@phdthesis{oai:tsukuba.repo.nii.ac.jp:00008276,
 author = {北原, 正彦 and Kitahara, Masahiko},
 month = {},
 note = {The search for patterns is an important component of ecological research, and has probably been the most important step in the development of community ecology. Up to the present, biological communities have been analyzed mainly from the viewpoint of biotic interactions such as interspecific competition and predation, niche theory, and species diversity. However, few attempt has been made to analyze natural community structure in terms of life-history strategies (e.g., generalist and specialist strategies) of its component species. On the other hand, the verification of the generalist/specialist (r/K) concept has been examined mainly among populations of a species, or of closely related species. However, little attempt has yet been made to verify the concept among many species at the community level. In the present study, I examine temperate butterfly communities subjected to different degrees of human disturbance, and apply the generalist/specialist concept to analyze butterfly community structure. Butterflies are the most suitable organisms for the study of the structure and dynamics of biological populations and communities. The adults can be identified easily in the field and their life-histories are already well known in Japan. The series of my studies are composed of the following four chapters. Chapter 1: Patterns in the structure of butterfly communities along an environmental gradient of human disturbance based on the concept of generalist vs. specialist strategies
I monitored nine butterfly communities with varying degrees of human disturbance by conducting a census twice a month during 1980 by the line transect method in and around Tsukuba City, central Japan. I analyzed the community structures using the generalist/specialist concept. The site (community) order based on decreasing human disturbance was positively correlated with butterfly species diversity (H'), species richness (the total number of species), and the number of specialist species in a community, but not with the number of generalist species. The number of generalist species was rather constant, irrespective of the degree of human disturbance. Thus, both the butterfly species diversity and species richness were more dependent on the specialists than the generalists. The analyses also showed that the generalist species were distributed widely over the communities, and they maintained high population densities, resulting in high rank status in abundance in a community, with more spatial variation in density per species. Specialist species showed the opposite trends. These results demonstrate that the generalist/specialist concept is a powerful tool applicable to analyze structure of natural communities. Chapter 2: Reexamination of the butterfly community patterns based on the generalist/specialist concept along a grassland habitat gradient of human disturbance in a different region In the previous chapter, I analyzed the butterfly communities along a gradient of human disturbance by applying the generalist/specialist concept. Butterfly species were classified into generalist or specialist species based on their voltinism (seasonal time dimension) and potential larval resource breadth (food dimension). The community structure and species composition showed the systematic changes along the gradient. To verify the generality of those trends, I chose another butterfly community data (Sei 1986a,b) obtained by monitoring five grassland butterfly communities with varying degrees of human disturbance twice a month during 1985 by the line transect method at the foot of Mt. Fuji, central Japan, and analyzed their structure in a manner similar to that employed in the previous chapter. Most results were consistent with the patterns recognized in the previous chapter. The route (community) order based on increasing human disturbance was strongly and negatively correlated with butterfly species richness but with neither butterfly species diversity (H' ) nor evenness (J'). Also, the degree of human disturbance was significantly and negatively correlated with the number of specialist species, but not with that of generalists, in a community. Butterfly species richness was more strongly correlated with the number of specialist species than with that of generalists. The analyses also showed that the generalist species were distributed more widely over the communities than were the specialists. However, in contrast to the trend revealed in the previous chapter, there was no significant difference in the population densities and in the spatial population variability between the two species groups. As a whole, the present analyses confirmed the consistency of most community patterns detected in the previous chapter. The causes of the inconsistencies in some patterns were thought to be mainly due to the present habitat conditions with a relatively short growing season at high altitudes. Chapter 3: Analyses of butterfly community structure and composition through multivariate approaches and the concept of generalist/specialist strategies, and conservation implications In this chapter, I analyzed the community structure of the butterfly data in Chapter 1 using multivariate analyses. The present analyses are in sharp contrast to my previous analyses (Chapter 1) in which each