Numerical Taxonomic Techniques Research Articles

Associations of Plants and Insects in Deciduous Forest

Phytophagous insect larvae (Lepidoptera and Tenthredinoidea) in the leaf—chewing guild were sampled from accessible parts of all the 18 species of common woody plants in a mature upland forest in New York State, in early and in late June. Among the procedures used to describe patterns of associations between plants and insects are principal components analysis and the clustering techniques of numerical taxonomy, whereby the faunistic similarity among plant species is described. We introduce a weighting factor for cluster analysis that weights a character (in this case the density of an insect species on a plant) in proportion to its deviation from the norm. During June, both the abundance and variety of insects declined on most plant species, which vary greatly in both these aspects of their fauna. Insect species that are either quite highly specialized or generalized in diet appeared to be prevalent; those with intermediate diet breadth were less so. Clustering plant species by faunal similarity revealed some clusters of taxonomically related species, but the correspondence between faunal similarity and taxonomic affinity is not strong. Few plant species have a highly distinct, specialized fauna. The complexity of distribution of insects over plants indicates that insects are responding to many factors that differentiate plant species.

Fatty acid composition of Rhizobium spp.

The fatty acid composition of 42 isolates belonging to the major plant affinity groups of Rhizobium has been determined and found to vary reproducible with culture age. Numerical taxonomic techniques applied to the 15 major fatty acid components of log-phase cultures of comparable physiological age showed that the rhizobia constitute a uniform group. However, two clusters comprising soybean-cowpea isolates and pea-bean isolates were evident. These observations, based on a simple analysis of only one group of chemical components, indicate relationships among rhizobia which differ from the conventional plant-affinity groupings but which are consistent with other proposed relationships established using a variety of biochemical and physiological criteria.

Performing Land-Capability Evaluation by Use of Numerical Taxonomy: Land Use and Environmental Decisionmaking Made Hard?

Researchers have attempted to incorporate environmental variables into the land-use planning process by use of several ranking and mapping formulations. Most of these are based on some type of classification scheme. More recently, multivariate statistical techniques have been utilized to classify land areas into groups with similar suitability for urban development. A test was made of one of these numerical taxonomic techniques on a data set from Medford Township, New Jersey, and the results analyzed in terms of the pros and cons of these methods. A group of 484 forty-acre grid cells with forty-two environmental variables was collapsed into a ten variable set for ten groups of grid cells having like characteristics. The analysis involved two steps, factor analysis followed by a euclidean-distance-classification algorithm. The results show that numerical taxonomy can greatly facilitate the analysis of large environmental data sets and can help to identify the ecological relationships in quantitative terms. However, the complexity of the statistical methods involved greatly limits the wide application of these techniques, and the use of numerical taxonomic results in land-capability evaluation cannot release the researcher from making many judgmental decisions.

The Mite Genus Dermanyssus: A Survey, with Description of Dermanyssus Trochilinis, N. Sp., And a Revised Key to the Species (Acari: Mesostigmata: Dermanyssidae)

Current research on Dennanyssus is surveyed briefly. Phenetic similarities among the 16 known species are analyzed using techniques of numerical taxonomy, including principal component analysis and multidimensional scaling. Dermanyssus trochilinis , n. sp. is described from a hummingbird nest in Peru. D. faralloni and D. antillarum are re-illustrated. The species are keyed. The status of D. passerinus and D. longipes remains unsettled.

FORTRAN program for lateral tracing of time-stratigraphic units based on faunal assemblage zones

