International Association for Ecology

Intraspecific Variation of Life History Parameters in the Terrestrial Isopod,

Armadillidium vulgare
Author(s): Ross H. Miller and Guy N. Cameron
Source: Oecologia, Vol. 57, No. 1/2 (1983), pp. 216-226
Published by: Springer in cooperation with International Association for Ecology
Stable URL: https://www.jstor.org/stable/4216949
Accessed: 17-01-2020 15:47 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Springer, International Association for Ecology are collaborating with JSTOR to digitize,
preserve and extend access to Oecologia

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 Oecologia (Berlin) (1983) 57:216-226 Oecologia
(c) Springer-Verla

Intraspecific variation of life history parameters

in the terrestrial isopod, Armadillidium vulgare
Ross H. Miller* and Guy N. Cameron
Program in Evolutionary Biology, Department of Biology, University of Houston, Houston, Texas 77004, USA

Summary. Individual populations of the terrestrial isopod, These studies have shown that adaptations within a species
Armadillidium vulgare, were studied in three Texas coastal exhibiting a broad geographic range may be as diverse as
prairie habitats: Chinese tallow forest, oak forest, and a between less geographically distributed species and that in-
Baccharis-grassland area. Time-specific life tables were traspecific differences may be genetically attuned to envi-
compared for each population to determine intraspecific ronmental differences (Ballanger 1979; Ferguson et al.
variation in life history parameters. Survivorship was grea- 1980; Krohne 1980).
test in the Baccharis area but density was lower and fluc- Armadillidium vulgare is a common North American ter-
tuated less in this area than in the tallow or oak forests. restrial isopod. Isopods are the only widespread crustacean
Density in the oak forest was higher than in the tallow to have successfully made the transition from aquatic to
or Baccharis habitat. Density in the tallow and oak forests terrestrial environments (Paris and Pitelka 1962). This has
increased sharply during spring and summer and declined been achieved by the evolution of behavioral and physiolog-
throughout the winter. Reproductive capacity of females ical responses which minimize dessication while allowing
in all three areas increased directly with body length but normal biological activities. Investigations of A. vulgare
did not differ significantly between habitats. Density and include studies of population dynamics (Paris and Pitelka
reproductive performance of isopods in the tallow and oak 1962; Paris 1963; Sorensen and Burkett 1977; Al-Dabbach
forests was asynchronous with autumn leaf fall; this asyn- and Block 1981), population regulation (Randolph 1968;
chrony may be a response to tannins leached from fallen Merriam 1970; Cromack 1967), energetics (Hubbell 1971),
leaves of these trees. Density and reproductive potential movement (Paris 1965), and reproduction (Lawlor 1976).
of isopods in the Baccharis-grassland, however, did not ex- Initital attempts to detail life table parameters for A. vulgare
hibit such marked fluctuations. Isopods in this latter habitat (Paris and Pitelka 1962) have been questioned on the
responded moderately to annual input of detritus during grounds of improper parameter estimation (Ayal and
the fall. Safriel 1979).
A. vulgare is primarily a reducer (=detritivore) organ-
ism although at times it consumes live plant material (Paris
1969; Cameron and LaPoint 1978). By virtue of its food
Introduction habits, A. vulgare occupies an important role in ecosystem
nutrient cycling. Previous investigations on the Texas coast-
A number of investigators have related life history strategies
al prairie demonstrated that A. vulgare was the predomi-
to patterns of environmental exigencies (see reviews by
nant terrestrial reducer organism (Cameron and LaPoint
Stearns 1976, 1977). Many workers classify an organism's
1978). This study demonstrated that feeding was inhibited
life history strategy by using r- and K-selection models as
by tannins in the leaves of the primary food plant (the
abbreviations for a suite of characteristics. Invertebrates
Chinese tallow tree, Sapium sebiferum) and that population
are typically considered r-selected due to combinations of
density was asynchronous with autumn leaf fall.
the traits of early maturity, large clutch size, large reproduc-
The purpose of the present study was to determine the
tive effort, small-sized young, and short life span; verte-
intraspecific variation in life history parameters of A. vul-
brates, on the other hand, are typically considered K-se-
gare on the Texas coastal prairie in three distinct habitats,
lected and possess characteristics of late maturity, small
including the tallow tree forest. These findings will form
clutch size, relatively small reproductive effort, nearly con-
a basis for future experimental studies elucidating the rea-
stant population size, and decreased death rates (Pianka
sons for specific life history variations among habitats.
1978; Stearns 1976, 1977; Nichols et al. 1976).
Utilizing life tables which include age-specific survival
and birth parameters, life history investigations have docu- Materials and methods
mented and compared species responses to environmental
heterogeneity (Stearns 1976). Recent studies have demon- Study area
strated intraspecific variation in these life history traits.
Three study sites representing the mosaic of different habi-
* Present address. Dept. of Entomology, Washington State Univ., tats on the Texas coastal prairie were chosen. Two of these
Pullman, WA 99164, USA sites, the Chinese tallow forest and the sea myrtle-grassland

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 217

area, were located on the University of Houston Coastal marsupia of gravid females and counting eggs or larvae.
Center, 56 km southeast of Houston, Texas. The third site, Females encountered with open marsupia containing nei-
a red oak forest, was located on the campus of the Universi- ther eggs or larvae were recorded but were not considered
ty of Houston at Clear Lake City, 44 km southeast of in computing the mean number of young for the various
Houston. size classes of females.
Vegetation in the tallow forest consisted predominately Several indices were computed from the basic reproduc-
of Chinese tallow trees (Sapium sebiferum) intermixed with tive counts for each habitat. Percent fertile eggs was com-
sparse hackberry (Celtis laevigata), and black willow (Salix puted for each size class by dividing the mean larvae per
nigra). Soil of the sampling area was composed of alluvial female by the mean number of eggs per female. The initial
silty clay. A 50 m2 grid, subdivided at 10-m intervals into number of offspring entering the population (No) was deter-
25 squares, was established; each 10 m2 square was num- mined by multiplying the total number of eggs produced
bered to permit randomization of sampling sites. in each size class by the percent fertility for that size class
Vegetation in the sea myrtle-grassland area consisted (to account for infertile eggs in the marsupium). This prod-
of sea myrtle (Baccharis halimifolia) interspersed with large uct was added to the total number of larvae produced in
clumps of Gulf cord grass (Spartina spartinae), dewberry each size class and summed over size classes. N. was multi-
(Rubus trivialis), bushy beardgrass (Andropogon glomera- plied by the frequency of females in each habitat (tallow=
tus), and little bluestem grass (Schizachyrium scoparium). 0.50, oak = 0.50, Baccharis= 0.55). Finally, the mean
The soil was deep alluvial clay. A grid identical to that number of female young produced per female (mj) was com
in the tallow forest was laid out in this habitat. puted by dividing No (expressed as female offspring enterin
The remaining study area was dominated by a stand the habitat) by the number of gravid females (females with
of southern red oak (Quercusfalcata) located on a plateau eggs plus those with larvae) dissected. Age at first reproduc-
of alluvial sand and clay. The sampling area was flat and tion was computed by regressing the mean number of
unobstructed, with a total area of approximately 2,500 m2. female young produced per female (m.) onto female age
Establishment of a marked grid was inpractical here so (age determined by body size, see below). However, ob-
natural landscape features were used to mark the grid pe- served age at first reproduction was estimated by assigning
rimeter and insure that successive sampling occurred within age to the smaller gravid females collected in the field.
the study area.

Life tables
Field sampling and laboratory analyses
Time-specific life tables were constructed for female isopods
Isopods were sampled biweekly in all habitats from in each habitat. Age for each size class was based on a
31 March 1978 through 23 March 1979. Ten 10-M2 squares unique growth rate determined for each habitat from suc-
were randomly chosen from the grids in the tallow and cessive field samples. In computing a growth rate for
Baccharis area; each 1 0-M2 square was considered as a pos-isopods, frequency distributions of the various age classes
sible sampling site during each sampling interval. A 0.1 --m2 were found to have unequal class intervals. Geometric
circular quadrat was randomly positioned within each means were therefore computed for each sample of isopods
I m02 square. This quadrat was forced through the plant (Cromack 1967) as:
litter layer into the uppermost layers of the topsoil, and
all isopods inside were collected. Ten random O.1_-M2 log x, = I/N E ni log x,-E at log xi
samples were also taken in the oak forest. Due to high
isopod densities in this area, all litter and the uppernost where
3 cm of soil were collected from each quadrat and sealed x=the geometric mean of the length of isopods caught
in a plastic bag. Individual quadrat samples from the oak by all sieves,
forest were pooled. N= the number of isopods sized by screening in each
Isopods were sorted and counted within 3 days after sample,
sampling with U.S. Standard Testing Sieves (W.S. Tyler, ni= the number of isopods retained by sieve i,
Inc.); mesh sizes utilized were No. 5 (4.00 mm), No. 6 xi= the mean length of individual isopods caught by sieve
(3.36 mm), No. 7 (2.83 mm), No. 10 (2.00 mm), No. 14 i, and,
(1.41 mm), No. 25 (0.71 mm), and No. 35 (0.50 mm). ai = ni/N.
Isopods placed in the uppermost sieve of the largest mesh
Mean length of isopods retained by each sieve (xi) was
diameter crawled downward (due to negative phototaxis)
determined from Cromack (1967, his Appendix I). The vari-
through the sieves until a mesh size sufficiently small pre-
cluded further descent; individual isopods were prodded ance of sieve means (xi) was homogenized by using the
geometric mean of isopod length (Snedecor 1956). The
with a soft brush to insure that they could not fit through
pooled variance for all sieves was then computed as
the last sieve. Isopod size was assigned as the smallest sieve
(Cromack 1967):
through which they could crawl. Individual isopods were
preserved in a solution of 95% ethanol and glycerin for V(5w)= /N-1 [5 _ l X/N(Z x)2].
reproductive analyses.
Reproductive adults were sexed; males were distin- Once a weighted mean had been computed for each sample
guished by the presence of elongated endopodites on the (Miller 1979), the relationship between increase in body
first two pairs of pleopods (Sutton 1972). During the breed- length and time was determined by linear regression, and
ing season, female isopods carry eggs and larvae in a ventral age was assigned to each size class uniquely for each habitat.
marsupium (Paris and Pitelka 1962; Lawlor 1976). Repro- Survivorship curves were estimated for females within
ductive information was easily collected by dissecting the each habitat. Age was assigned to the size classes and an

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
218

exponential curve fitted to the plot of ages observed in 00- a. Bacchoris

each habitat. The number of females present at 50-day in- 80-

tervals was predicted from the curve fitted to the plotted

60-
points of female age in the field. The oldest age class present
40-
in the field was estimated from survivorship curves and
20-
by examining females collected in all three habitats and
assigning a maximum age to the longest female observed.
0-
From this information, an age-specific survivorship (Ix) 1600-
schedule was computed for females in each habitat and
then combined with the age-specific female reproductive b. Oak
1200/
output per female (mr) to obtain a net reproductive rate
(RO). The intrinsic rate of increase (r) for female isopods cqJ

in each habitat was calculated by iteration of Euler's equa- 3- 800-

tion and verified by graphic techniques outlined in Caugh- uLJ
ley (1977); the finite rate of increase (A), generation time
400-
(I), and mean life expectancy of an isopod alive at time
z
interval x(er) were computed with standard methods (Mertz
1970; Krebs 1978).

Results 300- c. To low

Population demography 200,

Annual total density fluctuations were significantly differ-

ent between all three habitats (Fig. 1; Wilcoxon matched- 100-
pairs signed-rank test: tallow vs. Baccharis, z = -4.55, P <
0.01; tallow vs. oak, z= -3.04, P< 0.01; oak vs. Baccharis,
z= -4.28, P<0.01). Density in the oak forest rose much
M A M J J A S O N D J F M
higher and generally remained higher than the tallow or
Fig. 1 a-c. Population density for Armadillidium vulgare in a Bac-
Baccharis areas during the summer and fall. Density in the
charis, b oak, and c tallow habitats. *-. = total population;
tallow forest was higher than the oak or Baccharis areas
o o =juvenile population
during spring and remained higher than the Baccharis habi-
tat throughout the year. Density in Baccharis was con-
sistently lower than in either the oak or tallow forests. served changes in density (F11 15=2.53, P<0.05, mean
Isopod density was greatest in the tallow forest from monthly juvenile density; F1,1, =2.28, P>0.05, mean
June through September, ranging from 75 to over 250/M2. monthly adult density; F1, ,15=0.76, P>0.05, mean
Density decreased to about 60/M2 during late fall and early monthly total density). Juveniles were recruited into the
winter (November-February) and ranged from a low of adult population in early fall (October-November) and were
33/M2 to a maximum of 102/M2. Density increased during
present during the winter in low numbers (Fig. 1). A com-
May and reached a maximum in August; this density in- parison of density in mid-summer and early fall indicated
crease was significant (F26, 243=4.98, P<0.01). The isopod that a lower percentage (13%) of juveniles was recruited
population was composed primarily of juveniles (57%; into the oak population than into the tallow forest (20%)
?4.4 mm in length) from June through September; these or Baccharis (20%) populations. Notwithstanding the dra-
were initially released from the marsupium in mid-May. matic summer increase, isopod densities in the oak habitat
Toward the end of summer (August and September), juve- were about equal to those in the tallow and Baccharis habi-
niles were recruited into the adult population and were in- tats during winter. The sex ratio of reproductive adults re-
distinguishable by size from reproductive adults. Juveniles mained equal throughout the sampling period (X2= 1.485,
were also present in the population during the winter but P> 0.05). Juvenile sex ratios were assumed to be the same
in insignificant numbers. Annual fluctuations in juvenile as adult sex ratios.
(F26,243=6.63, P<0.01) and adult (F26 243=4.78, P< Isopod density in the Baccharis grassland habitat dis-
0.01) densities were significant. Sex ratios were equal played significant seasonal variation in total (F26,243 =
throughout the sampling period (X2 =0.181, P>0.05) for
2.22, P<0.01), adult (F26 243=2.42, P<0.01), and juvenile
reproductive adults (>5.77mm in length). Sex could not (F26, 243=3.58, P<0.01) density. Total density was signifi-
be determined among juveniles but was presumed to follow cantly less than the other two habitats. Density reached
adult ratios. a maximum of 100/M2 during early fall and ranged from
Isopod densities in the oak forest fluctuated similarly 11 /m2 to 50/M2 during the rest of the year. Breakdown
to those in the tallow forest. Density increased rapidly dur- of the population into juveniles and adults revealed greater
ing late May and reached a peak during summer (1,397/m2 variation in population structure than found in the other
in mid-June and 1,422/M2 in August). Density decreased habitats. Non-reproductive juvenile isopods (?4.4 mm in
rapidly during September and October and remained fairly length) initially appeared in the population during early
constant at 60/M2 during late fall and winter (November- June and were recruited into the adult population by late
February). Non-reproductive juveniles (?4.4 mm in length) October. The percentage of juvenile recruitment was the
comprised roughly 95%/ of the total population during the same as that in the tallow forest (20%). Maintenance of
summer breeding season and were responsible for the ob- adult density during the winter may have been due to the

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 219

(a) Bocchoris (b) Oak

705 3/31 56 7/25 45 _ 12/1 59 100 3/31 7/25 969 12/1 131
75
50- 50-
25 25-

4/14 77 - 8/4 42 12/15 24 _ 4/14 55 8/4 1422 12/15 80

4/25 22 8/26 I _ 12/29 34 _ 4/25 49 8/26 132 12/29 99

(n- 5/9 33 9/9 39 1/13 Ia (1_ 5/9 94 9/9 461 1/13 89

- 5/18 5 9/23 100 1/26 _ 5/S _o73 9/23 446 1/26 40

z
- 5/31 12 10/7 61 2/9 42 - 5/31 I 10/7 420 2/9 23

0 -
w 6/9 34 10/21 17 2/23 41 W 6/9 588 10/21 57 2/23 60
m

Z 6/19 33 11/3 16 3/9 50 Z 6/19 524 11/3 7 3/9 71

6/29 24 11/17 34 3/23 39 6/29 1397 11/17 227 3/23 57

7/12 21 +5 5 6 7 10 14 25 +55S 6 7 10 14 25 7/12 489 +5 5 6 7 10 14 25 +5 5 6 7 10 14 25

+5 5 6 7 10 14 25 +5 5 6 7 10 14 25

TOTAL LENGTH (SIEVE SIZE) TOTAL LENGTH (SIEVE SIZE)

(c) Tallow

100 3/31 69 - 7/25 253 12/1 96

75
50
25 - --
4/14 89 - 8/4 258 12/1S 69

4/25 70 8/26 199 12/29 102

0-0

n 5/9 19 9/9 129 1/13 50

>_ 5/18 3 1 9/23 255 1/26 67

z
O - 5/31 2 4 10/7 134 2/9 81

W 6/9 171 10/21 33 2/23 60

m
Fig. 2a-c. Size class distribution of A. vulgare for biweekly sam-
Z 6/19 13S 11/3 42 3/9 103 pling intervals. The proportion of individuals in each size class
is plotted for a Baccharis, b oak, and c tallow habitats during
each sampling period. Date of sample is in upper left corner and
6/29 7S 11/17 83 3/23 200
sample size in upper right corner for each histogram. Size categories
are given below each column; + 5 is the largest and 25 the smallest
7/12 81 +5 5 6 7 10 14 25 +5 5 6 7 10 14 25 size category. Reproduction in these populations is evident by size
class 7 (see Table 3). Solid bars = 1978 cohort; open bars = 1979
cohort
+5 5 6 7 10 14 25

TOTAL LENGTH (SIEVE SIZE)

presence of adult isopods a year or more in age. As in Size distribution

the tallow and oak forests, juveniles were collected during
the winter in low numbers. The sex ratio of adult isopods Body length was assigned to the various sieve classes ac-
was skewed toward females, the ratio of females to males cording to the values reported by Comack (1967) and fre-
being 0.554 (X2= 11.045, P<0.05). Juveniles were presumed quency distributions of body length for individual samples
to be the same as adult sex ratios. of isopods in all three habitats were constructed (Fig. 2).

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
220

Table 1. Regression of the geometric mean of body length onto Table 2. Adjusted age estimates (in days) and their associated 95%
time in days confidence intervals (L, and L2)

Tallow forest Oak forest Baccharis Sleve Mean Tallow Oak Baccharis
class length
Y =2.9796+0.0162X Y = 1.7083+0.0214X Y =3.6174+0.0146X (mm)
r=0.9611** r=0.9742** r=0.8629**
df =26 df= 25 df=26 + 5 10.63 472.25 416.90 480.32
LI = 389.79 LI = 359.27 LI= 332.98
L2= 556.23 L2=479.33 L2= 663.90
ANOVA 5 8.57 351.26 325.31 346.07
L1=297.06 LI=269.90 L1= 191.11
Source df SS MS Fs L2= 484.60 L2= 384.18 L2= 522.03
6 7.44 275.33 267.84 261.82
Among 2 8.4031 4.2016 7.33* L1=207.34 L1=213.15 L1= 118.73
Within 77 44.1210 0.5730 L2= 388.45 L2= 325.11 L2= 416.33
7 5.77 172.29 189.90 147.44
L, = 82.00 LI = 119.77 L1= -0.84
A posteriori testing of regression coefficients (b) * L2= 261.84 L2= 230.21 L2= 294.15
10 4.40 87.68 125.78 53.60
Oak forest Tallow forest Baccharis LI = 21.79 L1 = 70.68 L1 =-103.02
L2 =159.07 L2=181.38 L2 = 197.98
0.0214 0.0162 0.0146

* P<0.05; ** P<0.01; *** Values connected are not sign

different at P<0.05

Young isopods were first collected on 18 May in the tallow Ages for the sieve size classes in each habitat were com-
habitat and on 31 May in the oak and Baccharis habitats. puted from growth rates and mean isopod length for each
Separation and overlapping of generations are obvious for sieve class (Cromack 1967) (Table 2). Equal-sized individ-
all three habitats, but are particularly distinct for tallow uals in sieve class +5 were oldest in Baccharis, youngest
and oak. Boundaries between generations are not as well in oak, and intermediate in the tallow forest. In the smallest
defined in the Baccharis habitat (Fig. 2). Growth of the sieve class for which age could be determined (sieve
current year generation (which entered the population in class 10), equal-sized individuals were oldest in the oak
late May) probably continued until the following August, forest, youngest in Baccharis, and intermediate in the tallow
yielding an overlap of about 3 months. This overlap is forest. There was no significant difference, however, in ages
somewhat shorter in oak (about 2 months). between habitats for either the large or small sieve size
Throughout the year there were significant differences class.
in +5 sieve class densities between the habitats (F2, 80 =
7.70, P<0.01), with the oak forest averaging 10.7+ 5 sieve
Reproduction
class isopods over the year to 3.8 in the tallow forest and
1.7 in Baccharis. During the nonbreeding season, however, Breeding began earlier in the tallow and Baccharis areas
the Baccharis population contained a greater percentage than in the oak forest. Gravid females were first encoun-
(4.3%) of larger, and thus older, isopods (+5 sieve class) tered in tallow and Baccharis in mid-March; gravid females
than did either the tallow (2.1%) or oak forest (1.5%). were not found in the oak forest until April (Fig. 3). A
In April and May, the disparity between habitats in large, shorter, secondary breeding peak occurred in the tallow
mature isopods was greatest, with 124 collected in the oak forest in August and September and in Baccharis in Sep-
forest compared to only 42 in the tallow forest and 9 in tember. This pulse involved only small gravid females
Baccharis. (170.25 days old in tallow and 147.44 days old in Bacchar-
is). These females may have hatched at the onset of the
spring breeding season (mid-Feburary or March) and grew
Growth and age
to maturity by late summer. No secondary breeding pulse
Growth rates for isopods in each habitat were determined was noticed in the oak forest.
from the size distributions (Fig. 2). The growth rate of Results of the egg and larval counts from dissections
isopods in each habitat was obtained by plotting geometric of female marsupia from each habitat are summarized in
means of body length for each sample over time. Size fre- Table 3. While mean egg and mean larval production were
quencies for a single generation over one full year were consistently greater in the largest size classes, the percentage
used. The zero point for each growth line was determined of fertile eggs decreased for these classes. Females in sieve
by setting the initial day of appearance of the new genera- class 5 (8.67 mm) were the most efficient in all three habi-
tion at zero and plotting each subsequent sample mean tats, producing larvae from greater than 90% of their eggs.
along the abscissa as a positive deviation from that zero Smaller and larger females were less efficient. Females in
point. Size was regressed onto time; all regressions were the tallow forest had a higher percentage of fertile eggs
significant (Table 1). A posteriori testing of the regression for all sieve classes than did females in the other habitats,
coefficients showed the growth rates in tallow and Baccharis while females in the oak forest had the lowest percentage
were not significantly different from one another, but both of fertile eggs.
were significantly lower than oak. Differences in mean egg and larval production per

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 221

W 80- tion (Table 3). No varied across habitats with new recruits
in the oak forest being the most numerous, followed by
new recruits in the tallow and Baccharis habitats. It was
assumed, for purposes of computing No, that all larvae
were successful in leaving the marsupium and entering the
J
LI) 40-
w habitat.
w
The mean number of eggs produced per female regressed
z onto female size was significant for all habitats (Table 4).
<20-
w Slopes of the regression lines were not significantly different
between habitats, although females in Baccharis produced
a higher number of young in the largest size class than
w
-J 60- in the other habitats. There was a significant relation be-
tween isopod size and larvae number for all habitats (Ta-
w ble 5). When the mean number of larvae per female was
LL40
N, 4 2 M - regressed onto female length, the slopes of the regression
lines in the tallow and oak forests were not different from
20- one another but were significantly less than that of Bacchar-
is (Tables 3 and 5). This relationship is clarified when the
z
mean number of female young produced per female (mr)
is compared across habitats (Table 3). There is little varia-
X. M AM J J A SON D J F M
tion between habitats in the mean young per female (pooled
Fig. 3. The mean number of eggs and the mean number of larvae
across size classes; F2, 9=0.008; P>0.05); however, a
per female over all sampling periods. Size classes are pooled and
female measuring 10.63 mm (+ 5) in Baccharis averaged
means for each habitat are plotted separately: * = tallow, o = oak,
34.35 female young per brood while the same sized females
and m = Baccharis. Dashed line indicates a missing sampling inter-
val in the tallow and oak forests produced 28.5 and 29.08
female young per brood, respectively. Females smaller than
10.63 mm in Baccharis generally produced fewer female
female were not significantly different across habitats (egg young than their counterparts in the tallow and oak forests.
production + 5, F2,70=2.56, P>0.05; 5, F2,92=2.94. P>Thus, increased larval production in Baccharis contributed
0.05; 6, F2,11=2.76, P>0.05; larval production: +5, to increased mx in +5 individuals. Additionally, after as-
F2, 37=0.13, P>0.05; 5, F2 63= 1.42, P > 0.05) (Table 3). signing age to each size class in each habitat, a regression
However, isopods in the oak habitat produced significantly of the mean female young per female (mx) onto age revealed
more eggs (F2 205=25.91, P<0.01) and larvae (F2 138= no significant differences between habitats (Table 6).
6.30, P<0.01) (Fig. 3). Females of sieve class 6 in the tallow Computed ages of first reproduction indicate that
forest produced significantly more larvae on the average females in Baccharis should begin reproducing at
than did those of the same size in Baccharis (F1, 12 =13.81, 137.90 days, behind females in the other habitats (Table 6).
P<0.01). However, there were no sieve class 6 females Computed age of first reprodction for females in the tallow
found with larvae in the marsupia in the oak forest. forest was 103.85 days and in oak was 120.07 days. How-
The initial number of young recruited into the popula- ever, as no females smaller than sieve class 7 were reproduc-
tion (No) was computed from total egg and larvae informa- tive in the field, observed ages at first reproduction were

Table 3. Reproductive parameters from marsupia dissections for the three coastal prairie habitats. No =initital number of new female
offspring entering the population; mx= mean number of female young produced per female

Sieve Total Total No. of Y x eggs/9 % fertile Total No. of Y x- larvae/9 No mx

class eggs with + SE eggs larvae with + SE
eggs larvae

Tallow +5 56 1,024 17 60.24+ 13.88 94.62 798 14 57.00+27.02 883.45 28.50

5 212 1,870 46 40.65 +10.11 97.12 987 25 39.48 + 10.35 1,401.57 19.74
6 250 247 8 30.88? 7.66 84.55 235 9 26.11 + 7.85 221.92 13.05
7 470 283 22 12.86? 2.55 81.03 219 21 10.42? 4.77 224.16 5.22
Pooled 988 3,424 93 36.04+ 18.40 89.33 2,239 69 32.45+22.11 2,731.10 16.63

Oak +5 141 3,560 51 69.80+17.97 82.11 1,322 22 60.09+17.30 2,122.56 29.08

5 181 1,231 30 41.03 +10.01 89.50 1,143 31 36.87 +12.50 1,122.37 18.40
6 140 65 2 32.50+ 7.79 - 0 0 - 32.50 16.25
7 444 20 1 20.00 55.00 11 1 11.00 11.00 5.50
Pooled 906 4,876 84 58.05 + 21.29 75.59 2,476 54 45.85 +19.01 3,288.43 17.31

Baccharis + 5 29 523 7 74.71 + 16.79 82.99 248 4 62.00 + 21.49 377.85 34.35
5 141 698 20 34.90+ 9.16 93.70 327 10 32.70+ 4.76 543.49 18.12
6 141 87 4 21.75+ 3.86 66.67 87 6 14.50+ 2.43 80.34 8.03
7 136 7 1 7.00 - 0 0 - 3.88 3.88
Pooled 447 1,315 32 41.09+21.75 81.12 662 20 33.10+ 19.25 1,005.56 16.10

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
222

Table 4. Regression of the mean number of eggs per female onto

Table 7. Life table for A. vulgare in three Texas coastal prairie
female length in millimeters habitats

a) Tallow forest
Tallow forest Oak forest Baccharis

Y = -42.4022 Y =- 41.7273 Y = -78.041 1 Age I. l/m. ex xlXmin

x (days)
+ 9.663X + 10.1I642X + 13.8584X
r =0.9988* r =0.9799* r =0.9764*
0 1.0000 0 0 138.45 0
df=2 df=2 df = 2
50 0.5807 0.6849 0.3977 152.32 19.885
100 0.3926 3.7794 1.4838 151.34 148.838
ANOVA 150 0.2655 6.8739 1.8250 149.85 273.750
200 0.1795 9.9684 1.7893 147.69 357.860
250 0.1214 13.0629 1.5858 144.44 396.450
Source df Ss Ms Fs
300 0.0821 16.1575 1.3265 139.65 397.950
350 0.0555 19.2520 1.0685 132.61 373.975
Among 2 20.5726 10.2863 1.94 ns
400 0.0375 22.3465 0.8380 122.27 335.200
Within 8 42.4856 5.3107
450 0.0254 25.4410 0.6462 106.69 290.790
* P<0.05 500 0.0172 28.5355 0.4908 83.72 245.400
550 0.0116 31.6300 0.3669 50.00 201.795
3,041.893

Ro= 11.8185
Table 5. Regression of the mean number of larvae per female onto female offspring/female/generation
r = 0.009595 female offspring/female/50 days
female length in millimeters T= 257.3840 days

Tallow forest Oak forest Baccharis

b) Oak forest

Y=- 44.7848 Y =- 44.4135 Y - 96.3702

Age I mx Ixm ex xlxmx
+ 9.6093X ? 9.5110OX + 14.8957X
x (days)
r=0.9994** r =0.9987* r=0.9999**
df=2 df=2 df=2
0 1.0000 0 0 122.30 0
50 0.4897 0.7243 0.3511 149.16 17.555
ANOVA
100 0.3253 4.1612 1.3536 147.76 135.360
150 0.2182 7.5981 1.6579 145.74 248.685
200 0.1464 11.0350 1.6155 142.69 323.100
Source df Ss Ms Fs
250 0.0983 14.4720 1.4226 138.05 355.650
300 0.0659 17.9089 1.1802 131.34 354.060
Among 2 121.7530 60.8765 42.27*
350 0.0442 21.3458 0.9435 121.27 330.225
Within 3 4.3236 1.4412 400 0.0297 24.7827 0.7360 106.06 294.400
450 0.0199 28.2196 0.5616 83.67 252.720
500 0.0134 31.6565 0.4242 50.00 212.100
A posteriori testing of regression coefficients (b) *
2,523.8S5
Oak forest Tallow forest Baccharis Ro= 10.2462 female offspring/female/generation
r = 0.009447 female offspring/female/50 days
9.5110 9.6093 14.8957 T=246.3211 days

* Values connected are not signif-icantly different (P < c) Baccharis

0.05); ** P<0.01
Age I. mx lxm. ex xIxmx
x (days)
Table 6. Regression of mean female young produced per female
0 1.0000 0 0 143.66 0
(m.s) onto age in days
50 0.5114 0 0 183.14 0
Tallow forest Oak forest Baccharis 100 0.3752 2.5374 0.9520 181.48 95.200
150 0.2753 6.0893 1.6764 179.19 251.460
200 0.2019 9.6411 1.9465 176.15 389.300
Y = -.8.0851 Y =- 12.3252 Y =- 13.0947
250 0.1481 13.1930 1.9539 171.98 488.475
+ 0.0779X + 0.0985X + 0.0950X
r =0.9993* r =0.9863* r =0.9735* 300 0.1087 16.7449 1.8202 166.19 546.060
df=2 df=2 df=2 350 0.0797 20.2967 1.6176 158.47 566.160
400 0.0585 23.8486 1.3951 147.78 558.040
450 0.0429 27.4004 1.1755 133.33 528.975
ANOVA
500 0.0315 30.9523 0.9750 113.49 487.500
550 0.0231 34.5042 0.7970 86.58 438.350
600 0.0169 38.0560 0.6431 50.00 385.860
Source df Ss Ms Es
4,735.380
Among 2 10.4909 5.2455 1.059 ns
Ro = 14.9523 female offspring/female/generation
Within 6 29.7330 4.9555
r = 0.008541 female offspring/female/SO days
T= 316.6991 days

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 223

summer, and another buildup in late summer-early fall.

Table 8. Regression of the logarithm (base 10) of female isopod
survivorship (I,x) onto time in days Density in both habitats decreased during late fall and re-
mained low until the following spring. These density trends
Tallow forest Oak forest Baccharis were similar to those described for A. vulgare in habitats
where environments were very different from the Texas gulf
Y=0.8960-0.0079X Y==0.8009-0.0083X Y = 0.7699 - 0.0064X
coast (Paris and Pitelka 1962; Paris 1963; Sorensen and
r _-0.9997** r= -0.9981 ** r=0.9975**
Burkett 1977; Al-Dabbach and Block 1981). Densities in
df=10 df=9 df=11
Baccharis were much lower and showed less fluctuation;
the most noticable increase was during late summer with
ANOVA a smaller increase during early spring. Sex ratio was approx-
imately equal in the tallow and oak habitats, but was
Source df SS MS Fs skewed toward females in Baccharis.
Juveniles entered the population at about the same time
Among 3 0.7600 0.3800 64.41 ** in oak and tallow forests, but were several weeks later in
Within 31 0.1829 0.0059 Baccharis. Juveniles represented the smallest proportion in
the Baccharis population suggesting adults were longer
lived. Correspondingly, generations were least separated in
A posteriori testing of regression coefficients (b) *
Baccharis (and overlapped by at least 3 months). The pro-
Oak forest Tallow forest Baccharis portion of juveniles in the oak and tallow habitats was
highest during summer (95 and 60%). Recruitment of juve-
0.0079 0.0083 0.0064 niles into the population was during late fall for all habitats,
but a higher proportion were recruited in the tallow habitat
* Values connected are not significantly different (P< (20%).
0.05); ** P<0.01 Growth rates were fastest in oak. These differences may
reflect differences in the length (age) of isopods leaving
the marsupium caused by differential marsupia growth
estimated as 172.29 days in tallow, 189.80 days in oak, and rates or variation among habitats in the length of time an
147.44 days in Baccharis (Table 2). egg or larva spends in the marsupium. However, on the
average, equal-sized individuals (+5 age class) were oldest
in Baccharis and youngest in oak. While these differences
Life table parameters
were not significant, the presence of younger-aged isopods
There was little difference in the intrinsic rate of increase in oak was consistent with the smaller generation overlap
(r) between habitats, but r was slightly lower for isopods and greater growth rate in this habitat. The presence of
in the Baccharis habitat (Table 7). However, Ro, T, and older isopods in Baccharis also was consistent with the lack
ex were greater in Baccharis than oak or tallow forests dueof density fluctuations and increased overlap of genera-
to increased survival in Baccharis (Fig. 5; see below). Breed- tions. Growth rates were not directly comparable to those
ing was initiated during the 50-day cohort in the tallow of other investigators because their sampling intervals and
and oak forests but was delayed until the 100-day cohort age class designations differed (Paris and Pitekla 1962;
in the Baccharis habitat. Saito 1969; Sorensen and Burkett 1977).
Survivorship curves were computed for each habitat Reproductive phenology was also different among habi-
where lx was the proportion of females alive at time t-1 tats. Continuous breeding during the year was not observed
which survived to time t. The logarithm of lx was regressed in any habitat, nor has it been reported from other A. vul-
onto time in days to test for differences in survival (Table 8). gare studies. Gravid females were found in early fall and
A posteriori testing of the resultant regression coefficients isolated group of juveniles were collected during the winter,
revealed no significant differences between tallow and oak but these were few in number and did not greatly affect
forests but both were significantly larger than Baccharis. size class distributions. Reproduction was slightly later in
Increased survival in Baccharis and the corresponding in- the oak forest, and there was no second reproductive pulse;
creased generation time (Table 7c) resulted in a higher Ro both the tallow and Baccharis habitats had a late fall repro-
than the other habitats in spite of the delay in breeding. ductive burst (involving only small gravid females). The
The linearity of the plot of days against the logarithm of mean number of eggs and larvae per female was lower in
survivorship indicated an equal rate of mortality with tallow and Baccharis than oak; however, the percent fertile
respect to age for female isopods within a given habitat was higher in these habitats than oak. The number of new
(Deevy 1947). females entering the population (No) was much lower in
Baccharis due to the combination of low average number
of larvae per female and low percent fertile. However, when
D'iscussion
corrected for population density (females only), reproduc-
Populations of A. vulgare exhibited different life history tive output (ms) was not significantly different among habi-
patterns in each of the three habitats examined during this tats.
study. Isopod populations in the oak and tallow forest ex- The relationship between isopod size and number of
hibited many similarities, but populations from Baccharis larvae produced was significantly greater in the Baccharis
were strikingly different. Density patterns were similar in habitat. Larger females in Baccharis produced more female
the tallow and oak habitats except that numbers were much young than similar-sized females in the other habitats, yield-
higher in the oak forest. Density trends in both habitats ing greater mx for large-sized female isopods in Baccharis.
reflected an early summer increase, a trough during mid- Fewer of these large females were present in the population

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
224

Table 9. Life table parameters for various animals. r =intrinsic rate of natural increase; R= net reproductive rate; = finite rate of
increase (r = ln A)

Species r Ro Authority
Per female per female per female
per day per generation per day

Tribolium castaneum* 0.101 275.0 1.110 Leslie and Park 1949

Calandra oryzae* 0.109 113.6 1.114 Birch 1948
Pediculus humanus* 0.111 30.9 1.118 Evans and Smith 1953
Microtus agrestis* 0.0125 5.9 1.013 Leslie and Ranson 1940
Rattus novegicus* 0.0147 25.7 1.015 Leslie et al. 1952
Porcellio scaber* 0.0101 47.4 1.0101 Brereton 1955
Armadillidium vulgare 0.000055 1.01 1.00005 Paris and Pitelka 1962
Armadillidium vulgare* 0.0214 28.0 1.0202 Cromack 1967
Armadillidium vulgare
Tallow forest 0.00019 11.82 1.00019 present study
Oak forest 0.00019 10.25 1.00019 present study
Baccharis 0.00017 14.95 1.00017 present study

