Click here for the original journal page (in Acrobat pdf format).

The text below is grayed out because it is not intended to be read. It is a necessarily imperfect OCR of the original and is only used by a search engine.

Journal of The Lepidopterists' Society

Volume 48                                    1994                                    Number 2

Journal of the Lepidopterists' Society 48(2), 1994, 85-105

SAMPLING STRATEGIES FOR ESTIMATING MOTH

SPECIES DIVERSITY USING A LIGHT TRAP IN A

NORTHEASTERN SOFTWOOD FOREST

A. W. Thomas

Canadian Forest Service, Natural Resources Canada, P.O. Box 4000, Fredericton, New Brunswick E3B 5P7, Canada

AND

G. M. Thomas

University of New Brunswick, Fredericton, New Brunswick E3B 6C2, Canada

ABSTRACT. A 22-watt black-light trap was operated for 29 nights within a forest canopy in the Maritime Lowlands Ecoregion of the Acadian Forest. The species-abundance frequency distribution (pattern of species abundance) was a good fit to the log series model and this model was used for subsequent data analysis. No single-night sample adequately estimated the log series alpha index of diversity based on the total catch; some sampling effort was required each night. Each night's catch was separated into 16, 30-minute samples. The alpha index of diversity for the summed catch for each time-period was compared with the overall alpha based on the total catch. A strategy that involved operating the trap for just a 1-hour period each night had no effect on the pattern of species abundance and gave a value for alpha equal to that obtained by operating the trap for an 8-hour period each night. This strategy reduced the catch from 6088 to 971 moths and the number of species from 255 to 161. Processing costs associated with the larger sample and any possible negative effect on the moth population caused by removal trapping were greatly reduced. This new sampling strategy is thus useful for comparing indices of species diversity between several sites when data are collected simultaneously, but is of limited use for species-inventory studies.

Additional key words: species-abundance distribution, 30-minute samples, log-series model, partial-night sampling.

In recent years, the challenge to maintain biodiversity on this planet has become a major public concern. Most attention focuses on Neotropical ecosystems (Mares 1992). However, the importance of maintaining Canada's biodiversity was addressed in Environment Canada's Green Plan (Hyslop & Brunton 1991), and the launching, in 1991, of "Canadian Biodiversity" produced by the Canadian Centre of Biodiversity at the Canadian Museum of Nature lends credence to the recent

86

Journal of the Lepidopterists' Society

national interest in biodiversity. The values of maintaining current biodiversity have been stated by many authors, and summarized by Ehrlich (1990) into ethical, aesthetic, economic, and 'ecosystem services/ Salwasser (1990) added the legal obligation for conserving biological diversity. Intimately linked with the concept of maintaining biodiversity, and especially protection of areas rich in species, is the need for a "quick and dirty survey to chart biodiversity of the planet'' (Roberts 1988), a view reiterated by Ehrlich (1992). The 'quick and dirty' approach does not advocate poor science; rather, it recognizes that the scope of diversity from individual gene systems through populations of species, communities, ecosystems, and ultimately all life in the biosphere (Wilson 1988) cannot be addressed in the short-term. It suggests that studies should be focused on certain taxonomic groups over an extensive area. The hope is that areas with many species or high endemism in the selected groups will reflect similarly high values for other groups (Roberts 1988). Because of logistic and knowledge constraints, the number of species within a community can be determined for only a limited number of taxonomic groups.

This study addresses just one segment of biodiversity, i.e., the diversity of moths in a single ecosystem. Diversity is used here to mean the number of species and their relative abundance (Magurran 1988), and to prevent ambiguity we will always use 'species diversity' where appropriate. Relative abundance is considered in the form of species-abundance frequency distributions, which show the relationship between the abundance of individuals and the number of species possessing that abundance (May 1975); abbreviated in this paper as the pattern of species abundance. The ecosystem studied is one locality in the Maritime Lowlands Ecoregion of the Acadian Forest (Loucks 1962).

The use of the moth community, in the 15 families used in this study (see Appendix), as an exemplar of the species diversity of this ecosystem has advantages that include the relative ease of identification at the species level, the somewhat standardized sampling methodology (Williams 1951, Williams et al. 1955, Taylor & French 1974, Bowden 1982), and the high correlation of insects, in general, with the spatial, architectural, and taxonomic diversity of plants (Southwood et al. 1979).

No community consists of species of equal abundance (Magurran 1988). It is normally the case that the majority of sp>ecies are rare while a number are moderately common with the remaining few species being very abundant (Williams 1964, May 1975, Pielou 1975, South-wood 1978, Magurran 1988). Within this general distributional form, communities have characteristically different patterns of species abundance which remain stable despite changes in species composition (Pielou 1975, May 1976, Kempton 1979). The pattern of species abundances

Volume 48, Number 2

87

at a site allows for comparison with similar sites that have different mixes of species, and a change in the pattern of abuindance at one site has been shown to be a useful indicator of environmental disturbance (Kempton & Taylor 1974, Taylor et al. 1978, Kempton 1979).

Four main species-abundance models (the geometric series, the logarithmic series, the log normal distribution, and MacArthur's broken stick model) have been developed to describe species diversity in terms of an 'index parameter' as well as the pattern of species abundance. In addition there are several non-parametric indices based on the proportional abundances of species (May 1975, Southwood 1978, Magurran 1988).