INTRODUCTION Many methods are available for multivariate analysis of faunal zones (see Brower, Millendorf, and Dyman, 1978 for discussion). Especially popular are the techniques of numerical taxonomy, such as cluster analysis, factor analysis, principal-components analysis and principal coordinates. Generally in such analyses all samples from all stratigraphic sections are analyzed simultaneously. However, this approach is reasonable only when the purpose of the analysis is to depict ecological zones or biofacies. If, on the other hand, the results are to portray time-stratigraphic correlations, an alternative approach is recommended; namely, lateral tracing. At a given time of deposition, environmental conditions and thus faunal composition change in any particular geographic area. This variation generally is continuous, with faunal elements gradually being replaced by others both laterally (geographically) and vertically (stratigraphically). In the assemblage zone methods mentioned (e.g. the usual practice of Hazel, 1970, 1977; Brower, Millendorf, and Dyman, 1978), all samples from within the study area generally are analyzed together. This approach ignores the effects of faunal gradation within an isochronous unit with respect to geographic position. Thus, if the faunal composition of such an isochronous unit changes across the study area, samples taken from distant points in the unit might be dissimilar enough not to cluster. Simply, the greater the lateral variation and the larger the distance between them, the less similar are two isochronous samples. This difficulty can be ameliorated by utilizing a priori knowledge of the geographic distribution of the stratigraphic sections under study. Lateral tracing uses this knowledge which allows gradation of stratal character without affecting stratigraphic resolution. This is accomplished simply by performing the analysis in a stepwise fashion between pairs of adjacent stratigraphic sections. The lateral tracing algorithm was proposed originally by Sneath (1975a, 1975b) with an intended use for lithostratigraphy utilizing petrological and geochemical data. A

A Multivariate Assessment of Phenetic Relationships Within the Feather Mite Family Eustathiidae (Acari)

Moss, W. W., P. C. Peterson, and W. T. Atyeo (Department of Entomology, Academy of Natural Sciences, 19th and The Parkway, Philadelphia, Pennsylvania 19103, Department of Biological Sciences, Youngstown State University, Youngstown, Ohio 44555, and Department of Entomology, University of Georgia, Athens, Georgia 30602). 1977. A multivariate assessment of phenetic relationships within the feather mite family Eustathiidae (Acari). Syst. Zool. 26:386–409.—Techniques of numerical taxonomy were applied in combination to morphometric data based on 111 characters from 61 OTUs, representing 61 species of the feather mite family Eustathiidae. Smaller subsets of taxa were analyzed to clarify the relationships and status of certain species. A statisfactory classification was obtained, complementing a monographic revision of the family in press. Supraspecific taxa were indicated by the placement of species on minimum spanning trees superimposed upon ordination diagrams. The latter were obtained by nonmetric multidimensional scaling applied to taxonomic distances. An initial configuration for multidimensional scaling was taken from a principal component analysis of character correlations. In every case except one the placement of species into genera by numerical methods matched our impressions of generic entities obtained by traditional approaches, although between- and within-generic relationships were clarified. The combination of Q- and R-mode analysis, including the use of nonmetric multidimensional scaling, provided an effective approach to the analysis of taxonomic structure. Multidimensional scaling is strongly recommended for similar studies as a technique to improve the accurate visual representation of taxonomic relationships. Only moderately strong host-parasite relationships occur within the Eustathiidae, although the group as a whole is restricted to the Apodi. No member of the Eustathiidae occurs on the other suborder of the Apodiformes, the Trochili. This emphasizes the distinction between Apodi and Trochili, and points again toward the question of whether or not these taxa of birds should be placed together. Host-parasite relationships within the Eustathiidae may be associated with resource tracking, but further data on feeding habits and life histories of the mites are needed to answer this question. Finally, this study represents a collaborative approach to the solution of a taxonomic problem. We feel that the collaborative approach has considerable potential in the resolution of difficult relationships within the Acari and other groups of organisms. [Numerical taxonomy; phenetics; ordination; multidimensional scaling; cluster analysis; Acari; Eustathiidae; feather mites; Apodi; swifts.]

A numerical taxonomic study of the Mallophagan genera Cummingsiella (=Quadraceps),Saemundssonia (Ischnocera: Philopteridae), and Austromenopon (Amblycera: Menoponidae) from alcids (Aves: Charadriiformes) of the northwest Atlantic with reference to host–parasite relationships

Specimens of Cummingsiella, Saemundssonia, and Austromenopon recovered from the alcids of the northwest Atlantic were analyzed using the techniques of numerical taxonomy. Twenty-one morphological characters common to both sexes plus four genital characters in the males were measured and used in the analyses. Principal axis factor analysis (PAFA) and four clustering techniques were used to determine the phenetic relationships at various taxonomic levels.The results supported the familial and generic classifications established by conventional taxonomists. The diagnostic value of genital characters at generic and specific levels were evaluated. The results from the analysis of each genus are presented and compared with existing classifications. With the exception of Austromenopon, the species groups formed within each genus were very similar for both sexes. Host–parasite relationships based on parasite interrelationships were investigated.

Taxonomic Position of Hydropus floccipes and Allied Species: A Quantitative Approach

SUMMARYA previously described numerical taxonomy technique is used to compare the subsection Floccipedes (in current taxonomy classified under Hydropus or Mycena, sect. Floccipedes) with the genus Hydropus (without subsection Floccipedes) and Gerronema subgenus Gerronema. The technique was developed for assigning genera to families, and there are some differences, which are noted, when it is applied to this lower level. Further comments are made on the general method, including response to discussions of the method by a Swiss mycologist and an American statistician. The taxonomic result of our study confirms that subsection Floccipedes should be attached to Hydropus rather than Gerronema.

Taxonomic Position of Hydropus Floccipes and Allied Species—A Quantitative Approach

A previously described numerical taxonomy technique is used to compare the subsection Floccipedes (in current taxonomy classified under Hydropus or Mycena, sect. Floccipedes) with the genus Hydropu...

Evaluation and Optimal Combination of TLC Systems for Qualitative Identification I: Sulfonamides

A mathematical criterion for the evaluation of chromatographic analysis procedures is given by the information content as derived from Shannon’s equation. This information content yields a numerical value representative of the merits of each chromatographic separation and thus allows selection of the optimal systems. In most cases, however, one analysis is not sufficient to allow the qualitative identification of the sample. Therefore, several chromatographic systems are combined. Two approaches allow the desired combination; one either calculates the information content of several procedures as one mathematical value or classifies the systems according to mutual resemblance by numerical taxonomy techniques. From the resulting groups of dissimilar systems, one optimal system can be chosen per group according to the information content. The results obtained by these mathematical procedures are illustrated with a practical example: the selection and evaluation of systems for the TLC analysis of sulfonamides.

Evidence of microevolution in an escaped pear population

A population of 87 wild pears (Pyrus communis L.) of fruiting age, growing in a relatively homogeneous old field environment, was found to be highly variable, with coefficients of variation up to 40%. Generally, variability was greater in size than in meristic characters, in meristic characters than in shape characters, and floral traits were less variable than vegetative. Tree trunk core samples showed the population to have been founded circa 1905 to 1908, by three trees. Establishment of progeny began about 19 years later. The population is high in 1975 and still expanding logarithmically. Numerical taxonomic clustering techniques using morphological characters showed the population to contain two very diverse groups. Trees of both groups are of all age classes and are distributed randomly over the site. Trees phenetically like and unlike the oldest trees were analysed as to growth rates. Those trees similar to the oldest trees were significantly slower in growth rate than trees that were morphologically dissimilar. This suggests that natural selection is operating to remove the less fit genotype from the population.

Comparison of Numerical Taxonomic Methods Used to Estimate Flavonoid Similarities in the Limnanthaceae

The Limnanthaceae is a small family of North American herbs with uncertain internal relationships. Taxa of the family were compared on the basis of 46 flavonol glycosides occurring in all plant tissues or in petals only. The flavonoid data were analyzed by 3 numerical taxonomic techniques: 1) clustering by the Weighted Pair Group method considering positive and negative matches (Using Simple Matching coefficient) ; 2) clustering by the Weighted Pair Group method considering only positive matches (using Jaccard coefficient) ; 3) Varimax Factor Analysis with rotation. No significantly different results were produced by the two clustering methods. Factor analysis provided a clearer indication of relationships among taxa of the Limnanthaceae than did conventional clustering analysis. Apparently, ordination of taxa in three dimensions is necessary to accurately express relationships between divergent evolutionary lines of the family since much of the expressed variation in flavonoids cannot be accounted for by a one-dimensional clustering technique.

An Electrophoretic and Serological Investigation of Seed Proteins in Galeopsis tetrahit L. (Labiatae) and Its Putative Parental Species

Seed proteins from the tetraploid Galeopsis tetrahit L. and its putative parental species G. pubescens Bess. and G. speciosa Mill., were examined using disc gel electrophoresis (PAGE), Ouchterlony double diffusion, and Immunoelectrophoresis. