MSU

LIBRARIES
‘—

 

 

RETURNING MATERIALS:
P1ace in book drop to
remove this checkout from
your record. FINES wi11
be charged if book is
returned after the date
stamped be10w.

 

 

 

THE ROLE OP PERIPHYTON IN IHE FEEDING, GROWTH
AND PRODUCTION OF STEIONEIA SPP.
(EPHEHEROPIERA: HEPIAGENIIDAE)

By

Kevin Moore Webb

0

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Entomology

1985

ABSTRACT
THE ROLE OF PERIPHYTON IN THE FEEDING, GROWTH,

AND PRODUCTION or STENONEMA SPP.
(EPHEMEROPTERA: HEETKOERTTDAE)

By

Kevin Moore Webb

Field and laboratory studies examined-the role of peri-

phyton in the feeding, growth, and productivity of Stenonema

 

mayflies. Edeld.studies compared these variables between
natural pOpulations living under differing primary produc-
tion regimes in a 2nd and a 4th order stream. Three

species, §L exiguum, §; modestum, and §; vicarium, were

 

 

found in both streams, and all exhibited univoltine, slow
seasonal life history patterns. The proportion of diatoms
in gut contents was related to body length, but did not
differ greatly between streams, reflecting the similarity of
periphyton standing crOps. Growth rates did not differ
significantly between streams, and were rapid (7.5 %/d) in
summer, but much lower in the fall (1.1 %/d). The producti-
vity of each species in each stream was determined largely
by the suitability of available substrates, not stream order
as hypothesized. Laboratory studies compared the growth of
g; vicarium on diets of dark-conditioned leaves and natural
periphyton. As hypothesized, growth was more rapid on
periphyton than on leaves, although seasonal changes in
periphyton quality, and possibly endogenous control of

growth, may also have influenced growth rates.

To Paula. Thanks for being patient.

ii

ACKNOWLEDGEHENTS

Special thanks to my major professor, Richard W.
Merritt, for unfaltering encouragement and support
throughout my degree program. Sincere thanks also to
commitee members Jean Stout, Tom Burton, Jim Miller, and Ed
Grafius for many useful comments and suggestions. I am
greatly indebted to Dan Lawson for his patience in
explaining to me the intricacies of the MSU computer system.
Mark Oemke provided many insights and ideas, especially
during the early stages of the design of this study.
Technical assistance and moral support were provided by Bill
Taft, Mike O'Malley, Gary Whelan, and fellow graduate
students Gail Motyka, Mike Kaufman, Dave Cornelius, Dave
Gesl, and Mary Ufford. Species identifications were
verified by William L. Hilsenhoff (U. of Wisconsin,
Madison). Last, but not least, I thank my parents for
encouraging me to explore my curiousity.

This research was supported in part by Department of

Navy contract NOOOB9 - 81 — C0357.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . . . . . . .
INTRODUCTION AND LITERATURE REVIEW . . . . . . .
CHAPTER 1: FEEDING, GROWTH, AND PRODUCTION OF
STENONEMA SPP. IN TWO MICHIGAN STREAMS . . . .
Introduction . . . . . . . . . . . . . . .
Site Descriptions . . . . . . . . . . . . .
Ford River . . . . . . . . . . . . . . .
Schwartz Creek . . . . . . . . . . . . .
Materials and Methods . . . . . . . . . . .

Stenonema Populations . . . . . . . . . .

 

Ford Riffle . . . . . . . . . . . . . .
Snags . . . . . . . . . . . . . . . . .
Length/Weight Relationship . . . . . . .
Characterization of Periphyton Communities
Colonization . . . . . . . . . . . . .
Ambient Periphyton Communities . . . .
Feeding Habits . . . . . . . . . . . .
Growth and Production . . . . . . . . . .
Results . . . . . . . . . . . . . . . . . .

Stenonema P0pulations . . . . . . . . . .

 

Ford Riffle . . . . . . . . . .

Ford Snags . . . . . . . . . . . . . .

Schwartz Snags . . . . . . . . . . . .
Length/Weight Relationship . . . . . . .

iv

vi

. vii

1O
12
12
14
14
14
15
16
17
18
18
19
19
20
22
22
22
22
26

29

Characterization of Periphyton Communities

Feeding Habits . . . . . . .
Growth and Production . . . . . . . .
Discussion . . . . . . . . . . . . . . .

Stenonema Populations . . . . . .

 

Periphyton Communities . . . . . . .
Feeding Habits . . . . . . . . . . . .
Growth and Production . . . . . . .
General Discussion . . . . . . . . .
CHAPTER 2: THE INFLUENCE OF DIET ON THE GROWTH
STENONEMA VICARIUM (WALKER) . . . . . . . .

 

Introduction . . . . . . . . . . . . . .
Materials and Methods
Results . . . . . . . . . . . . . . . . .

DiscuSSion O O O O O O O O O O O O O 0

LITERATURE CITED . . . . . . . . . . . . . .

29
33
33
38
38
4o
42
42
46

51
52
34
58
61
68

NNN

.1.

LIST OF TABLES

Physical and chemical data for Schwartz Creek and
the Ford River. All values are mean and (range). 13

Characterization of ambient periphyton communities.
All values are mean i SE. . . . . . . . . . . . 32

Calculation of production for the Schwartz
1984 cohort. . . . . . . . . . . . . . . . . . . 36

Summary of production estimates. . . . . . . . . 37
MANOVA table for log-transformed data. . . . . . 6O
Feeding habits of nymphs in each treatment. . . . 62

Correlation between individual weight and growth
rate in each treatment group. . . . . . . . . . . 62

Comparison of methods and results. . . . . . . . 65

vi

p- \n

.10

.11
.12

LIST OF FIGURES

Size distribution of nymphs, Ford riffle, 1983. . 23
Size distribution of nymphs, Ford riffle, 1984. . 24
Size distribution of nymphs, Ford snags, 1984. . 25

Size distribution of Stenonema nymphs in Schwartz
Creek, 1984. Solid, Open, and stippled bars denote
female, male, and unsexed nymphs, respectively.
Arrows denote emerging individuals of the indicated
sex. Proposed separation of 1983 and 1984 cohorts

is indicated with a dashed line. . . . . . . . . 27

 

Combined distribution of Stenonema spp., Schwartz
Creek, 1984. Dashed line indicates separation of
1983 and 1984 cohorts. . . . . . . . . . . . . . 28

 

Length/weight relationship for S; vicarium. . . . 3O

Periphyton colonization in Schwartz Creek (---)
and the Ford River ( ), August, 1984.. .. .. 31

 

Regression of % diatoms in gut contents (+ 95%
CI for regression line) in: A) Ford River;
B) Schwartz Creek. . . . . . . . . . . . . . . 34

Growth regression for the combined populations. . 35
Life cycle diagram for Stenonema spp. Shaded area
denotes size range of nymphs present as they

progress from eggs (bottom line) to emergence

(top line). The sampling period for the present
study is indicated. 0 O O I O O O O O O O O C 0 O 39
Model used in hypothesis formation. . . . . . . . 47
Improved model of productivity.. .. ..... . 49

Mean growth rate (i 95% CI) of each
treatment group. . . . . . . . . . . . . . . . . 59

vii

INTRODUCTION AND LITERATURE REVIEW

The concept of trophic structuring of biological commu-
nities has been one of the central themes of ecological
thought since its introduction by Thienemann (1920). As
such, it has served as a conceptual basis for studies of
community structure, nutrient cycling, and species interac-
tions, and is closely linked to the niche concept of Elton
(1927). Thus, biologists in the early part of this century
began to view their subjects not as isolated units but as
members of larger communities and ecosystems, complexly
interacting with other organisms and the physical environ-
ment. The principle of energy flow through biotic communi-
ties, first recognized by Lindemann (1942), is one of the
most important derivatives of the trophic structure concept.

Determining energy flow through communities involves
three basic components: rates of energy exchange between
trophic levels, rates of energy loss from and between
tr0phic levels, and rates of energy storage within trophic
levelss(e.g. Odum 1957). Energy storage within a trophic
level is defined as net production and is expressed as units
of biomass or energy gained/unit area/unit time. Producti-
vity of lower trephic levels limits the rate of energy

transfer to higher trophic levels, and therefore is an

1

important determinant of community structure. Furthermore,
production combines in one measurement two population para-
meters of major ecological significance: individual growth
and population survivorship (Benke 1984). Therefore,
production is important to the understanding of both popula-
tion.and community level dynamics. .Productivity has also
been shown to be sensitive to the effects of pollution (e4;
Kimerle and Anderson 1971; Cairns 1977; Uutala 1981).
Historically, much of the early interest in aquatic
invertebrate productivity is attributable to fisheries
managers who were interested in estimating the amount of
"fish food" available in any body of water (eug. Allen
1949,1951; Boysen-Jensen 1919). Interest in benthic produc—
tivity blossomed in the early 1970's largely as a result of
the International Biological Programme (IBPL. One of the
main purposes of the IBP was to study the productivity of
terrestrial and aquatic ecosystems to attain a more rational
exploitation of the world's biotic resources (Winberg 1971).
Particularly important to the study of benthic produc-
tivity was the development by Hynes and Coleman (1968) of
the size-frequency method of estimating secondary production
from field samples (later corrected and modified by Hamilton
1969, Benke 1979, and Kreuger and Martin 1980). This method
estimates the average number of individuals that reach each
of a series of size classes over the course of a year, thus
creating an apparent survivorship curve by collapsing time
(Benke 1984). 13180 doing, the size frequency method does

not require recognition of individual cohorts as do other

methods (see reviews by Waters 1977 and Benke 1984 for
descriptions of these methods). The average deveIOpmental
time, or Cohort Production Interval (CPI) of Benke (1979) is
the only life history parameter that is required. There—
fore, production may be estimated for populations having
non-seasonal life history patterns without the need for
additional laboratory growth studies (eug. Cooper 1965).
Laboratory growth studies may be necessary for fast-growing
species for which CPI is difficult to determine from field
samples (e.g. Menzie 1981). With the derivation of a vari-
ance estimator for the size frequency method by Kreuger and
Martin (1980), specific hypotheses concerning secondary
production may be tested.

The relative contribution of allochthonous vs. autoch-
thonous (iJL exogenous vs. endogenous) energy sources in
stream community metabolism has been one of the long
standing debates in stream ecology. Many early workers
hag. Minckley 1963; Minshall 1967; Hynes 1970; Kaushik and
Hynes 1971; Cummins 1973) stressed the importance of alloch-
thonous detritus (e.g. leaves) in stream community metabo-
lism. However, Minshall (1978) pointed out a'bias in the
literature towards low—order, deciduous forest streams where
autumnally-shed leaves are the most obvious feature. Thus,
some workers (Minckley 1963; Minshall 1967) concluded on the
basis of standing crops of detritus and periphyton that
autochthonous primary production is of little importance to

stream communities. However, such a conclusion erroneously

4
assumes that periphyton and detritus are similar in nutri-
tive value and productivity.

In general, living food resources are nutritionally and
calorically superior to detritus (Lamberti and Moore 1984).
Since protein is considered to be important in insect nutri-
tion as the principle source of amino acids (Chapman 1982),
carbon to nitrogen (C:N) ratios have been used to estimate
the protein content of food resources (see review by Cummins
and Klug 1979). C:N ratios for periphyton range between
3.7:1 to 7ih1 (McMahon et al. 1974; McCullough et al.
1979). For comparison, C:N for pure protein averages
3.25:1 “McMahon et al. 1974). ChN'ratios for detritus are
much higher, with some reported measurements as high as
1340:1 (Cummins and Klug 1979). These high values for
detritus reflect its high content of lignin, cellulose and
other refractory substances (Lamberti euui Moore 1984).
Microbes associated with detritus may greatly enhance its
nutritional value via biochemical alteration or by supplying
metabolites (Lawson et al. 1984). However, reported assimi-
lation efficiencies for algal diets range between 30 to 60%
(Trama 1972; McCullough et al. 1979), whereas efficiencies
on detrital diets generally range between 10 and 30% (Berrie
1976; Benke and Wallace 1980).

McIntire (1973) demonstrated with a computer simulation
that a small (low standing crop) but highly productive
periphyton community is capable of supporting a much larger
standing crop of consumers. Elwood and Nelson (1972) found

that grazing rates of snails in Walker Branch, Tennessee

approached net primary productivity, indicating that
McIntire's model is probably a realistic simulation.
Furthermore, Lamberti and Resh (1983) showed that grazing
maintains periphyton communities at a low standing crop but
highly productive state. Lamberti and Moore (1984) proposed
a graphic model which synthesizes the results of several
studies on the effects of grazing on periphyton communities.
Their model suggests that periphyton standing crop is
inversely related to grazing pressure, but that primary
productivity is maximized at intermediate grazing pressures.
AutotrOphy may therefore be more important in stream
ecosystems than had been previously assumed. By tabulating
data on energy sources for a number of stream ecosystems,
Minshall (1978) showed that autochthonous inputs either
approach or surpass allochthonous inputs in 9 out of 12
reported studies. However, the streams listed comprise an
orderly series, ranging from those with negligible autoch-
thonous inputs to those with negligible allochthonous
inputs. Minshall (1978) also showed that mean annual gross
primary productivity in open-canopied streams generally
exeeds input rates of allochthonous detritus, while the
opposite is generally true in closed—canopied streams.
Vannote et al. (1980) proposed a model that attempts to
eXplain these patterns. Briefly, this River Continuum
Concept states that predictable changes occur in stream
communities parallel to longitudinal changesixiphysical

conditions from headwaters to mouth. These community

6

changes should be evidenced by: 1) community Production/Res-
piration (P/R) ratios less than 1 in low and high order
sections and maximum P/R >1 in mid-order sections, due to
changes in shading and detrital inputs along a stream's
course; and 2) changes in benthic invertebrate community
structure which parallel P/R ratio changes. Although the
River Contiuum Concept has been criticised as being applica-
ble only to Northern Temperate streams (Winterbourn et al.
1981; Benke et al. 1984), it does appear to hold for streams
in that latitude. For example, Naiman and Sedell (1980)
showed that predictions concerning community metabolic para-
meters are accurate for several streams in the Pacific
Northwest. In the same set of streams, Hawkins and Sedell
(1981) showed that longitudinal changes in benthic inverte-
brate communities generally follow the predictions of the
Concept.

As stated, the River Continuum Concept predicts longi-
tudinal changes in benthic invertebrate community structure
in terms of the functional feeding groups of Merritt and
Cummins (1984). More fundamentally, the Concept predicts
changes in the relative contribution of algal vs. detrital
food resources to total benthic secondary production. Since
most benthic invertebrates are opportunistic generalists
(Chapman and Demory 1963; Cummins 1973), feeding habits of
species may vary over life stages, seasons, and stream
orders. Functional feeding groups can therefore be taken as
only crude approximations of feeding habits. For example,

if species "A" is labelled a grazer in a 5th order stream,

implying that periphyton is its major food resource, does it
ingest less periphyton when it occurs in headwater streams
where periphyton is less available? In view ofifluanutri-
tional superiority of algae over detritus, what are the
consequences of such differences in diet for the growth and
production of species'UVfl and furthermore, for the entire
benthic community?

One way to approach these questions would be to apply
the "trophic basis of production" method of Benke and
Wallace (1980) to entire benthic invertebrate communities at
two different points along a stream's length. This method
combines production, feeding, and bioenergetics to estimate
the amount of invertebrate productivity attributable to
animal, algal, and detrital food resources. Benke and
Wallace (1980) appliedthis method to a guild of filter-
feeding caddisflies and showed that although detritus was
the predominant food resource, 80% of caddisfly production
was attributable to animal food, due to a higher estimated
assimilation efficiency for animal food (70% vs. 10% for
detritus).

To my knowledge, the "trOphic basis of production"
method has never been applied to entire benthic communities
in the context of the River Continuum Concept. Without such
studies, we can only speculate as to the effect of reduced
primary productivity in headwater streams on community
energy budgets. Obviously, such a study carried out on two

entire benthic communities would entail an enormous amount

of effort.
This study examined the feeding and production response

of one species, Stenonema vicarium (Walker) (Ephemeroptera:

 

Heptageniidae) and its congeners, to different primary
production regimes in two Michigan streams, Schwartz Creek
(2nd order) and the Ford River (4th order). Since aquatic
consumers may enhance their growth rates on poor food
resources by increasing consumption rates (Cummins and Klug
1979), a laboratory experiment was also performed to
determine whether inclusion of periphyton in the diet of S;
vicarium results in increased growth. The overall objective
of this research was to determine the role of algae in the

feeding, growth, and production of Stenonema spp.

 

CHAPTER 1

FEEDING, GROWTH, AND PRODUCTION OF STENONEHA SPP.

 

IN TWO MICHIGAN STREAMS

INTRODUCTION

The River Continuum Concept (Vannote et al. 1980)
presents a holistic view of stream ecosystems in which
biological communities are predicted to respond to a
continuum of physical conditions from headwaters to mouth.
One of the predictions of the Concept is that grazers should
be less abundant in headwater streams due to resource limi-
tation. This prediction assumes that grazers are special-
ized periphyton feeders, and are unable to efficiently
utilize alternate food resources. Anderson and Cummins

(1979) showed that pupal weights of Glossosoma were posi-

 

tively correlated with community P/R in some Michigan
streams, but the effects of P/R were confounded with temper-
,ature. Since algae is nutritionally superior to detritus
(Lamberti and Moore 1984), reduced productivity of "special-
ized" grazers in headwaters would be expected. However,
most aquatic insects are opportunistic generalists (Chapman
and Demory 1963; Cummins 1973), and may increase their
ingestion rates to compensate for poor food resources
(Cummins and Klug 1979). Differences in resource availabi-
lity may therefore have little effect on the growth and
productivity of generalists.

This study examines the feeding, growth, and producti-

vity'of the mayfly Stenonema vicarium (Heptageniidae) and

 

its congeners under different primary production regimes.
S; vicarium was chosen because: 1) it is widespread through—

out Michigan and occurs in both headwater and mid-order

lO

ll

streams (Flowers and Hilsenhoff 1978; Bednarik and McCaffer-
ty 1979); and 2) reported feeding habits indicate that it is
an opportunistic generalist (Shapas and Hilsenhoff 1976;
Edmunds 1984), and therefore is likely to show some response
in feeding under different conditions of resource availabi—
lity.

The overall objective of this study was to determine
the importance of periphyton in the feeding, growth, and

production of Stenonema vicarium and its congeners under

 

differing primary productiOn regimes in 2 Michigan streams:
closed canOpied Schwartz Creek (2nd order) and the open
canopied Ford River (4th orderL. The specific hypotheses

were:

1) Stenonema spp. ingest a greater proportion of algae

 

in the Ford River than in Schwartz Creek, due to
greater resource availability in the Ford.

2) Growth rates of Stenonema are higher in the Ford

 

River than in Schwartz Creek, due to a nutritional-
ly superior diet.

3) Production of Stenonema is greater in the Ford

 

River than in Schwartz Creek due to faster growth.

SITE DESCRIPTIONS

Ford River

 

The Ford River is a low gradient, hard water brook
trout stream in Michigan's Upper Peninsula (Lake Michigan
drainage). Sampling was conducted in a single riffle and
adjacent snags (woody debris) in a 4th order (Strahler 1957)
section of the river (Dickinson 00.; T43N R29W sec.14;
= site FEX of Burton et al. 1984). The stream channel at the
study site was 10-12 m wide and 20-50 cm deep under midsum-
mer baseflow conditions. Riparian vegetation was dominated

by Tag Alder (Alnus rugosa), Red-osier Dogwood (Cornus

 

stolonifera), and Balm of Gilead (Populus gileadensis).

 

Discharge during summer normally ranged between 0.5 and
1.0 m3/s, but reached up to 7 m3/s during spring and autumn

(Burton et. al 1984). Mid-channel current speeds averaged
about 20 cm/s. Substrates in the river were predominantly
sand, although pebble and cobble (phi -5 and -6, respective-
ly; Hynes 1970) riffles comprised about 30% of the river's
bottom in the vicinity of the study site. Snags provided an
additional .16 1112 surface area/m2 stream bottom for inverte-
brate habitat. Water temperatures from June to August 1984
ranged between 12.5 and 22.500 (mean = 18°C; mean daily
range = 2.?30). Other physical and chemical data are

presented in Table 1.1.

12

13

Table 1A.
and the Ford River.

Physical and chemical data for Schwartz Creek
All values are mean and (range).

 

Schwartz Ford Ford
1983 1983 1984
pH 7.9 8.0 7.9
(7.4 - 8.2) (7.7 - 8.3) (7.8 - 8.2)
Hardness 133 160 163
(mg CaCO /1) (79 - 217) (124 — 185) (161 - 167)
3
Alkalinity 117 144 151
(mg CaCOB/l) (67 - 143) (114 - 159) (147 - 155)
Conductivity .195 239 261
(umhos/cm) (101 - 256) (185 - 270) (204 - 270)
D.O. 8.3 8.9 8.8
(m8 02/1) (7.7 - 9.5) (8.6 - 9.4) (8.5 - 9-8)
Turbidity 3.2 1.8 1.3
(NTU) (2.0 - 3.5) (1.6 - 1.9) (1.1 - 1.4)

 

Data from Burton et al.

(1984)-

14

Schwartz Creek

 

Schwartz Creek is a 2nd order stream in the Escanaba
River (Lake Michigan) drainage in northern Dickinson Co.,
MI. Sampling was conducted in a heavily shaded 50 m section
of the stream (T44N R28W sec. 10) approximately 200 m down-
stream from a marshland. The study site was located in a

dense White Cedar (Thuja occidentalis) swamp with patches of

 

Tag Alder (Alnus rugosa) also occurring along the stream-

 

banks. The stream channel was 5-7 m wide and 20-50 cm deep
under midsummer baseflow conditions, with a bottom composed
entirely of sand. Current speed.at the study site was ca.
10-15 cm/s. large qnsntities of snags (.68 m2 surface
area/m2 stream bottom) constituted the only stable substrate
for invertebrate activity. Water temperatures from June to
August 1984 ranged between 11.5 and 21.330, with a mean
temperature of 170C. Data collected during summer, 1983
indicated that Schwartz Creek was physically and chemically

similar to the Ford River (Table 1.11

MATERIALS AND METHODS

Stenonema Populations

 

Stenonema spp. were sampled on both snag and stone

 

substrates in the Ford River, but only on snags in Schwartz

Creek as no other fixed substrates were available.

15

Ford Riffle - Invertebrate populations iniflmaFord riffle
were sampled as part of the study of Burton et al. (1984).
Artificial substrates consisting of semicylindrical plastic
baskets (28 cm L x 18 cm W x10 cm D) lined with 60 pm mesh
netting were filled with stream sediments and buried in the
stream bed. From June to October, 1983, and May to October,
1984, 5 replicate samples were collected at monthly inter-
vals with replacement of samplers after processing (approx.
30 d colonization periods). Similar methods were used during

winter months, but Stenonema did not occur in these samples.

 

Sample collection proceeded as follows: Each sampler
was lifted from the stream and its contents emptied into a
bucket of water. Larger rocks in the sample were indivi-
dually scrubbed and discarded. The remaining sample mater-
ial was then washed and decanted several times through a 250
pm mesh soil seive. Seive contents were preserved in 70%
ethanol until further processing. Aquatic insects were
picked from sample debris in the laboratory under a dissec-
ting microsc0pe and separated for later identification and
enumeration.

Species of Stenonema were identified using the keys of

 

Bednarik and McCafferty (1979). Nymphs were measured to the
nearest<lJ mm length; those less than£L5 mm could not be
confidently identified to species. Final instar nymphs were
identified by their dark wingpads (Burks 1953).

Since substrate surface area limits the availability of
periphyton to consumers, surface area of stone substrates in

the Ford riffle was estimated to determine effective habitat

l6

(Resh 1979). Twenty 50 cm transects were established by
dropping a meter stick into the riffle, with the position
and orientation of the meter stick upon retrieval defining
the transect, and endpoints taken as the 0 and 50 cm marks.
For each transect, the distance between the endpoints was
measured directly on the substrate surface using a flexible
tape measure, and effective habitat A (m2 surface area/m2

quadrat) was estimated as:
2
A=(2ds)

where dS = the distance between points measured directly on
the substrate surface. This method is similar to that of

Fricke and Thum (1976).

Snags - Three replicate snag samples were collected at
biweekly intervals from each stream between June 18 and
August 27, 1984, and a final set of replicates was collected
on October 19. To minmize sample variance, sampling was
limited to snags 5 to 10 cm in diameter and in an advanced
state of decay (id% no bark and wood soft and pitted;
equivalent to class 3 of Dudley and Anderson 1982).
Sampling was further limited to snags in current speeds
similar to that flowing over the Ford riffle substrate
samplers.

Snag sampling proceeded as follows: The sample snag was
wrapped with a 250 Pm mesh screen to prevent invertebrates
from escaping, then cut off approximately 30 cm from the

end. The snag piece was transferred to a plastic bag, which

17

was then filled with stream water. The sample was then
carbonated (using 100 g baking soda and 250 ml vinegar) to
prevent the animals from regurgitating when formaldehyde was
added as a preservative (M.P. Oemke, MSU; pers. comm.), thus
enabling analysis of feeding habits.

In the laboratory, each snag was placed in a white
enamel pan and rinsed under running water. The surface of
the snag was examined and any remaining animals were
removed. All water used in collecting and processing the
sample was washed through a 250 pm mesh soil seive and the
seive contents preserved in 70% ethanol. Further processing
continued as described above for the Ford riffle samples.

The surface area of each snag piece was estimated by
multiplying its length times average circumference. Adjust-
ments were made as needed to account for branches and

crevices. Sample sizes ranged between 330 and 840 cm2.

Length/Weight Relationship

 

On Oct. 23, 1984, 20 Stenonema vicarium nymphs ranging

 

from.2.9 to 12.2 mm in length were collected from Sloane
Creek, Meridian Township, Ingham Co., MI. In the laboratory
the nymphs were killed in hot (70°C) water, then body length
(anterior edge of head capsule to posterior end of 10th
abdominal tergite) and head capsule width were measured to
(the nearest OUInmn Nymphs were then blotted dry on tissue
paper for 5 s, weighed on a Cahn 27 electrobalance, dried

for 24 hr at 500C, desiccated for 1 hr, and reweighed. The

18

data were loge transformed and individual dry weight
regressed against length and head capsule width. The
length/weight relationship derived for S; vicarium in Sloane
Creek was assumed to be applicable to the Ford River and
Schwartz Creek, since Smock (1980) found that the relation-
ship forfi_annexum did not differ sigificantly between 2

North Carolina streams.

Characterization 2f Periphyton Communities

 

 

Periphyton communities were characterized in each

stream to guage resource availability.

Colonization - Accumulation of periphyton on glass micro-

 

scope slides was followed in each stream as a rough measure
of primary productivity. Colonization data for the Ford
River are from Burton et alJ (1984). Periphyton samplers
consisted of plexiglass slide racks which held 8 standard
microscope slides in a vertical position. Racks were tied
to bricks and placed in both streams on July 30, 1984 with
the slide edges facing into the current. Half of the slides
in each stream were removed at 14 and 28 d intervals and
frozen until. later analysis for biomass (Ash Free Dry
Weight), chlorophyll-a, and phanphytin-a following standard
procedures (APHA 1980). Numbers of replicates for chloro-
phyll and phanphytin were 20 in the Ford, 4 in Schwartz;
for biomass, 10 in the Ford, 4 in Schwartz. Data were
analyzed by comparing the slopes of the regression lines of

biomass and chlorphyll-a against days elapsed.

19

Ambient Periphyton Communities - Chlorophyll-a and phanphy-

 

tin-a were measured on natural substrates as indicators of
periphyton standing crop and physiological condition.
Biomass estimates were not possible due to the difficulty in
separating periphyton from substrate material. On Aug. 24,
1984, 3 replicate 15 cm2 periphyton samples were scraped
from snag upper surfaces from each stream and extracted in
75 ml acetone for 24 hr at 4°C. Data on chlorOphyll-a and
phanphytin-a on stone substrates in the Ford River were
collected by Dr. T.M. Burton (MSU). On Aug. 23, 1984, 5
rocks from the Ford riffle were immersed in separate acetone
baths for 24 hr at 40C. Prior to extraction, the stones used
in these analyses had spent approximately 1.5 hr in clear
plexiglass circulating production/ respiration chambers.
Analyses of chlorophyll—a.and phanphytin—a proceeded as
above. The surface area of each rock was estimated by
wrapping it with aluminum foil, and then measuring the area
of the foil with an electronic area meter (Licor model LI-
3000). Periphyton was assumed to cover half of the total
surface area of each stone. Results were analyzed by 1-way
ANOVA and designed orthogonal contrasts (Schwartz snags vs.

both Ford sources; Ford snags vs. Ford rocks).

Feeding Habits

 

Midgut contents of Stenonema spp. nymphs from each

 

population were examined using the method of Cummins (1973).

After teasing out the midgut in distilled water, the peri-

20

trOphic membrane and its contents were pipetted into a clean
depression slide. Gut contents were emptied into the slide,
broken up with forceps, then washed into a clean beaker.
After sonicating for 5 s, the contents of the beaker were
vacuum filtered (5 psi) through a 0.45 pm Millipore filter.
The filter was dried.for’10 min.at 50°C, cleared in Kleer-
mount mounting medium (Carolina Biological Supply, Burling-
ton,lHU, and then slide mounted with a coversldjn Midgut
contents of individuals < 5.0 mm length were combined within
size classes and species to increase particle density.

Proportions of diatoms and detritus particles on each
slide were estimated by a line-transect technique using a
phase contrast microsc0pe at 4oox. Three 9 mm transects of
30 fields each (300 Pm/field) were taken across each slide.
For each transect, the prOportion of each particle type was
estimated as the total number of ocular micrometer units
intersected by that particle type, divided by total units
for both particle types. The mean of the 3 transects was
taken as the overall estimate for the slide.

Individuals less than 1.5 mm length were whole mounted
in Hoyer's mounting medium since they were difficult to
dissect. Percent diatmms in the gut was determined by
estimating the area of diatoms relative to the area of the

entire gut.

Growth and Production

 

Instantaneous growth rates were estimated for the 1984

cohort of each population as the slope of the regression of

21

loge maximum individual dry weight on each sample date
against days elapsed. Maximum weights rather than mean
weights were used because continuous recruitment occurred in
all 3 populations (see results below). Continuous influx of
small individuals into a population would lead to low mean
weight estimates, thus underestimating growth rate (Waters
1966). Separate estimates were derived for summer (June
through August) and autumn (September and October) since
other studies have indicated that growth of Stenonema
decreases during autumn (Richardson and Tarter 1976; Barton
1980; Kreuger and Cook 1984).

Using the above growth rates, production was estimated
for each sample interval by the Instantaneous Growth method
(Waters 1977):

P = G x B
where P==production.in mg dry weight/mz/days elapsed; G =
growth in mg = instantaneous growth rate times days elapsed;
B = average biomass in mg dry weight/m2, estimated as the
arithmetic mean of the biomass estimates at the beginning
and end of the interval. Total production was the sum of
the interval estimates (see example of calculations in Table

1.3).

RESULTS

Stenonema Populations

 

Ford Riffle - Effective habitatixlthe Ford riffle(&;95%
CI) was estimated tolna1.45;t.07 m2 substrate surface/m2
quadrat. All following references to the riffle population
are expressed in terms of effective habitat.

Stenonema vicarium (Walker) [=fuscum.(Clemens)J pre-
dominated in the Ford riffle in both 1983 and 1984. Only a
few specimens of S; modestum (Banks) L: rubrum (McDunnough)J
were collected in either year. Nymphs were not found in
samples between June and November in either year, although

S; vicarium had been observed in winter leaf pack samples

 

from the Ford River (pers. dst. Population density was
greater in 1984 than in 1983 (mean density = 473/m2 and
136/m2, respectively), but individuals were more evenly
distributed over all size classes in 1983 (Figure 1J).
Nymphs larger than 5.0 mm did not occur in the 1984 samples
(Figure 1.2). The presence of individuals in the smallest
Size class from mid-June through mid-October of both years
'(Figures 1(1,1.2) indicated that egg hatching;was continu-
ous throughout the summer. Mature nymphs were collected in

June, 1983, indicating that adult emergence occurred then.

Ford Snags - Numbers of Stenonema spp. on Ford snags were

 

 

extremely lOW'On all sample dates, with most individuals
less than 2.0 mm in length (Figure 1.3). As observed in the

riffle samples, individuals in the smallest size class

22

'5 Total Number

:21

Figure 1.1.

 

 

 

23

 

80'
. 1— June 17 (rt-30)
4o«
6‘ 9
—1 71—. ,q—q'
601
_

 

 

f

July 18 (nos 1)

 

 

Aug. 17 (fl-152)

 

 

 

 

 

 

Sept. 1 7 01-52)

 

 

 

 

 

 

 

 

 

60'

 

 

 

Oct. 20 (11.9)

 

4O

 

 

Nov. 13 (nu-3)

 

 

 

 

 

”-1

T V I T

4 6 8 10
Length (mm)

Size distribution of nymphs, Ford riffle,

1983.

24

801 r

June 15 (fl-12)

40‘

 

 

 

 

 

 

 

 

 

 

 

60'
‘ July 15 (name)
4 L
5 40 _1
.1
3 T
E 1 .
é! Aug. 14 (n-509)
a 20"
fl
0
1- L
a? _
40-
(—
H o
201 Sept.12 (M137)

 

 

 

 

 

 

 

60‘ .
Oct. 13 (n35)

 

 

 

 

4fl2r4 6

Length (mm)

Figure 1.2. Size distribution of nymphs, Ford riffle,
1984.

25

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9
June 18 (m9)
r'“!
40 July 2 (n-13)
801 1
July 16 (M7)
40-
8
g [—1
:3
Z 8
: ”1 '1
1' ‘ July 29 (n-102)
8' 4o.
”Fl—g
501 H
L Aug. 14 (nu-58)
501 .

 

Aug. 28 (n-10)

 

Q4

. *é . ‘3
Length (mm)

Figure 1.3. Size distribution of nymphs, Ford snags, 1984.

26

occurred throughout the summer. No Stenonema were found in

 

the October samples. All individuals large enough to be

identified to species were Stenonema exiguum Traver (=guin-

 

qgespinum Lewis), although some individuals less than 2.5 mm

 

length had mouthparts typical of S; modestum or S; vicarium.

Schwartz Snags - Three species of Stenonema were found in

 

 

Schwartz Creek: S; exiguum comprised approximately 53% of
the total population, imodestum 35%, and S_._ vicarium 12%.

 

 

The size frequency distribution of each species taken alone
(Figure 1.4) did not differ significantly from the distribu-
tion of the combined populations (Figure 13% contingency
tables; P > CL05 for all species). Therefore, the combined
size frequency distribution in Figure‘h51was taken to be
representative of the life cycles of all 3 species, because
estimation of growth rates from Figure 1.4 would not have
been possible.

