.5": \

‘\.u\
‘l\‘>.;k‘:
Wu“. . mu
“-0‘ o. g‘w ..M-
l‘.‘
4' ;’\n. l. , 7‘0
“my nan”
. ' Q

8. L.

WI VI

,.
f

réwir L

A

4,... .
..’..,

N.
;
~ 5'" may
17“

£1!

__'m‘;
r hum;
?.1'

1'

Ae . .J‘
.aln,<' u
y

MM
*p. hm

' SHEA
y, .

.,;..:.‘:.-;
. ("m-In;
’ u‘n- -s ‘ ”5.
.. ""1." at».
x- m. «7.3.x Tum ..EZr.
4‘ A,

. .‘ ‘53.!- gar; m.
.3;-
' K'?

‘

-#.
auvn
~ 6‘

’ ‘ o Q: ’~
. .x . . . .. .
“\"A .u n l m :u
‘v. r . 32...." ‘:mn iii: cunn:€-‘{;D|ml~\t'¢ :
~_~ 4- I...
v u- . J 2? J
' ' X

.o ‘
u
I 1).!

u

w... A;
'.‘.. .., am.

u
“1.:

i .‘n.
‘n Va

.5“:
“1-145,th
.. k: ..

 

 

 

 

z
pr“,
0-... *

.! .. .z‘,
'.'..”- .3...
J

«'.'..7».’
' ..4 fin"
. . 1"
""Hl!l-ll‘
...,.,

”..'.. M
...',c _

, . _ 5’6
. . ".'.-.-.- .
. w. . . ,...~ , . .U: .
.. , . .. ..,
........9.' r,.:,..,,,}.,
'...

1,1" ., many
- (-w .1“ '1 .“. .

.,,. v;:r.'
" 'I‘

'.VIthV'-.M

 

.-.,. ..
0,11vau y'ur

\
:v-lpr’. I" “f. . .‘....

. . '-
v . "'W .9
...4. “3:” . ! ,.,.

1"”:2 ., " c‘..’.,...~:i:..-....
v'o 7'.“'-- H?"

m
a".
CU-O ”no 7r. . 1-; ~u~¢

.. a». a:-

”.1.
n. ,. _ ‘ ,
up fi:‘J'-D" "P. AI4D"
.- . wo'v-"-'~;:-:'n_- r- :‘m': ‘4 .
- J K. 47;-" '1“! . 13'»
""'*":'...- ' .' ‘ “7.. c .
-. _ . .
‘” ' ..'tu- ..L-Lfi‘
v:~':-4~p.m‘ “4:...
' 3'4”."

‘ ::'.'I"I'III:
I'i

 

MICHIGAN STATEU UNI IERV

I IIIIIZIII IIIIII IIIIIIIIIIIIIIIIIIIIIIIIII

00901 7363

II

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

THE POPULATION DYNAMICS AND FOOD HABITS OF
THE CRAYFISH ORCONECTES PROPINQUUS (GIRARD)
IN NORTHERN MICHIGAN STREAMS

 

presented by

Robert Scott Stelzer

has been accepted towards fulfillment
of the requirements for

M o S 0 degree in Zoology

Z 1 ,0! E

Major professor

Date 1/1" /;' 7/

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

__._.—.A _____._. _.—.—, .7 , 7* 7. 7

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before due due.

DATE DUE DATE DUE DATE DUE

I

   
 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I

MSU in An Affirmative Action/Equal Opportunity inetitution
GWWMI

 

 

 

THE POPULATION DYNAMICS AND FOOD HABITS OF
THE CRAYFISH ORCONECTES PROPINOUUS (GIRARD)
IN NORTHERN MICHIGAN STREAMS

BY

Robert Scott Stelzer

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

MASTER OF SCIENCE

Zoology

1991

ABSTRACT

THE POPULATION DYNAMICS AND FOOD HABITS OF
THE CRAYFISH ORCONECTES PROPINOUUS (GIRARD)
IN NORTHERN MICHIGAN STREAMS

BY

Robert Scott Stelzer

Two populations of Q. propinquus were studied from the
Ford River (hard water) and Baraga Creek (soft water) in
northern Michigan. Mean crayfish abundance and total net
production were greater in the Ford River (9.3 one year old
individuals/m2 vs 3.0 one year old individuals/m2 in 1989,
8.9 g/m2 vs. 5.8 g/m2 during 7/14/89—7/14/90). Individual
growth was greater in Baraga Creek. Early instar Ford River
crayfish experienced considerable mortality from the pleopod
egg stage through their first three weeks of independence.
Some early instar crayfish appeared to change the kind of
microhabitat they used as they matured. Diet studies
revealed Ford River Orconectes propinquus ate predominantly
diatoms, filamentous green algae, and insects, with insects
more abundant in juveniles. The diatoms Navicula and
Gomohonema achieved different relative abundances when each
genus was compared between juvenile and adult crayfish. Some
evidence suggests that Q. propinquus was selecting

Chironomidae from the environment.

This Master's Thesis is dedicated to my parents
Robert William Stelzer and Dolores Jane Stelzer,
whose love and nurturing has allowed me to
pursue a career in biology.

iii

ACKNOWLEDGMENTS

There are several people that helped with this work. I
especially thank Dr. Thomas Burton, Michigan State
University, for his guidance, helpful criticism, and general
willingness to discuss the progress of my research at any
time. He also introduced me to the streams of northern
Michigan and first suggested a hard water/soft water
comparison study.. Dr. Darrell King and Dr. Donald Hall, both
of Michigan State University, provided helpful insight at
several key junctures during the study. Dr. Hall was the

impetus for my study of early instar Q. propinguus. Ms.

 

Susan Eggert, Ms. Deborah Repert, and Ms. Melissa Treml
provided much appreciated assistance in the field. Dennis
Mullen allowed me the use of his block nets for the duration
of my research. I kindly thank Dr. Horton H. Hobbs Jr.,
Smithsonian Institution, for help with species
identification. Mrs. Jennifer Molloy, Michigan State
University, contributed helpful guidance with regard to
diatom identification. Lastly, Dr. Charles Cress, Michigan
State University, provided useful suggestions regarding

statistical analysis.

iv

PREFACE

Due to the strong differences among the kind of questions
asked and addressed in this study, the thesis was divided
into two chapters: 1) Population studies of Q. propinquus in
two northern Michigan streams. 2) Diet Analysis of Q.
propinquus in a northern Michigan stream. These two chapters
each have separate introductions and were attempted to be
made self-sufficient. Themes from each chapter are addressed

in a common conclusion at the end of this report.

TABLE OF CONTENTS

List of Tables p. vii
List of Figures p. ix
Chapter 1: Population Studies of Q. propinouus in p. 1

 

Two Northern Michigan Streams

Chapter 2: Diet Analysis of Q. Dropinquus in a p. 67
Northern Michigan Stream

Conclusions and Recommendations p. 88

List of References p. 92

vi

10.

11.

12.

13.

LIST OF TABLES

Summer and fall 1989 chemical and physical data for
matched sites (p.5).

Mean depth and velocity of latitudinal habitats from the
Ford River during early summer 1990 (p.10).

Mean individual dry weights, mean densities, and
capture totals for Q. propinquus in Baraga Creek between
7/7/89-7/14/90 (p.22).

Mean densities of Q. propinquus early instars from
different habitats in the Ford River (p.24).

Mean individual dry weights, mean densities, and
capture totals for Q. propinquus in the Ford River
between 7/1/89-8/31/90 (p.28).

Net production of Q. propinquus in the Ford River between
7/1/89 and 8/31/90 (p.31).

Net production of Q. propinquus in Baraga Creek between
7/7/89 and 7/14/90 (p.32).

Annual life table for female Q. propinquus from the Ford
River (p.38).

 

Results of Chi-Square Goodness of Fit Tests comparing
numbers of adult male and female Q. propinouus (p.41).

 

AKOVA table for a 3-factor analysis of variance on
densities of early instar Q. propinquus from the Ford
River (p.45).

Ford River water temperature data at the end of the YOY
growing season and beginning of the yl growing season for
1988 and 1989 Q. propinquus cohorts (p.64).

Relative abundances of small and medium size diatom
genera in stomachs of Q. propinquus from the Ford River
(p.78).

Relative abundances of large size diatom genera in the
stomachs Q. propinquus from the Ford River (p.79).

 

vii

14.

15.

ANOVA tables for a 2-factor analysis of variance on the
relative abundance of diatom genera in Q. propinquus
stomachs (p.80-81).

 

Relative abundance of various insect groups in the Ford
River in 1983 (p.85).

viii

10.

ll.

12.

13.

14.

LIST OF FIGURES

Carapace length-dry weight regressions of female and male
Q. propinduus from the Ford River (p.14).

Carapace length-dry weight regressions of female and male
Q. propinquus from Baraga Creek (p.15).

Length-frequency histograms of male Q. propinquus from
the Ford River showing yl, y2, and YOY size classes
(p.16).

An example of Harding’s graphical method used to separate
cohorts in a size frequency histogram (p.18).

Density of yl Q. propinquus from the Ford River and
Baraga Creek (p.20).

Density of YOY Q. propinquus from the Ford River and
Baraga Creek (p.21).

Change in mean density of Q. propinquus early
instars from the Ford River (p.25).

Change in mean individual size of Q. propinquus in 1988
and 1989 cohorts from the Ford River and Baraga Creek
(p.26).

 

Change in individual weight of Q. propinquus YOY from the
Ford River and Baraga Creek (p.29).

 

Carapace length-pleopod egg regression for Q. propinguus
from the Ford River on 5/6/90 (p.34).

 

Regression of mean pleopod egg diameter on carapace
length of Q.propinquus from the Ford River on 5/6/90
(p.35).

Regression of total pleopod egg volume on carapace length
of Q. propinquus from the Ford River on 5/6/90 (p.36).

Size comparison of male and female YOY Q. propinquus from
the Ford River (p.42).

 

Size comparison of male and female y1 & y2 Q. propinquus
from the Ford River (p.43).

 

ix

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Mean depletion densities of Ford Q. propinduus early
instars from run-T, riffle, and run-H habitats beginning
6/29/90 (p.46).

Mean depletion densities of Ford Q. propinduus early
instars from middle and side habitats beginning 6/29/90
(p.48). ‘

Relationship between water depth and density of early
instar Q. propinquus from the Ford River (p.49).

 

Relationship between water velocity and depletion density
of early instar Q. propinquus from the Ford River (p.50).

Revised survivorship curve of Q. propinquus early instars
from the Ford River in 1990 (p.57)

Ford River Q. propinquus captured per kicking pass
regressed on cumulative capture (p.62).

 

The percentage of Q. propinquus from the Ford River with
various food items (p.71).

Mean number of insects in Q. propinquus from the Ford
River (p.73).

Mean relative abundance of insect types in the stomachs
of Q. propinquus from the Ford River (p.74).

Mean relative abundance of diatom genera in Q. propinquus
from the Ford River (p.75).

Relative abundance of small, medium, and large size
diatom genera in Q. propinquus from the Ford River
(p.76).

Chapter 1: Population studies of Q. propinquus
in two northern Michigan streams

INTRODUCTION

Crayfish are often very numerous in freshwater ecosystems
and can dominate the invertebrate biomass of lakes (Momot et
a1 1978) and streams (Webster and Patten 1979, Huryn and
Wallace 1987). Total net production of crayfish in streams
however, is often less than that of insects (Huryn and
Wallace 1987), which usually have faster turnover times.
Nevertheless, crayfish often play important roles in
communities. They occupy multiple trophic levels feeding as
primary consumers, secondary consumers, and detritivores
(Vannote 1963, Momot et al 1978). Their role as detritivores
in streams often involves the breakdown of allochthonous
material, such as leaves, which contributes to nutrient
cycling (Momot et a1 1978, Webster and Patten 1979, Huryn and
Wallace 1987). Crayfish are significant prey items for fish
in some locations, in some cases representing the dominant
item in their diet (Vannote 1963, Fedoruk 1966, Stein 1977).
Several comparison studies have been done examining crayfish
growth and other population factors between contiguous lakes
(Goldman and Rundquist 1977, Momot and Gowing 1977a, Momot
and Gowing 1977b, Jones and Momot 1981, Momot and Gowing
1983, France 1983). Although Corey (1988) compared the
growth of two Q. propinquus populations from different
rivers, I am not aware of any study comparing growth,
abundance and production between crayfish populations from

different streams. Numerous life history and growth studies

however, have been conducted on Orconectes propinduus
(Creaser 1934a, Van Deventer 1937, Bovbjerg 1952, Vannote
1963, Fielder 1972, Capelli 1975, Lodge et al 1986, Corey
1987a, Corey 1987b, Corey 1988).

Calcium is a large component of crayfish exoskeletons and
is important in their metabolism and physiology (Robertson
1941). Growth of crayfish increased when individuals were
reared in experimental ponds with increased calcium (de la
Bretonne et al 1969). Although Capelli (1975) failed to find
a statistically significant correlation (r =.450) between
calcium concentration and crayfish abundance in Wisconsin
lakes with crayfish, low calcium concentrations may have
restricted crayfish from certain lakes. Albeit improbable,
the potential may exist for calcium to influence both the
growth and abundance of crayfish in streams.

An important dimension of a crayfish population is the
young-of-the-year (YOY) cohort. Mortality is often quite
high in this stage of a crayfish’s life. The survivorship of
the YOY cohort can contribute to the increase or demise of an
entire crayfish population. Besides Capelli’s (1975) work,
data on the survivorship of early crayfish instars is lacking
and no researcher to my knowledge has attempted a detailed
analysis of this sort for a crayfish population. In addition
to learning about the survival of early crayfish instars, it
is also important to determine how they use the habitat
available to them and if their use of microhabitats changes
as they grow. Besides Rabeni (1985) no investigator to my
knowledge has performed a detailed habitat analysis of early

instar crayfish.

‘ A

l! I)

'4

'u

The objectives of this work were:

(i) to compare the growth, abundance, and net production or
Q. propinquus in the hard water Ford River and the soft
water Baraga Creek in order to determine if these factors
are correlated with water hardness within the two
systems. Growth, abundance, and net production of
Q. Dropinquus was hypothesized to be greater in the Ford
River owing to its higher water hardness.

(ii) to determine the stability of the Ford River population
over the course of two summers using life table analysis
and year to year comparisons of growth and abundance.

(iii) to measure the survivorship of the YOY instars of
Q. propinquus in a detailed manner during their first
month of independence from the mother.

(iv) to analyze the habitat of Q. propinquus early instars
and to determine if their habitat preference changes
during their first month of growth, to gain further
insight into the general ecology of crayfish at this

life stage.

STUDY SITES

The streams studied were in the Lake Michigan drainage of
the upper peninsula of Michigan. One study site each was
chosen on the Ford River and Baraga Creek. The 800 m site on
the Ford River was located in Dickinson County (Sec 18, T. 43
N., R. 29 W.). The stream averaged 5.3 m in width and 17.5
cm in depth at this location. Speckled alder (Alnus ruqosa),

balsam poplar (Populus balsamifera) and quaking aspen

(Populus tremuloides) dominated the riparian vegetation, and
the stream catchment was largely early successional hardwood
forest (Burton 1991). The site’s substrate was predominantly
cobble. The site on Baraga Creek was also I 800 m long and
was located in Marquette County (Sec 15, T. 49 N., R. 30 W).
The streams average width (7.4 m) and depth (17.8 cm) were
similar to the Ford River’s. Baraga Creek drained a small
lake, and was bordered by dense stands of shrubs, largely
Algus rugosa. The site was less shaded than the Ford River
site however, as most of the taller trees were set back 4-6 m
from the stream bed. The Baraga stream catchment, like the
Ford's, was also mainly forested but was made up of a larger
percentage of conifers. Cobble was also the dominant
substrate in Baraga Creek. Both sites were virtually void
of aquatic macrophytes but did contain an abundance of
periphyton. Baraga Creek had water that was much darker in
color than the Ford River, suggesting that organic acids were
more abundant in Baraga Creek (Burton 1991). The chemical
and physical characteristics of the two streams were quite
similar (Table 1). The streams differed considerably
however, in total water hardness (and conductivity). The
Ford River had a mean water hardness of 207 ppm compared to
25 ppm CaCO3 in Baraga Creek. The two sites will be referred
to as the Ford River or Baraga Creek for the rest of this
paper unless otherwise stated, with the understanding that
any conclusions drawn from this study are based on
information collected at the individual sites and cannot

necessarily be extrapolated to other areas of each stream.