* Values are based on observations under optimal conditon

during the spring breeding season, but their survival was habitat and did not begin until the 100-day cohort whereas
extended compared to females in the other habitats. This reproduction began with the 50-day cohort in the other
equalized the reproductive value of females in like age habitats. The increased output of larger females in the Bac-
classes among habitats. Females in the Baccharis habitat charis habitat, combined with increased longevity
bred sooner (younger) than females in the other habitats, (Table 7c), resulted in a much longer generation time.
whereas females in the oak habitat bred at the oldest age Isopods in the Baccharis habitat had a life span of about
and did not have a secondary fall breeding pulse. Life expec- 600 days while isopods in the oak and tallow forests lived
tancies for females, when considered with age of initial re- about 500 and 550 days, respectively. Paris and Pitelka
production, indicated that females in all three habitats may (1962) estimated California populations of A. vulgare lived
breed more than once during their lifetime. Females in Bac- up to 4 years whereas Saito (1969) and Sorensen and
charis who initially bred during the secondary breeding Burkett (1977) estimated a longevity of 2 and 3 years, re-
pulse in late summer may live to breed during the onset spectively.
of the following spring pulse. Females in the tallow and Values of Ro greater than 1.0 and r greater than 0.0
oak forest may breed successively during the sustained indicated that populations in all three habitats were expand-
breeding pulses observed in those habitats. ing slightly. The intrinsic rate of increase (r) may also serve
Laboratory populations of A. vulgare maximize their re- as an indicator of habitat suitability and can thus be used
productive performance at high densities by increasing the to compare different populations of the same species under
percentage of gravidity in large (sieve class 5) females (Ran- different environmental conditions (Birch 1948). Since r was
dolph 1968). At lower densities there was also a high per- so similar among habitats, the three populations of isopods
centage of gravidity among females but, because fewer of in this study were either responding to different microenvi-
the gravid females were of the largest size class, total repro- ronmental parameters in basically the same manner or were
ductive performance was less. Similarly, densities of large responding identically to a few macroenvironmental factors
breeding females were higher in the oak forest than in either characteristic of the Texas coastal prairie (among habitats).
the tallow forest or Baccharis habitat; however, the mean Table 9 compares life history parameters for A. vulgare with
production of female young per female (ml) was less in several other organisms. While r computed for A. vulgare
the oak forest. This suggests that increased density during in this study was smaller than rm values computed by
summer and fall in the oak forest may skew the distribution Cromack (1967), it was similar to that computed for the
of females toward the larger size classes. Alternatively, peri- field data of Paris and Pitelka (1962). Paris and Pitelka's
ods of prolonged drought (two weeks or more without rain) (1962) study was conducted in a California grassland habi-
during the summer or fall may result in mortality among tat where summer aridity is a major limiting factor to
all age classes through dessication (a similar mechanism isopod growth. Their populations were nearly stable (r:
for summer population depression has been reported by 0.0; Ro 0 1.0) whereas our values indicated expanding pop-
Paris and Pitekla 1962). The sandy clay soil of the oak ulations.
forest drained more rapidly than the clay soil of the tallow In spite of the apparent similarity among life table pa-
or Baccharis habitats. This situation might explain the rameters, population responses were sufficiently different
sharp decrease in juvenile recruitment in the oak forest inamong habitats to suggest isopod populations in the tallow
comparison with the tallow forest. and oak habitats may be responding to different factors
Life table parameters revealed additional differences than isopod populations in the Baccharis habitat. Popula-
among habitats. Survival was significantly higher in the tions in the Baccharis habitat maintained a low, relatively
stable population density during the year with a slight re-
Baccharis habitat; the resultant increased mx of the larger-
productive increase during the summer. Generations in this
sized females led to a larger Ro for the Baccharis population
and an r nearly the same as for the other habitats habitat showed more overlap than the other habitats, indi-
(Table 7c). Reproduction was also delayed in the Baccharis viduals lived longer, reproduction was delayed, and older

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
 225

females had increased reproductive output. All of these Acknowledgements. We thank R.D. Akre, E.H. Bryant, D.L.
characteristics suggest that the Baccharis habitat may be Jameson, and L.J. Lester for reviewing the manuscript. W.B.
more stable environmentally. Kincaid, T. McCombs, S.R. Spencer, and L. Williams helped with

Several factors may explain the differences in life history various phases of this project. Financial assistance from the Uni-
versity of Houston Department of Biology, the University of
responses among these coastal prairie habitats. Environ-
Houston Coastal Center, and the University of Houston Student
mental factors, particularly temperature (Hubbell 1971;
Research Program was appreciated. Computer time was provided
McQueen 1976a, 1976b; McQueen and Carnio 1974) and by the University of Houston Computing Center.
seasonal drought (Paris 1963), affect terrestrial isopod pop-
ulations. Al-Dabbach and Block (1981), however, found
that while weather affected individual growth rates, it was References
not adequate to account for population responses. Shachak
(1981) demonstrated that isopod species living in severe Al-Dabbach KY, Block W (1981) Population ecology of a terrestri-
al isopod in two breckland grass heaths. J Anim Ecol 50:61-77
environments have traits of r-selected species whereas
Ayal Y, Safriel U (1979) On the misuse of life tables in determining
species living in less severe environments have traits of K-
demographic parameters of natural field populations. Bull Ecol
selected species. Miller (1979) analyzed environmental vari-
Soc America 60:12-16
ables on the Texas coastal prairie, and concluded that Ballanger RE (1979) Intraspecific variation in demography and
A. vulgare in Baccharis and oak habitats responded more life history of the lizard, Sceloporus jarrovi, along an altitudinal
to rainfall than temperature and that low temperature in gradient in southeastern Arizona. Ecology 60:901-909
oak litter may affect breeding. The importance of these Birch LC (1948) The intrinsic rate of increase of an insect popula-
trends must be verified experimentally before the contribu- tion. J Anim Ecol 17:15-26
tion of microenvironment to life history patterns is under- Brereton J Le G (1955) A study of some factors controlling the
stood. Alternatively, isopods could respond to the annual population of some terrestrial isopods. PhD dissert., Oxford
Univ., Oxford, England
pulse of energy (fallen leaves) in these habitats. The re-
Cameron GN, LaPoint TW (1978) Effects of tannins on the decom-
sponse to the input of detritus is not a major factor differen-
position of Chinese tallow leaves by terrestrial and aquatic in-
tiating the habitats because grasses in the Baccharis habitat vertebrates. Oecologia (Berl.) 32:349-366
also die during the fall, but there was not a dramatic sea- Caughley G (1977) Analysis of vertebrate populations. John Wiley
sonal population increase as in the other habitats. and Sons, London
Another explanation for the life history differences may Cromack K (1967) Some consequences of food quality for Armadil-
be related to the fact that tallow and oak are both decidu- lidium vulgare populations (Isopoda: Oniscoidea). MS thesis,
ous, tannin-containing trees whereas Baccharis is an ever- Univ. Texas, Austin
Deevey ES (1947) Life tables for natural populations of animals.
green, non-tannin containing, shrub. A. vulgare will not eat
Quart Rev Biol 22:283-314
tallow leaves containing tannins, but will consume leaves
Feeny PP (1970) Seasonal changes in oak leaf tannins and nutrients
from which tannins have been leached (Cameron and
as a cuase of spring feeding by winter moth caterpillars. Ecology
LaPoint 1978). Since these leaves form the major food 51:565-581
source it is not surprising to see that major isopod density Evans FC, Smith EF (1953) The intrinsic rate of natural increase
increases in the oak and tallow habitats are asynchronous for the human louse, Pediculus humanus (L.). Amer Nat
with leaf fall; leaf fall in tallow forests is during October- 86:299-310
November (Cameron and LaPoint 1978) and in red oak Ferguson GW, Bohlen CH, Woolley HP (1980) Sceloporus undula-
forests is in November-December (Harcombe, pers. comm.) tus: comparative life history and regulation of a Kansas popula-
tion. Ecology 61:313-322
while isopod populations increase dramatically the follow-
Hubbell SP (1971) Of sowbugs and systems: the ecological bioener-
ing spring and summer (Fig. 1). Similarly increases in repro-
getics of a terrestrial isopod. In: Patten BC (ed) Systems analy-
duction are evident in the oak and tallow habitats the spring
sis and simulation ecology, vol 1. Academic Press, New York,
following leaf fall and after winter rains have leached p 269-324
tannins from the leaves; hence, an asynchronous response Krebs CJ (1978) Ecology, the experimental analysis of distribution
between leaf fall and isopod population increase ensues. and abundance. Second edition. Harper and Row, New York
Population fluctuation is much less dramatic in Baccharis; Krohne DT (1980) Intraspecific litter size variation in Microtus
in fact, the major increase in this habitat is during the fall. californicus. II. Variation between populations. Evolution
The resultant population characteristics in tallow and oak 34:1174-1182