butterfly species was assigned, prior to data analyses, either generalist, intermediate, or specialist based on their larval food plant and voltinism in accordance with the generalist/specialist concept. The present results of principal components analysis (PCA) showed that the butterfly community was composed of two species groups (I and II). This species grouping was also supported by the results of the cluster analysis (unweighted pair-group method using arithmetic average (UPGMA)). Comparing the present results with species classification used in Chapter 1, the butterfly community was found to be composed of five species groups (specialists, intermediates I and II, and generalists I and II), which differed from each other in their ecological characteristics. Through the comparisons of their characteristics among the five species groups, it was suggested that, in the butterfly community, the positions of the specialist group and the generalist group II are the endpoints on the generalist-specialist selection spectrum, and the three other groups are positioned between these two extremes. The results of the multivariate analyses showed that, among the specialist, intermediate, and generalist groups classified a priori in the previous study, only the specialist group could be discriminated well in one group, and the generalist and intermediate groups were both divided further into two subgroups (I and II) with different characteristics. The multivariate approaches also succeeded in extracting a typical generalist group (i.e., generalist II) from the butterfly community. I propose and recommend the synergetic application of the generalist/specialist (r/K) concept and multivariate approaches to the detailed analysis and deeper understanding of community structure and composition. The present results also suggest that the approach employed in this study is much effective and helpful to identify and find out priority or target species for local butterfly conservation. Chapter 4: Application of the community analysis based on the generalist/specialist concept to the environmental evaluation of habitat islands: The island biogeographical analysis of butterfly community structure in the newly designed parks of Tsukuba City I analyzed the butterfly communities in the newly designed city parks (area C), "newly opened habitat islands", of Tsukuba City, central Japan. The area constituted a natural ecological experiment on the mainland for clarifying the pattern and process of faunal immigration. I compared butterfly communities in area C with those in two other areas in the light of the theory of island biogeography and the concept of generalist/specialist. The results showed the following: (1) Fewer species were found in area C than in other areas, due largely to the absence of many specialist types, restricted and habitat specialists, and/or low density species in the area. Generalist types, widespread and habitat generalists, and/or high density species predominated in area C. (2) The difference in the species numbers among the three sections within area C could be explained by the habitat structure in and around the respective sections. (3) The densities of many species were low in area C, probably due to its man-modified habitat structure. In particular, several species occurred at extremely low densities in area C, but at high densities in other areas. (4) The internal structure of the habitat island butterfly community in area C was almost perfectly consistent with that of "quasi-equilibrium" communities that appear during the colonization of an island. These results demonstrate that the synergetic application of the generalist/specialist concept and the island biogeography theory is effective for the understanding of the patterns and structures of habitat island communities. Overall, I conclude that (1) the patterns in the structure of butterfly communities along an environmental gradient of human disturbance based on the generalist/specialist concept found in the present studies are both robust and general ones, probably common to various types of habitats, environments, and ecosystems, (2) even relatively stable environments usually have heterogeneous structure with stability and unstability, and therefore, support the mixture of specialists and generalists and the resultant high species richness and diversity, and (3) the analytical approach based on the generalist/specialist concept is a highly operational method with good predictive and explanatory powers much applicable to the analysis of various types of biotic communities, in spite of a variety of criticism of the concept.
Key words: Butterfly communities, Community patterns, Community structure, Conservation priority, Environmental disturbance, Generalist-Specialist selection continuum, Generalist and Specialist strategy, Habitat island, Human disturbance, Island biogeography theory, Island patterns, Lepidoptera, Principal components analysis (PCA), Route order, Species characteristics, Species classification, Species composition, Species distribution, Species richness, Temperate butterfly species., 2004, Includes bibliographical references},
 school = {筑波大学, University of Tsukuba},
 title = {The analysis and understanding of butterfly community structure based on the concept of generalist vs.  specialist strategies : a new approach to biological community analysis},
 year = {2004}
}