that individual South Branch trout ate more than
did Main Stream trout of the same size. Therefore to understand the
relationship between population size and the amount of food per fish,
I compared the total amount of food consumed by the trout in each
river in the following way:
volume of Illil Il- average number
food eaten ber volume of days
per acre for Of consumed in grow- (Equation 1)
the growing fish × Peć fish X ing
Se 2. SOIl 3. Clºe day Se 3 SOIl
37
38
Table 12. --Number of fish per acre for the South Branch and Main
Stream Au Sable River, September 27-November 9, 1976.
Inch
Brown trout
I"OUI South Main
g p Branch Stream Branch Stream Branch Stream Branch Stream
Brook trout
South Main
Rainbow trout
South Main
Total
/
South Main
:
7
10
11
12
13
14
15
16+
9
:
19.
12.
7
.
:
.
48.
139.
101.
33.
78.
85.
59.
56.
60.
28.
17.
.
::
:
7. 2 79. 7
95. 7 120. 7
77.8 12.0
13. 9 12. 2
8. 8 15. 8
11. 0 12. 1
4.0 3.4
0.2 1.2
218. 6 257. 1
47.
44.
2.
1.
1
º
5
4
2
2
4
16. 8 176.
304.
115.
46.
102.
171.
175.
31.
14.
30.
16.
7
.
:
7
.
Total
288.
716.
0. 0 118.
506.
6 1091.
104.
65.
60.
62.
28.
18.
.:
.
39
The present study yielded values for the number of fish per acre
and for the instantaneous stomach volume. To convert the latter into
volume consumed per day (daily ration), I used the method described by
Alexander and Gowing (1976). They analyzed stomach and scale samples
Of brook, brown, and rainbow trout from many lakes and streams in
northern lower Michigan. They estimated that the growing season lasted
180 days. To determine the average volume consumed per fish per day
or daily ration they used the following equation:
annual food -
daily ration = { weight X conversion) + 180 days (Equation 2)
gain ratio
They cited findings by Ball (1948), who determined that for stomach
contents, 1 ml was equivalent to 1 g wet weight. The food conversion
ratio which they used was 5.0. The data which I used to compute daily
ration are shown in Table 13. The "mean weight" values in Table 13
were derived by using the length-weight relationship of fish sampled in
SUIrºn Imler” and the mean length of fish sampled in 1ate fall. I assumed that
the average weight did not change between late fall and early spring.
The "mean summer stomach volume" values represented the volumes
which were present in the upcoming growing season. For example,
the stomach volume for fish which were age I* in early spring (age I
in the coming summer) was 0.430 for South Branch fish. Most of the
ratios of daily ration to stomach volume fell between 3.05 and 4. 37.
I obtained much higher values for the higher-aged fish, but I disregarded
40
Table 13. --Annual weight gain (g), mean stomach volume (ml), esti-
mated daily ration (m1) and ratio of daily ration to stomach volume for
brown trout from the South Branch and Main Stream, Au Sable River.
Age in Annual Estimated Mean Ratio of daily
early Mean weight daily STUIII) IQ.6 (" ration to
& weight ſº * stomach stomach
Spring galn ration
volume volume
SOUTH BRANCH
Newly-
hatched O 10. 9 0.339 0. 101 3. 36
I-k \ 10. 9 53. 1 1. 475 0. 430 3.43
II* 64. 0 276. 3 7. 675 0. 83.9 9. 15
III-k 340. 3
MAIN STREAM
Newly-
hatched O 8. 9 0.247 0.081 3. 05
I-k 8. 9 3.1. 8 0.883 0. 202 4. 37
II* 40. 7 112.8 3. 133 0.813 3. 85
III-k 153. 5 242. 8 6. 744 0. 631 10. 69
41
them on the basis of the biased summer sampling as explained earlier.
Therefore I averaged the daily ration: stomach volume values for South
Branch fish Of the first two ages and for Main Stream fish Of the first
three ages, and obtained values of 3.40 and 3. 76, respectively. The
same calculations for the data of Alexander and Gowing (1976) yielded
values of 2.17 (for trout in streams) and 2.50 (for trout in lakes).
On the basis of the entire population of trout broken down by
inch group, equation 1 was computed for each river (Table 14). A major
assumption made was that brook and rainbow trout ate similar amounts
as brown trout of the same size. The instantaneous Stomach volume
values for the 3- to 5- and 7 – to 9-inch classes were taken from the
observed data of Figure 4. Values for all other inch classes were taken
from extrapolations of the curves in Figure 4. The Main Stream trout
population consumed 260, 580.6 ml of food, almost twice as much as the
South Branch population, which consumed only 133, 871. 4 ml.
The different levels of food consumption by trout populations of
the two rivers prompted questions about the effects of grazing on the
benthos, about which very little research has been done. In Coopes
(1974), Quigley sampled the benthos at some of the same sites on the
South Branch and Main Stream used in my study. The mean number of
organisms per square foot was 412 for two stations on the South Branch
and 144 for three stations on the Main Stream. Because the composition
of organisms from the two rivers was very similar, I assumed that the
weight of organisms per square foot was greater for the South Branch
than for the Main Stream, and corresponded to the 412: 144 ratio for
42
Table 14. --Estimated volume of food eaten (ml) per acre during the 1976
growing season (May–October) for the South Branch and Main Stream,
Au Sable River.
South Branch
Main Stream
Inch Mean Volume Volume Mean Volume Volume
group Ston- per per acre Storm- per per acre
ach a C re per grow- ach 3.C. Cºe per grow-
volume per day ing season volume per day ing season
2 0.02 1. 14 205. 2 0.02 13. 24 2, 383. 2
3 0.08 46. 70 8, 406. 0 0.06 68. 70 12, 366.0
4 0. 16 95. 53 17, 195.4 0.11 47. 65 8, 577. O
5 0.32 34. 16 6, 148.8 0. 18 31. 74 5, 713. 2
6 0.37 18. 62 3, 351. 6 0.30 115. 28 20, 750. 4
7 0.41 42.94 7, 729. 2 0.41 160.48 28, 886.4
8 0. 73 116. 16 20, 908. 8 0. 79 193. 08 34, 754.4
9 1. 11 27. 55 4, 959. 0 0.88 199. 19 35, 854. 2
10 1. 38 27. 21 4, 897. 8 1. 15 269. 39 48, 490. 2
11 1. 73 43. 53 7, 835.4 1. 44 155. 39 27, 970. 2
12 2. 08 49. 50 8, 910. 0 1. 72 121. 58 21, 884.4
13 2. 50 41.65 7,497. 0 2.04 28. 38 5, 108.4
14 2. 98 87. 14 15, 685. 2 2.42 10. 92 1, 965. 6
15 3. 62 35. 69 6, 424. 2 2.82 24. 39 4, 390. 2
16+ 4. 67 76. 21 13, 717. 8 3. 66 8. 26 1, 486. 8
Total 743. 73 133, 871.4 1,447. 67 260, 580. 6
43
the number of individuals. This meant that the standing crop of benthos
from the South Branch was almost three times that of the Main Stream,
which probably reflected the intense grazing on the benthos by the Main
Stream trout population. This may continually depress the benthic
community in this part of the Main Stream. The overall effect of this
situation, in comparison to the South Branch, is a lower standing crop
of invertebrates at any given time with perhaps a higher yearly
productivity and turnover rate.
In a study of brown trout from the Horokiwi, Allen (1951) found
a negative relation between the standing crop of benthos and the amount
consumed yearly by the trout population. He concluded that the bottom
fauna played an important role in regulating the population of trout by a
self-regulating mechanism which tended to maintain the weight of the
stock at a constant level by increasing or decreasing the trout's growth
rate. Trout in the Main Stream and South Branch also demonstrated
the same density-dependent process. Apparently, trout in the Main
Stream consumed nearly twice the amount of food per acre than did
South Branch trout, and as a result the standing crop of benthos in the
Main Stream was lower than in the South Branch. Furthermore, the
growth of individual trout in the Main Stream was slower to compensate
for the lower standing crop of benthos.
The slower growth of Main Stream trout may not have been
caused solely by competition for food. Brown (1945) found that in
overcrowded brown trout populations, growth was retarded by mutual
mechanical disturbance in addition to competition for food. Apparently
44
the high level of disturbance caused an increase in activity and resulted
in higher food requirements. Within every group of fish, she found a
social hierarchy based on size. Removal of the larger, faster-growing
fish allowed the growth rates of remaining fish to increase. This effect
was attributed solely to behavioral factors and not to competition for
food.
DISCUSSION
Population size was the outstanding difference between the South
Branch and Main Stream trout, and consequently led to inequalities in
diet, condition, and growth. These factors and several others were
treated separately in this study. Tying them all together presents a
useful picture of the two trout populations.
The South Branch population, about half the density of the Main
Stream population, consisted of more brown than brook trout. The
Main Stream had quite a few more brown than brook trout, plus a
significant number of rainbow trout. The South Branch brown trout
were in better condition than Main Stream browns, and grew faster.
These differences arose from the diet of individual fish--Over all
lengths of trout sampled (3.0-9.9 inches), South Branch trout ate more
than Main Stream trout of the same size. Although they ate different
total amounts, the types of food eaten were very similar, with
Trichoptera and Ephemeroptera heading the list. Very few adult
insects were found in the stomach samples. The difference in types.
of food eaten was that South Branch trout ate more Diptera and Main
Stream trout ate more Isopoda and Amphipoda.
The Main Stream trout populations are denser than those of the
South Branch because of better spawning success. Both rivers have
abundant substrate suitable for spawning, however the amount of
45
46
groundwater percolation is much less on the South Branch (Alexander,
personal communication). Because percolation is very important to
egg development, many more eggs spawned in the Main Stream survive
to the hatching of fry than survive in the South Branch. Less ground-
water in the South Branch probably allows temperatures in winter to
fall below optimum. This causes high egg mortalities and results in a
low density of trout fry.
Benefitting from the sparse populations which result, individual
South Branch trout find plenty of suitable food and grow quickly. On
the other hand, early survival in the Main Stream is high, which allows
for dense trout populations, thus causing slower growth from competition
for the limited food supply. South Branch trout overtake Main Stream
trout in condition at a length of about 6 inches. From this point on,
Main Stream fish undergo intense competition for food. This situation is
aggravated by the special regulations which protect 10- to 12-inch fish
from anglers. As a result, the Main Stream now has a large population
of smaller, thinner fish because of the longer time necessary for trout
to attain legal size in comparison to trout of the South Branch. Although
this section of the Main Stream was put under special regulations for the
purposes of reducing exploitation and producing trophy-size fish,
apparently overcrowding prevents many tiºn from reaching legal size.
Harvesting brown trout of 10 to 12 inches or smaller might result in a
population of faster-growing, plumper fish.
SUMMARY
Stomach samples were analyzed from 243 and 255 brown trout,
respectively, from the South Branch and Main Stream of the
Au Sable River, Michigan. Readings were made on scale samples
from these fish plus those from 537 additional brown trout
sampled during the Fish Division's fall population estimates.
The mean stomach volumes for small fish (mean total length =
4. 56 inches) were 0.180 ml and 0.125 ml for the South Branch
and Main Stream, respectively. Total volumes for large fish
(mean total length = 8.81 inches) averaged 0.885 ml and 0.768 ml
for the South Branch and Main Stream, respectively.
The major diet components for both rivers were Trichoptera,
Ephemeroptera, Diptera, and Mollusca. Also, Main Stream
fish benefitted from an abundance of Isopoda.
There was a marked shift between the diet of young and old fish.
Small food items such as Ephemeroptera and Isopoda which
were so important to small fish were replaced in the diet of
larger fish by higher announts of Mollusca, Decapoda, Odonata,
and especially by more fish.
A great difference in diet was observed between different months.
Total stomach volume and the volumes of Trichoptera, Isopoda,
and Amphipoda were highest in the early part of the growing
47
48
season, especially in May. In contrast to this were the volumes
of Mollusca, which were highest late in the season.
South Branch trout were in better condition and grew faster than
Main Stream fish. This difference probably resulted from the
greater volume of food per individual trout of the South Branch.
The population of trout per acre for the Main Stream was more
than twice that of the South Branch. As a result, intense
competition for food probably accounted for the lower volumes
of food per fish observed for the Main Stream.
The estimated total volume of food eaten per acre for the entire
growing season was 133, 871.4 ml for the South Branch and
260, 580, 6 ml for the Main Stream.
REFERENCES CITED
Alexander, Gaylord R., and Howard Gowing. 1976. Relationships
between diet and growth in rainbow trout (Salmo gairdneri),
brook trout (Salvelinus fontinalis), and brown trout (Salmo
trutta). Michigan Dep. Nat. Res., Fish Research Rep. 1841,
41 pp.
Allen, K. R. 1951. The Horokiwi Stream. New Zealand Marine
Dep. Fish. Bull. No. 10, 231 pp.
Ball, Robert C. 1948. Relationship between available fish food,
feeding habits and total fish production in a Michigan lake.
Michigan State Coll. , Agr. Exp. Sta. Bull. 206: 1-59.
Brown, Margaret E. 1945. The growth of brown trout (Salmo trutta
Linn.). I. Factors influencing the growth of trout fry.
J. Exp. Biol. 22: 118-129.
N
Caucci, Al, and Bob Nastasi. 1975. Hatches. Woodside, New York
Comparahatch, Ltd. 320 pp.
Coopes, Gary F. 1974. Au Sable River Watershed Project Biological
Report (1971-1973). Michigan Dep. Nat. Res., Fish Manage.
Rep. 7, 296 pp.
Cooper, Edwin L. 1953. Periodicity of growth and change of
condition of brook trout (Salvelinus fontinalis) in three
Michigan trout streams. Copeia 2: 107-114.
Cooper, Edwin L. 1961. Growth of wild and hatchery strains of
brook trout. Trans. Am. Fish. Soc. 90: 424-438.
Hanson, Arthur J. 1972. The role of prior feeding and temperature
in regulation of food uptake by brook trout. PhD thesis,
The Univ. Mich. , 167 pp.
Lorz, Harold W. 1974. Ecology and management of brown trout in
Little Deschutes River. Oregon Wildlife Commission,
Fishery Res. Rep. No. 8, 49 pp.
Steel, R. G. D., and J. H. Torrie. 1960. Principles and procedures
of statistics. McGraw-Hill Co., Inc., New York, 481 pp.
49
50
White, R. J. , C. DeJong, and J. Gosse. 1975. Growth of wild
brown trout in the Main Branch of the Au Sable River,
Michigan. Draft copy of a report to the Research Committee
of the Michigan State Council of Trout Unlimited. Mich.
State Council of Trout Unlimited. Mich. State Univ.,
Dep. Fish. Wildl., East Lansing, 20 pp.
ERSITY
MMIM

}O15 003

--- « » （« ºz. --• • • • • ••• • • • • • • •••••••••••••（~~~~*~ſe yº，…，-3，…， ………………………
************************ 28% ſą gaeaeaeaegsææ;&######
•■ ■ ■ssaer - -§§~--~ ：：· º aea ） • .
Ü
： ~~~~ae ： ， ！---~|- -·:。 、、、」 sæs， æ， ø ± ø §！ º aes sae saev • • • • • • • • • • • • • • •=. • • • • •
：w& ***
ae§： -،· -· -~· <！--
* * * * * * * * * * · · · · · · -t.. ae-<！-- * * * -- w • • ► ► ► ► ► ► ► ► ► ► ► ► ► ► ► ► 3 ºſºț¢ © ® ° ≈ ≠ ≤ ≥ ± − ×：،-： （…* * * * * .r. ）， ºº … :-）
§ ø ± • • • • • • • • • • • . . . .-·：·~--~* * * *）！ |-●~~ ~ ~~ºs=} b.
----~--~•* •~--~~~~--~•·ºs · * * * * · * * * * * * · * * * * * · * * * * * · *· · * * * * * * * * * * * =： e- → <<+> <<+ ºs）: ** * * * * *º<.›› ‹› ‹› ‹› ‹› ‹› ‹› ‹›;*** • • • • • • • • • • 9. * * · *， **， ** ** =<！--* * *, ******, ** ********，****************************** • ** *************-**** ***************************************ærwºesº！ ！！ sv：， ， ， ， ， ， , , ， wa waer，* ， ， ）=º（ae ae → ∞, ∞, ∞; ∞， ∞; ∞， ∞; ∞， √æ√æeſſ se-yº~#~#~ſº ſ x， ſi
Caes（aer）（）;
***** ** ** ** * * · ***
&& ！ ！ ！ ！ ！
§.
*** * * *
，，,，，,，fºs
*********** - e