 

Amvmmo.ofimhm.o Aevm.oHo.m AmVN.NHo.mH Amvafimm Amvfi.ofim.b Amvaflmm amoumm

Aevmbo.onom.o Amvm.on.m AmvH.NHw.mH Amvmafimom AmVH.on.m Amvofifihom boom
863mg: 385 loot 23253 Amoomo 53

moumnomflo NO .mmflo .QEwu 0mm >ufl>fiuodocoo mm mmwcoumm muflm

 

mum mwSHm>

.mwmmnucmLMQ Ca cw>flo ma gonads deEmm .m.m H mcmoE

.mmuflm omzooma How comp Hmoflmmnm pom HMOHEmno mama HHMw can umEEdm La maome

 

5‘
V».

C“:
Uv~
.

l<:_)

 

G .

I".
y.

‘ts

\

'(J

U)
11'

I)

9
r

’r.)
«r:

r-h

THE STUDY ORGANISM

Q. propinquus is a member of the subfamily Cambarinae in

 

the family Astacidae. It is relatively small compared to
most North American crayfish. Its range extends west and
south into northwest Missouri and southern Illinois, east
into New York, and north into much of Ontario (Crocker and

O. propinquus is clearly the dominant crayfish

Barr 1968).
species in Baraga Creek and the Ford River. Although

Q. virilis is also present, less than 10 individuals were
encountered in both streams during the two year study.
References in this report to crayfish in the Ford River and

Baraga Creek imply Q. prooinguus exclusively.

METHODS

Population sampling

Populations of Q. propinouus were sampled at two week

 

intervals throughout the summer beginning in June 1989 for
both Ford and Baraga populations. After fall and spring
sampling trips, sampling was terminated in June 1990 in
Baraga Creek and continued through August 1990 in the Ford
River. A trip was cancelled in December 1989 due to ice
cover on both streams. Five different subsites were utilized
per sampling period at each stream, and to avoid harvesting
from recently disturbed habitat, no subsite was sampled twice

within a six month period. Subsites were marked with dated

flags to indicate that sampling had occurred recently. At
both the Ford River and Baraga Creek, crayfish 2 7 mm in
carapace length were collected by two persons kick seining
abreast into a downstream block net (0.25 cm2 mesh size).
The area sampled per subsite ranged from 2 to 13 m2. Larger
areas were chosen at Baraga Creek in order to harvest a
statistically sufficient number of individuals for mean size
determination. A removal collection method (Zippen 1958) was
used which involved making three kicking passes at each
subsite and collecting crayfish from the block net between
trials. Assuming that the number of individuals captured on
any given pass was a linear inverse function of the
cumulative total, the X-intercept gave the estimated total
number of crayfish. I will from here on refer to this
X-intercept total as the depletion total (or depletion
density when the area sampled is considered). Carapace
length of individual crayfish (the distance from the tip of
the rostrum to the posterior edge of the carapace along the
dorsal midline) was measured to the nearest millimeter using
vernier calipers (Total length, which is approximately twice
carapace length, is occasionally used in this report). The
total number of individuals captured per kick seining trial
was noted as well as any evidence of recent molting and the
sex of each individual. Male and female crayfish were
readily distinguishable. The first and second pleopods are
modified as intromittent organs in males, while these
pleopods in females are much like the third, fourth, and
fifth pairs of pleopods. Females also possess a prominent

annulus ventralis, or seminal receptacle, on their ventral

surface between their fourth and fifth periopods. In
addition, males typically have larger chelae and a narrower
abdomen than females. Year classes were distinguished by
applying Harding’s (1949) graphical method to length
frequency histograms constructed for each sampling date. The
cohorts appeared to meet the method’s assumptions of normal
distribution. Separate histograms were made for male and
females as the size range of a given year class was sometimes
different for each sex.

Early instar Q. propinquus (< 7 mm carapace length) were
sampled in the Ford River in 1990 from the time they were
recruited into the population (late June) until they could be
sampled with the adults using the block net method described
(late July). Early instar crayfish were sampled twice a week
at twenty-five randomly chosen subsites, each with a 0.5 m2
area, using hand held kick seines made with window screen
(0.023 cm2 mesh size). No subsite was sampled twice during
the entire early instar sampling period. Three kicking
passes were made at each of the 25 subsites as part of
Zippen’s (1958) removal collection method. Early instar
crayfish were collected from the kick screen between kicking
passes using fine forceps. Early instar density was
determined using Zippen's (1958) method in the manner applied
to adult crayfish. The rationale for using kick screens
instead of attempting to sample these early instars with the
block net was that the kick screen's smaller mesh size
prevented escape and this method reduced the amount of area

required to be sorted for early instars. The following

information was recorded at each subsite: (1) number of
crayfish obtained in each kicking pass, (2) individual
carapace length (some individuals were alternatively measured
in the laboratory), (3) stream depth, (4) current velocity at
0.6 x depth, using a Gurley Pygmy meter, (5) latitudinal
position of the stream (side or middle; the center 3/5 stream
area was designated as middle, and the 1/5 area adjacent to
each bank, side), and (6) longitudinal position (riffle,
run-T, or run-H). Riffles were defined as those areas of the
stream where surface turbulence was visible. Areas between
adjacent riffles were designated as runs. Runs were
subdivided into upper and lower sections as follows. The
area zero to five meters immediately downstream from a riffle
was considered the head portion of the run (designated as
run-H). The portion of the run greater than five meters
downstream from a riffle was regarded as the tail portion
(run-T). Runs were subdivided on the basis that an unusually
high concentration of early instar crayfish were collected in
areas immediately downstream from riffles (zero to five
meters) early in the sampling regime. The average depth and
velocity of these habitats during the early instar sampling

period is given in Table 2.

Densitv

The mean density of individuals in the 800 m study reach of
each stream, large enough to be sampled by the block net
method, was obtained as follows: the total number of

Q. prooinquus was estimated from each subsite using the

10

Table 2. Mean depth and velocity of latitudinal habitats
from the Ford River during early summer 1990.
Standard errors are given in parentheses.

 

 

Habitat Depth (cm) Velocity (m/s)
riffle 12.0 (0.5) 0.305 (0.022)
run-H 18.6 (1.0) 0.163 (0.013)

run-T 21.1 (1.1) 0.112 (0.009)

11

removal collection method (Zippen 1958). The densities were
then partitioned into year class densities based on the
numerical ratio of individuals of different year classes at
each subsite. The average of the five subsite partitioned
densities for each year class was used in equations 2 and 3.
These are the densities used throughout this paper unless
otherwise stated. The overall density of early instar YOY
individuals sampled with kick screens was determined by
weighting the mean densities from the longitudinal habitats
based on the relative abundance of those habitats over the

entire Ford River site.

Growth and Production

In order to calculate growth and production, carapace
length was converted to dry weight using length-weight
regressions for each sex. Dry weights were obtained by
weighing individuals after oven drying them for 24 hours at
60 C. Growth in each cohort was measured using the

instantaneous growth equation (Ricker 1958):

G = ln -2 - ln'w (eq 1)

where G = instantaneous growth rate per day

a: = average dry weight of an individual from a given

cohort at time 2.

12

average dry weight of an individual from a given

W1 =
cohort at time 1.
t = number of days between time 1 & 2.

Net production of each cohort was obtained using the
following equations based on Ricker (1946) and Allen’s (1949)
work. For the rest of this paper net production will be

referred to simply as production.

B1 = D1 * W1 (eq 2)

(eq 4)

 

ml
||
ltIJ
+
if?

P = G *'B * t (eq 5)

where Bl = Grams of dry crayfish biomass per square meter
for each cohort at time 1.
82 = Grams of dry crayfish biomass per square meter for
each cohort at time 2.

number of individuals per cohort/m2 at time 1.

U
H
(I

D2 = number of individuals per cohort/m2 at time 2.
B = mean grams of dry crayfish biomass per square

meter for each cohort between sampling dates.

13

P = production of each cohort between sampling dates

in grams/m2.

The production/biomass or the P/B ratio, an approximation of
the biomass turnover rate (Waters 1969), was calculated by
dividing total annual production by the grand mean of

weighted mean biomasses determined throughout the year.

RESULTS

Length-weight regressions

Length-weight regressions constructed separately for male
and female Q. propinquus in the Ford River and Baraga Creek
are given in Figures 1-2. A logarithmic function described
the relationship between carapace length and dry weight in
each of the crayfish subgroups considered. Ford regressions
(R2 > 0.98) fit the empirical data slightly better than the
Baraga regressions (R2 > 0.96). Male crayfish tended to have
a slightly larger dry weight than female crayfish of the same
20 mm carapace length (Figures 1-2). In addition, Ford
Q. propinquus weighed more than Baraga individuals of the

same sex at 20 mm (Figures 1-2).
Size-frequency histograms
Representative length-frequency histograms for male

Q. propinouus in June and July 1990 are given in Figure 3.

The YOY size class, first appearing in the block net on

14

FEMALE

 

v = 7.265E-6 . x3-8“ R2 = 0.981

DRY WIGHT (G)

 

 

 

 

 

 

 

 

 

 

o .
0 3O
CARAPACE LENGTH (MM)
MALE
4 -
‘Y = 4.0648-6 . x4-055 R2 = 0.986
A
El
3 3~
{D
E 2‘
>1
3
A 1‘
o , .
0 30 4O
CARAPACE LENGTH (MM)
Figure 1. Carapace length-dry weight regressions of
female 8: male Q. propinquus from the Ford River. 20 mm
carapace length female = 0.734 g; 20 mm c.l. male

= 0.793 g.

15

FEMALE

 

"Y = 3.0733-5 . x3-357 R2 = 0.957 a

DRY WEIGHT (G)
M

 

 

 

 

 

 

 

 

o . . .
0 10 2O 30 4O
CARAPACE LENGTH (MM)
MALE
4
‘ Y = 9.6128-6 . x3-753 R2 = 0.960
A
8 39
EB :0
Q o
E 2‘
9.. ‘ a3 E
m .. .
A ‘ EE'
0 if u r 1 .
10 20 30 4O
CARAPACE LENGTH (MM)
Figure 2. Carapace length-dry weight regressions of

female and male Q. propinquus from Baraga Creek. 20
mm c.l. female = 0.716 g; 20 mm c.1. male = 0.734 g.

16

 

 
 
 

 

 

 

 

 

 

 

 

 

E 6111/90
E 20
0
I ‘0‘
O
I41618202224262830323436384042444648505254
:srmrnmuM(m
6122/90
.3 20
E .
3 10-4
E ° 9

I4161820222426283032343638404244464850525456
IMMILM(II)

717/90

 

20

10‘

 

 

 

14161820222426283032343638404244464850525456586062
EMMEM(U)

7/27I90

  
  

20

IO

 

mu 0' cums!

film——

141618202224262830323436384042444648505254
RWMILM(II)

Figure 3. Length-frequency histograms of male 9. W
from the Ford River shoving yl, 72, and YOY size classes.

0

 

 

17

7/27/90, was easily recognized as they were clearly smaller
than year 1 (yl) individuals. The separation of y1 and year
2 (y2) individuals was usually not clearly demarcated. In
some cases however, Harding’s (1949) method enabled yl and y2
size classes to be distinguished (Figure 4). Still, because
the age of larger relatively rare individuals was not known
with certainty, estimation of the abundance of the y2 class

was not considered definitive.

Age estimation

The mean age of a year class at a given time of the year
was determined based on the assumptions that the members of
each cohort hatched from eggs synchronously on a mean date of
June 17, and that this date was stable from year to year.
June 17 was set as an approximate hatching date based on the
fact that ovigerous females from the Ford River were carrying
exclusively unhatched eggs on 6/11/90 while the majority of

ovigerous females were carrying hatched young on 6/22/90.

Abundance

Some variation in density of y1 and y2 Q. propinquus from
the Ford River between 1989 and 1990 was suggested (10.7/m2
vs 7.3/m2) although a t-test showed densities were not
significantly different between the two years (P = 0.10).
Young-of—the-year densities from the Ford River were more
similar between 1989 and 1990 (10.8/m2 vs 10.4/m2, t-test

P = 0.88) than y1 and y2 densities between the two years

18

4t 56: 40:

 

manor?“ 7235.; so 56 52 :8 « «a
tour. m (as)

Figure 4. An example of Harding's graphical method
used to separate cohorts in a size frequency
histogram. The graph shows the separation of YOY (40%
of total), Y1 (56%), and Y2 (4%) size classes of male
Q. propinquus from the Ford River on 8/4/89. The
inflection points on the nonlinear cumulative
frequency graph for all size classes indicates the
separation points between individual size classes.

19

(Figures 5-6). In 1989, density of Ford yl’s showed a
continuous and rapid decline from the beginning of July
through October, with an average density of 9.3/m2 during the
period (Figure 5). In 1990, Ford y1 densities remained
relatively stable from the beginning of May through August.
Densities of crayfish 1.15-1.2 years old from each year were
similar, with overlapping mean t SE intervals. When compared
over the same age span, the Ford YOY densities from each year
showed a comparable pattern of variation, although the
sizable standard errors particularly in 1990 should be noted
(Figure 6). Densities of Q. propinquus from Baraga Creek
during 1989 were considerably lower than Ford values from
either year when like size classes were compared ( t-tests

P < 0.02, Figures 5-6). The average yl density was 3.0/m2
and mean YOY density 1.2/m2. The exceptionally low density
of 0.1 YOY/m2 on 7/30/90 (Table 3) was not included in the
summer average, as this was the first time YOY appeared in
the block net at Baraga Creek and a considerable number of
individuals may have been overlooked.

Crayfish large enough to be collected in the block net (3 7
mm carapace length) were generally homogenously distributed
in each stream. Although the standard errors in Figures 5-6
suggest a sometimes clumped distribution, Q.prooinguus
greater than 7 mm in carapace length did not seem to prefer
particular areas within the study stretches. Densities of
both YOY individuals, and yls and y28 pooled together were
not significantly different between runs and riffles in each

stream (t-tests, P > 0.436).