The 1983 and 1984 cohorts overlapped considerably, with
both egg hatching and adult emergence occurring throughout
the sampling period (Figures 1.4, 1.5). Interspecific
differences in egg hatching periods were not likely, since-
members of the 1984 cohort of each species were first
observed on the same sample date (Figure 1.4). Adult emer-
gence occurred mainly in June and July (Figure 1JH. Sex
ratios were not significantly different from 1:1 (Chi-square
tests; P > CL50 for all species). Mean length at emergence
for females (9.0 mm) was greater than that of males

(7.5 mm), as is typical of mayflies (Burks 1953).

27

.ocfla common m nuHS noumoaocw ma muuonoo vmma cam mama
uo cowumummom venomoum .xmm Umumoaccw on» NO mamsoa>floCH
mcflmuoEo ouocoo m3ouud .ham>auoommou .mnmE>: coxomco can
.onE.onsom ouocoo mama noammwum cam .como .oflaom .vmma
.xomuo Nuum3nom ea mnmaxc maococoum mo cowuonwuumao wuam .v.H ousmwm

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

8330? .m 8:332: .w 5.59.8 .w.
or w o v RN or o o v 9N Op 0 o v RN
GF\OP fl. 0 "u mm . / $.22”
, /
z
./ /
/
NN\Q / Emu / %
x /
/ /
z ,
mpxm Alta «fl, .zul
/ /
/ M1 9
ooxh / mfl II / m
, / ”mm
d
U.
9
or: I. Iall
«x» 1lsmfll
38 .I .U I

2

 
 

‘11 Total Number

Figure 1.5.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

V Y j

25. June 18 (11293)
4
Fri—1" “—1-:
I
50‘ 1 July 2 (n48)
l
‘1 .F‘Lfi
501 1
1 July 16 (n-95)
‘ 1
1
7 A - j 1 W
1
801 F 1‘
‘1 July 30 (nu-79)
401 ‘\
\\
_}_l_J
\ Aug. 13 (mm)
\ -
\
lfi—r—l
\
. \ Aug. 27 111-182)
J hl—l—u-r—‘l \
\
, - \
\
25: \
\\ Oct. 19 (rt-72)
\
. , . J ‘ . 4—:
2 4 8 8 10

Length (mm)

Combined distribution of Stenonema spp.,
Schwartz Creek, 1984. Dashed line indicates
separation of 1983 and 1984 cohorts.

 

29

Length/Weight Relationship

The relationship between body length and dry weight for
Stenonema vicarium is presented in Figure 1.6. Length was a
better predictor of dry weight (r2 = .975) than was head
capsule width (r2: .936). The relationship in Figure 1.6

was assumed to hold true for all Stenonema species found in

 

Ford and Schwartz, since there were no obvious differences

in body proportions among these species.

Characterization g: Periphyton Communities

 

 

As expected, accumulation of periphyton was more rapid
in the Ford River than in Schwartz Creek, in terms of both
biomass and chlorophyll-a (Figures 1.7-A, B; t-tests;
P <CLO1 for both). Chlorophyll/phaeophytin ratio
(Chl/phaeo) increased linearly in both streams (Figure
1.7-C), and was not significantly different between streams
on either sample date (t-tests; P > 0.05 in both cases).
Autotrophic Index (AI = mean AFDW / mean chlorophyll-a) was
greater in Schwartz on both sample dates (Figure 1.7-D).

No difference in chlorOphyll-a density on natural sub-
strates was detected between the 2 streams, suggesting that
their algal standing crops were similar (Table 1.2). Phaeo—
phytin-a density was significantly lower and chl/phaeo
significantly higher on Schwartz snags than on either Ford
River substrate (Table 1.2; One-way ANOVA” orthogonal
contrasts; P < 0431 for both). Phaeophytin-a and chl/phaeo

did not differ among Ford substrates.

3()

8.0
-

’61
.2

E v- 0.0027 x3 5
Vg.‘ 2
._. r - 0.90
:1:
9.
Lu
3

‘2.
>0
0:
o

9.

 

 

 

220 4:0 020 020 16.0 11.0 13.0

LENGTH (mm)

Figure l.6. Length/weight relationship for S. vicarium.

31

ouocun mama Hmowuno>

.mm H

.33 Susana .Alv 003m Boa

unu cam Altlv xomuo Nuum3som cw coaumNHcoaoo couznmfluom

c.:a0nxm.0 n>eo

 

 

MWl. «F
H 11111 me
o.—
m.—
0 od
ouanoaxm_o m>ao
on v.—
1111 m .\\
\ \\\
a . \\u\ e.
\
\ a
\\ om
e\\\ .
M
ed

0.?

(000 l +) IV

(aw/6w) 9-1110

ouaoonxm.o a>eo
an RF

A

.h.H

 

 

 

musmflm

ONNdINO

(aw/b) MOdV

Table 1.2.

communities.

32

All values are mean t SE.

Characterization of ambient periphyton

 

ChlorOphyll-a Phaeophytin-a Chl/Phaeo

SOURCE (mg/m2) (mg/m2)

SCHWARTZ
SNAGS 18.15 _ 3.53 3.94 : o.96** 4.78 O.39**
(n=3)
FORD
SNAGS 18.39 3.53 6.69 i 0.96 2.62 0.39
(n=3)
FORD
(ROCKS) 19.14 2.74 8.55 i 0.74 2.26 0.30
n = 5

 

** = significantlydifferent from other values in same
column; P < 0.01, one-way ANOVA, orthogonal contrasts.

33

Feeding Habits

 

The prOportion Of diatoms in guts of Stenonema spp. was

 

positively correlated with body length in all 3 populations
(Figure 1.8-A, B). Interspecific differences in feeding
were not observed in Schwartz (t-tests; P > 0.30 for all
combinations), so the length/diet regression in Figure 1.8-B
was calculated using the combined data. Length-specific
ingestion of diatoms was significantly greater in the Ford
riffle than on Schwartz snags (t-test; P1<CL05), but Ford
snags did not significantly differ from either of these

(t-tests; P >(L01 for both).

Growth and Production

 

Growth rates from June through August, 1984 did not
significantly differ among the 3 populations (t-tests;
P > OJMD for all combinations). Therefore, the growth rate
for all 3 populations during this period was estimated as
the slope of the regression line for the combined data sets
(Figure 1.9; .0749 mg/mg/d). The autumn growth rate
estimated from the combined Schwartz and Ford riffle data
was much lower (.0110 mg/mg/d; Figure 1.9), as had been
expected. Growth rates estimated using the 5 largest
individuals on each sample date were similar (.0730 and
.0118 mg/mg/d for suumer and autumn, respectively), so that
using only 1 individual per sample date did not bias the the
growth rate estimates. Growth rates could not be estimated
for the 1983 Ford riffle population because cohorts were not

clearly distinguishable (Figure 1.1%

Figure 1.8.

 

 

34

 

  
   
 
   
  

 

 

9
D!
N / A
/
/ /4
q // //
2‘ / /
O / /
/ / o
/ /
/ /
m O'J // O / I,
2 N / °// //
C)“ "
p.
s °F0rd tlflle
Q o
-+ y-0.30 + 1.571:
N 0
r2- 0.70
¢:_4 11 Ford 00000
' y-1.14 + 1.3011
12- 0.40
O
o’ r r 1 I I
0.0 2.0 4.0 8.0 8.0 10.0
LENGTH (mm)
°.
C—
N 0 S. exlguum B
(D S. modestum
9 A S. vlcerlum I!)
2‘ X Stenonema spp.
(I) c: y - 0.00 . 1.55x ‘9
2 N4 2
O -‘ 7 =0.75
1—
s
D O
--1
N O
94
V
O
c; I 1 7 1
0.0 2.0 8.0 8.0 10.0

4.0
LENGTH (mm)

Regression of % diatoms in gut contents (1 95%
CI for regression line) in: A) Ford River;
B) Schwartz Creek.

35

mum-5.20 + 0.011 d
12- 0.50

A Ford riffle

max. Individual dry wt. (pg)

0 Ford snags

' Schwartz snags

 

 

 

T' ‘r’ ' I! ' ' rr '
20 40 60 80 100 120
J A S 0

days elapsed

Figure 1.9. Growth regression for the combined populations.

36

hm.m n «H.5a \ hm.m: u m\m

 

 

 

 

 

 

e mmfi\me\me oeem.me u d me\me msmfi.efl u m
Hom>.m: mH\0fi
ommm.ma ommm.o oaao. :oow.mm
bnon.mm hm\m
«hon.mfi mmzo.fi msbo. Hmbm.:fi
:mmm.w mfi\m
:bmm.: wmsorfi mzbo. :mmfi.:
:hmb.fi om\h
ombm.m omzo.fi 0:50. bmmfi.m .
mmbm.m ma\b
5mmm.a omso.fi mzbo. mmbm.a
mmmm. m\h
mmmm. mm:o.a mzho. monm.
mbmfi. maxm
flm Mo "am pm ".0 Ac\wawwev m mmmeomm A E\wev sumo
1 owmpo>< mmmeofim oaaemm
a
.m

.uuonoo vmma uuumznom ecu How :ofiuooboum mo c0wumflsoamu .m.H wands

37

Table 1.4. Summary of production estimates.

* * **

POPULATION COHORT P B P/B DAILY P

SCHWARTZ 1983 + 1984 538.1 69.0 7.8 4.37

 

 

 

 

SCHWARTZ 1984
S exi uum 21 1 9.0 2.3 .172
E: modestum 11 9 4.1 2.9 097
§_ vicarium 11 0 4.0 2.8 089
TOTAL 44 O 17 1 2.6 .357
FORD RIFFLE 1984
S; vicarium (> 95%) 34.8 6.6 5.3 .290
S; modestum (< 5%)
FORD SNAGS 1984
S; exiguum 6.2 0.8 7.8 .050
other species?
2
* = mg dry weight / m
2

** = mg dry weight / m / day

38

Total production of Stenonema spp. was greatest in

 

Schwartz Creek, regardless of whether or not the 1983 cohort
was considered (Table 1.4). Comparing species individually,
S; exiguum and S; modestum were most productive on Schwartz
Creek snags, whereas production of S; vicarium was greatest
in the Ford riffle. S; modestum accounted for less than 5%

of the total Stenonema production in the Ford riffle. The

 

production of each species in Schwartz was approximately
proportional to its relative abundance in the total popula-
tion. Turnover ratio (P/B) was high (7.8) for the combined
Schwartz cohorts and for Ford snags, intermediate (5.3) for
the Ford riffle population, and lowest for the Schwartz 1984
cohort (mean = 2.6; Table 1.4).

DISCUSSION

Stenonema Populations

 

The size freqency distributions of Stenonema Spp.

 

observed in this study (Figures 1.4, 1.5) are similar to
those reported for other northern streams (Coleman and Hynes
1970; Richardson and Tarter 1976; Flowers and Hilsenhoff
1978; Barton 1980; Kreuger and Cook 1984). The apparent

similarity of life cycles among Stenonema spp. in Schwartz

 

Creek is supported by Flowers and Hilsenhoff (1978), who

concluded that life cycles of Stenonema.spp. in Wisconsin

 

39

.noumoaccw ma xpsum ucomoum on» How nowuom mcflamsmm
one .Aucwa mouv mocomuoEu ou AOGHH Eouuonv mmmo Eoum
mmmumoum >02» mm uaomoum mama»: no money mnwm mouocoo

mono boomsm .mmm mEococoum MOM Emummflo.mao>o omen

ecoeaom

 

 

uum

 

 

 

 

:32

 

noted OEEEem

 

.OH.H museum

0219 10110111“.

40

are similar. These life cycles conform to the univoltine
"slow seasonal" pattern of Hynes (1970) (Figure 1.10): Eggs
hatch after a short incubation period, nymphs grow rapidly
until late fall, resume growing in spring when water temper-
atures rise, and emerge as adults primarily in late spring
and early summer. Extended egg hatching as observed in this

study has not been reported previously for Stenonema,

 

although the broad size frequency distributions reported in
other studies suggest that this phenonmenon may be wide—
spread in the genus (Richardson and Tarter 1976; Flowers and
Hilsenhoff 1978; Kreuger and Cook 1984).

Disappearance of Stenonema nymphs from Ford River

 

samples during late autumn was coincidental with rising
discharge (Burton et al. 1984). Disappearance of S; vicarium
nymphs during winter was also noted by Coleman and Hynes
(1970), who suggested that the nymphs may migrate into the

hyporheic zone during this period.

Periphyton Communities

 

The results of the periphyton colonization study
suggest that primary productivity was greater in Ford than
in Schwartz, since periphyton accumulation was more rapid in
the unshaded Ford River (Figure 137L Diatom colonization
studies carried out in these streams during 1982 and 1983
yielded similar results (M.P. Oemke, MSU; pers. comm.).
Periphyton accumulation rates can not be interpreted as net
primary productivity, since they represent a composite of

colonization and growth (Wetzel 1966). Most important to

41

consumers though, is that the Ford periphyton community
regenerates more quickly, and thus may be able to withstand
greater grazing pressure.

Autotrophic Indices (AI) and Chlorophyll/phaeophytin
ratios (chl/phaeo) are general indicators of the physiologi-
cal condition of periphyton communities (Weber 1973; APHA
1980). AI roughly corresponds to the prOportion of the
total periphyton community (including bacteria, fungi,
protozoa, and detritus) comprised by algae, and normally
ranges between 50 and 200 for "healthy” (predominantly auto-
trOphic) periphyton communities and higher for heterotrOphic
communities (APHA 1980). Therefore, the greater AI in
Schwartz (Figure 1.7) suggests that detritus and heterotro-
phic organisms comprise a greater portion of its periphyton
community, although algal standing crop is similar to that
in the Ford (Table 1th Chl/phaeo roughly corresponds to
the ratio of living (actively photosynthesizing) to senes-
cent algal cells, since phaeophytin-a is the major breakdown
product of chlorophyll-a. Thus, the low chl/phaeo on
Schwartz Creek snags indicates a periphyton community in
better physiological condition to that in the Ford River.
This finding is surprising, since physiological condition
would be expected to be highly correlated with productivity.

In summary, the results suggest that while algal
colonization of substrates is initially more rapid in the
Ford River than in Schwartz Creek, both periphyton communi-

ties eventually attain a similar standing crop. The high

42

chl/phaeo observed in Schwartz may suggest that this peri-
phyton community was in an actively growing state, whereas
low chl/phaeo in the Ford River may indicate self-limitation
(i.e. senescence of cells due to competition for light and

nutrients; see Lamberti and Resh 1983, 1985)-

Feeding Habits

 

Lack of major differences in the feeding habits of the
3 populations (Figure 1.8) is not surprising, in view of the
similarity 1J1 periphyton standing crops between the 2
streams (Table 1.2). Since the regression lines of the snag
inhabiting populations were almost identical (Figure 1.8),
habitat appears to have been a more important factor in the
feeding habits of these populations than stream order.
Presence of wood particles in the guts of snag-inhabiting
nymphs suggests that they ingested substrate material in the
process of scraping periphyton. Despite the significant
difference in feeding observed between the Ford riffle and
Schwartz snags, Hypothesis #1 (algal consumption in the Ford
River > Schwartz Creek) is rejected, since it was formulated
with the expectation that resource availability would differ

between the 2 streams.

Growth and Production

 

The summer instantaneous growth rate of 7.5 %/d (.0749
mg/mg/d) is higher than most growth rates reported for
mayflies living in similar temperature regimes (see Table 6

of McCullough et al. 1979). Higher growth rates have been

43

reported only for species of Tricorythodes: 12.6 %/d for

 

1h_pipptus (McCullough et al. 1979) and 15.3 %/d for

.T; atratus (Hall 1975).1Tricorythodes spp..are small, multi-

 

voltine mayflies with short developmental times. Rapid
growth for these species may be an adaptation for producing
several generations per year (high "r"). On the other hand,

the rapid growth of Stenonema during summer may be necessary

 

to compensate for the extensive period of its life cycle
during which there is little growth (Figure 1.10).
Rejection of Hypothesis #2 (Growth in the Ford >
Schwartz) could have been caused by several factors: First,
Figure 1.8 is only a crude assessment of nymphal feeding
habits, particularly with respect to the nutritional quality
of the diets. It is possible that part of the "detritus"
fraction is actually diatom cell contents. Second, there is
no way to determine whether diatoms were alive or dead when
ingested. The higher chl/phaeo in the Ford (Table 1.2)
suggests that the nymphs ingested proportionally more
senescent diatoms than their counterparts in Schwartz Creek,
so that ingestion of viable cells could have been comparable
in the 2 streams. Third, even if feeding habits were accu-
rately assessed, the diets of the 3 populations probably did
not differ enough to produce detectable differences in
growth rates. Fourth, nymphs in Schwartz Creek may have
increased their ingestion rates to compensate for poor food
quality (Cummins and Klug 1979). Temperature probably was

not a factor, because the Ford River was only a mean of 1°C

44

warmer than Schwartz Creek. Based on the temperature-growth
relationship for the heptageniid Ecdyonurus dispar (Humpesch
1981), such a minor difference in temperatures probably
would not have had a detectable effect on growth.

The daily production estimates for 1984 cohorts (Table
1.3) suggest acceptance of Hypothesis #3 (Production in the

Ford > Schwartz) only with respect to S; vicarium, since

 

production of S; exiguum and S; modestum was greatest in

 

Schwartz Creek. However, the hypothesis can not be accepted
because acceptance is dependent on the existence of faster
growth in the Ford River. Absence of detectable differences
in growth rates indicates that production was determined
largely by factors influencing population density and
standing stock (eug. recruitment and mortality).

Among these factors, the substrates available in each
stream undoubtedly had major consequences for species compo-

sition and population density. Stenonema exiguum occurred

 

exclusively on snags, even in the Ford River where alternate
substrates were available. The opposite was true of Steno-

nema vicarium, which was dominant in the Ford riffle in bOth

 

1983 and 1984, but absent from adjacent snags.' Stenonema

 

vicarium was also the least abundant of the 3 Species on

 

Schwartz snags. S; modestum occurred on both stone and snag
substrates, but attained its greatest numbers on Schwartz
snags. Thus, each species was associated with a specific

habitat: S; exiguum and S; modestum with snags, and S;

 

vicarium with stone substrates. Other workers have pointed

out similar habitat associations for these species (Leonard

45

and Leonard 1962; Lewis 19743 Flowers and Hilsenhoff 1978;
Bednarik and McCafferty 1979), although S; modestum is often
associated with stone substrates.

The P/B (turnover) ratios in Table 1.3 can provide
valuable insights into the life history characteristics of
each population (Benke 1984). Since P ==(} * B by the
Instantaneous Growth method, then cohort P/B should be equal
to cohort G. However, this relationship holds true only
when both growth and mortality are exponential; inequality
of cohort P/B and G therefore indicates deviation from
exponential growth and/or mortality (Waters 1966, 1983). It
has already been established that growth was not exponential
over the entire sampling period of the present study (Figure
1.9). Reduced growth rates in the autumn decreased P/B
relative to G, because large, slow-growing individuals
increased the average biomass of the cohort (B) without
contributing substantially to production (Waters 1983). On
the other hand, continuous recruitment and high mortality
(or drift) increased P/B relative toClby'reducing average
biomass (Winterbourne 1974). Calculation of cohort G for

these Stenonema pOpulations would be difficult because the

 

sampling period did not cover an entire cohort. However,
since all 3 populations were subject to the same conditions
of non—exponential growth (Figure 1.9) and continuous
recruitment (Figures 1.2 - 1.5), the P/B ratios in Table 1.3
indicate the relative magnitude of the biomass losses from

each population. iflmahigh P/B ratios observed on snags in

46

both the Ford and Schwartz (combined 1983 and 1984 cohorts)
indicate that these populations experienced the greatest
biomass losses. Obviously, adult emergence was a major
source of losses from Schwartz snags (Figure 1.5L. Consi—
dering only the 1984 cohorts on snags, biomass losses were
much greater from snags in Ford than Schwartz. The disap-

pearance of Stenonema nymphs from the Ford River during

 

autumn undoubtedly accounted for much of the biomass loss
from snags, but the magnitude of P/B for Ford snags relative
to the riffle indicates that mortality was quite severe.

Therefore, both biomass and "survivorship" of Stenonema were

 

greatest:on Schwartz snags, suggesting that snags in the 2
streams did not provide equally suitable habitat. One
obvious physical difference between them was their degree of
lighting; nymphs living on the more brightly-lit snags in
the Ford River may have been more susceptible to predation,

especially by fish.

General Discussion

 

The main conclusion that can be drawn from this study
is that habitat, not stream order, is the primary factor

controlling the productivity of Stenonema spp. in the Ford

 

River and Schwartz Creek. Certainly, the lack of major
differences in periphyton availability between streams
invalidated the initial hypotheses, since the expected
conditions of differing resource avilability were not met.
Even if the streams were to differ in periphyton standing

crop, it would have been difficult to unequivocally connect

47

 

lsrnem ORDER

 

 

   

 

 

TEMPERATURE I RESOURCE AVAILABILITY J

 

 

DIET

 

0110er RATEJ

 

 

 

| PRODUCTIVITY J

 

Figure 1.11. Model used in hypothesis formation.

48

these conditions to the productivity of a single species or
group of species. Examination of the model used in hypothe-
sis formulation for this study (Figure 1.11) clarifies this
point: stream order was hypothesized to indirectly influence
secondary productivity via its effects on resource availabi-
lity, feeding, and growth. In retrospect, this model is
admittedly an extremely simplistic view of stream ecosystems
since it fails to account for habitat, competition, preda-
tion, etc.; in short, it considers only those factors of
interest as if in a controlled laboratory setting.

A more realistic model (Figure 1.12), based on the
results of this study and the reviews of Benke (1984)
Sweeney (1984), and Minshall (1984) highlights the complex-
ity of factors influencing a species' productivity.
Although this model is by no means exhaustive, it is quite
clear that comparisons of the productivity of a single
species are inappropriate, since the results will. be
confounded by uncontrollable factors, such as substrate and
community structure. Community structure may have played an
important role in this study: several potential competitors

for Stenonema exist in the Ford River, including 5 other

 

heptageniid species (Burton et al. 1984). In contrast, the
invertebrate fauna in Schwartz Creek is much less diverse,
with only 2 other heptageniid species occurring in extremely
low numbers (pers. Ost. It is possible then that the high

productivity of Stenonema spp. in Schwartz Creek is partial-

 

ly caused by the absence of interspecific competition.

49

 

 

WATERSHED CHARACTERISTICS l

 

 

STREAM ORDER [—45-—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

      
    
 

 

 

 

 

 
    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SUBSTRATE
OTHER 1 1 -
LIMNOLOGICAL COMMUNITY
FACTORS STRUCTURE
[TEMPERATURE] RESOURCE
e AVAILABILITY
SPECIES
INTERACTIONS
SEASONAL
FACTORS
GROWTH RATE
NUMBERS
I FECUNDITY BIOMASS
[PRODUCTIVITY

 

Figure 1.12. Improved model of productivity.

50

In conclusion, substrate availability had major conse-

quences for the distribution and productivity of Stenonema

 

spp. in Schwartz Creek and the Ford River. Since many
benthic invertebrates are generally associated with snag
habitats (Dudley and Anderson 1982), the distribution and
abundance of snags along a stream's course may govern
species distributions. Schwartz Creek is typical of head-
water streams in the study area, with snags and overhanging
vegetation providing the only fixed substrates. Snags are
less abundant in downstream reaches as the stream channels
widen. Similar snag distributions have been reported in
other North American streams (see Wallace and Benke 1984).
This pattern of habitat distribution has several implica-
tions for the River Continuum Concept (Vannote et al. 1980):
First, headwater invertebrate communities may be based on
heterotrOphy because of ingestion of substrate material, not
because of reduced primary productivity as suggested by the
Concept. Second, the absence of "specialized" grazers (egg

Glossosoma) in headwaters may only be a result of the

 

unsuitability of snag habitats for their existence (e.g.

absence of pGCer case building materials for Glossosoma).

 

Third, utilization of periphyton by species associated with
snags may be inefficient, shifting the resource base of the
community farther towards heterotrOphy than suggested by
resource availability; While the River Continuum Concept may
be generally accurate, the distribution of habitats along a

streamfs course must also be considered.

CHAPTER 2

THE INFLUENCE OF DIET on THE GROWTH
or STENONEMA VICARIUM (WALKER)

 

INTRODUCTION

Food quality is often cited as having pronounced
effects on growth and other life history characteristics of
aquatic invertebrates in the laboratory (Willoughby and
Sutcliffe 1976; Anderson and Cummins 1979; Cianciara 1980;
Fuller and Mackay 1981; Sutcliffe et al. 1981; Bird and
Kaushik 1984; Sweeney and Vannote 1984). However, the
effect of food on growth rates in natural populations is
still uncertain, since temperature and other factors are
also involved (Sweeney 1984). Separation of the effects of
temperature and food is especially important in the context
of the River ContinuUm Concept (Vannote et al. 1980) because
longitudinal changes in shading often result in a positive
correlation between primary productivity and temperature in
mamy Northern Temperate streams. Thus, 'the effects of
resource availability and temperature may be confounded with
respect to stream order (Anderson and Cummins 1979).
Unequivocal separation of the effects of these two factors
may be possible only in controlled laboratory experiments
(Sweeney and Vannote 1984).

Living food resources, such.as animal tissue, algae,
and fungi, are generally superior to detritus in nutritional
quality (Lamberti and Moore 1984).F%n'example, carbon to
nitrogen (C:N) ratios for periphyton range between 3.7:1 to
738:1 (McMahon et a1. 1974; McCullough et al. 1979), indica-
ting a high protein content. In contrast, C:N ratios for

detritus may range as high as 1340:1, reflecting a high
52

53

content of cellulose, lignin, and other refractory substan-
ces (Cummins and Klug 1979; Lamberti and Moore 1984).
Assimilation efficiencies for algal diets range between 30-
60% (Trama 1972; McCullough et al. 1979), whereas efficien-
ciescnldetrital diets generally fall between 7—25% (sum—
marized in Berrie 1976). Detritivores may compensate for
poor food quality by increasing ingestion rates (Cummins and
Klug 1979), ingesting small quantities of animal tissue
(Hanson.et al. 1983),cn'by utilizing byproducts of micro-
bial metabolism (Lawson et al. 1984).

Growth rates produced by various food resources have
Often been used to assess their relative nutritional quality
(Willoughby and Sutcliffe 1976; Cianciara 1980; Fuller and
Mackay 1981; Sutcliffe et al. 1981; Bird and Kaushik 1984;
Sweeney and Vannote 1984A Only 4 studies (McCullough et al.
1979; Cianciara 1980; Bird and Kaushik 1984; Sweeney and
Vannote 1984; the latter 2 published since this experiment
was conducted) have directly addressed the connection
between diet and growth in mayflies. A fourth (Trama 1972)

described the bioenergetics of Stenonema pulchellum, (Hepta-

 

geniidae) feeding on diatoms, but did not compare this
information with other diets.
This experiment was designed to investigate the role of

algae in the growth of Stenonema vicarium (Walker), a common

 

heptageniid mayfly in streams of the Great Lakes region
(Flowers and Hilsenhoff 1978; Bednarik and McCafferty 1979).

The objective of this study was to determine the relative

54

efficiency of utilization of natural stream periphyton vs.

detritus by i vicarium in terms of its growth rate on each

 

food resource. The specific hypothesis was that growth
would be greater on a diet that includes algae than on a

diet of leaf detritus.

MATERIALS AND METHODS

The experiment was conducted in 2 thermally controlled
artificial stream channels (Frigid Units, Toledo, OH). In
each of 3 runs, one channel was manipulated to simulate
autotrophic conditions (AUT), while the second was manipu-
lated to simulate heterotrophic conditions (HETL. Treat-
ments were randomly assigned to channels for Run 1,
switched for Run 2, and randomly assigned for Run 3, so that
treatment effects could be separated from Channel effects
(e.g. residual effects from previous usage of the stream
channels). At the beginning of each run, both channels were
filled with 160 l of water from the Red Cedar River (RCR),
Meridian Township, Ingham Co., MI. Approximately 1/3 of the
water in each channel was replaced every 4 d to prevent
nutrient limitation. Each channel was provided with 15
plastic cages (16 x 16 x 10 cm, 1 mm mesh screened sides)
for individual growth chambers. Both channels were run at
11°C (: 1°C) under a 12:12 hr photoperiod during all runs.

Channels were thoroughly drained and cleaned between runs.

55

Run 1 began on Sept. 29, 1984, Run 2 on Oct. 17, and Run 3
on Nov. 8.

Treatments were conducted as follows: Natural periphy—
ton growing on stones was provided for food in AUT. Stones
of about 100 cm2 upper surface area were collected from
Sloane Creek, a tributary of RCR. Stones were replaced
every 4 d to prevent food limitation. Residual periphyton on
the stone surfaces at each change indicated that food
limitation was not a factor in AUT. Care was taken to remove
all macroinvertebrates from the stones prior to use in AUT.
Lighting over AUT was augmented by flourescent "grow
lights", also set at a 12:12 photoperiod, and suspended 40
cm over the water's surface. Dried, autumn scenescent White

Ash.(Fraxinus americana) leaves were provided for food in

 

HET. Leaves were conditioned in the dark for 14 d prior to
each run in aerated RCR water at 18°C. To stimulate fungal
and bacterial growth on the leaf surfaces, the culture water
was supplemented with 25 g KHZPO , 6.5 g NaCl, 18 g MgSO4,
3 g CaCl 2(H20), and 37 8 KNO3 (total volume = 37 l; Lawson
et al. 1984). Fungi, bacteria, and protozoa were observed
on leaf surfaces cultured in this manner; however, algae was
never observed. Each growth Chamber was provided with
approximately 20 entire leaves,.and a stone of approximately
the same size as those provided in AUT cages. These stones
were collected from a gravel pit and washed prior to their
use in each run. jLack of extensive skeletonizing of leaf

surfaces at the end of each run indicated that food limita-

56

tion did not occur in HET. All water used in HET was
filtered through compressed glass wool to remove algae.

Stenonema vicarium nymphs were collected from Sloane
Creek on the day preceding the start of each run, and kept
without food overnight in the dark at 10°C. At the start of
each.runq 30 nymphs were blotted dry on tissue paper for 5
s, then weighed to the nearest 0.1 mg on a Sartorius 1207
MP2 electrobalance. Nymphs were then randomly assigned to
cages within each treatment. At the end of 14 days, nymphs
were removed from their cages, reweighed, and preserved in
70% ethanol. Exuviae in each cage were also preserved as
secondary evidence of growth.

Instantaneous growth rate (% wet weight/d) was calcu-

lated for each nymph as per Sutcliffe et al. (1981):
G = [ln(We/Wb)/tj x 100% (1)

where We = wet weight at the end of the run, W = wet weight

b
at the beginning of the run, and t = elapsed time in days.

Growth rate estimates were loglo transformed to correct
for heterogeneous variance (Cochran's C; P = 0.053), and

analysed using the multivariate model:

G + D. + F

ijkl = P + Ci 3 k + Dij + eijkl (2)

where: G 1 = an individual growth rate

ijk

P = overall mean growth rate

0
II

fixed effect of stream channel i (i = 1,2)

U
II

fixed effect of starting date j (j =1,2,3)

fixed effect of food source k Ui==1,2)

 

57

DF interaction between starting date andfood

jk‘
eijkl = random individual error

Since preliminary analysis indicated that channel
effects were not significant CPDnSO), Ci was deleted from
model (2) to eliminate confounding of the interaction term

(14% each channel received only one treatment during each

run). Thus, the model was simplified to:

ijl = P + Dj + Fk + Dij + ejkl (3)

where all terms are defined as in (2).

To confirm that the treatments produced the desired
nymphal diets, midgut contents of 3 or 4 nymphs in each
treatment group were dissected and mounted on microscope
slides using the method of Cummins (1973). Approximate
proportions of diatoms and detritus were determined by
taking a line transect across each slide using a phase
contrast microscope at 4oox. The prOportion of each
particle type was estimated as the total number of ocular
micrometer units in 30 fields (300 um each field) intersec-
ted by each particle type, divided by the total micrometer

units for both particle types.

 

RESULTS

Mortality was very low or nonexistent in all treatment
groups, with only one death each in AUT Runs 1 and 2, and
HET Run 1. During Run 3, about 1/2 of the nymphs in each
treatment did not molt or grow significantly (Figure 2.1).
All other nymphs molted at least once during the eXperi-
ments. Non-molting individuals were excluded from all
statistical analyses.

The overall effect of diet on growth rates was highly
significant (P< 0.01,Tablr>2.1),with nymphs in AUT growing
an average of 0.22 %/d faster than those in HET (Figure
2A). However, only Run 1 was significant when treatment
means were compared within each run (P = 0.001, 0.14, 0.09
for Runs 1,25 and 3 respectively; orthogonal contrasts).
The starting date of each run also had a highly significant
effect (P <<L01; Table 2A), with growth rates generally
decreasing over the course of the 3 runs (Figure 2.1). The
overall effect of interaction was not significant (P = 0231;
Table 2.1), although interaction approached significance
(P = 0.06) between Runs 1 and 2.

InSpection of midgut contents of nymphs in each treat-
ment group verified that nymphs in HET did not ingest
diatoms (Table 2.2). Guts of HET nymphs contained primarily
amorphous organic material with fragments of fungal hyphae
and mineral particles. Guts of AUT nymphs contained

diatoms, mineral particles, and amorphous organic material.

58

3.0-

2.04

Growth Rate (%ld)
'35

 

59

n-14

 

 

 

 

 

 

 

 

Figure 2.1.

molted not

I II ||| molted '

Run Number

Mean growth rate (: 95% CI) of each treatment
group.

 

 

60

Nb ofihfim.b a<&o&

 

 

meefio. we m::oe.a modem

cam. mmmma.fi Romeo. m mommo. eoeeooeoueH
moo. moumm.oe mmfima. A mmfima. ooou
moo. mmHEm.o emmofi. m modem. oboe
Hmmmm.mmm :maee.m a :mfiee.m bemomeoo

d a mm<aom z<m2 we mmmeaom 22m one<Hm<> do momaom

.mwmc Upon ERZORw coELommcmRRIOHon Rom CHAMP <>oz<z .fi.m CHAMP

 

61

The proportion of diatoms in gut contents of AUT Run 3
nymphs was much higher than in previous runs (Table 2.2).
Diatoms ingested by AUT nymphs were primarily Pinnularia,

Cocconeis, Gomphonema, and Navicula.

 

 

DISCUSSION

Although the overall effect of diet was highly signifi-
cant, the results only partially support the initial hypo-
thesis that §L vicarium grows better on a diet of periphyton
than on leaf detritus. Lack of significant differences in
growth rates on the 2 diets during Runs 1 and 2 could have
been caused by: 1) contamination by algae in HET Runs 2 and
3, but examination of gut contents (Table 2.2) indicates
that this was not the case; 2) changes in the nutritional
quality of either of the diets over the course of the
experiment; or 3) a greater consumption rate by HET nymphs.
Cummins and Klug (1979) suggested that aquatic consumers may
increase their consumption rate to compensate for poor food
quality, although this parameter was not considered in the
present study.

Since periphyton was collected for AUT during a period
of rapid seasonal changes in Sloane Creek, it is possible
that changes in nutritional quality occurred over the course
of the experiment. Qualitative changes in periphyton during

autumn, such as an increase in C:N ratio (McMahon et al.