can be likened to an r-selected species or a species in an Lawlor LR (1976) Molting, growth, and reproductive strategies
in the terrestrial isopod, Armadillidium vulgare. Ecology
unstable habitat; conversely, population characteristics in
57:1179-1194
Baccharis, where there is no seasonal flush of tannins, are
Leslie PH, Park T (1949) The intrinsic rate of natural increase
similar to a K-selected species or a species in a more stable of Tribolium castaneum Herbst. Ecology 30:469-477
habitat (Fig. 1). Similar responses to tannins (as well as Leslie PH, Ranson RM (1940) The mortality, fertility and rate
other secondary plant compounds) have been reported for of natural increase of the vole (Microtus agrestis) as observed
other systems (Feeny 1970; Rhoades 1979). in the laboratory. J Anim Ecol 9:27-52
A. vulgare is an introduced but nonetheless widespread Leslie PH, Venables UM, Venables ISV (1952) The fertility and
species in North America. Its occurrence in different geo- population structure of the brown rat (Rattus norvegicus) in
corn-ricks and some other habitats. Proc Zool Soc London
graphic localities would suggest modifications in life history
122:187-238
attributes as responses to local environmental conditions.
McQueen DJ (1976 a) Porcellio spinicornis Say (Isopoda) demogra-
Our results indicate that such responses occur among popu-
phy. II. A comparison between field and laboratory data. Cana-
lations of terrestrial isopods on the Texas coastal prairie. dian J Zool 54:825-842
Further experimental work is needed, however, to clarify McQueen DJ (1976b) Porcellio spinicornis Say (Isopoda) demogra-
and characterize the role of microenvironmental and chemi- phy. III. A comparison between field data and the results of
cal factors in life history responses by the terrestrial isopod. a simulation model. Canadian J Zool 54:2174-2184

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms
226

McQueen DJ, Carnio JS (1974) A laboratory study of the effects Pianka ER (1978) Evolutionary ecology. Second edition. Harper
of some climatic factors on the demography of the terrestrial and Row, New York
isopod (Porcellio spinicornis Say). Canadian J Zool 52:599-611 Randolph JC (1968) The effects of density and the relationship
Merriam HG (1970) Sensitivity of terrestrial isopod populations to cover in Armadillidium vulgare populations (Isopoda: On-
(Armadillidium) to food quality differences. Canadian J Zool iscoidea). MS thesis, Univ. Texas, Austin
49:667-674 Rhoades DF (1971) Evolution of plant chemical defense against
Mertz DB (1970) Notes on methods used in life-history studies. herbivores. In: Rosenthal GA, Janzen DH (eds) Herbivores:
In: Connell JH, DB Mertz, Murdoch WW (eds) Readings in Their interaction with secondary plant metabolites. Academic
ecology and ecological genetics Harper and Row, New York, Press, New York, p 3-54
p4-17 Saito S (1969) Energetics of isopod populations in a forest of
Miller RH (1979) A comparison of life-history strategies and envi- central Japan. Res Popul Ecol 11: 229-258
ronmental effects upon populations of Armadillidium vulgare Shachak M (1980) Energy allocation and life history strategy of
in three habitats. MS thesis, Univ. Houston, Houston, Texas the desert isopod H. reaumuri. Oecologia (BerI.) 45:404-413
Nichols JD, Conley W, Batt B, Tipton AR (1976) Temporally Snedecor GW (1956) Statistical methods applied to experiments
dynamic reproductive strategies and the concept of r- and K- in agriculture and biology. Iowa St. Univ. Press, Ames
selection. Amer Natur 110:995-1005 Sorensen EMB, Burkett RD (1977) A population study of the
Paris OH (1963) The ecology of Armadillidium vulgare (Isopoda, isopod. Armadillidium vulgare, in northeastern Texas. South-
Oniscoidea) in California grassland. Ecol Monogr 33:1-20 west Nat 22:375-388
Paris OH (1965) Vagility of P32-labeled isopods in grassland. Ecol- Stearns SC (1976) Life-history tactics: a review of the ideas. Quart
ogy 46:635-648 Rev Biol 51:3-47
Paris OH (1969) The function of soil fauna in grassland ecosystems. Stearns SC (1977) The evolution of life history traits: a critique
In: Dix RL, Beidleman RG (eds) The grassland ecosystem: of the theory and a review of the data. Ann Rev Ecol Syst
a preliminary synthesis. Range Science Dept, Science Series 8:145-171
No. 2, Colorado State University, p 331-360 Sutton SL (1972) Woodlice. Ginn and Co., Ltd., London
Paris OH, Pitelka FA (1962) Population characteristics of the ter-
restrial isopod Armadillidium vulgare in California grassland.
Ecology 43:229-248 Received September 23, 1982

This content downloaded from 198.102.159.147 on Fri, 17 Jan 2020 15:47:31 UTC
All use subject to https://about.jstor.org/terms