20

.xmmuu mmmumm n m £93m week .I. m .m.m H
mumoflocfl mumn uouum .xmwuo mmmumm paw uw>flm whom
may Eoum masocflaoua ..q in mo huflmcmn .m musmflm

 

Ammefimwv m0<

.94 mm.— N._ _._ O; 0.0

 

 

mam: m III—Ill

82 a ....9.:
89 l .Ial

 

 

[OF

 

 

 

 

om

 

H191 Him

1111.211 HHYIIOS

21

mama m
omma h
mmma .m

.m.m H wumoflonfi mumn
wouum .xwmuo wmmumm paw um>flm Ouom may
EOum mssdafiaoud ..0. NO» NO huflmcwa .m 9.398

 

Amy/NOV m0<

o: 09 00_ on 00 CV om
- h -

 

lllll

.I+|l IO—

 

 

IT

 

 

 

 

 

 

 

 

om

 

H391 33m

KHZ]! ZHHOS

22

.UOUMUHUCH OUMU OSU MOM mOSHm> EOCMCS O>HUHmOQ £UH3 wHHOO OUMOHUCH mmCHH COCWMU -A

.mqu mafia co m u z 0:0 .o mmm.m

uflmflmz cme .NE\mm> m.o wumz mumnu -9

 

 

 

 

hm 0.0 HmN.N mm m.H mam.o --- --- --- om\vfi\b
Hm N.H boo.fi hm m.o mmH.o «om\m\m
m N.o mvm.m mm H.H omo.H mm v.0 omfi.o mm\mm\0H
NH m.o Hem.m Hm m.m mom.o be ¢.N HHH.o mm\m\m
m N.o hmm.m mHH b.m mHH.H ma v.0 omo.o mm\ma\m
m m.o mmm.m ooa o.m meo.fi m H.o omo.o mm\OM\b
e H.o orm.m mHH >.m va.H --- --- --- mm\b\h
z e 3 z e 3 z ime\nve lava
..... m»----- -----Hs----- -----wo»----- memo
.om\¢a\b-mm\b\b cwaDwn xmmuo amoumm Ce masocflQOuQ .d
qu sz mamuou quDQmo cam .AUV mmfluflmcwo some .A3V moQOHwS muc H05©H>flocfl c002 .m magma

23

Early instar abundance

Early instar densities from longitudinal habitats
were weighted according to the relative abundance of a given
longitudinal habitat over the entire area of the Ford River
site because densities from some of the longitudinal habitats
were significantly different for some of the sampling dates
(Table 4). Densities from latitudinal habitats were not used
in the calculation of overall early instar density because
they were not significantly different for any of the sampling
dates (Table 4). Initial density of Ford YOY crayfish in
1990 was determined by the number of pleopod eggs/m2 on
5/6/90. Considerable mortality occurred from the egg stage
on 5/6/90 up until the time the vast majority of instars were
independant of their mothers, 54 days later (Figure 7).
Early instar densities stabilized for the next 3 weeks (54-76
days in Figure 7). An apparent increase in density of the
cohort was realized when the YOY began being sampled with
larger crayfish as part of the block net method (2 82 days in
figure 7). The distribution of Ford Q. propinquus early
instars in the various microhabitats will be discussed in

more detail later.
Growth
Crayfish growth was not continuous over time but

incremental, since they must molt to grow. Thus, smooth

growth curves were deceiving (Figure 8). Year 1 and y2

24

.Amo.o
mcflm: ucmuwwwflp haucmoflwflcmflm Doc wuw3 umuuma mEMm wzu Luez mono pom

mndawv wocwumMWHc ocmoflwflcoflm ommma cou00u0ud m.um:mflm

AcofluHMOQ

HmcHUSDHuma no HmcflojbflmcoHv muommumo umbeomc wEmm gnu Eouu mcmoE 03o >cm -A

 

 

 

 

ohm.o H bH.H mmm.o H mH.m nmm.o H mm.o nmm.o H om.o woo.o H oo.m om\am\n
0mm.o H om.o wom.o H wo.a QoH.o H mm.o 90mm.o H om.o wom.o H HH.N om\mfi\>
8mm.o H mm.fi cam.o H bm.a woo.H H mm.m woa.o H mm.o web.o H om.a om\ma\h
0mm.o H ma.m woe.o H mH.H 0mm.o H om.a who.o H oo.H amm.o H oo.m om\0fi\b
who.a H oo.N omm.o H am.H wom.o H mm.a wom.o H om.o omm.o H hm.m om\m\b
0mm.fi H mm.N mme.o H mH.H wem.a H mv.m wav.o H hm.o who.a H hm.m om\v\b
MmH.m H mm.e mum.fi H mm.v omo.m H mq.m mmm.m H mm.v amm.o H hm.a om\mm\m
moflm waooflz m-csm mHMMHm B-c5m mono
coauwmom chwozuwumq coauflmom HMGHUSDHOGOA
.m.m H mcmwe mum wwSHm> .pw>flm
Upon 0:» CH mumuflnmn ucmumMMHo EOHM mucumcfl haumw monocflo0u0.d mo mmfluflmcmo cow: .8 maflwfi

NUMBER PER SQUARE METER

25

 

 

 

 

 

 

lfl'Tfl'lfI‘T'l

65 68 73 76 82 87 103 117

 

DAYS SINCE 5/6/9 0

Figure 7. Change in mean density of Q. propinquus early
instars from the Ford River in 1990. Data point at zero
days represents mean pleopod eggs/m2. Data points at
54-76 days represent the grand mean of mean longitudinal
densities (Figure 16) weighted for habitat relative
abundance at the Ford site. Data points at 82-117 days
represent mean densities obtained with the block net
method. Error bars indicate i S.E. of means unweighted
for habitat relative abundance on a given date.

26

uuonoo
uuonoo
unonoo

unonoo

mmm
mmm
mmm
mwh

.m.m H wumoHpGH mumn

wouum .xwmuo momuwm cam H0553” whom 93
60.5 muuonoo mmmH w mmma CH 3 .d
mo wNHm HMSUHKQUGH Game CH 00:98 .m musmfim

Ammmfimwv m0<

 

 

 

 

 

 

(9) LEBIIA LEO:

27

Q. propinquus from the Ford River and Baraga Creek grew
during the molting period of spring and early summer and
stayed at a fixed size for the remainder of the year (Tables
3 & 5). Ford juveniles grew steadily during their first
summer and fall and continued growth to maturity in their
second summer (Figure 8). No appreciable growth occurred in
YOY, yl, or y2 age classes from 10/28/89 to 5/6/90 in either
population (t-tests P > 0.05, Tables 3 & 5). As with
density, year-to-year differences in size (and by implication
growth) were apparent among yl individuals in the Ford River.
Mean dry weight of 1990 yl’s was 0.472 g compared to 0.729 g
for 1989 yl's over the same interval (Figure 8). Growth
comparisons of the two size classes could not be made because
sampling began in the summer of 1989 after a sizeable amount
of the growth in 1989 y1 individuals had already occurred.
The mean size of y2 individuals in both populations was
highly variable between samplings, probably a function of the
small numbers collected (Tables 3 & 5). Comparisons between
growth of YOY classes from the Ford River between years was
Possible. Individuals in 1990 grew faster over a common
interval (G = 0.0538/day) than 1989 individuals

(G = 0.0419/day) (Figure 9). Baraga 1989 and 1990 y1

Q. propinquus achieved a larger size (1989 mean weight =
1.077 9) than Ford individuals from corresponding years and
Y1 1990 spring growth was clearly higher in Baraga Creek
(Figure 8). Baraga YOY were also larger than Ford 1989 YOY
and showed greater growth during their first year (Figure 9).
Between 7/30-10/28/89 Baraga YOY growth rate was 0.0199/day

compared to the interpolated Ford YOY growth rate of

.pmumoflocfl mono on» uOm mosam> czocxc: m>HuHmom Suez mHHmo wumoflocfl mocHH cosmmo -A
.mump menu Co .v u z w .m moH.N n ucmflw3 cme .NE\mm> N.o muwz mums» -:

 

28

 

m.o mmH.H we m.m mme.o NNH m.oa mmo.o om\am\m
N.H ham.fi or m.m Hme.o omfi m.mH moo.o om\>H\m
>.H ome.fi Hm e.m ewe.o em m.> Hmo.o om\fi\m
>.H nmm.a mo m.o mme.o oefl o.eH omo.o om\>m\>
--- --- --- --- --- Ne o.qH bmoo.o om\Hm\h
--- --- --- --- --- mm e.oa mHHo.o om\mfl\>
--- --- --- --- --- em m.HH mmoo.o om\ma\w
--- --- --- --- --- mm >.mH smoo.o om\OH\e
m.H eem.a MOH m.m mme.o --- --- --- om\b\m
.-- --- --- --- --- hm m.mH --- om\m\>
- --- --- -.- --- --- me m.eH mmoo.o om\e\e
--- --- --- --- --- «Ha H.mm maoo.o om\mm\m
m.m mmm.a mma m.w Hem.o om\mm\m
m.m mmo.fi Nae ©.m Hefi.o om\HH\m
N.e mmm.o Had v.b emo.o som\m\m
H.o Hmv.a Ne o.m mem.o HQH m.oa mmo.o mm\mm\OH
m.o mem.a HHH m.m om>.o mma e.oa meo.o mm\fi\m
m.o mmm.H mmH H.w emm.o Hmm e.eH mmo.o mm\mfi\m
m.o mom.m mmfi v.m mmb.o pea m.m mmo.o mm\e\m
m.o mmm.fi Hmm m.mH mm>.o ea n.o MHo.o mm\am\b
H.H eav.m mmv m.mH Nmm.o --- --- --- mw\H\b
e 3 z e 3 2 85:36 A93
..... mm----- -----H>----- -----wow----- mama

 

.om\Hm\m-mm\H\b cwmzuwn um>flm ouom 0:» CH mmmddwdmmm .d
uOm sz mamuou monummo com .on mmHuHmcwo came .ABV munmflwz moo HmsoH>HUGH c002 .m menus

29

ommH m

mmma m
mmma m

{

IIIOIIII

%

0v—

.Ammfioév mm m

.xemee.ee mm .H ”m~\oH-OM\e “lemme.o0 om an
.Amavomv mm m “H\m-HN\b umwdamew macho
.m.m H wumofiocfl mumn nouum .xmmuo mmmuwm
can uw>Hm ouch mnu Eoum NO» mangawqoua .10.
mo unmflmee Hospfiwflocfl CH 005.50 .m musofim

 

Am><n3 m0<

09 00_ 00 00 on
p b p p

b . p _ p p -

0m

 

 

 

1H

 

0.0

 

m6

(9) LHOIZA Lac

3O

0.0130/day over the same interval. YOY growth was similar in
both populations until 0.7 years of age. Ford crayfish grew
very little during the fall of 1989 however, while Baraga

individuals maintained nearly linear growth (Figure 9) .

Production

The greater size and abundance of Ford yl individuals in
1989 resulted in almost a threefold larger biomass than was
Present in 1990 (Table 6) . Since early growth of
Q. gropinguus was not sampled in 1989, production estimates
from the Ford River were based primarily on 1990 data from
Spring and summer. Year 1 individuals contributed the
greatest amount to total Ford crayfish production in 1990,

followed by y2 and YOY age classes (Table 6) . Q. propinquus

 

Production in the Ford River was about 1.5 times higher than
in Baraga Creek (5.0 g/mZ vs. 3.0 g/m2) from 7/14/89-7/14/90
(Tables 6-7). Some interpolation was required, and
Production of Ford early instars and eggs was not included
for comparative purposes. Total production in the Ford River
over the same interval was 5.2 g/m2 when egg and early instar
prOc31.1ction was considered, so the comparative data may be
underestimated by 4 %. Although crayfish in the Ford River
were over three times more numerous than in Baraga Creek, the
la‘l‘ger size and faster growth of Baraga individuals resulted
in Sixty percent of the crayfish production in the Ford

River. Production/biomass ratios were 1.04 and 1.11, for the

31

.Uwuwoflpcfl Uoflumm mEflu mnu uOm mwSHm> czocxc: m>fluflmom zoflz maamo mumoflpcfl mmcfla Uwzmmp -A
GCHHQEmm umumcfl >Humm CH pmcHMDQo mmSHm> Seuw vmumEHumm -«
NE\u£0Ho3 >00 0 CH wofluma ocflusc COHDUSUOuQ uwc -o

NE\u£0Ho3 >00 0 CH wofl

pod 0cflu50 mmmEofln mowew>m -Q

UoHqu mEHo mcflozw >00 pom £03000 mjomcmbc0umcfl -0

 

 

m.0 0 0.0 0 0 0.m 0 m.0 0.0 0m0.0 00\Hm\0-mfi\0
m.0 0 m.H 0 0 0.m 0 m.0 0.0 meo.0 00\>H\0-H\0
H.0 o m.m 0 0 m.m 0 H.0 m.0 0Ha.0 om\H\0->m\0
0.0 0 H.m 0 0.0 0.0 000.0 9H.0 ---- ---- om\0m\0-n\n
H.H 0 0.m 0 H.H m.H 0m0.0 9H.0v ---- ---- 0m\0\0-mm\0
0.H 0.0 0.m 0H0.0 0.0 m.H Hm0.0 9H.0v ---- ---- om\mm\0-fia\0
0.H H.a e.m 000.0 0.0 0.0 0m0.0 9H.0v ---- ---- om\HH\0-0\m
H.0v 0 m.0 0 0 e.m 0 H.0v 0.0 H00.0v 0m\0\m-0m\0H
H.0v 0 m.0 0 0 m.m 0 H.0v 0.0 H00.0 m0\0m\0H-H\m
m.0 0 0.0 0 0 H.m 0 m.0 0.0 «No.0 m0\a\m-0fi\0
m.o o 0.0 o o 8.0 o m.o 0.0 Heo.o @0\0H\m-e\0
fi.0 0 m.0 0 0 H.0 0 H.0 H.0 m00.0 m0\e\m-fim\>
0.0 o m.H o a.o o.oe 000.0 ---- ---- ---- am\em\e-e\e
m m 0 m m 0 0m 9m 00
0 H0009 ...... mm ........... a» ............ wow ..... onuma wees

 

 

.0m\Hm\0 new m0\H\h cww3uwn um>flm Upon wzu CH mSSUCHQOpQ

.O 00 COHDUSGOuQ 002 .0 manmh

32

 

0.m m.H w.e HHo.o 0.H m.o mmo.o ---- ---- ---- om\efi\e-m\m
0 0 0.0 0 0 m.H o 0 H.0 0 0m\0\m-0m\0H
H.0 0 0.0 0 0 0.H 0 H.0 m.0 0H0.o mm\0m\0H-0\m
H.0 0 0.0 0 0 0.0 0 H.0 H.0 0N0.0 m0\0\m-mfi\0
H.0v o 0.0 o o e.e o H.oV H.0v meo.o m0\mfi\m-om\e
---- o w.o o o H.m o ---- ---- ---- @0\OM\e-e\e
a m o a m o a m o
0 Hence ..... mw .......... Hw ........... wow ..... eoeema mews

 

.0m\0a\0 0cm m0\0\> cmm3umn x0000 mmmumm CH masocHooum .O

 

mo cofluoswoua uwz

.e seems

33

Ford and Baraga streams respectively, during the 7/14/89-

7/14/90 interval.

Reproduction of O. propinquus from the Ford River

Ford Q. propinquus were collected for pleopod egg counts
on 5/6/90 as part of routine block net sampling. A carapace
length/pleopod egg regression was constructed to estimate
fecundities of all berried females (N = 14) sampled on 5/6/90
(Figure 10). Fifty-eight percent of mature females, or 8 %
of the entire population, were reproductive on 5/6/90. The
mean number of pleopod eggs per female declined from 85.5 on
5/6/90 to 66.5 on 6/11/90, although these means were not
significantly different when subjected to a t-test (P =
0.0737), probably due to the low sample number and high
variability of values on both dates.

Ten pleopod eggs each were chosen randomly from several
berried females on 5/6/90 for egg diameter determination.
Eggs were measured to the nearest 0.1 mm with vernier
calipers. Mean Ford Q. propinquus egg diameter was 2.0 mm.
The relationship between carapace length and average egg
diameter of this species is shown in Figure 11. The best
fit in these reproduction regressions was realized when total
egg volume (estimated from mean egg diameter) was regressed

on carapace length (Figure 12).