62

Table 2.2. Feeding habits of S; vicarium nymphs in each
treatment group.

PERCENTAGE IN GUT

Treatment Run n Diatoms Detritus
AUT 1 4 33 67
2 3 28 72
3 3 76 24
HET 1 4 0 100
2 4 0 100
3 3 O 100

Table 2.3. Correlation between individual weight and growth
rate in each treatment group.

Starting weight (mg. wet wt.)

 

_____________________________ *
Run x + sd range r P
AUT 1 “-15.0 E 6.0 6.8 —-3o.3 :520 _ 0.06
2 17.2 i 8.1 7.9 - 31.9 -.216 0.46
3 30.2 i 12 3 10.6 - 51.5 -.282 0.54
HET 1 12.8 i 5.3 4.1 - 22.7 -.436 0.12
2 17.9 i 10.4 4.8 - 37.2 -.596 0.02
3 26.7 i 8.2 17.3 - 42.9 -.603 0.11

* Correlation between individual starting wei ht and
growth rate, and significance of r (t-tests§

63

1974) and decreasing fatty acid content (Moore 1975), could
have an adverse effect on consumer growth. In addition,
Hornick et al. (1981) showed that an autumnal decline in
periphyton productivity in.zn1 Appalachian MCuntain stream
was most highly correlated with changes in lighting, flow,
and dissolved organic inputs. In contrast to the
susceptability of AUT to seasonal changes, food for HET was
prepared in the laboratory under identical conditions, and
thus was homogeneous across all 3 runs. Thus, over the
course of the 3 runs, the nutritional quality of the peri-
phyton may have decreased relative to that of the leaves.

The greater ingestion of diatoms by Run 3 AUT nymphs
may also reflect seasonal changes in the periphyton commu-
nity in Sloane Creek. Clumps of the filamentous alga CladO-
phppa partially covered rocks used in AUT Run 1, but were
less prevalent during Run 2 and absent during Run 3.
Although CladOphora fragments were never observed in gut
contents, these clumps may have entrained detritus or
hampered the nymphs' ability to scrape diatoms from the rock
surfaces.

While the general downward trend in growth rates over
the course of the 3 runs could have been caused in part by
the increase in starting weight of nymphs used in successive
runs, the correlation between individual biomass and growth
rate was not significant in most treatment groups (Table
243% Since treatment conditions were identical throughout

the experiment, factors external to the experimental condi-

64

tions, ite. preconditioning 0f the nymphs, may have had a
significant impact on growth rates. This is supported by
the fact that many of the Run 3 nymphs in both treatments
failed to molt (Figure 2J). Several workers have concluded

that growth of Stenonema spp. slows or stops during winter

 

months (Richardson and Tarter 1976; Barton 1980; Kreuger and
Cook 1984; Chapter 1 of the present study). Thus, the
seasonal pattern of growth rates observed in the laboratory
paralleled that commonly observed in the field. This

suggests that growth of S; vicarium is not controlled

 

entirely by the direct effects of temperature, diet, etc. on
metabolism, but may also be controlled endogenously using an
environmental timing cue (qu. temperature or photoperiod).
Such mechanisms for the timing of life history events have
been demonstrated in terrestrial insects (Ricklefs 1973;
Chapman 1982), but unfortunately have never been examined in
aquatic insects (Sweeney 1984). In view of the low Net
Production Efficiency (NPE = Growth/ Assimilation) reported

for Stenonema pulchellum (Trama 1972), slower growth during

 

winter may be necessary for survival of the members of this

genus. The metabolism of Stenonema may be so low during

 

winter that virtually all assimilated energy must be chan—
nelled into maintenance.

Table 2.4 compares the methods and results of this
study with those of other studies that compare growth of
aquatic insects on diatoms vs. leaf detritus (Fuller and
Mackay 1981; Bird and Kaushik 1984; Sweeney and Vannote

1984). In all of these studies, growth rates were greater

65

 

 

 

 

 

 

 

 

 

 

 

.m0>moa .m> omwam so 0009 cpzopw Cw pEoEo>0RaEw & n ommoeosfl w 1 a
Aammav moxouz oer Leased 1 :
.MCDR m Ham mo momma Dem mouma Sysopm ”hogan ucomopm 1 m
.sospu oeoooo "flammac xaeuoox oer seam . m
.mRo>w>Rom on n mz «Armmfiv Opoccm> pom >oc003m 1 a
flame EORBHCV
oo: o.fi «.0 assays: Dado: com aa :.mam onozwmopchm
Ac0p>naaaoav Aomcwwsowmuaomv
ma a.m m.fi amazes: nm< fie mEDHRBOM> mEocosoum
Ac0p>naaaoav Ammo«HH0RoEO£mmv
Ham m.m m.H Hmeopmc cams: ma Newpm>ndm maaopoeonam
>.: mz Ameoumwov OH Aomowpommv
:fi fi.bm m.mm copsuaso Shoxowm mm Hisaoumwc coooano
.ommopocH w omua< mo>mom oopoom Beacomm Aoove moaoomm
r Ae\wv oboe nezooo Amma< who;

 

.mpadmos cam moonpoe mo comwpmaeoo .:.m canoe

 

66

on diatoms than on leaves, but the observed improvement in
growth differs markedly between studies (Table 2.4; %
increase). These discrepancies in results may be explained
by several differences in experimental methods. For example,
Bird and Kaushik (1984) may have imposed food limitation in
their leaf treatment, since only 15 leaf discs (size and
species not given) were provided to groups of 10 "early
instar" mayfly nymphs every 4 days. Leaf rations were much
greater in all other studies. (Grafius and Anderson (1980)
showed that Lepidostoma unicolor larvae increase their
consumption and growth rates as food availability increases.
Therefore, in experiments of this type it is prudent to
supply all food resources in excess to remove the confoun-
ding effects of limitation; i.e. food resources are best
compared by the maximum possible growth rates they produce.
Food limitation does not appear to have been a problem in
any of the other studies, since consumption never exceeded
food supply.

The sources of both algal and detrital food resources
may also partially explain the differences in results. The
ash leaves used in this study were probably of much higher
nutritional quality than leaves used hithe other studies
(Kaushik and Hynes 1971; Peterson and Cummins 1974). McCul-
lough et al. (1979) found that the assimilation efficiency

of Tricorythodes minutus is greater on pure diatom cultures

 

(approximated by Fuller and Mackay 1981, and Sweeney and

Vannote 1984) than on mixed cultures (as in Bird and Kaushik

67

1984, and this study).

Unfortunately; these variations in experimental tech-
nique obscure interspecific differences in resource utiliza-
tion efficiency. Assessment of such interspecific differen-
ces may lead to clearer definition of the importance of
various food resources to stream ecOsystems. While measure-
ments of assimilation efficiency or protein content may be
more accurate (subject to less variance), the influence of
diet on growth (and its consequent effects on other life
history features) is ultimately most important to the
individual, and consequently to the population and communi-
ty. HOwever, it is still unclear how minute variations in
diet influence growth in natural populations. Growth
experiments to date generally use diets that are highly
artificial, since food resources rarely occur in isolation
in nature (e4; diatoms colonize leaf surfacesfi Further
refinement of techniques may produce results that are more

comparable to natural situations.

LITERATURE CITED

LITERATURE CITED

Allen, K.R. 1949. Some aspects of the production and
cropping of fresh waters. Report, Sixth Science
Congress, 1947. Trans. r. Soc. N. Z. 77: 222-228.

Allen, K.R. 1951. The Horokiwi Stream. A study of a trout
pOpulation. Bull. N. Z. Mar. Dept. Fish. 10: 1-238.

Anderson, N.H. and K.W. Cummins. 1979. Influences of diet
on the life histories of aquatic insects. J. Fish.
Res. Ed Can. 36: 335-342.

APHA (American Public Health Association). 1980. Standard
methods for the examination of water and wastewater,
15th edition.

Barton, DJL 1980. Observations on the life histories and
biology of EphemerOptera and Plecoptera in northeastern
Alberta. Aquatic Insects 2: 97-111.

Bednarik, A.F. and W.P. McCafferty. 1979. Biosystematic
revision of the genus Stenonema (Ephemeroptera:
Heptageniidae). Can. Bull. FiSh. Aguat. Sci. 201: 1-73.

 

Benke, A.C. 1979. A modification of the Hynes method of
estimating secondary production with particular
significance for multivoltine populations. Limnol.
Oceanogr. 24: 168-171.

Benke, A. C. 1984. Secondary production of aquatic insects.
In: v.11. Resh.and D.li. Rosenberg (eds.), The ecology
of aquatic insects. Praeger, New York. 625 pp.

Benke, A.C. and J.B. Wallace. 1980. Trophic basis of
production among net-spinning caddisflies in a southern
Appalachian stream. Ecology 61: 108-118.

Benke, A.C., T.C. VanArsdall, Jr., D.M. Gillespie and F.K.
Parrish. 1984. Invertebrate productivity in a sub-
trOpical blackwater river: the importancecufhabitat
and life history. Ecol. Monogr. 54: 25—63.

Berrie, A.I).1976.1Detritus, micro- organisms, and animals
in fresh water. pp. 323— 338 In: J. M. Anderson and A.
McFadyen (eds. ), The role of terrestrial and aquatic
organisms in decomposition processes. Blackwell
Scientific, London. 474 pp.

Bird, G. A. and N. K. Kaushik. 1984. Survival and growth of
early instar Ephemerella subvaria fed various diets.
Hydrobiologia 119: 227- 233.

68

 

 

 

 

69

Boysen-Jensen, P. 1919. Valuation of the Limfjord. I.
Studies on the fish-food in the Limfjord 1909-1917, its
quantity, variation and annual production, Rep. Dan.
biol. Stn 26: 3-44.

Burks, BQD. 1953. The mayflies, or EphemerOptera, of

Burton, T.M., R.W Merritt, R.J. Stout, W.W. Taylor, M.P.
Oemke, M. O'Malley, G. Whelan, D. Cornelius, D. Gesl
and K. Webb. 1984. ELF communications system
ecological monitoring program annual report for aquatic
ecosystems - Tasks 5.8, 5.9, 5.10. Illinois Institute
of Technology Research Institute report no. AE-031.

Cairns, J. Jr. 1977. Quantification of biological integrity.
pp. 171-187 In: ELK. Ballentine and InJ. Guarraia
(eds.), The integrity of water. U.SJ&PhA. Office of
Water and Hazardous Materials, Washington, D.C. 230 pp.

Chapman, D.W. and R.L. Demory. 1963. Seasonal changes in
the food ingested.by aquatic insect larvae and nymphs
in two Oregon streams. Ecology 44: 140-146.

Chapman, RJF. 1982. The insects. Structure and function. 3rd
edition. Harvard Univ. Press, Cambridge, Mass. 919 pp.

Cianciara, S. 1980. Food preference of Cloeon dipterum (LJ
larvae and the dependence of their development on the
type of food. Pol. Arch. Hydrobiol. 27: 143-160.

 

Coleman, M. J. and H. B. N. Hynes. 1970. The life histories of
some Plecoptera and EphemerOptera in a southern Ontario
stream. Can. J. Zool. 48: 1333- 1339.

Cooper, W.E. 1965. Dynamics and production of a natural
population of a fresh-water amphipod, Hyallela azteca.
Ecol. Monogr. 35: 377-94.

 

Cummins, K.W. 1973. Trophic relations of aquatic insects.
Ann. Rev. Ent. 18: 183-206.

Cummins, KLW. and M.J. Klug. 1979. Feeding ecology of
stream invertebrates. Ann. Rev. Ecol. Syst. 10: 147-172.

Dudley, T. and NJL Anderson. 1982. A survey of inverte-
brates associated with wood debris in aquatic habitats.
Melanderia 32: 1-21.