The log series model was the first to describe the pattern of species abundance (Fisher 1943). Since then it has been found to have a wide application for catches of many invertebrates, e.g., moths in light traps (Williams 1943, 1945, 1964, Taylor & Brown 1972, Taylor & French 1974, Kempton & Taylor 1974, Taylor et al. 1976, 1978, Taylor 1986), Ichneumonidae (Owen & Chanter 1970), cockroaches (Wolda 1983), Psocoptera (Broadhead & Wolda 1985), Hymenoptera (Noyes 1989), and the community of phytophagous arthropods on apple (Brown & Adler 1989). Its wide applicability is because it is based on the abundances of the species with medium abundance rather than the very abundant and very rare species (Taylor et al. 1976, Kempton 1979, Brown & Adler 1989).

The log series is a simple two-parameter model, with two denning multispecies population parameters, chi and alpha. Chi is devoted to sample characteristics and varies with sample size as it is a function of the mean number of individuals per species. Alpha is independent of sample size and characterizes the required population quality (Kempton & Taylor 1974). Fisher's (1943) initial suggestion was that alpha might be useful as a measure of 'species richness' when comparing samples. Williams (1943) suggested that the parameter alpha be known as a community's 'index of diversity.' Later he recognized that this term was applicable to other functions having the same properties and referred to Fisher's alpha as 'diversity calculated on the basis of the logarithmic series' (Williams 1964). The log series model can be derived from two statistics, S, the total number of species, and N, the total number of moths. It is a discontinuous frequency series with an infinite number of terms:

n1? riix/2, nlX2/S, n1x3/4,

where nx is the number of species with 1 individual and successive terms with 2,3,4, etc. individuals, and x (chi) is a constant < 1 (Williams 1947).

88

Journal of the Lepidopterists' Society

The log normal model was compared with the log series model by Kempton and Taylor (1974) in an analysis of moth catches from light traps at 18 sites for four successive years in an attempt to quantify intuitively recognized properties of habitats. This comparison found that samples from stable environments were best fitted by the log series whereas those from highly perturbed sites better fitted the log normal. Their overall conclusion was that alpha of the log series was the superior diversity discriminant, which they defined as a population parameter that behaves consistently within a stable population and responds to changes within, and to differences between, environments (see also Taylor et al. 1976).

The Simpson-Yule diversity statistic and the Shannon-Weaver information statistic (both non-parametric indices) were compared to the log series alpha index of diversity by Taylor et al. (1976) using 10 years of light-trap data at one site. Although the log series model was not the ideal description of the pattern of species abundance, the site's environmental stability was better reflected by alpha than by either of the other two statistics.

One constraint with using the moth community as an exemplar of species diversity for an ecosystem is the logistics of sorting, counting, and identifying all the individuals in the sample (Taylor 1979). For example, a one-night catch from one trap in Kenya exceeded 6.7 kg (Taylor et al. 1979); 26,300 moths were captured in one light-trap during a nine-month period at Rothamsted (UK) (Williams 1964); 113,256 moths were taken in one light-trap in one year in Kansas (USA) (Williams 1945); 6088 moths were taken in one trap in one month (this study). Methods for reducing the size of catches were detailed by Taylor and Brown (1972), and for subsampling from large catches by Taylor et al. (1979). The objectives of this study were: (1) to describe the species-abundance frequency distribution and determine the log series alpha index of diversity, for moths captured in a light-trap in a within-canopy site of a predominantly balsam fir forest during the flight season of the major forest pest, spruce bud worm (Chorutoneura fumiferana (Clemens) (Tortricidae)), and (2) to develop a sampling strategy that reduced the catch to a minimum without causing significant loss of information, measured as no change in the pattern of species abundance and a reduction in the alpha value of 5% or less.

Methods

Moth collection and identification. Beginning on 21 June 1990 (day 1) and ending on 30 July (day 40), one 22-watt black-light trap (Universal Light Trap, Bioquip Products, California) was operated in the Peter Brook study area of the Acadia Forest Experiment Station near

Volume 48, Number 2

89

Fredericton, New Brunswick, Canada. For a variety of reasons, full-night trap data are available for only 29 of the potential 40 nights. Intensive studies on the population dynamics of spruce budworm have been in progress at this site since 1986. The physical characteristics and vegetation of the site have been described (Lethiecq & Regniere 1988). Briefly, the study area is composed of 77% balsam fix, Abies balsamea (L.) Miller (Pinaceae), 12% red maple, Acer rubrum L. (Aceraceae), and eight other tree species. However, the surrounding area is heterogenous and within a 10-km radius contains mixed forest, lakes, streams, sphagnum bogs, large clear-cuts, and roadsides.

The trap, with the lamp at 6.4 m above the ground, was on a platform, 3 x 1.5 m, on a tower within the closed crowns of balsam fir trees; the otherwise touching branches were trimmed to leave a clearing of 3 x 1.5 m. A blue plastic sheet, 1.8 x 2.4 m, was stretched above the platform at a height of 2.4 m above the lamp. This sheet made direct observation of the light impossible from above, although the reflection of the light off of the foliage of the adjacent trees gave a glow to the immediate area which was obvious from the ground.