URNlHAU

Int

Huh-

I \sta

\ OW—

34

cm

VN

0m00.0
man 5000

m
madmchoum .d

000 @0900 HQ - nu mama wommmumo

A220

mm

.0303 co 0023
wow COHmmwuomu

. 0H 0050.3

EBOZmA mo<m<m<0

0N

Upon

m:

0_

 

 

 

 

 

tho.m + dab. Pb-

N

0v

00

00

00—

ON—

5993 (IOdOZ'Id 30 11381101!

35

ON

Nmm0.0 Hm .0m\0\m
c0 uni/Hm whom 93 Seen mdeacflmowd ...Q

mo numcwa wommwumo co umqumHU 000
pomomflm came .00 Gofimmmummm .HH musmflm

 

3):): mBOZMA m0<m <m<0

vm mm 0m 0_

 

 

G

 

 

mane 0 mm xm.mo~m.m + 304 n w

 

Wm

(1m) ammm 9:): H31!

36

HH00.0 H m

A220

vm mm

b p p p

5303 so 003m 000.0 0:0 5000
mammdflmoum .10. mo nuucma wommmumo co meadow.

000 oomowdm Hmuou mo GOHmmmumwm .NH mudmfim

ZBOZWA m0<m<m<0

 

 

 

 

 

806 u mm

Xme.mH + mm.m0N-

000

com

00m

(sun) M'IOA 993 11.1.00.

37

Life Table for Ford River O. propinquus

A life table was constructed for Q. prooinguus from the
Ford River to evaluate the population’s stability (Table 8).
The lX column was constructed by estimating the number of
females surviving to different ages, assuming each age class
began as a brood equal to the total number of female age 0
individuals on 5/6/90. (An assumption of this analysis was
that the number of age 0 females remains relatively constant
from year to year.) This latter number was derived from the
total number of age 0 individuals/m2, or the number of
pleopod eggs/m2 on 5/6/90. By assuming a 50:50 sex ratio of
age 0 individuals, halving the number of pleopod eggs/m2 gave
an estimate of the number of female age 0 individuals on
5/6/90. Survivorship of age 0.5, 1.5, and 2.5 females was
estimated using densities obtained in population sampling on
5/6/90. The column P indicates the proportion of females
found to be reproductive in each age class. One can not
assume all mature females extruded eggs and contributed to
the succeeding generation. The mx values represent the
empirically determined mean number of female pleopod eggs
produced per female at each age class (assuming a 1:1 sex
ratio). Data in the lXPmX column (Table 8), indicated that
y2 females were responsible for 93 % of the young production.
This high proportion reflected the numerical dominance of y2
females relative to y3 females in the Ford River (Table 5).
R0 represents the reproductive rate per generation, or the
number of eventual reproductive females produced per female.

The R0 of a population over many generations should be close

38

Table 8. Annual life table for female Q. propinquus from the

 

Ford River.

 

 

 

Age in years lX P mX lXPmX

0.0 1.000 0

0.5 0.090 0

1.5 0.055 0.411 41.0 0.927

2.5 0.001 1.0 57.3 0.057
R0 = 0.984

lX = probability that a female survives to the indicated age

based on the number of age 0 females & the number of
females of the given age class on 5/6/90.

P = proportion of females reproductive at a given age.

mX = potential number of female eggs an individual would

produce if she were alive at the age indicated.

R0 = ZlXPmX = the reproductive rate per generation.

39

to 1, providing the population is stable (Momot and Gowing
1977a). A RO well above or below one over time probably
reflects an overall population increase or decrease,
respectively. The Ford Q. propinquus population appeared to
be stable with a R0 of 0.984; however, the population would
have to be monitored for several years to determine if it is
truly stable. A life table was not feasible for Baraga Creek
due to the small number of berried females captured (N = 4)
in the spring of 1990, a number insufficient for reliable

fecundity estimates.

Life history

The following segment documents the life history of
Q. propinquus in the Ford River. Mature males and females
were observed mating in the stream on 7/13/90,7/27/90, and
8/1/90. (Males were considered sexually mature adults after
they had successfully passed through their juvenile instar
stages and attained the form I condition. Sexual maturity in
females was recognized in females observed copulating with
males in the stream, and in females bearing eggs and young in
the spring). Mating was not observed in 1990 prior to
7/13/90. Sampling in the Ford River was terminated on
8/31/90 so it is unknown if Q. propinquus mates in the fall
in the Ford River. No appreciable growth occurred between
10/28/89 and 5/6/90 (Table 5). Females were first observed
in berry in 1990 on 5/6/90, the first sampling trip of the

year, and remained in berry through 6/1/90. On 6/22/90, the

40

eggs of the majority of ovigerous females had hatched and
instars remained attached to the female pleopods. The
smallest ovigerous female observed was 18 mm in carapace
length. This indicates that individuals did not reach sexual
maturity until after their first spring and did not mate and
reproduce until into their second year, as yls of both sexes
were all under 18 mm during the entire spring of 1990.

Mature males apparently molted synchronously twice a year.
Mature males molted from form I to their nonsexual form, form
II, between May and June, as all mature males examined on
5/6/90 were in form II, and 75 % of individuals examined in
June and July of 1989 and 1990 were in form II. By August,
mature males had molted back to form I. All individuals
examined on 8/5/89 were in form I. Most individuals died
after their second year (survivorship of y2’s to 7/7/90 was
1.6 %). A few crayfish apparently lived to their third
spring (survivorship of y37s to 5/6/90 was 0.2 %).

Year 1 and y2 Q. propinquus from the Ford River generally
maintained a 1:1 sex ratio throughout the year (Table 9). A
noticeable exception occurred on 7/1/89 when males
outnumbered females nearly 3:2. Males dominated the yl and y2
sample more strongly and regularly in Baraga Creek than in
the Ford River (Table 9). A small size difference between
the sexes existed among both YOY and y1 & y2 Ford
Q. propinguus. Young-of-the-year males were 4.4% larger in
total length on the two occasions they were compared (t-test,
P < 0.05, Figure 13). Year 1 and y2 males were 4.3% larger
on one occasion (t-test P < 0.05) and only 1.8% larger on

another (t-test P > 0.05, Figure 14).

 

41

 

 

Table 9. Results of Chi-Square Goodness of Fit Tests
comparing numbers of yl and y2 male and female
Q. propinquus.
Site Date #Males #Females Chi-Square Significance
value

Ford 7/1/89 307 227 11.985 P<0.001
7/21/89 131 131 0 NS
8/4/89 100 96 0.082 NS
8/16/89 77 82 0.157 NS
9/1/89 58 58 0 NS
10/28/89 20 23 0.209 NS
5/6/90 35 35 0 NS
6/11/90 24 27 0.176 NS
6/22/90 33 29 0.258 NS
7/7/90 16 8 2.667 NS
7/27/90 43 39 0.195 NS
8/1/90 48 28 5.263 P<0.05
8/17/90 34 49 2.711 NS
8/31/90 16 30 4.261 P<0.05

Baraga 7/7/89 80 39 14.126 P<0.001
7/30/89 67 42 5.734 p<0.05 '
8/12/89 82 39 15.281 P<0.001
9/8/89 56 37 3.882 P<0.05
10/28/89 26 18 1.455 NS
5/5/90 61 8 40.710 P<0.001
7/14/90 43 40 0.108 NS

42

male
female
male
female

I
E
I

 

 

 

 

 

 

 

 

 

 

 

 

 
  

  

      

 

 

 

 

 

40

)
(
All/f/I/IIIIIAAII
H \\\\\\\\\\\\\\\\\\
I/I/I/I/I/l/x/III
I»///////// T
\\\\\\\\\\\\ G
//////////// 2 N
2 E \\\\\\\\\\\\\\\
III/ll/I/l/l/l/
L \\\\\\\\\\\\\\\
Alt/4l/I/IAAAII/A/IA
\\\\\\\\\\\\\\\\\\\\
//////////\/\/ /\/\////\/\/\ 0 L
/////////
2 A \\\\\\\\\\
T III/III/I
\\\\\\\\\\ T
III/Il/I/
\\\\\\\\\\ 8
1 D
l. T
6 M
1| I
. S
o
4
1 B
< d s q
H u -
0 O O O
0 O O 0 3 2 1
3 2 9|

mmHhhde ho “Hmflpl MMHhhde ho ”NMHDI

 

8
2

 

24

18 2O 22
EST IMATED TOTAL LENGTH

16

14

(MM)

Size comparison of male and female

8/17/90,
males

Am
8/31/90,

B.

 

r

e

v

i

R

d
r .
0,6
P.1l
0
60
mo.
=.1
9
m o
r010
f o
0

sri
use:
u.9
QsiP
n 6
.ml 1
r
meme
.D m
B 4..a
0.7.1
m 89%
u e
n9v1174
oloa .
Fvim.4

43

 

 

 

 

 

 

 

 

 

 

 

 

            

 

    

A .
30
In .
In
H
a 1
N :i
U ,1
h 1 I. I. I male
0 : :1 female
’1 l‘
a 10 - \ Z; 1:
g 3
: I: I:
\ 1‘ i1
0
28 30 32 34 36 38 40 42 44 46 48 50 S2 54 S6
ESTIMATED TOTAL LENGTH (M)
B .
30
In
In
E x
«I 20- §
8 V
3 : I I male
N I: j I‘ female
H ‘0 -‘ :‘ \ :1
‘ \ 1
g , g i 1‘ W 1‘ g
l‘ \l I‘
O I l‘ \ ’i \ N

 

 

 

 

30 32 34 36 38 4O 42 44 46 48 50 52 54 56 58
ESTIMATED TOTAL LENGTH (MM)

Figure 14. Size comparison of male and female

adult Q. propinquus from the Ford River. A. 7/21/89,
males 1.8 % larger, P = 0.1933. B. 8/4/89, males
4.3 % larger, P = 0.0103.

44

Ford early instar distribution

A three factor analysis of variance was performed to
examine the effects of time (with seven levels corresponding
to sampling dates), longitudinal position, and latitudinal
position on early instar crayfish density in the Ford River
(Table 10). No source in the ANOVA table sufficiently
explained the variance in densities to be significant at the
95% confidence level. Time, with a P value of 0.112, seemed
to be responsible for some of the variance. Densities
generally decreased with time as one might expect in an
organism that produces many eggs. That the three factor
interaction had a low P value (0.0793) was a bit surprising.
High densities early in the sampling regime at habitats that
would later lose prominence in "attracting" young instars may
have been attributable. Mean densities from latitudinal
habitats on the same sampling date did not differ
significantly (P > 0.05) when subjected to a multiple range
test using Fisher’s protected least significant difference
(Table 4). Mean densities from longitudinal habitats on the
same sampling date were significantly different (P < 0.05) on
two occasions however, using the same statistical test (Table
4).

Considering longitudinal habitats, run-Hs had the highest
overall mean density of early instars during the study
(2.48/m2), followed closely by run-Ts (2.26/m2), with riffles
being the least preferred habitat (1.06/m2). Abundance in

run-Ts was the most stable over the period (.igure 15).

45

 

 

 

Table 10. ANOVA table for a 3-factor analysis of variance on
densities of early instar Q. propinquus from the
Ford River.
Source df Sum of Mean square F-test P value
squares
time(A) 6 146.402 24.400 1.734 0.118
long. pos.(B) 2 39.052 19.526 1.388 0.2533
AB 12 164.418 13.701 0.974 0.4772
lat. pos.(C) 1 1.851E-4 1.851E-4 1.315E-5 0.997
AC 6 37.341 6.224 0.442 0.8494
BC 2 31.793 15.896 1.13 0.3262
ABC 12 282.751 23.563 1.674 0.0793
error 133 1871.645 14.073

46

Iucse

2t?
pics;

.mm H mumOHch mums wouum
00330: 0-5: new .030:

mSSUCflQOHQ .O UHOE HO mwflUflmCOU C602

.om\mm\0 935309“
30.55 60.5 museums: waumm

. m H mudmflm

 

om\mm\0 mosz mwco

0m 0m 0_

 

 

 

-'.'-
1" I

I

 

 

 

Hid HHEHflH

EHEELEBI EHEYTIUE;

47

Densities in run-Hs and riffles were high initially,
eventually declining to < 1/m2 after 18 days, at which time
nearly all individuals collected were from run-Ts. Overall
mean densities from side (2.41/m2) and middle (1.87/m2)
portions of the stream were quite similar. A decline in
density over time was witnessed for both habitats (Figure
16). The crossing graph lines in Figure 16 suggest a
possible lateral migration from side to middle positions, but
because early instar crayfish densities were not
significantly different between the two microhabitats,
conclusions are speculative.

Other physical aspects of the stream environment were
considered in their relationship to early instar abundance.
No relationship existed between stream depth and crayfish
density in the Ford River (Figure 17). Water velocity also
was not significantly correlated with density (Figure 18),
although crayfish generally achieved higher numbers in slower
water and no individuals were found in water with a velocity

greater than 0.440 m/sec.

DISCUSSION

Q. propinquus growth and fecundity traits in the Ford
River compared well with values reported for other
populations of the species in the northern part of its range
(Capelli 1975, Corey 1987). Growth, production, and
fecundity were lower than most values reported for more

southern populations of Q. propinquus (Van Deventer 1937,

48

.m.m H 38038 meme .3er .3330
mcflcdflmwn mumuflnmn moflm new mHUpHE Eoum muMumGH hauwm
a; .0 Upon we meuHmch cmmz .0H musowm

3330 8sz mwao

 

 

 

0m 0m 0_
A : a o
..v .. ..
{I .49 A
.4
mfiufim ll 1
3835 lull ..

 

 

 

Hid film

2111.311 ZHYIIOS

 

 

NUMBER PER METER SOU ARED

49

 

 

 

 

 

 

30 1 A i A A - g L A 1 J i 4 g
j o
254
204 O L
15‘ o _
101 . .-
5W 0 s o . o s . o . as o
‘ . C C . E“... .: ... .. .. . . .
O A ‘ loo. 0‘“; 0. Li. .1... ’_ o_2“ - g - 0A 0 c :
L
i
-5 I’ 1 1 I T I t I T I t 1 1 I— 1 I t
0 5 10 15 20 25 30 35 4O 45
DEPTH (CM)

Figure 17. Relationship between water depth and density
of early instar Q. propinquus from the Ford River.
R2 = 5.253 E-3.

NUMBER PER METER SOU ARED

50

 

 

 

 

 

 

15 - ee. 1 - . es .1 - - - 1 -
1
144 9 -
1 1
12‘ _
< ,
10. .
1 O
8« 9
9 C C C
6 O C O C .3 b
o o o o
4 o o o 9
( o o s o o o o o o
2-1 0 o o o o o o o o o o 1-
‘ o o o o o o s o o o 0 o o o o >
O 3 33 3 333 3 3 3 3 333 3 3 3 3 3 3 3 3 3 3 3 3 3 3
1 L
'2 . 1 r 1 v I . I . r -3
- 1 0 1 .4 5 6

.2 .3
VELOCITY (M/S)

Figure 18. Relationship between water velocity and
density of early instar Q. propinquus from the Ford River.
R2 = 0.022.