Edmunds, G. F. Jr. 1984. EphemerOptera. In: R. W. Merritt and
KI.W. Cummins (eds.), An introduction to the aquatic
insects of North America. Kendall- Hunt, Dubuque, Iowa.

722 pp.

7O

Elwood, J.W. and D.J. Nelson. 1972. Periphyton production
and grazing rates measured with a P-32 material balance
method. Oikos 23: 295-303.

Elton, 0.8. 1927. Animal Ecology. MacMillan, New York.
209 pp.

Flowers, R.W. and WhL.IHilsenhoff. 1978. Life cycles and
habitats of Wisconsin Heptageniidae (Ephemeroptera).
Hydrobiol. 60: 159-171.

Frike, A.B. and A.B. Thum. 1976. A distance-measuring wheel
for estimating substrate surface area. Trans. r. Soc.
S. Afr. 42: 185-191.

Fuller, R.L. and R.J. Mackay. 1981. Effects of food quality
on the growth of three Hydropsyche species (Trichop-
tera: Hydropsychidae). Can. J. Zool. 59: 1133-1140.

 

Grafius, E. and NJL Anderson. 1980. Population dynamics
and role of two species of Lepidostoma.(Trichoptera:
Lepidostomatidae) in an Oregon coniferous forest
stream. Ecology 61: 808-816.

 

Hall, R.I. 1975. Life history, drift and production rate of
the stream mayfly Tricorythodes atratus McDunnough in
the headwaters of the Mississippi’River. Pth. thesis,
U. Minnesota. 228 pp.

 

Hamilton, AJL 1969. (hi estimating annuaJ. production.
Limnol. Oceanogr. 14: 771-782.

Hanson, B.J., K.W. Cummins, A.S. Cargill II and R.R. Lowry.
1983. Dietary effects on lipid and fatty acid composi-
tion of Clistoronia magnifica (Trichoptera: Limnephil-
idae). Freshwat. Invertebr. Biol. 2: 2-15.

 

Hawkins, C.P. and J.B. Sedell. 1981. Longitudinal and
seasonal changes in functional organization of macro-
invertebrate communities in four Oregon streams.
Ecology 62: 387-397-

Hornick, L.E., J.R. Webster and E.F. Benfield. 1981. Peri-
phyton production in an Appalachian Mountain trout
stream. Am. Midl. Nat. 106: 22-36.

Humpesch, UJL 1981. Effect of temperature on larval growth
of Ecdyonurus dispar (Ephemeroptera: Heptageniidae)
from two English Lakes. Freshwat. Biol. 11: 441-457.

 

Hynes, HgBJL 1970. The ecology of running waters. Univ.
Toronto Press, Toronto. 555 pp.

71

Hynes, HJLN} and MgJ. Coleman. 1968. A simple method of
assessing the annual production of stream benthos.
Limnol. Oceanogr. 13: 569-573.

Kaushik, NIK. and HJLN} Hynes. 1971. The fate of dead
leaves that fall into streams. Arch. Hydrobiol. 68: 405-
515.

Kimerle, ILA. and NQH. Anderson. 1971. Production and
bioenergetic role of the midge Glyptotendipes barbipes
(Staeger) in a waste stabilization lagoon. Limnol.
Oceanogr. 16: 646-659.

Kreuger, CAL and ELF. Cook. 1984. Life cycles, standing
stocks, and drift of some Megaloptera, EphemerOptera,
and Diptera from streams in Minnesota, UJLA. Aquatic
Insects 6: 101-108.

Kreuger, C.C. and F.B. Martin. 1980. Computation of
confidence intervals for the size-frequency (Hynes)
method of estimating secondary production. Limnol.
Oceanogr. 25: 773-777.

Lamberti, G.A. and J.W. Moore. 1984. Aquatic insects as
rimary consumers. In: V.H. Resh and D.M. Rosenberg
Feds). The ecology of aquatic insects. Praeger,

New York. 625 pp.

Lamberti, G.A. and V.H. Resh. 1983. Stream periphyton and
insect herbivores: an experimental study of grazing by
a caddisfly population. Ecology 64: 1124-1135.

Lamberti, G.A. and V.H. Resh. 1985. Comparability of
introduced tiles and natural substrates for sampling

lotic bacteria, algae, and macroinvertebrates.
Freshwat. Biol. 15: 21-30.

Lawson, D.L., M.J. Klug and R.W. Merritt. 1984. The influ-
ence of the physical, chemical and microbiological
characteristics of decomposing leaves on the growth of
the detritivore Tipula abdominalis (Diptera:Tipulidae).
Can. J. Zool. 62: 2339-2343.

 

Leonard, J.W. and F.W. Leonard. 1962. Mayflies of Michigan
trout streams. Cranbrook Inst. Sci., Bloomfield Hills,
Mich. 139 PP.

Lewis, P.A. 1974. Taxonomy and ecology of Stenonema
mayflies (Heptageniidae: EphemerOptera). UKS.E.PAA.
Environmental Monitoring Series Rept. EPA-670/4-74-006.

81 pp.

 

Lindeman, ILL. 1942. The trophic dynamic aspects of ecology.
Ecology 23: 399-418.

72

McCullough, D.A” (LW. Minshall and C.E. Cushing. 1979.
Bioenergetics of a stream "collector" organism,
Tricorythodes minutus (Insecta: EphemerOptera). Limnol.
Oceanogr. 24: 45-58.

McIntire, C.D. 1973. Periphyton dynamics in laboratory
streams: a simulation model and its implications. Ecol.
Monogr. 43: 399-420.

McMahon, RJF” ILD. Hunter and'WJL Russell-Hunter. 1974.
Variation in Aufwuchs at six freshwater habitats in
terms of carbon biomass and of carbon:nitrogen ratio.
Hydrobiologia 45: 391-404.

Menzie, C.A. 1981. Production ecology of Cricotopus
sylvestris (Fabricius) (Diptera: Chironomidaefin a.
shallow estuarine cove. Limnol. Oceanogr. 26: 467-481.

 

Merritt, R.W. and K.W. Cummins (eds.). 1984. An introduc-
tion to the aquatic insects of North America“ Kendall-
Hunt, Dubuque, Iowa. 722 pp.

Minckley, W;L. 1963. The ecology of a spring stream, Doe
Run, Meade County, Kentucky. Wildl. Monogr. 11: 1-124.

Minshall, GdW. 1967. Role of allochthonous detritus in the
trOphic structure of a woodland springbrook community.
Ecology 48: 139-149.

Minshall, (LW. 1978. AutotrOphy in stream ecosystems.
Bioscience 28: 767-771.

Minshall, GuW. 1984. Aquatic insect-substratum relation-
ships. In: V.H. Resh and D.M. Rosenberg (eds.). The
ecology of aquatic insects. Praeger, New York. 625 pp.

Moore, J.W. 1975. Seasonal changes in the proximate and
fatty acid composition of some naturally grown fresh-
water chlorOphytes. J. Phycol. 11: 205-211.

Naiman, R.J. and J.R. Sedell. 1980. Relationships between
metabolic parameters and stream order in Oregon. Can.
J. Fish. aquat. Sci. 37: 834-847.

Odum, H.T. 1957. TrOphic structure and productivity of
Silver Springs, Florida. Ecol. Monogr. 27: 55-112.

Peterson, R.C. and K.W. Cummins. 1974. Leaf processing in a
woodland stream. Freshwat. Biol. 4: 343-368.

Resh, V.H. 1979. Sampling variability and life history
features: basic considerations in the design of aquatic
insect studies. J. Fish. Res. Bd Can. 36: 290-311.

73

Richardson, M.Y. and DAL Tarter. 1976. Life histories of
Stenonema vibarium (Walker) and S. tripunctatum (Banks)
in a west Virginia stream (EphemeropteraJHeptageniidae)
Am. Midl. Nat. 95: 1-9.

 

Rickle6fs, R.E. 1973. Ecology. Chiron Press, Portland, OR.
8 1 pp.

Shapas, T.J. and W.L. Hilsenhoff. 1976. Feeding habits of
Wisconsinfs predominant lotic Plec0ptera, EphemerOptera
and Trichoptera. Great Lakes Ent. 9: 175-188.

Smock, L.A. 1980. Relationships between body size and
biomass of aquatic insects. Freshwat. Biol. 10: 375-
383.

Strahler, AJL 1957. Quantitative analysis of watershed
geomorphology. Am. Geophys. Union Trans. 38: 913-920.

Sutcliffe, D;W” GLR. Carrick and.InG. Willoughby. 1981.
Effects of diet, body size, and temperature on growth
rates in the amphipod Gammarus pulex. Freshwat. Biol.
11: 183-214.

 

Sweeney, B.W. 1984. Factors influencing life history pat-
terns of aquatic insects. ‘In: V.H. Resh and D.M.
Rosenberg (eds.), The ecology of aquatic insects.
Praeger, New York. 625 pp.

Sweeney, B;W. and RnL. Vannote. 1984. Influence of food
quality and temperature on life history characteristics
of the parthenogenetic mayfly, Cloeon triangulifer.
Freshwat. Biol. 14: 621-630.

 

Thienemann, A. 1920. Untersuchungen uber die Biziehungen
zwischen Sauerstoffgehalt des Wassers und der
Zusammensetung der Fauna in norddeutschen Seen. Arch.
Hyrobiol. 12: 1-65.

Trams” FEB. 1972. Transformation of energy by an aquatic
herbivore (Stenonema pulchellum) EphemerOptera. Pol.
Arch. Hydrobiol. 19: 113-121.

 

Uutala, AnJ. 1981. Composition and secondary production of
the chironomid (Diptera) communities in two lakes in
the Adirondack Mountain region, New York. In: R. Singer
(ed.), Effects of acidic precipitation on benthos.
North American Benthological Society, Springfield, IL.
154 pp.

Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell and
CIE. Cushing. 1980. The river continuum concept. Can.
J. Fish. aquat. Sci. 37: 130-137.

74

Wallace, J.B. and A.C. Benke. 1984. Quantification of wood
habitat in subtrOpical coastal plain streams. Can. J.
Fish. aquat. Sci. 41: 1643-1652.

Waters, TQF. 1966. Production rate, population density, and
drift of a stream invertebrate. Ecology 47: 595-604.

Waters, TJL 1977. Secondary production in inland waters.
In: A. McFadyen (edJ, Advances in ecological research,
vol. 10. Academic Press, New York. 177 pp.

Waters, TJA 1983. The P/B dogma. Invited editorial, North
American Benthological Society Newsletter, February.

Weber, C.I. 1973. Recent developments in the measurement of
the response of plankton and periphyton to changes in
their environment. _I_n: G.E. Glass (ed.), Bioassay
techniques and environmental chemistry. Ann Arbor
Science Publ., Ann Arbor, MI. 499 pp.

Wetzel, ILG. 1966. Techniques and problems of primary
productivity measurements in higher aquatic plants and
periphyton. In: C.R. Goldman (ed.), Primary producti-
vity in aqua—trio environments. Univ. Cal. Press,
Berkley, CA. 464 pp.

Willoughby, L.G. and D.W. Sutcliffe. 1976. Experiments on
feeding and growth of the amphipod Gammarus pulex (LJ
related to its distribution in the River Duddon.
Freshwat. Biol. 6: 577-586.

 

Winberg, GAL (ed.L.1971. Methods for the estimation of
aquatic animals. Academic Press, London. 175 pp.

Winterbourne, M.J. 1974. The life histories, trOphic
relations, and production of Stenoperla prassina
(Plecoptera) andeeleatidium sp{TEphemeroptera)ixia
New Zealand river. Freshwat. Biol. 4: 507-524.

 

 

Winterbourne, M., J.S. Rounick and B. Cowie. 1981. Are New
Zealand streams really different? N. Z. J. mar.
Freshwat. Res. 15: 321-328.

MICHIGAN STnTE UNIV. LIBRQRIES
1|11|WW1||111|1|1111|11|||1H|W|”VHIIIVHHNHIWI
31293015936622