The trap was equipped with an automatic time-interval collecting device (King et al. 1965, Smith et al. 1973). Each night's total catch consisted of 16, 30-minute sequential samples, beginning with time-period 1 from 2130-2200 h and ending with time-period 16 from 0500-0530 h. On 21 June, day 1, sunset was at 2120 h and sunrise the following morning at 0536 h; on 30 July, day 29, sunset was at 2058 h and sunrise the following morning at 0606 h. At the latitude of New Brunswick, the sky is noticeably lighter about 30 min before sunrise and remains light for 30 min after sunset.

The moths were killed with 1,1,1 trichloroethane. Moths were stored at — 17°C until identified and counted. Most specimens were identified with the aid of the literature and confirmed by consulting the Forest Insect and Disease Survey (FIDS) Reference Collection, Canadian Forest Service, Fredericton, which contains specimens identified by the Biological Resources Division (BRD) of the Centre for Land and Biological Resources Research, Ottawa. Genitalia mounts of specimens were made when identification was uncertain. A further 52 species of geometrids were identified by Klaus Bolte and 81 species of noctuids by Don Lafontaine, both at BRD. All moths in the following families were identified to species and counted: Hepialidae; Sesiidae; Cossidae; Limacodidae; Thyatiridae; Drepanidae; Geometridae, except for Eu-pithecia; Lasiocampidae; Saturniidae; Sphingidae; Notodontidae; Arc-tiidae; Lymantriidae; and Noctuidae. In addition, all specimens of spruce budworm (Tortricidae) were counted. Moths belonging to other families were not identified or recorded. Publications used for species identi-

90

Journal of the Lepidopterists' Society

fication were Forbes (1954), McGuffin (1967, 1972, 1977, 1981), Rock-burne and Lafontaine (1976), Ferguson (1978), Morris (1980), McCabe (1980), Covell (1984), Laplante (1985), Lafontaine (1987), and Lafontaine and Poole (1991).

Species-abundance frequency distribution. The numbers of species having abundances of 1, 2, 3, . . ., 724 moths (based on the total catch) were compared with the expected numbers from the log series model (Williams 1947) for goodness-of-fit, using the cfoi-square test (Owen & Chanter 1970, Kempton & Taylor 1974, Taylor et al. 1976, Broadhead & Wolda 1985, Magurran 1988, Noyes 1989, Basset & Kitching 1991). The observed abundances covered a large range, 1-724 moths per species, and because many of these 724 abundance classes were zero (e.g., abundance classes 31 and 36 each had two species, but no species had just 32, 33, 34, or 35 moths and thus classes 32-35 were zeros) the abundance classes were grouped into 10 new abundance classes of approximately equal range on the logarithmic (base 2) scale (Kempton & Taylor 1974, Kempton 1975, Taylor et al. 1976). Because the abundance class having >511 moths had an expected frequency of <1 species, this class was pooled with the preceding class to give an expected frequency of > 1 species; resulting in just nine abundance classes. This grouping and pooling of abundance classes (see Table 2) resulted in the data set meeting the requirements for the c/n-square analysis in that no more than 20% of the classes had an expected frequency of <5 species (1 out of 9 did) and no expected frequency was <1 (Zar 1984).

Index of diversity. For the purpose of this study, the 29-night sample from the trap was taken to be the population being sampled. The log series alpha index of diversity was determined after rearranging equations (7) and (8) of Williams (1947) to obtain:

[1]                           (Sx/-ln(l - x)) - N(l - x) = 0

and solving for x using MathCad (1991), and then solving [2] for alpha:

[2]                                      alpha = N(l - x)/x

This value based on the single 29-night sample was termed 'the overall alpha.'

Strategies to reduce sample size. Three data manipulations were employed to determine a strategy that would reduce the size of the sample and thus reduce processing costs and lessen the possible effect of removal trapping on the moth population.

Single-night samples. The first attempt at a sampling strategy was to determine alpha for each night's catch and to compare each value with the overall alpha. Such a strategy would certainly reduce sample size, but it was not known how representative such an alpha based on

Volume 48, Number 2

91

one night's catch would be of the overall alpha basesd on the total 29-day catch.

'Replicated9 single-time-period samples. As each night's catch consisted of 16 sequential 30-minute samples, there were 16 single-time-period samples, with each sample 'replicated' for 29 nights. The alpha index of diversity was calculated for each pooled time period (e.g., all the moths trapped during time period 1 were pooled) and compared with the overall alpha. If an index equivalent to the overall index could be estimated from a single 30-minute sample taken each night for 29 nights, significant saving in processing costs would occur, i.e., 1 x 29 = 29 samples instead of 16 x 29 = 464.

Truncated samples. This strategy was based on the results of the single-time-period analysis. As certain time periods gave low alpha values, it was argued that these time periods could be eliminated (thus reducing the number of samples, the number of moths, the processing costs) without significant loss of information. Two sub-strategies were employed. The first, termed early truncation, was to discard cumulative sequential time periods from the entire data set beginning with all 29 samples from time period 1, then all 58 samples from time period 1 + time period 2, etc. After 15 truncations only the data set from time period 16 remained. The alpha index of diversity was calculated from the data set remaining after each truncation and compared with the overall alpha to determine the percentage change. Also after each truncation, the pattern of species abundance was compared with that from the log series model using the deviance chi-squ&re values (Kemp-ton & Taylor 1974). The second sub-strategy, termed late truncation, was similar to early truncation except that all 29 samples from time period 16 were first discarded, then all 58 samples from time periods 16 + 15, etc. Combining selected data sets that remained after early and late truncation (effectively a double-ended truncation) gave several sampling strategies that met the goal of reducing sample size without compromising the value for alpha or the pattern of species abundance. The durations for these sampling strategies are shown in Table 1.