51

Vannote 1963), and in larger crayfish species (Langlois 1936,
Momot 1967, Weagle and Ozburn 1972, Momot et a1 1978, and
Corey 1987). The sequence and timing of life history events
in the Ford River were similar to those reported for

Q. propinquus in northern latitudes (Crocker and Barr 1968,
Capelli 1975, and Corey 1988). Others who have studied

9. propinquus in the southern reaches of its range found
developmental events to occur earlier in the year (Van
Deventer 1937, Bovbjerg 1952). For instance, Bovbjerg (1952)
reported juveniles recruited into the population in late May
and early June, while instars were not independent of their
mothers in the Ford until late June. Q. propinquus usually
takes two years to mature over its entire range (Van Deventer
1937, Capelli 1975, and Corey 1988), while crayfish species
in the southern United States tend to reach maturity within
their first 8 to 16 months (Penn 1943). The absence of
winter growth found for Q. propinquus in the Ford River has
been found in other temperate crayfish populations as well
(Vannote 1963, Hopkins 1967, Flint 1975, Pratten 1980, France
1985, Corey 1988).

The deviation from the 1:1 sex ratio seen in the early
summer Ford River population (Table 9) may have reflected a
period of female seclusion during their postreproductive
molt. A pattern of more males than females early in the
growing season was observed in another Q. propinquus
population and was attributed to inadequate sampling of
reclusive ovigerous females (Fielder 1972). Svardson (1948)
found the sex ratio of Astacus fluviatilis to deviate

significantly from one in early June, with males being more

52

abundant. He found females to outnumber males in July,
presumably related to the male molting period. The two
occasions in August when sex ratios differed significantly
from one in the Ford River (Table 9), did not coincide with
molting periods. The deviant ratios may have been a product
of small sample size. Sex ratios of Q. propinquus in the
Baraga Creek population were more consistently different from
one, also in favor of males (Table 9). The May and early June
occurrences may be explained by isolation of reproductive
females. The general consistency of the pattern throughout
the summer however, may reflect a real difference in the
numbers of males and females in the population.

Aspects of fecundity such as the percentage of ovigerous
females in a population, size at first reproduction, number
of eggs produced per female, and the amount of egg mortality
can all influence the intrinsic rate of increase in a
population. The values for these components of fecundity in
Q. propinquus from the Ford River were generally similar to
those found in other Q. propinquus populations at similar
latitudes (Capelli 1975, Corey 1987, Corey 1988). Data on
egg mortality were not given by these workers however. The
figure of 57.7% of mature females ovigerous in the Ford River
falls within the range of 44.4-78.3% reported in other
species (Mason 1975, Woodcock and Reynolds 1988). The Ford
value of 8.1% ovigerous out of the entire population is at

the lower end of the 8.0-34.6% range given for Q. propinquus

 

by Corey (1988).
The smallest gravid female in the Ford population was 18 mm

in carapace length. The smallest gravid female in Capelli’s

53

(1975) Wisconsin Q. propinquus population was also 18 mm in
length. Corey (1988) found the species to bear eggs at a
smaller size (12.8-14.3 mm) in four Ontario river
populations. This smaller size may have reflected her method
of measuring carapace length from the base of the eyestalk
instead of the tip of the rostrum.

The mean number of pleopod eggs per reproductive female in
Ford Q. propinquus (85.5) was within the range of 60-107.5
reported for the species at similar northern latitudes
(Capelli and Magnuson 1974, Corey 1987). Van Deventer’s
(1937) mean of 132 for the species was exceptionally high and
may have reflected a longer growing season in Illinois,
compared to the shorter growing seasons in Wisconsin,
Michigan, and Ontario.

Mixed reports of egg loss were found in the literature.
Mason (1970) witnessed a 15-60% percent loss of the original
brood of eggs in Pacifastacus trowbridgii and attributed it
to cannibalization by the mother. Brown and Bowler (1977)
also reported a considerable loss of eggs in A. pallipes.
Conversely, Rhodes and Holdich (1982) found no relationship
in the above species between egg number and time of
collection during the brooding period. A trend towards lower
egg number over time occurred in the Ford River but was not
significant.

Q. propinquus annual production in the Ford and Baraga
populations was lower than values found for more southerly
Q. propinquus populations and larger crayfish species
(Vannote 1963, Momot 1967, Momot and Gowing 1977b, Shimuzu

and Goldman 1981). Vannote (1963) reported considerably

54

higher annual production figures of 41.5 g wet weight/m2
(~13.8 g dry weight) for Q. propinquus in southern Michigan,
perhaps related to a longer growing season. The larger
Q. virilis had annual production figures in Michigan lower
peninsula lakes ranging from 6.02-20.72 g dry weight/m2
(Momot 1967, Momot and Gowing 1977b). Annual
productionzbiomass ratios of Ford and Baraga Q, propinquus
were similar to those reported for other short lived crayfish
species. Vannote (1963) reported a 0.9 P/B ratio for
Q. propinquus. Momot (1967) and Momot and Gowing (1977b)
found a 0.94-1.53 range of P/B values in their Q. virilis
studies. These ratios tend to be larger then the 0.32-0.90
range reported for longer lived crayfish species (Mason 1963,
Shimizu and Goldman 1981, Brewis and Bowler 1983, Huryn and
Wallace 1987, Cukerzis 1988). Longer lived species are often
quite large and usually experience slow relative growth,
lending to the lower P/B ratios.

The instantaneous mortality rate (Ricker 1975) of Ford
early instars between 5/6/90 and 7/6/90 was 0.0531 / day

~

(eggs hatched 6/17/90). Instantaneous mortality rates of

~

YOY crayfish from 6/20/90 to 7/14/90 (eggs hatched 6/25/90)
calculated from Capelli’s (1975) data on a lentic population
of Q. propinquus were lower than my value at 1 meter depths
(Z = 0.0123 / day) but higher at 3 meters (Z = 0.0561 / day)
and 5 meters (2 = 0.1630 / day). No researcher to my
knowledge has attempted a detailed analysis of the

survivorship of the earliest instars in a crayfish

population. Capelli's (1975) estimate of mortality was based

55

on data from two sampling dates 25 days apart. Some have
estimated the density of YOY individuals later in the growing
season when crayfish were at a more catchable size (Momot
1967). Using this method and fecundity data, Momot and Gowing
(1977b) found survivorship of YOY Q. virilis to the end of
their first growing season to be 5-40 % over different lakes
and years. Survivorship of Ford YOY to 8/31/90 was within
this range at 13.9 %. Others have solely used pleopod egg
counts to estimate the abundance of the YOY size class
(Shimizu and Goldman 1981, Cukerzis 1988).

The survivorship curve constructed for Ford early instar
Q. propinquus in 1990 was clearly an atypical survivorship
curve (Figure 7). The initial steep decline of the curve,
reflecting high initial mortality, was not unexpected. The
increase in instar density at 82 days commensurate with the
time early instars began being sampled in the block net
however, seemed counterintuitive. One would expect the total
numbers of a cohort to decrease with time as individuals
succumbed to various forms of mortality. Two to five times
more early instar crayfish were colleCted using the block net
approach than were collected using kick screens. There is
reason to believe that the kick screen method used to sample
early instars during the first weeks of their lives was less
efficient than the block net method. There was likely an
increased probability of escape by YOY crayfish around the
sides of the kick screen and underneath the bottom. Because
of the nonpliable nature of window screen and the rough,
mostly cobble substrate of the Ford River, it was sometimes

difficult to position the bottom edge of the kick screen

56

flush with the stream bed. The block net, conversely,
spanned the entire width of the stream and was equipped with
lead weights on its bottom edge which were supplemented with
rocks to effectively seal the net to the bed. Thus,
potential escape of crayfish around or underneath the block
net was largely prevented. The fact that Q. propinquus YOY
were quite small (< 7mm carapace length) and often difficult
to locate among the mass of dislodged periphyton,
invertebrates, leaf material, and other debris in the kick
screen may have also contributed to the comparitive low
numbers of crayfish obtained using the kick screen method.

On 8/1/90 the kick screen method and the block net method
were compared in order to determine a correction factor,
Factor E, for the reduced catch-per-unit-effort of the kick
screen method. The kick screen approach produced an
unrealistically low density figure of 0.83/m2 however, so the
methods were compared using a kick screen effort on 7/21/90
and a block net sampling on 7/27/90. For purposes of
computing the correction factor, zero mortality was assumed
between 7/21/90 and 7/27/90. Figure 19 is a modification of
the survivorship curve shown in Figure 7 in which densities
derived from kick screen sampling have been multiplied by
Correction Factor E (3.99). This modified survivorship curve
is more typical of the survivorship curve exhibited by a
highly fecund short lived organism such as a crayfish.

Various workers have examined the distribution of juvenile
and adult crayfish in different microhabitats within a single
population (Flint and Goldman 1977, Gore and Bryant 1990).

Rabeni (1985) found that juveniles of two sympatric

57

 

 

 

 

 

 

Ci
5
t:
O!
In
E
01
H
9.1
m
E;
OI'T‘TII'I‘I'T'I‘I'I'I'fi
0 S4 59 6165 68 73 76 82 87103117
DAYS SINCE 5/6/90
Figure 19. Revised survivorship curve of

Q. propinquus early instars from the Ford River in
1990. Data point at zero days represents mean pleopod
eggs/m2. Data points at 54-76 days represent the
grand mean of mean longitudinal densities

(Figure 16) weighted for habitat relative abundance

at the Ford site and adjusted by Correction Factor E.
Data points at 82-117 days represent mean densities
obtained with the block net method. Error bars
indicate : S.E. of means unweighted for habitat
relative abundance on a given date.

58

Orconectes species used different microhabitats to some
extent, and invoked competition as a mechanism explaining the
differences. Ford Q. propinquus early instars may be using
different microhabitats during their first weeks of life.
Because distribution of early instar crayfish tended to be
rather clumped in all microhabitats examined, standard errors
of mean densities for a given microhabitat were rather high.
High standard errors were the main reason most densities were
not significantly different between microhabitats. Keeping
in mind that densities between microhabitats were not
significantly different for most sampling dates, speculation
will be made on the abundance trends observed. High
densities of crayfish early instars in riffles and run-Hs
initially (Figure 15) may have reflected the position of
their mothers at the time the young became independent. Gore
and Bryant's (1990) data on the habitat preference of adult
Q. neglectus in an Ozark stream suggested that while adults
preferred backwater areas during most of the year, ovigerous
females opted for riffle areas. This behavior of berried
females presumably meets the high oxygen demands of their
pleopod eggs, and could explain why Ford Q. propinquus
instars were initially the most abundant in riffles and
run-Hs, which may have been more rich in oxygen. Ovigerous
females in the Ford River were collected more commonly from
runs than riffles during block net sampling however, a fact
inconsistent with the preceding explanation. The decline over
time in early instar densities from riffle and run-H habitats

was probably due to various causes of mortality in the

59

cohort, but may have reflected some migration into run-T
habitats, since run-T densities stayed fairly stable over
time while overall densities declined (Figures 7 & 15).
Rabeni (1985) reported his data as mean densities pooled over
the course of his study, so there was no indication if the
early instars changed their position in the stream as they
grew.

There was considerable year-to-year variation and between
stream differences in Q. propinquus abundance. It is not
clear why densities of y1 individuals from the Ford River
were so high initially in 1989 (Figure 5). It may have been
that 1988 was an exceptional year for young production.
Reasons for the steady decline in 1989 Ford River yls are
also unclear. Predation by fish or raccoons may have been
responsible, or possibly intraspecific competition, in light
of the initial high densities (> lO/mZ). Densities from 1989
were similar to those of most lotic crayfish populations that
have been studied. Slack (1955) reported autumn crayfish
densities in two Indiana streams between 23.2-29.2/m2. Momot
et al (1979) lists literature reports of stream crayfish
density as high as 33/m2, but it is not clear if this density
represents only adults or all size classes. Others have
reported annual mean adult densities in streams between 0.5-
8.9/m2 (Vannote 1963, Shimuzu and Goldman 1981, Price and
Payne 1984, Rabeni 1985). Even though many of these latter
densities represent mean stream densities over a wider range
of habitats than were used in the Ford River study, 1989 Ford
crayfish densities were quite high, possibly high enough to

foster intraspecific competition. The 1990 yl size class in

60

the Ford River may have been at low enough numbers initially
to prevent the effects of predation or competition, as
densities were fairly stable over the 1990 sampling period.
Baraga Creek 1989 yls, which were never more abundant than
5.7/m2, also experienced rather stable densities over the
summer (Figure 5). Low densities on 10/28/89 in both Baraga
and Ford populations (Figure 5), may have been partially due
to decreased sampling efficiency of crayfish, since they
tended to be sluggish due to low stream temperature.

The sampling efficiency of the block net method in general
was determined using the results of a mark and recapture
study in an area adjacent to the Ford site on 9/13/90. The
study used an Adjusted Peterson Estimate of N (Ricker 1975).
After partitioning a 10 m2 area with upstream and downstream
block nets, three kicking passes were made in the manner
described earlier. Individuals were marked by clipping their
left fifth periopod, released back in the enclosed area, and
allowed to relocate for two hours. The recapture run
consisted of three kicking passes. The total number of
crayfish captured from the three initial kicking passes
divided by the Peterson N represented the efficiency of the
block net method. The block net method had a 58% efficiency
in sampling crayfish 2 7 mm in length. The 95% confidence
interval on the block net method effiency was quite large
(79% > 58% > 41%). The inverse of this efficiency value
represented Correction Factor D (1.73) and could be applied
to densities to result in adjusted densities. In adjusting
crayfish densities using Correction Factor D, I would have

preferred to use densities derived from the sum of the three

61

kicking passes, instead of depletion densities obtained using
removal collection (Zippen 1958). This would have avoided
adjusting densities that had already been adjusted (i.e.
potentially compounding errors). Nevertheless, the
percentage difference between crayfish totals derived from
the sum of the three kicking passes and totals obtained using
removal collection (Zippen 1958) was generally small. The
mean difference over all the Ford and Baraga sampling dates
was 4.7 i 0.7 percent. Even though this difference becomes
8.1 i 1.2 percent when multiplied through by Correction
Factor D, the figure was within the error range of the block
net efficiency. Figure 20 represents the two extremes for
all 1989-90 Ford and Baraga subsites in the discrepancy
between the two kinds of totals (labeled raw total and
depletion total in the graph).

Q. propinquus densities from the Ford River adusted using
Correction Factor D were clearly on the high end of densities
reported for other stream crayfish populations. For
instance, the mean adjusted density of yls in the Ford River
between July and October of 1989 was 16.2/m2. Crayfish
densities in Baraga Creek were adjusted using Correction
Factor D obtained in the Ford River mark and recapture study.
The assumption was made that Baraga Creek crayfish were
sampled with the same efficiency as Ford individuals based on
the similar substrate, stream width and depth, and flow
regime of each stream. The adjusted mean density for Baraga

Creek yls during 1989 was still fairly low at 5.2/m2.

62

 

 

 

 

 

 

 

 

 

 

A.
40
. Y = 53.796 - 0.649X R2 = 0.931
In
In
4
Di
3
III
a.
5'3
Id
0
O Y I ' I ' T ' I r I
30 4O 50 60 70 80
CUMULATIVE CRAYFISH CAPTURED
B.
80
01 Y = 350.0 - 4.500x R2 = 0.954
W a
'1 6O
91
H
[II
94 40‘
5.3
3 208
1 e
1 El
0 1 1 .
60 7O 8O
CUMULATIVE CRAYFISH CAPTURED
Figure 20. Ford River _0_. propinquus captured per
kicking pass regressed on cumulative capture. A. 7/27/90:
raw total = 64, depletion total = 83. B. 8/31/90: raw

total = 78, depletion total = 78.

63

Differences in size (and therefore growth) of Ford y1
individuals between 1989 (1988 cohort) and 1990 (1989 cohort)
may have been related to variable climates between the two
years. Aiken (1967) found that molting (and therefore
growth) in Q. virilis ceased when water temperatures fell
below 10 C. Therefore, mean water temperatures in the Ford
River were examined for those periods during the crayfish
growth season that could potentially be low enough to limit
growth: The 1988 cohort experienced higher mean water
temperatures and more days when mean water temperatures 2 10
C, for both the period at the end of its YOY growth season
and the period of spring growth the following year, than the
1989 cohort during comparable intervals (Table 11). This
data does not rule out other potential factors that may have
been responsible for the year-to-year size differences
observed. A paucity of data points makes conclusions about
year-to-year size differences in Baraga y1 Q. propinquus
difficult, although 1989 individuals appeared to be slightly
larger.

There was an inverse relationship between Q. propinquus
density and individual growth of YOYs in both Baraga Creek
and the Ford River perhaps mediated by intraspecific
competition. Growth of YOY individuals considered over an
entire season was inversely correlated with the density of yl
individuals (Figure 9, Tables 3 & 5). In addition, growth of
yl individuals was substantially higher in the Baraga
population compared to the denser Ford population in 1990
(Figure 8). If density was solely controlling individual

growth in these crayfish populations one would have expected

64

Table 11. Ford River water temperature data1 at the end of
the YOY growing season and beginning of the yl
growing season for 1988 and 1989 Q. propinquus

 

 

cohorts.
Period Mean temperature Days2 > 10 C
(C) i” S.E.
Sept9-Oct9 88 12.29 i 0.45 24
Sept9-Oct9 89 11.57 i 0.59 21
.May 89 11.54 i 0.71 21
May 90 9.87 i 0.49 15

 

1- data from Burton et al (1988-89) and Burton et a1
(1989-90).

2- number of days during time interval when the daily mean
water temperature exceeded 10 C.

65

Ford 1990 yls to have been larger than the more numerous 1989
yi individuals. This deviation might be explained by the
lower stream temperatures in Fall 1989 and Spring 1990,
however, and a case of physical factors overriding the
influence of biological ones. Inverse relationships between
density and individual growth in crayfish have been noted by
many other investigators (Svérdson 1948, Abrahamsson 1966,
Hopkins 1966, Momot 1967, Flint and Goldman 1977, Momot
1977b, France 1985, Sokol 1988, Hanson et al 1990).

Although water hardness could not be ruled out as a factor
contributing to the greater abundance of Q. propinquus in the
Ford River compared to Baraga Creek, water hardness was

apparently not affecting O. propinquus growth in either of

the populations studied. If water hardness, and specifically
calcium, was an important factor in growth, one would have
expected individual growth to have been higher in the Ford
River where calcium was plentiful (~20 ppm) compared to
Baraga Creek where calcium was less abundant (~10 ppm).
Although growth of Procambarus clarkii increased with
hardness up to 200 ppm (de la Bretonne et al 1969) there are
reports of individual crayfish in systems with equal or less
amounts of calcium than Baraga Creek, growing as well as
crayfish subject to much higher amounts of calcium (Capelli
1975, Goldman and Rundquist 1977). Capelli (1975) also
determined experimentally that survivorship of Q. propinquus
juveniles in a lake with 2.8 ppm Ca++ was as high as that in

a lake with three times more calcium. Conversely, de la

Bretonne (1969) found that survivorship of Procambarus

 

clarkii in plastic ponds was slightly lower in water with 50

66

ppm CaCO3 (higher than the 25.0 ppm CaCO3 in Baraga Creek)
than in water with 100 ppm CaCO3. Thus, although it seems

unlikely that calcium was limiting growth of Q. propinquus in

 

Baraga Creek, it may have been affecting survivorship.

Chapter 2: Diet analysis of Q. propinQuus
in a northern Michigan stream

INTRODUCTION

A considerable number of diet studies have been performed
on crayfish indicating that they are omnivorous generalists.
Because of crayfish's wide feeding preferences they may act
as primary consumers, secondary consumers, and as
detritivores; in the latter case contributing to nutrient
cycling by initiating the breakdown of allochthonous
materials (Momot et al 1978, Huryn 1987). Although crayfish
are omnivorous, plant material including vascular plants,
algae, and detritus usually dominate their diet (Bovbjerg
1952, Vannote 1963, Momot 1967, Prins 1968, Capelli 1980, and
Sokol 1988). Animal material usually represents a smaller
percentage of their total diet. Exceptions are evident
though, such as a population of Astacus leptodactvlus
cubanicus in the River Don in Russia which primarily consumed
amphipods (Tcherkashina 1977). As crayfish often dominate
the macroinvertebrate biomass of streams (Webster and Patten
1979, Momot et al 1978, Huryn and Wallace 1987), there is
potential for them to regulate the abundance and distribution
of various organisms. Experimental work has shown that
crayfish can significantly reduce macrophyte (Lodge and
Lorman 1987) and macroinvertebrate biomass (Hanson et al
1990), and reduce or increase periphyton production at
high and low crayfish densities respectively (Flint and
Goldman 1975). Different size classes of crayfish may affect
community structure in different ways as evidence suggests

that a crayfish’s diet changes with age (Creaser 1934b,

67

68

Vannote 1963, Abrahamsson 1966, Mason 1975, Lowery and
Holdicn 1988).

Diatoms compose varying percentages of a crayfish’s diet in
different systems (Prins 1968, Momot et al 1978, Capelli
1980). The kinds of diatoms eaten by crayfish however, have
not been examined in any detail (but see Flint and Goldman
1975). Preliminary observation of Q. propinquus diet in
the Ford River indicated that diatoms were the predominant
component. The objectiVe of this study was to conduct a
detailed diet analysis of Q. propinquus to determine whether
crayfish age and/or season affected the kind and relative
amount of diatoms they consumed. Other components of

Q. propinquus’ diet were analyzed with the same intent.

METHODS

Crayfish were collected for diet analysis during the summer
of 1990 in a ~200 meter stretch of the Ford River
approximately 200 meters downstream from the population
sampling site (see site description, Chapter 1). Young-of-
the-year (YOY) were collected on seven sampling dates
beginning in late June and ending in early September. Adults
were collected on three occasions between late June and early
September. All individuals were obtained using hand held
kick seines. Crayfish were placed on dry ice for 1-3 hours
within an hour of collection, after which they were kept in a
freezer until the contents of their stomachs were examined.
All specimens were analyzed within four months of collection.

Frozen crayfish were allowed to thaw in ethyl alcohol for

69

at least 30 minutes before their stomachs were removed and
examined. Stomachs were subjected to a sequential analytical
regime. A Bausch and Lomb dissecting microscope (10-45x) was
used initially to examine the total contents of a stomach for
insects, larger algal material, and detritus. The stomach
contents were then examined with a Leitz Laborlux 11 compound
light microscope under 150x. This step allowed smaller items
such as insect head capsules to be identified and counted.
Semi-permanent wet mount slides were made by placing a
subsample of the total stomach material (prior to any
analysis) of a crayfish and a few drops of water on a glass
slide, homogenizing the material by stirring it with the tip
of a pipet, and securing a coverslip on the slide with
household nail polish. These slides were also examined under
150x for insects and other material. Finally, the slides
were examined for diatoms under 1500x using oil immersion.
The first 50 valves encountered during horizontal searching
passes across the slide were identified to genus (Prescott
1978, Wujek and Rupp 1980, Germain 1981) and recorded.

Slides with less than 50 diatom valves were not included in
the data analysis. All insects encountered during the
various analysis steps were identified to order, or family
when possible (Wiggins 1977, Coffman and Ferrington 1984,
Edmunds 1984, Harper and Stuart 1984). The ability of a
crayfish’s gastric mill (Pennak 1978) to grind up food items
made insect identification difficult. Some mangled insects
were classified simply as unknown insects. A total count of
all insects that could be identified as such was made for

each stomach examined. For counting purposes, a piece of

70

insect was sufficient to conclude that at least one insect
was present in a stomach. The assumption was made that a
piece of insect in a stomach represented the remains of a
whole insect ingested by the crayfish. The presence of two
or more whole insects or insect pieces obviously not from a
single individual (e.g. two head capsules), was necessary to
confirm that two or more insects were ingested. Care was
taken not to count the same insect or piece of insect twice.
A qualitative note acknowledging the presence or absence of

filamentous green algae, was made also for each stomach.

RESULTS

Diatoms, filamentous green algae, insects, detritus, and
unicellular green algae were the major items found in both
juvenile and adult Q. propinquus stomachs. With the
exception of adults on 9/7/90, filamentous green algae and at
least 50 diatom valves each attained over 70 % incidence in
crayfish stomachs examined for all dates during 1990 (Figure
21). Insects became increasingly abundant in YOYs as they
grew over the summer (Figure 21). Well over 50 % of adult
stomachs were empty on 9/7/90, in sharp contrast to earlier
sampling dates. Because of difficulties in identifying
fragmented algal filaments usually missing their
chloroplasts, filamentous green algae were grouped together
in a single category and reported as such. Unicellular green

algae found in crayfish stomachs included Closterium,

 

Staurastrum, Cosmarium, Scenedesmus, and Merismopedia. Sand

 

II 090(fl

E] (flat U)

E!
El

ins.(j)
figa.(a)

E] diat.(a)

71

ins.(a)

 

 

 

S$§§§S§$§

 

 

 

 

 

"”1“““11111”

 

V. .1111! l 1 ”11”“

 

 

 

 

$§§§§§§§R§§§§§$§§§§fi§

 

 

 

“1.11.11.19.1"111"-11““)”1.91.1“ 01011111100001 1

 

 

 

 

.IUIUHHUUW”1W!!!

    
 

 

1,111.11111111.!1

   

 

 

 

 

 

 

 

 

 

 

 

SHDYHOLS 10 ZSYLHHDHHJ

7/26/90

9/7/90

8/15/90

7/4/90 7/10/90 7/18/90

6/29/90

_DATE

unaningnns from

various food items.

Figure 21.

f.g.a.
2 50 diatom
a = adult .

diat. =
= juvenile.

The percentage of Q.
= insects. j

filamentous green algae.
ins.

the Ford River with

valves.

72

grains and detritus, probably consisting mostly of leaf
material, were also found in many stomachs.

Insects were considerably more abundant in juveniles than
adults (Figure 22). Peak insect abundance occurred on
8/15/90 (47 days in Figure 22) in both age classes. Crayfish
age and time of sample were both significant (ANOVA, P =
0.0001) in explaining the variation in insect abundance in
stomachs when compared over 7/26/90 and 8/15/90. Diets on
9/7/90 were not included in the comparison because of the low
incidence of food items in adults (Figure 21). Chironomidae
dominated the insect fauna ingested in both age classes
(Figure 23). Although juveniles contained more Trichoptera
than adults, adults tended to eat more large insects
including Plecoptera, Ephemeroptera, and insects in the
"unknown" category.

Several diatom genera attained at least 7 % mean relative
abundance in juveniles and adults when data was pooled over
the summer. These included Achnanthes, Cocconeis,
Cvclotella, Cympella, Fragilaria, Gomphonema, Navicula, and
Rhoicosphenia (Figure 24). A number of other genera occurred
only rarely, never exceeding 3 % relative abundance. These

included: Amphora, Diatoma, Epithemia, Eunotia, Frustulia,

 

Gyrosigma, Meridion, Opephora, Stauroneis, and Synedra.
Achnanthes, Cocconeis, and Navicula were more prominent in
juvenile stomachs while Cvclotella, Cvmbella, and
Rhoicosphenia were more common in adults when summer data was
pooled (Figure 24). The kinds of diatoms prevalent in
stomachs changed over the summer. Figure 25 shows that

certain genera such as Achnanthes dominated the diatoms

 

73

:58 '9'

2.533 Iall

.mm H ucmmwumwu mama wouum .uwxfim whom 9.3 60.5

.mdnddfidad .d :H muommGH 00 0085:” :00: .NN wuzmfim

00

om\mm\0 mosz mwmo

 

 

 

 

 

 

HONOLS I SIDHSII :10 11mm:

74

JUVENILES

Chironomidae
Trichoptera
Ephemeroptera

Plecoptera

DRIED

unknown

Chironomidae
Trichoptera
Ephemeroptera
Plecoptera

unknown

DEIED

 

Figure 23. Mean relative abundance of insect
types in the stomachs of Q. mingms from
the Ford River; data were pooled for the entire
1990 summer period.

75

 

 

 

ADULTS

 

 

Figure 24. Mean relative abundance of diatom genera
in.Q4 propinquus fro. the Ford River; data vere pooled
for entire 1990 summer sampling period.

76

SHALL SIZE

 

30

-—Hr—- anmhnnaa'
-——e—- acumen»:
--<>--!quluda

 

IHHMEEH:CHE YUHTEES

 

 

 

 

 

 

 

 

 

 

 

 

80
DAYS SINCE 6729l90
EEDIUH SIZE
E
p.
N .
0 -—e—— Cumduu
a ----- 9----- Cyclotella
E;
0 20 40 60 80
DAYS SINCE 6I29I90
LARGE SIZE

E

E

h -—<}—-I&nuMIis

O -« - Immune

fl --1k--Zflnmauphnda

g:

0 l r ' I ' ' I ' ' 1 ‘ ‘
0 20 40 60 80

DAYS SINCE 6129(90

Figure 25. Relative abundance of small, ncdiun, and large
size diato- genera in,Q; propinggus from the Ford River. The
Y axis represents the nunher of diaton valves of a particular
genus out of 50 total counted. Error bars indicate 1 S.E.

77

ingested early in the life of YOYs while others such as
Fragilaria and Cvclotella increased in prominence in YOYs
over the summer. Adults and juveniles generally appeared to
be eating the same kinds of diatoms when compared over
7/26/90 and 8/15/90 (Table 12-13). When crayfish age and
time of year were considered in a two factor ANOVA as
possible influences on the relative abundance of common
genera however, some significant differences were found

(Table 14). Gomphonema was more abundant in adults (P =

 

0.0021) while Navicula was more common in juveniles (P
=0.0289). Additionally, Achnanthes was more common in
juveniles but this difference was not significant at the 0.05
level (P = 0.0633). The ANOVA also showed that numerous
diatom genera experienced significant changes in their
relative abundance within the diet of a single crayfish age

class between 7/26/90 and 8/15/90 (Table 14).

DISCUSSION

 

Although the relative proportions of different food items
in Q. propinquus stomachs were not quantitatively determined,
diatoms appeared to be the most abundant food item by volume
in both age classes. Filamentous green algae and insects
were the second and third most common items in adults.
Juveniles seemed to favor insects over filamentous green
algae.

It is not clear why such a high proportion of adult
Q. propinquus stomachs were empty on 9/7/90 (Figure 21).

High water at the time of sampling may have seen crayfish

bel a

C

S.E.

i

 

Cvclotella

Values are means*

 

Gomphonema

 

Fragilaria

 

propinquus from the Ford River.
Achnanthes

 

N

0.

Relative abundances of small and medium size diatom genera in the stomachs

Age

 

 

Table 12.
Date

0.7

+|

4.7

+l

0.2
0.3
0.4

0.9

H

0.7

+l

3

1.4
1.9

17 19.5

J
J
J
J
J
J
J
A
A

6/29/90
7/4/90

+l

0.7
0.9

H

20.9

15
17
17

0.4
0.7

2.8 i 0.4
9.4 i 1.3
5.5 i 1.1
3.9 i 0.6

+1

0.2

+1

+l

8.3
11.1

+l

1.6
1.3

16.9

7/10/90
7/18/90
7/26/90

H

1.1
0.6
2.9 i 0.7
0.9 i 0.5

8.7
5.1

+|

10.9

+l

+|

15
14

0.7
2

6.6
11.7

1.3

10.8 i 2.3

+|

8/15/90
9/7/90

.4

i 0.9

3.0

15

3

1.