Results

Totals of 6088 individual moths representing 255 macrolepidoptera species in 15 families were identified from the 29-night catch (see Appendix).

Species-abundance distribution and index of diversity. The pattern of species abundance is shown in Table 2. In general, the number of species in the abundance classes decreased as the abundance increased. Most species (52) were in the first abundance class, making this the

92                                                 Journal of the Lepidopterists' Society

commonest class. The apparent paradox is that members of these species were rare with just one moth in each species (see Appendix). The fewest species (3) were in the largest abundance class making this the rarest class but members of these species were abundant (>255 moths in each, see Appendix). Also shown in Table 2 are the frequencies expected from the log series model. The similarity between observed and expected appears close and is confirmed as being a good fit by the deviance chi-square value of 8.6. The 5% critical value of the c/ii-square distribution with 7 df is 14.1 indicating that the log series model provides a good description of the data. The overall alpha index of diversity was 54. Single-night samples. The number of moth species and individuals trapped in a single night ranged from a low value of 30 moths in 18 species to a high value of 548 moths in 88 species. Values for alpha

Table 2. Species abundance frequency distribution of a moth catch in the Acadia Forest Experiment Station compared with expected frequencies from the log series model.

Total chi-square = 8.6, P < 0.5, P > 0.1, df = 7. Last abundance class pooled with previous class to meet requirements of chi-square test (see Methods).

Volume 48, Number 2

93

CD

O

cr

80

TO

60

50

30

i ' i ■ i ' i ' i ' i ' i ""n ' i ■ i ' i ' i |—i ^r^

2 4 6 8 10 12 14 16 18 20 22 24 26 28 SAMPLE NIGHT

Fig. 1. Alpha index of diversity for single-night catches as a percentage of the overall alpha based on the total catch.

fluctuated wildly between 12 and 40 with no meaningful trend and never closely approaching the overall alpha (Fig. 1). It was apparent that no single-night sample could be used to estimate the index of diversity and thus no pattern of species abundance was determined.

'Replicated' single-time-period samples. For any single 'replicated' time period (consisting of 29, 30-minute samples) the total number of moths trapped ranged between 48 and 627 and the total number of

110

^ 100

i— o

•—       90

u_

O

lu       80

CD

■<

^        70

30

4

r

7 IME

n i r-

8 9 10 PERIOD

11 12 13 14 15 16

Fig. 2. Alpha index of diversity for single-time-period catches, averaged over 29 nights, as a percentage of overall alpha based on the total catch. Time periods are sequential 30-minute periods starting at 2130-2200 h and ending at 0500-0530 h.

94

Journal of the Lepidopterists' Society

i i i i i i i i i i i-------1-------1-------r~

2 3 4 5 6 7 8 9 10 11 12 13 14 15 TIME PERIOD AT START OF TRAPPING

16

i r~

2 3

TIME

~i--------1 i--------r

4 5 6 7

PERIOD AT

—i-------1 i-------1------1------r

8 9 10 11 12 13

END OF TRAPPING

14 15

FlG. 3. Percentage change in the alpha index of diversity compared to the overall alpha: (A) when trapping starts with successively later time periods and ends with time period 16 (0500-0530 h); (B) when trapping starts at time period 1 (2130-2200 h) and end at successively later time periods.

species trapped ranged between 23 and 132. The values for alpha for the replicated' single-time-period samples started low in the first part of the night, rose rapidly to a maximum during the middle part of the night and then decreased towards dawn (Fig. 2). For time period 5 the value for alpha was 102% that of the overall alpha. However, this datum was an outlier that did not follow the trend and it was not thought

Volume 48, Number 2

95

7000 6000 5000

u_ 4000 O

llj 3000 m

2000 1000

4 0

-4

\— "8

O

LJJ CD

O

~~i i i r

2 3 4 5

7

~l-----1-----1-----1-----1-----[

9 10 11 12 113 14

SAMPLING STRATEGY

Fig. 4. Percentage change (empty rectangles) in the alpha index of diversity relative to the overall alpha (strategy 1), and number of moths trapped (solid rectangles) for the various sampling strategies. See methods and Table 1 for explanation of sampling strategies.

prudent to accept this single time period as representative of the overall alpha.

Truncated samples—early truncation. Discarding the data in time periods 1 through 6 had no significant effect on alpha determined from the remaining data set (Fig. 3A). That is, if the light trap had begun operating at 0030 h, start of period 7, and had run until 0530, alpha would have been within 5% of the value obtained by starting the light trap at 2130 h. Also, early truncation of time periods 1 through 6 had no effect on the pattern of species abundance in the remaining data set (time-periods 7-16), chi-square = 10.2, 7 df (P > 0.1).

Late truncation. Discarding the data in time periods 16 through 9 had no significant effect on alpha based on the remaining data set (Fig. 3B). That is, if the light trap had begun operating at 2130 h and had run until 0130 h, the end of period 8, alpha would have been within 5% of the overall alpha. Also, late truncation had no effect on the pattern of species abundance in the remaining data set (time periods 1-8), c/n-square = 7.7, 6 df (P > 0.1).

Double-ended truncation. Several combinations of early- and late-truncation provided 13 sampling strategies that reduced the sampling period and reduced the number of moths trapped These strategies (Table 1) had no significant effect on alpha and did not compromise the pattern of species abundance. No calculated c/ii-square value, comparison between observed pattern of species abundance and expected

96

Journal of the Lepidopterists' Society

pattern from the log series model, was significant (P > 0.05). When the sampling strategies were arranged in a sequence of decreasing sampling periods (Fig. 4), the downward trend in the number of moths trapped and the insignificant effect on alpha became obvious. The most cost-effective strategy was a 1-h sample obtained nightly from 0130-0230 h (strategy 14, Fig. 4) that resulted in a total sample of 971 moths in 161 species giving an alpha value of 55.

Discussion

The inadequacy of a single-night sample to estimate accurately the alpha index of diversity for moths caught during a one-month period was observed by Williams (1943, 1964) in England. Nightly samples during the month of July gave alpha values that varied from 42-81% of the overall alpha based on the total catch for the whole month, with no evidence of any regular trend (Williams 1964, Table 67). Taylor and Brown (1972) presented data from two traps for nine days in July in Kenya. Single-night alpha values ranged from 30.5-80% of the two overall values. Our data showed a similar random pattern with nightly values varying from 22-74% of the overall alpha value. Even when Williams (1964, Table 67) calculated diversity on a weekly basis, the average weekly value for alpha was only 77% of the monthly value. These data support our conclusion that some sampling effort is required nightly throughout the duration of the calendar dates of interest.

Taylor (1979) commented on the cost-efficiency of sampling insects and the advantages of an attractant trap, such as a light trap, in selecting specific taxa. He also noted that, when used to control pest-species, light traps have as an objective the removal of as large a proportion of the population as possible. However, when used as a monitoring tool, the objective is to affect the population as little as possible compatible with obtaining adequate numbers for analysis. As mentioned in the introduction, large samples have problems associated with the cost of sorting, identifying, counting, and data handling. Reducing sample size by subsampling from a larger sample has drawbacks (Taylor et al. 1979). Taylor and Brown (1972) tried several methods to decrease the size of the moth catch in light traps that included obscuring the light with black paint, changing the source of illumination (different bulb types), and changing the direction of illumination. These methods reduced the size of the catch, but had no effect on the alpha index of diversity. They did not examine the effect on the pattern of species abundance.

Our technique of a short-time-period 'replicated' nightly sample to determine the alpha index of diversity without changing the pattern of species-abundance is new. Because it results in a relatively small sample, it has the advantage of affecting the moth population much

Volume 48, Number 2

97

less than a full-night sample. It appears to be of use for determining the moth species diversity of several sites simultaneously which otherwise could not be considered because of processing costs associated with the usually large catches in light traps.

There are no alpha index of diversity values from eastern North American forests in similar latitudes with which to compare the alpha value obtained in this study. The long-recognized latitudinal and longitudinal gradients in species diversity (Pianka 1966, Smith 1980, see also refs. in Magurran 1988) preclude comparison of the alpha value from this study with alpha values for moth species diversity in two mid-west American states (Williams 1945) and England (Taylor et al. 1978).

Acknowledgments

We thank Georgette Smith for allowing access to the FIDS reference collection and helping with identification. We are especially grateful to Don Lafontaine and Klaus Bolte for their skill in the identification of many specimens that were in very poor condition.

Literature Cited

Basset, Y. & R. L. Kitching. 1991. Species number, species abundance and body length of arboreal arthropods associated with an Australian rainforest tree. Ecol. Entomol. 16:391-402.

BOWDEN, J. 1982. An analysis of factors affecting catches of insects in light-traps. Bull. Entomol. Res. 72:535-556.

Broadhead, E. & H. Wolda. 1985. The diversity of Psocoptera in two tropical forests in Panama. J. Anim. Ecol. 54:739-754.

BROWN, M. W. & C. R. L. ADLER. 1989. Community structure of phytophagous arthropods on apple. Environ. Entomol. 18:600-607.

Covell, C. V. 1984. A field guide to the moths of eastern North America. Houghton Mifflin, Boston. 496 pp.

EHRLICH, P. R. 1990. Habitats in crises: Why we should care about the loss of species. For. Ecol. Mngt. 35:5-11.

---------- 1992. Population biology of checkerspot butterflies and the preservation of

global biodiversity. Oikos 63:6-12.

FERGUSON, D. C. 1978. Noctuoidea: Lymantriidae. In Dominick, R. B. et al. (eds.),

The Moths of America North of Mexico, including Greenland. Fasc. 22.2. E. W.

Classey Ltd. & The Wedge Entomological Research Foundation, London. 110 pp. 9

plates + notes and indices. Fisher, R. A. 1943. A theoretical distribution for the apparent abundance of different

species, Part 3, pp. 54-57. In Fisher, R. A., A. S. Corbet & C. B. Williams (eds.), The

relation between the number of species and the number of individuals in a random

sample of an animal population. J. Anim. Ecol. 12:42-58. Forbes, W. T. M. 1954. Lepidoptera of New York and neighboring states. Part III

Noctuidae. Memoir 329, Cornell University Agricultural Experiment Station. 433 pp. Hyslop, C. & D. Brunton. 1991. Chapter 6. Wildlife: Maintaining biological diversity.

In Burnett, J. A. & P. Logan (eds.), State of Canada's environment. Ministry of Supply

and Services, Canada. Ottawa. KEMPTON, R. A. 1975. A generalized form of Fisher's logarithmic series. Biometrika

62:29-38.

---------- 1979. The structure of species abundance and measurement of diversity. Biometrics 35:307-321.

98

Journal of the Lefidopterists' Society

KEMPTON, R. A. & L. R. TAYLOR. 1974. Log-series and log-normal parameters as

diversity discriminants for the Lepidoptera. J. Anim. Ecol 43:381-399. King, E. W., C. D. Pless & J. K. Reed. 1965. An automatic sample-changing device

for light-trap collecting. J. Econ. Entomol. 58:170-172. Lafontaine, J. D. 1987. Noctuoidea: Noctuidae, Noctuinae (Part—Euxoa). In Dom-

inick, R. B. et al. (eds.), The Moths of America North of Mexico, including Greenland.

Fasc. 27.2. The Wedge Entomological Research Foundation, Washington. 237 pp. Lafontaine, J. D. & R. W. Poole. 1991. Noctuoidea: Noctuidae, Plusiinae. In Dom-

inick, R. B. et al. (eds.), The Moths of America North of Mexico, including Greenland.

Fasc. 25.1. The Wedge Entomological Research Foundation, Washington. 182 pp. Laplante, J.-P. 1985. Papillons et chenilles du Quebec et de Test du Canada. France-

Amerique, Montreal, Canada. 280 pp. Lethiecq, J-L. & J. Regniere. 1988. Comparative descriptions of the physical characteristics and vegetation of six sites used by the Canadian Forestry Service in the

study of spruce budworm population dynamics. Information Report LAU-X-83. Lau-

rentian Forestry Centre, Canadian Forestry Service. 46 pp. LOUCKS, O. L. 1962. A forest classification for the Maritime Provinces. Proc. Nova

Scotia Inst. Sci. 25:85-167. Magurran, A. E. 198&. Ecological diversity and its measurement. Princeton University

Press, New Jersey. 125 pp. Mares, M. A. 1992. Neotropical mammals and the myth of Amazonian biodiversity.

Science 255:976-979. MathCad. 1991. Version 3.0. MathSoft Inc., Cambridge, Massachusetts. May, R. M. 1975. Patterns of species abundance and diversity, pp. 81-120. In Cody,

M. L. & J. M. Diamond (eds.), Ecology and evolution of communities. The Belknap

Press of Harvard University, Cambridge, Massachusetts.

---------- 1976. Patterns in multi-species communities, pp. 142-162. In May, R. M. (ed.),

Theoretical ecology. Blackwell, Oxford.

McCabe, T. L. 1980. A reclassification of the Polia complex for North America (Lepidoptera: Noctuidae). New York State Museum, Bulletin 432. Albany, New York. 141 pp.

McGuffin, W. C. 1967. Guide to the Geometridae of Canada (Lepidoptera). I. Subfamily Sterrhinae. Entomol. Soc. Canad., Memoir 50. 67 pp.

---------- 1972. Guide to the Geometridae of Canada (Lepidoptera). II. Subfamily En-

nominae. 1. Entomol. Soc. Canad., Memoir 86. 159 pp.

---------- 1977. Guide to the Geometridae of Canada (Lepidoptera). II. Subfamily En-

nominae. 2. Entomol. Soc. Canad., Memoir 101. 191 pp.

---------- 1981. Guide to the Geometridae of Canada (Lepidoptera). II. Subfamily En-

nominae. 3. Entomol. Soc. Canad., Memoir 117. 153 pp.

MORRIS, R. F. 1980. Butterflies and moths of Newfoundland and Labrador—The macro-

lepidoptera. Ag. Canad., Ottawa. Publ. #1691, 407 pp. Noyes, J. S. 1989. The diversity of Hymenoptera in the tropics with special reference

to Parasitica in Sulawesi. Ecol. Entomol. 14:197-207. Owen, D. F. & D. O. Chanter. 1970. Species diversity and seasonal abundance in

tropical Ichneumonidae. Oikos 21:142-144. Pianka, E. R. 1966. Latitudinal gradients in species diversity: A review of concepts.

Amer. Nat. 100:33-46. PiELOU, E. C. 1975. Ecological diversity. John Wiley & Sons, New York. 165 pp. ROBERTS, L. 1988. Hard choices ahead on biodiversity. Science 241:1759-1761. Rockburne, E. W. & J. D. Lafontaine. 1976. The cutworm moths of Ontario and

Quebec. Ag. Canad., Ottawa. Publ. #1593, 164 pp. Salwasser, H. 1990. Conserving biological diversity: A perspective on scope and approaches. For. Ecol. Mngt. 35:79-90. Smith, J. S., J. M. Stanley & R. N. Gupton. 1973. Time-interval collecting device for

insect traps. Agriculture Research Service, USDA. Publication ARS-S-20, 5 pp. Smith, R. L. 1980. Ecology and field biology. 3rd ed. Harper & Row, New York. 835 pp.

Volume 48, Number 2

99

Southwood, T. R. E. 1978. Ecological methods. 2nd ed. Chapman & Hall, London.

524 pp. Southwood, T. R. E., V. K. Brown & P. M. Reader. 1979. The relationship of plant

and insect diversities in succession. Biol. J. Linn. 12:327-348 Taylor, L. R. 1979. The Rothamsted insect survey—An approach to the theory and

practice of synoptic pest forecasting in agriculture, pp. 148-185. In Rabb, R. L. &

G. G. Kennedy (eds.), Movement of highly mobile insects: Concepts and methodology

in research. North Carolina State University, Raleigh.

---------- 1986. Synoptic dynamics, migration and the Rothamsted insect survey. J. Anim.

Ecol. 55:1-38.

Taylor, L. R. & E. S. Brown. 1972. Effects of light-trap design and illumination on samples of moths in the Kenya highlands. Bull. Entomol. Res. 62:91-112.

Taylor, L. R. & R. A. French. 1974. Effects of light-trap design and illumination on samples of moths in an English woodland. Bull. Entomol. Res. 63:583-594.

Taylor, L. R., R. A. Kempton & I. P. Woiwod. 1976. Diversity statistics and the log-series model. J. Anim. Ecol. 45:255-272.

Taylor, L. R., R. A. French & I. P. Woiwod. 1978. The Rothamsted insect survey and the urbanization of land in Great Britain, pp. 31-65. In Frankie, G. W. & C. S. Koehler (eds.), Perspectives in urban entomology. Academic Press, New York.

Taylor, L. R., E. S. Brown & S. C. Littlewood. 1979. The effect of size on the height of flight of migrant moths. Bull. Entomol. Res. 69:605-609.

Williams, C. B. 1943. Results obtained by means of a light-trap at Rothamsted, Part 2, pp. 44-53. In Fisher, R. A., A. S. Corbet & C. B. Williams (eds.), The relation between the number of species and the number of individuals in a random sample of an animal population. J. Anim. Ecol. 12:42-58.

---------- 1945. Recent light trap catches of Lepidoptera in U.S.A. analysed in relation

to the logarithmic series and the index of diversity. Ann. Entomol. Soc. Amer. 38: 357-364.

---------- 1947. The logarithmic series and its application to biological problems. J. Ecol.

34:253-272.

---------- 1951. Comparing the efficiency of insect traps. Bull. Entomol. Res. 42:513-517.

---------- 1964. Patterns in the balance of nature. Academic Press, London. 324 pp.

Williams, C. B., R. A. French & M. M. Hosni. 1955. A second experiment on testing

the relative efficiency of insect traps. Bull. Entomol. Res. 46:193-204. Wilson, E. O. 1988. The current state of biodiversity, pp. 3-18. In Wilson, E. O. &

F. M. Peter (eds.), Biodiversity. National Academy Press, Washington, D.C. Wolda, H. 1983. Diversity, diversity indices and tropical cockroaches. Oecologia 58:

290-298. Zar, J. H. 1984. Biostatistical analysis. 2nd ed. Prentice-Hall, New Jersey. 718 pp.

Received for publication 10 June 1993; revised and accepted 9 October 1993.

100

Journal of the Lepidopterists' Society

Appendix. Species list with numbers of moths and extreme dates of capture.

Volume 48, Number 2

101

Appendix. Continued.

Xanthotype urticaria Swett                                25 June-21 July

Pero morrisonaria (Hy. Edw.)                            21 June-4 July

Nacophora quernaria (J. E. Smith)                     26 June-18 July

Campaea perlata (Gn.)                                       21 June-25 July

Tacparia atropunctata (Pack.)                            27 June

Tacparia detersata (Gn.)                                     21-26 June

Homochlodes fritillaria (Gn.)                             21-29 June

Metanema inatomaria Gn.                                 21 June-29 July

Metanema determinata Wlk.                             18-22 July

Metarranthis amyrisaria (Wlk.)                         21-28 June

Metarranthis hypocharia (H.-S.)                        25 June

Anagoga occiduaria (Wlk.)                                 21 June

Probole amicaria (H.-S.)                                      21 June-9 July

Plagodis serinaria H.-S.                                       27 June

Plagodis phlogosaria (Gn.)                                  26-29 June

Plagodis alcoolaria (Gn.)                                     21 June-4 July

Caripeta divisata Wlk.                                        25 June-29 July

Caripeta piniata (Pack.)                                     21 June-23 July

Caripeta angustiorata Wlk.                                17-24 July

Besma endropiaria (G. & R.)                              21-29 June

Sicya macularia (Harr.)                                       16-25 July

Eusarca confusaria Hbn.                                     16 July

Tetrads cachexiata Gn.                                       21 June-2 July

Nematocampa resistaria (H.-S.)                         17-29 July

Nemoria mimosaria (Gn.)                                   14-15 July

Cyclophora pendulinaria (Gn.)                          21 June-25 July

Scopula cacuminaria (Morr.)                              18 July

Scopula limboundata (Haw.)                              25 June-24 July

Dysstroma citrata (L.)                                        25-28 June

Dysstroma walkerata (Pears.)                             21 June-14 July

Dysstroma hersiliata (Gn.)                                  15-29 July

Eulithis explanata (Wlk.)                                    16-29 July

Ecliptopera silaceata albolineata (Pack.)           21 June

Hydriomena perjracta Swett                              21-25 June

Hydriomena renunciata (Wlk.)                          21 June-29 July

Hydria undulata (L.)                                          14 July

Rheumaptera hastata (L.)                                   16 July

Rheumaptera subhastata (Nolcken)                   26 June-8 July

Mesoleuca ruficillata (Gn.)                                 25 June

Spargania magnoliata Gn.                                  14 July

Perizoma basaliata (Wlk.)                                  25 July

Xanthorhoe abrasaria congregata (Wlk.)          25 June-13 July

Xanthorhoe iduata (Gn.)                                     12 July

Xanthorhoe ferrugata (Cl.)                                 21 June-4 July

Xanthorhoe lacustrata (Gn.)                               16 July

Hydrelia lucata (Gn.)                                          26 June-18 July

Hydrelia inornata (Hulst)                                   25 June-17 July

Eubaphe mendica (Wlk.)                                    17-20 July

Horisme intestinata (Gn.)                                  29 June

Lobophora nivigerata Wlk.                                 26 June-29 July

Lasiocampidae

Malacosoma disstria Hbn.                                    9-29 July

Malacosoma americanum (F.)                            15-25 July

5 13

4 12

1

8

5 11

4

3

1

1 15

3

7

4 78

9 22

6

3

1 39 39

2 47

1 36

2

4

3 55

1

2 792

2

1

2

1

1

1

8

1

3

1 11

9

4

1 63

136

27

102

Journal of the Lepidopterists' Society

Volume 48, Number 2

103

Appendix. Continued.

Cycnia tenera Hbn.                                             26 June

Ctenucha virginica (Esp.)                                   10-19 July

Lymantriidae

Dasychira plagiata (Wlk.)                                   26 June-25 July Leucoma salicis (L.)                                              4-19 July

Noctuidae

Idia americalis (Gn.)                                           21 June-29 July

Idia aemula Hbn.                                                 14-20 July

Idia rotundalis (Wlk.)                                         21 July

Zanclognatha pedipilalis (Gn.)                           18-24 July

Zanclognatha protumnusalis (Wlk.)                  12-22 July

Zanclognatha cruralis (Gn.)                                21 June-29 July Palthis angulalis (Hbn.)                                        8-29 July

Bomolocha baltimoralis (Gn.)                             26 June-20 July

Lomanaltes eductalis (Wlk.)                              25 June

Spargaloma sexpunctata Grt.                             21 June-21 July

Pangrapta decoralis Hbn.                                   21 June-20 July

Parallelia bistriaris Hbn.                                     21 June-29 July

Catocala sordida Grt.                                          24-25 July Chrysanympha formosa (Grt.)                             9-12 July

Autographa precationis (Gn.)                             29 June

Autographa mappa (G. & R.)                             26 June

Syngrapha altera (Ottol.)                                    26 June-21 July

Syngrapha octoscripta (Grt.)                               14 July

Syngrapha epigaea (Grt.)                                    15-21 July

Syngrapha viridisigma (Grt.)                              18-24 July

Syngrapha alias (Ottol.)                                      21 June-20 July Syngrapha cryptica Eichlin & Cunningham 24 July Syngrapha rectangula (W. Kby.)                         6-25 July

Syngrapha microgamma nearctica Fgn.           21 June

Plusia venusta Wlk.                                             17-19 July

Baileya ophthalmica (Gn.)                                  21 June

Lithacodia muscosula (Gn.)                                21 June-10 July Lithacodia synochitis (G. & R.)                           8 July

Lithacodia concinnimacula (Gn.)                       25 June-4 July

Lithacodia carneola (Gn.)                                   25 June-20 July

Leuconycta diphteroides (Gn.)                           21 June-19 July

Panthea acronyctoides (Wlk.)                            21 June-25 July

Panthea pallescens McD.                                    27 June-25 July

Charadra deridens (Gn.)                                     21 June-15 July

Raphia frater Grt.                                                21 June-29 July

Acronicta americana (Harr.)                              21 June-24 July

Acronicta dactylina Grt.                                      14-25 July

Acronicta lepusculina Gn.                                  29 June-4 July

Acronicta innotata Gn.                                       21 June-25 July

Acronicta tritona (Hbn.)                                     15-19 July

Acronicta grisea Wlk.                                          21 June-24 July

Acronicta superans Gn.                                       15 July

Acronicta hasta Gn.                                             25 June

Acronicta fragilis (Gn.)                                       21 June-25 July

Acronicta clarescens Gn.                                     25 June-25 July

Acronicta retardata (Wlk.)                                 26 June-25 July

69

7

50

11 1 3' 7 3 2

11 1 6

26 5 3

12 1 1 4 1 2 2

223 1

27 1 2 1 4 1 5

21

14

47

29

21 152

18 9 3

19 3

18 1 1

14 162

49

104

Journal of the Lepidopterists' Society

Volume 48, Number 2

105

Appendix. Continued.

Anomogyna elimata (Gn.) Anomogyna badicollis (Grt.) Anomogyna youngii (Sm.) Aplectoides condita (Gn.) Anaplectoides prasina (D. & S.) Anaplectoides pressus (Grt.) Eueretagrotis perattenta (Grt.) Eueretagrotis attenta (Grt.) Heptagrotis phyllophora (Grt.) Cryptocala acadiensis (Bethune) Noctua pronuba L.

1  Identification uncertain, may include or consist entirely of Hypagyrtis unipunciata (Haworth) (Geometridae).

2  Includes Hydriomena divisaria (Walker) (Geometridae).

3  Includes Syngrapha abstrusa Eichlin & Cunningham (Noctuidae).

4  Identification uncertain, may include or consist entirely of Xestia adela Franclemont (Noctuidae).