10.4

0.8
0.6

H

4.1 i 0.8
3.8 i 1.2

20

7/26/90
8/15/90

1.1

+|

+l

0.9

+1

16

- mean number of diatom valves per individual crayfish out of 50 valves counted.
juvenile crayfish.

 

A - adult crayfish.

*
J

79

 

 

b.H H N.HH h.o H c.m h.o H m m ma 4 om\ma\m
a.H H m.m m.o H m.q v.0 H m.« om 4 om\mm\b
b.a H m.m 0.0 H w.m H.N H v.m ma o om\>\m
N.o H H.@ h.o H ¢.m o.a H m.v «a h om\ma\m
m.o H v.m m.o H m.m m.o H m.m ma h om\mm\>
m.o H m.m m.o H N.m m.o H m.m ha o om\ma\h
0.0 H m.m m.o H m.m v.0 H m.m DH h om\oa\h
m.o H m.fi N.H H b.n m.o H q.w ma h om\w\c
m.o H H.m m.o H m.m H.H H o.o DH o om\mm\o
mwdwmmmmmwmmm MHmwANmZ . mflwcooooo 2 mod ouma

 

.m.m H mcwme mum mwdfim> ..umxrflm Chow wflu EOum mSSUCHQOMQ .0

mo mzomEoum on» CH mumcwo Eoumflo mNHm wanna mo moosmocsnm o>Humawm .mH manna

80

Table 14. ANOVA tables for a 2-factor analysis of variance
on the relative abundance of diatom genera in Q.
propinquus stomachs. An asterisk indicates
significance at the 0.05 level. Gomphonema had a
higher relative abundance in adults. Navicula had a
higher relative abundance in juveniles.

A. Achnanthes

 

 

Source df Sum of Mean square F-test P value
squares

age (A) 1 52.438 52.438 3.577 0.0633

time (B) 1 126.262 126.262 8.613 0.0047*

AB 1 7.545 7.545 0.515 0.4759

error 61 894.25 14.66

B. Cocconeis

 

 

Source df Sum of Mean square F-test P value
squares

age (A) 1 10.125 10.125 1.076 0.3037

time (B) 1 11.63 11.63 1.236 0.2706

AB 1 0.148 0.148 0.016 0.9005

error 61 574.014 9.41

C. Cvclotella

 

 

 

 

Source df Sum of Mean square F-test P value
squares

age (A) 1 3.821 3.821 0.162 0.6889
time (B) 1 177.989 177.989 7.535 0.0079*
AB 1 27.422 27.422 1.161 0.2855
error 61 1440.829 23.62

D. Cymbella

Source df Sum of Mean square F-test P value

squares

age (A) 1 1.799 1.799 0.144 0.7056
time (B) 1 548.931 548.931 43.965 0.0001*
AB 1 9.324 9.324 0.747 0.3909
error 61 761.629 12.486

E. Fragilaria

81

Table 14

(cont’d.).

 

 

 

 

 

 

 

 

 

 

Source df Sum of Mean square F-test P value
squares

age (A) 1 23.524 23.524 1.199 0.2778

time (B) 1 4.101 4.101 0.209 0.6491

AB 1 0.458 0.458 0.023 0.8791

error 61 1196.435 19.614

F. Gomphonema

Source df Sum of Mean square F-test P value
squares

age (A) 1 85.535 85.535 10.315 0.0021*

time (B) 1 190.709 190.709 22.999 0.0001*

AB 1 0.224 0.224 0.027 0.8700

error 61 505.812 8.292

G. Navicula

Source df Sum of Mean square F-test P value
squares

age (A) 1 46.001 46.001 5.008 0.0289*

time (B) 1 10.708 10.708 1.166 0.2845

AB 1 15.36 15.36 1.672 0.2008

error 61 560.312 9.185

H. Rhoicosphenia

Source df Sum of Mean square F-test P value
squares

age (A) 1 37.514 37.514 1.488 0.2272

time (B) 1 288.92 288.92 11.463 0.0012*

AB 1 40.001 40.001 1.587 0.2126

error 61 1537.516 25.205

82

secluded and not feeding, but this possibility does not
explain why the vast majority of juvenile stomachs were quite
full. A more likely reason for the high proportion of empty
adult stomachs may involve the time of day crayfish were
collected for analysis. Both juveniles and adults were
collected during daylight hours. Feeding periodicity data
collected during the summer of 1991 suggests that while
juvenile crayfish feed with roughly equal frequency during
daylight and after dark, adults show a strong preference for
feeding after dark. Thus, the high proportion of empty adult
stomachs on 9/7/90 may be the result of sampling adults
during their daytime feeding lull. One would not expect a
high percentage of empty stomachs in juveniles sampled during
the daytime, since they apparently readily feed during this
time. The preceding explanation does not account for why a
comparitively small percentage of adult crayfish stomachs
were empty of 7/26/90 and 8/15/90 (Figure 21).

Crayfish fragments were not seen in the 193 stomachs
examined. This was a bit surprising as many others have
reported crayfish remains in the stomachs of crayfish
(Vannote 1963, Prins 1968, Mason 1974, Capelli 1980, and
Crowns and Richardson 1988). Perhaps evidence of cannibalism
would have been found if adults were sampled for diet when
the very small 9. propinquus instars (4-5 mm carapace length)
were abundant in the stream. On 7/26/90, the earliest date
adults were sampled for diet, the early instar crayfish had
obtained a mean size of 8 mm carapace length, a size perhaps

at which they were less vulnerable to predation by adults.

83

Insects were clearly more abundant in juvenile stomachs
than adult’s on every sampling date when they were compared
(Figure 22). Considering that the ordinate axis in Figure 22
represents absolute abundance of insects and that juveniles
contained a substantially smaller total volume of material in
their guts, the relative abundance of insect material was
much greater in juveniles than adults. This may have
Kreflected a greater protein need of juveniles during a period
of active growth. Several others have also found juveniles
to eat a higher percentage of animal material than adults
(Creaser 1934b, Abrahamsson 1966, Mason 1975, Lowery and
Holdich 1988). The similar pattern of insect absolute
abundance in juveniles and adults over the 7/26-9/7/90 time
interval (Figure 22) may indicate seasonal changes in the
abundance of the insect community. The steady increase in
the percentage of juveniles eating insects over the summer
(Figure 21) could reflect an increase in the mobility and
size of juveniles, as they become more adept at a predatory
lifestyle. The fact that adult crayfish ate a higher
proportion of larger insects than juveniles (Figure 23) may
have been related to the larger mouth and chela size of
adults.

Chironomids were the most common type of insect in either
crayfish age class (Figure 23). Because the natural insect
community was not sampled in 1990 however, it is not known if
crayfish were actively selecting Chironomidae. Data from
1983 (Burton 1982-1983) on the summer insect community in the
Ford River at a site immediately downstream from the current

site suggests however, that Q. propinquus may have been

 

84

selecting particular insects to consume among a mixed
community (Table 15). The following assumption must hold if
conclusions about Q. propinquus selection in the Ford River
are to be valid: the relative abundances of insect groups
from the Ford River in 1990 at the current study site were
generally the same as the relative abundances in 1983 at the
downstream site. Aspects of stream morphology such as
substrate, stream width and depth, and flow regime were
similar when compared between 1983 and 1990 at the respective
sites. Chironomidae had a higher relative abundance in the
stomachs of juvenile and adult Q. propinquus than in the 1983
natural stream community (Figure 23, Table 15). Juveniles
may have been selecting Chironomidae, based largely on the
insect’s small size. It is not known why adult Q. propinquus
also apparently selected Chironomidae. A passive type of
selection, where Chironomidae associated with the periphyton
were ingested by grazing crayfish (Capelli 1980), may have
been operating. Trichoptera and Ephemeroptera were
underrepresented in the diet of Q. propinquus. Although the
larger size of these insects may have precluded their capture
and ingestion by juvenile crayfish, it is unclear why they
were not more common in adults.

There are a number of possible reasons for the differing
relative abundances of diatom genera in adult and juvenile
Q. propinquus (Table 12-13). Q. propinquus feed on the
periphyton by grasping material with the chela of their
periopods (personal observation). It seems possible that
benthic diatoms such as Navicula may be less common in the

diet of large crayfish (i.e. adult) than small crayfish (YOY)

85

Table 15. Relative abundance of various insect groups in the
Ford River in 19831. Values are percent relative

 

 

abundance.

Insect Percent abundance
Chironomidae 61.9
Trichoptera 14.1
Ephemeroptera 14.5
Coleoptera 1.9
Other 7.6

 

1- data from Burton (1982-1983).

86

because of their tendency to adhere flat to the substrate,
while an overstory form such as Gomphonema may be easier for
larger individuals to harvest and hence appear more
abundantly in their diets. Although Achnanthes, a benthic
form, had a greater relative abundance in juveniles than
adults (not statistically different however), Cocconeis,
another benthic form, showed no significant difference in
relative abundance between the two age classes (Table 14),
counter to the explanation proposed above. A second possible
explanation for the observed differences in diatom genera in
Q. propinquus’ diet involves active selection by crayfish for
particular kinds of diatoms. Other invertebrates, including
ciliates and snails, have been observed to selectively feed
on particular diatoms (Patrick 1970, Patrick 1977). While it
seems unlikely that adult Q. propinquus, due to their large
size, could have been actively selecting diatoms,

Q. propinquus juveniles, at 20 mm total length, may have been
small enough to actively select particular diatoms; and, in
the case of this study, may have selected Navicula. A third
possibility for the perceived differences in diatom relative
abundances in the diet of Q. propinquus involves adult and
juvenile crayfish feeding in different areas of the stream.
Diatom genera are known to favor particular substrates or
flow regimes in a stream (Patrick 1977). As many diatom
species attach to filamentous algae (Patrick 1977) there is
also the possibility that diatoms were "accidentally"
ingested by crayfish feeding on filamentous algae. Hence if

filamentous algae were selected as a food source, certain

87

diatom genera may have appeared in greater abundance in a
crayfish’s stomach than if filamentous algae were not being
eaten. As in the case of insects the natural diatom
communities were not analyzed in this study, so any
conclusions drawn about crayfish selecting particular diatom
genera are speculative.

Previous diet studies of Q. propinquus also found the
species to be omnivorous, with plant material generally
comprising the bulk of the diet. Diatoms and other algae
were the most frequent food items occurring in adults of a
population from a Wisconsin lake, although mayflies and
midges were also commonly eaten (Capelli 1980). Vegetation
was preferred to animal material in an Illinois stream
population of Q. propinquus (Bovbjerg 1952). Vannote (1963),
as in the current study, found Q. propinquus juveniles to eat

a larger percentage of animal material than adults.

CONCLUS IONS AND RECOMMENDATIONS

1) Q. propinquus was more abundant and experienced greater
net production in the hard water Ford River than in the
soft water Baraga Creek as I originally hypothesized (9.3
yl/m2 vs. 3.0 yl/m2 in 1989, 5.0 g/m2 vs. 3.0 g/m2 during
7/14/89 - 7/14/90).

2) Individual growth was greater in Baraga Creek than in the
Ford River for YOY and y1 age classes. This result was

contrary to my hypothesis that growth would be greater in

the hard water Ford River.

I” The Ford River population appeared to be fairly stable as
indicated by the RO (reproductive rate per generation) of

0.984 from life table analysis.

4) Mortality of early instar Q. propinquus from the Ford
River was considerable from the egg stage through the
first weeks of independence from the mother (Z =

0.0531/day from 5/6/90 to 7/6/90).

5) Early crayfish instars from the Ford River appeared to use
different microhabitats as they matured during their first
weeks of independence. Some seemed to shift from run-Hs

and riffles, where they were abundant early on, to run-Ts,

88

89

where most individuals were found on 7/18/90 and 7/21/90.

6).All size classes of Q. propinquus from the Ford River
predominantly ate diatoms during the summer of 1990.
Filamentous green algae and insects were also commonly

consumed.

7) Juvenile Q. propinquus from the Ford River ate more
insects than adults when compared over a common time

period (2.7 insects/stomach vs. 1.0 insects/stomach).

ED Although juvenile and adult crayfish from the Ford River
apparently consumed many of the same kind of diatoms in
similar relative proportions, Gomphonema had a greater
relative abundance in adult crayfish while Navicula had a

greater relative abundance in juveniles.

SH Q. propinquus of all age classes from the Ford River may be

selectively consuming Chironomidae.

Many of the conclusions reached in this study suggest
various avenues for further research in the systems. First
of all, it would be worthwhile to investigate the mechanisms
responsible for the individual size difference (or
growth) and the differing abundances of Q. propinquus in the
Ford River and Baraga Creek. There are several factors that
may be causing individual growth to be greater in Baraga
Creek. There may be an inverse relationship between crayfish

density and individual growth in these two systems. The

90

effects of crayfish density on growth could be examined
through the use of in situ cage experiments in which crayfish
are held in enclosures at predetermined densities along a.
continuum from low density (natural Baraga Creek crayfish
density or lower) to high density (natural Ford River
crayfish density or higher), and individual growth is
measured. If individual growth is density-dependent in these
streams, intraspecific competition for food may be involved.
Baraga individuals may have more food available to them on a
per individual basis. This possibility may be partially
addressed by sampling the environmental food supply to see
how it compares between the two rivers on a per crayfish
basis. It may also be instructive to quantitatively compare
the amount of food found in Q. propinquus from each stream.
The other related question one could ask is why
Q. propinquus is over three times more numerous in the Ford
River than in Baraga Creek. Because fecundity and
survivorship of young can have profound effects on population
size, these aspects should be compared between the two
populations. Although low water hardness may not
be affecting individual growth in this system, it may be
affecting Baraga crayfish on the population level by
interfering with reproduction or survivorship (France 1983).
Predation represents a biological mechanism that could be
affecting survivorship of Q. propinquus in either stream and
thus may ultimately be influencing population size. Various
exclosure experiments could be performed in the stream to
test this possibility. This type of experiment faces various

zoroblems however, one of which is controlling extraneous

91

variables. Habitat availability may be different between the
Ford River and Baraga Creek and this represents another
factor that could be limiting population size. This question
could be addressed by completing a detailed substrate
analysis in each stream.

In order to more adequately address the possibility that
Q. propinquus is selecting certain food types in the Ford
River a study could be undertaken where crayfish diet and the
biota are sampled at a single site over a common time period.
It would also be interesting to see if food selection is
taking place in Baraga Creek. This study could also separate
individuals based on their microhabitat to determine if
crayfish early instars are eating different things
in different microhabitats. This analysis may provide
insight as to the reasons for the apparent shifts in
microhabitat use by early instars. The microhabitat
distribution of adult crayfish could also be considered, to
determine if juveniles and adults are segregating spatially
in the stream. Finally, laboratory feeding studies could be
used in conjuction with data on Q. propinquus natural diet to
determine if the crayfish actively selects particular insect

and algal species among a mixed community.

LIST OF REFERENCES

LIST OF REFERENCES

Abrahamsson, S.A.A. 1966. Dynamics of an isolated
population of the crayfish Astacus astacus Linne. Oikos
17:96-107.

Aiken, D.E. 1967. Environmental regulations of molting and
reproduction in the crayfish, Orconectes virilis (Hagen) in
Alberta. Ph.D. thesis, University of Alberta, Edmonton,
Alta. 310 p.

Allen, K.R. 1949.“ Some aspects of the production and
cropping of fresh waters. Rpt. Sixth Sci. Congr. 1947.
Trans. R. Soc. N.Z. 77(5):222-228.

Bovbjerg, R.V. 1952. Comparative ecology-and physiology of
the crayfish Orconectes propinquus and Cambarus fodiens.
Physiological Zoology, XXV(1):34-36.

Brewis, J.M. and K. Bowler. 1983. A study of the dynamics
of a natural population of the freshwater crayfish,
Austropotamobious pallipes. Freshwater Biology 13:443-452.

Brown D.K. and K. Bowler. 1977. A population study of the
British freshwater crayfish Austropotamobius pallipes
(Lereboullet). In: O.Lindqvist (Ed.), Freshwater Crayfish
3:33-50. University of Kuopio, Finland.

Burton, T.M., R.W. Merritt, R.J. Stout, and W.W. Taylor.
1982-1983. ELF communications system ecological
monitoring program annual report for aquatic ecosystems.
Report to IIT Research Institute, Chicago, IL as part of
Contract N00039-84-C-0070 with U.S. Navy.

Burton, T.M., R.J. Stout, W.W. Taylor, D. Mullen, S. Marod,
S. Eggert, D. Repert. 1988-1989. ELF communications
system ecological monitoring program annual report for
aquatic ecosystems. Report to IIT Research Institute,
Chicago, IL as part of Contract N00039-88-C-0065 with U.S.
Navy.

Burton, T.M., R.J. Stout, W.W. Taylor, S. Winterstein, D.
Mullen, S. Marod, S. Eggert, D. Repert, J. Radhakrishnan.
1989-1990. ELF communications system ecological monitoring
program annual report for aquatic ecosystems. Report to IIT
Research Institute, Chicago, IL as part of Contract
N00039-88-C-0065 with U.S. Navy.

92

93

Burton, T.M., M.P. Oemke, and J.M. Molloy. 1991. The
effects of stream order and alkalinity on the composition
of diatom communities in two northern Michigan river
systems. (in press).

Capelli, G.M. 1975. Distribution, life history, and ecology
of crayfish in northern Wisconsin, with emphasis on
Orconectes propinquus (Girard). Ph.D. Dissertation,
University of Wisconsin. 214 p.

Capelli, G.M. 1980. Seasonal variation in the food habits
of the crayfish, Orconectes propinquus, in Trout Lake,
Vilas County, Wisconsin, USA. Crustaceana (Leiden)
38(1):82-86.

Capelli, G.M. and J.J. Magnuson. 1974. Reproduction,
molting, and distribution of Orconectes propinquus (Girard)
in relation to temperature in a northern mesotrophic lake.
In: J.W. Avault, Jr.(Ed.), Freshwater Crayfish 22415-422.
Louisiana State University.

Coffman, W.P. and L.C. Ferrington. 1984. Chironomidae.
Pages 551-652 in R.W. Merritt and K.W. Cummins (Eds.), An
introduction to the aquatic insects of North America.
Kendall Hunt Publishing Co., Dubuque, Iowa. 722 p.

Corey, S. 1987a. Comparative fecundity of four species of
crayfish in southwestern Ontario,Canada.
Crustaceana 52(3):266-286.

Corey, S. 1987b. Intraspecific differences in reproductive
potential, realized reproduction and actual production in
the crayfish Orconectes propinquus (Girard 1852) in
Ontario. The American Midland Naturalist 118(2):424-432.

Corey, S. 1988. Comparative life histories of four
populations of Orconectes propinquus (Girard 1852) in
southwestern Ontario, Canada. Crustaceana 54(2):129-138.

Creaser, E.P. 1934a. Age, growth, and sex ratios in the
crayfish, Faxonius propinquus. Pap. Mich. Acad. Sci.
19:581-585.

Creaser, E.P. 1934b. A biological survey of the Raquette
watershed. Suppl. to 23rd Ann. Rept. N.Y. Cons. Dept.
Biol. Surv. 8:158-163.

Crocker, W.C. and D.W. Barr. 1968. Handbook of the crayfish
of Ontario. University of Toronto Press. 158 p.

94

Cukerzis, J.M. 1988. Astacus astacus in Europe. In: D.M.
Holdich and R.S. Lowery (Eds), Freshwater Crayfish:
biology, management and exploitation. Timber Press,
Portland, Or. 498 p.

de la Bretonne, L.W. Jr., J.W. Avault Jr. and R.O.
Smitherman. 1969. Effects of soil and water hardness on
survival and growth of the red swamp crawfish, Procambarus
clarkii, in plastic pools. Pages 626-633 in Proceedings of
the 23rd Conference of the Southeastern Association of Game
and Fish Commissioners.

Edmunds, G.F. Jr. 1984. Ephemeroptera. Pages 94-125 in
R.W. Merritt and K.W. Cummins (Eds.), An introduction to
the aquatic insects of North America. Kendall Hunt
Publishing Co., Dubuque, Iowa. 722 p.

Fedoruk, A.N. 1966. Feeding relationship of walleye and
smallmouth bass. Journal of Fisheries Research Board of
Canada 23:941-943.

Fielder, D.D. 1972. Some aspects of the life histories of
three closely related crayfish species, Orconectes
obscurus, Q. sanborni, and Q. proninquus. The Ohio Journal
of Science 72:129-145. -

Flint, R.W. 1975. Growth in a population of the crayfish
Pacifastacus leniusculus from a subalpine lacustrine
environment. Journal of the Fisheries Research Board of
Canada 32(12):2433-2440.

Flint, R.W. and C.R. Goldman. 1975. The effects of a
benthic grazer on the primary productivity of the littoral
zone of Lake Tahoe. Limnol. Oceanog. 20:935-944.

Flint, R.W. and C.R. Goldman. 1977. Crayfish growth in Lake
Tahoe: effects of habitat variation. Journal of the
Fisheries Research Board of Canada 34:155-159.

France, R.L. 1983. Response of the crayfish Orconectes
virilis experimental acidification of a lake with special
reference to the importance of calcium. Pages 98-114 in
C.R. Goldman (Ed.), Freshwater Crayfish V. AVI Publishing
Co., Westport, Connecticut.

France, R.L. 1985. Relationship of crayfish (Orconectes
virilis” growth to population abundance and system
productivity in small oligotrophic lakes in the
experimental lakes area, northwestern Ontario. Canadian
Journal of Fisheries and Aquatic Sciences 42:1096-1102.

95

Germain, H. 1981. Flore des diatomées. Société Nouvelle
Des Editions Boubée, Paris.

Goldman, C.R. and J.C. Rundquist. 1977. A comparative
ecological study of the California crayfish, Pacifastacus
leniusculus (Dana), from two subalpine lakes (Lake Tahoe
and Lake Donner). Freshwater Crayfish 3:51-80.

Gore, J.A. and R.M. Bryant, Jr. 1990. Temporal shifts in
physical habitat of the crayfish, Orconectes neolectus
(Faxon). Hydrobiologia 199:131-142.

Growns, I.O. and A.M.M. Richardson. 1988. Diet and
burrowing habits of the freshwater crayfish, Parastacoides
tasmanicus tasmanicus Clark (Decapoda: Parastacidae).
Australian Journal of Marine and Freshwater Research
39:525-34.

Hanson, J.M., P.A. Chambers, and E.E. Prepas. 1990.
Selective foraging by the crayfish Orconectes virilis and

its impact on macroinvertebrates. Freshwater Biology
24:69-80.

Harding, J.P. 1949. The use of probability paper for the
graphical analysis of polymodal frequency distributions.
J. mar. biol. Ass. U.K. 28:141-153.

Harper, P.P. and K.W. Stewart. 1984. Plecoptera. Pages
182-230 in R.W. Merritt and K.W. Cummins (Eds.), An
introduction to the aquatic insects of North America.
Kendall Hunt Publishing Co., Dubuque, Iowa. 722 p.

Hopkins, C.L. 1966. Growth in the freshwater crayfish,
Paranephrops planifrons White. New Zealand Journal of
Science 9:50-56.

Hopkins, C.L. 1967. Growth rate in a population of the
freshwater crayfish Paranephrops planiformis White. New
Zealand Journal of Marine and Freshwater Research 1:464-
474.

Huryn, A.D. and J.B. Wallace. 1987. Production and litter
processing by crayfish in an Appalachian mountain stream.
Freshwater Biology 18:277-286.

Jones, P.D. and W.T. Momot. 1981. Crayfish productivity,
allochthony, and basin morphometry. Canadian Journal of
Fisheries and Aquatic Sciences 38:175-183.

Langlois, T.H. 1936. Notes on the habits of the crayfish,
Cambarus rusticus Girard, in fish ponds in Ohio. Ohio Dep.
Agric. Bull. No. 135, p. 3-6.

96

Lodge, D.M., T.K. Kratz and G.M. Capelli. 1986. Long-term
dynamics of three crayfish species in trout lake,
Wisconsin. Canadian Journal of Fisheries and Aquatic
Sciences 43:993-998.

Lodge, D.M., and J.G. Lorman. 1987. Reductions in submersed
macrophyte biomass and species richness by the crayfish
Orconectes rusticus. Canadian Journal of Fisheries and
Aquatic Sciences 44(3):591-597.

Lowery, R.S. and D.M. Holdich. 1988. Pacifastacus
leniusculus in North America and Europe with details of the
distribution of introduced and native crayfish species in
Europe. Pages 283-308 in D.M. Holdich and R.S. Lowery
(Eds.), Freshwater Crayfish: biology. management and
exploitation. Timber Press, Portland, Oregon. 498 p.

Mason, J.C. 1963. Life history and production of the
crayfish, Pacifastacus leniusculus trowbridgii (Stimpson),
in small woodland stream. M.S. thesis, Oregon State Univ.,
Corvallis. 204 p.

Mason, J.C. 1970. Maternal-offspring behavior of the
crayfish, Pacifastacus trowbridgii (Stimpson). American
Midland Naturalist 84:463-473.

Mason, J.C. 1974. Crayfish production in a small woodland
stream. Dep. of Environment, Fisheries and Marine Service.
Pacific Biological Stations. Nanaimo, B.C. 30 p. + 9 figs.

Mason, J.C. 1975. Crayfish production in a small woodland
stream. In: J.W. Avault (Ed.), Freshwater Crayfish 2:449-
479. Louisiana State University.

Momot, W.T. 1967. Population dynamics and productivity of
the crayfish, Orconectes virilis, in a marl lake. American
Midland Naturalist 78:55-81.

Momot, W.T. and H. Gowing. 1977a. Response of the crayfish
Orconectes virilis to exploitation. Journal of Fisheries
Research Board of Canada 34:1212-1219.

Momot, W.T. and H. Gowing. 1977b. Production and population
dynamics of the crayfish Orconectes virilis in three
Michigan lakes. Journal of Fisheries Research Board of
Canada 34:2041-2055.

Momot, W.T. and H. Gowing. 1983. Some factors regulating
cohort production of the crayfish, Orconectes virilis.
Freshwater Biology 13:1-12.

97

Momot, W.T., H. Gowing, and P.D. Jones. 1978. The dynamics
of crayfish and their role in ecosystems. American Midland
Naturalist 99:10-35.

Patrick R. 1970. Benthic stream communities. American
Scientist 58(5):546-549.

Patrick R. 1977. Ecology of freshwater diatoms and diatom
communities- Pages 284-332 in D. Werner (Ed.), The biology
of diatoms. University of California Press, Berkeley,
California. 498 p.

Penn, G.H. 1943. A study of the life history of the
Louisiana red-crayfish, Cambarus clarkii Girard.
Ecology 24:1-18.

Pennak, R.W. 1978. Freshwater Invertebrates of the United
States. J. Wiley and Sons, New York.

Prescott, G.W. 1978. How to know the freshwater algae. Wm.
C. Brown Company Publishers, Dubuque, Iowa. 293 p.

Price, J.O. and J.F. Payne. 1984. Size, age, and population
dynamics in an r-selected population of Orconectes
neglectus chaenodactvlus Williams (Decapoda, Cambaridae).
Crustaceana 46(1):30-38.

Prins, R. 1968. Comparative ecology of the crayfishes,
Orconectes rusticus and Cambarus tenebrosus, in Doe Run,
Meade County, Kentucky. Int. Rev. Gesamten Hydrobiol.,
53:667-714.

Rabeni, C.F. 1985. Resource partitioning by streamdwelling
crayfish: the influence of body size. American Midland
Naturalist 113:20-29.

Rhodes, C.P. and D.M. Holdich. 1982. Observations on the
fecundity of the freshwater crayfish, Austropotamobius
pallipes (Lereboullet) in the British Isles. Hydrobiologia
89:231-236.

 

Ricker, W.E. 1946. Production and utilization of fish
populations. Ecological Monographs 16:373-391.

Ricker, W.E. 1958. Handbook of computations for biological
statistics of fish populations. Bulletin of Fisheries
Research Board of Canada, No.119. 300 p.

Ricker, W.E. 1975. Computation and interpretation of
biological statistics of fish populations. Department of
Fisheries and Oceans, Ottawa. 382 p.

98

Robertson, J.D. 1941. The function and metabolism of
calcium in the invertebrates. Biol. Rev. 16:106-133.

Shimizu, S.J. and C.R. Goldman. 1981. Pacifastacus
leniusculus (Dana) production in the Sacramento River. In:
C. R. Goldman (Ed.), Freshwater Crayfish V. AVI Publishing
Co., Westport, Connecticut. 569 p.

Slack, K.V. 1955. A study of factors affecting stream
productivity by the comparative method. Invest. Indiana
Lakes and Streams, 4:3-47.

Sokol, A. 1988. The Australian Yabby. Pages 401-425 in
D.M. Holdich and R.S. Lowery (Eds), Freshwater Crayfish:
biology, management, and exploitation. Timber Press,
Portland, OR. 498 p.

Stein, R.A. 1977. Selective predation and optimal foraging
and predator-prey interaction between fish and crayfish.
Ecology 58:1237-1253.

Svardson, G. 1948. Stunted crayfish crayfish populations in
Swedan. Rept. Drottningholm Inst. Freshwater Res. 29:135-
145.

Tcherkashina, N. ya. 1977. Survival, growth, and feeding
dynamics of juvenile crayfish (Astacus leptodactvlus
cubanicus) in ponds and the River Don. Freshwater Crayfish
3:95-100.

Van Deventer, W.C. 1937. Studies on the biology of the
crayfish, Cambarus propinquus Girard. Ill. Biol.
Monographs 15. 67 p.

Vannote, R. 1963. Community productivity and energy flow in
an enriched warm-water stream. Ph.D. dissertation,
Michigan State University. 156 p.

Waters T. 1969. The turnover ratio in production ecology of
freshwater invertebrates. American Naturalist 103:173-185.

Weagle, K.V. and G.W. Ozburn. 1972. Observations on aspects
of the life history of the crayfish, Orconectes virilis
(Hagen) in northwestern Ontario. Can J. 2001. 50:366-370.

Webster, J.R. and B.C. Patten. 1979. Effects of watershed
perturbation on stream potassium and calcium dynamics.
Ecological Monographs 49:51-72.

Wiggins, G.B. 1977. Larvae of the North American caddisfly
genera (Trichoptera). University of Toronto Press,
Toronto.

99

Woodcock, B. and J.D. Reynolds. 1988. Reproduction in an
Irish lake population of the crayfish Austropotamobius
pallipes (Lereboullet). Freshwater Biology 19:79-86.

Wujek, D.E. and R.F. Rupp. 1980. Diatoms of the
Tittabawassee River, Michigan. A.R. Ganter Verlag K.G.,
Germany.

Zippen, C. 1958. The removal method of population
estimation. Journal of Wildlife Management 22:82-90.

9N STRTE UN V

G I . LIBRARIES
(I III) I) ()IIIVIHINIIIWUII(H)
9 17363

12 30090

HICHI
(W)
13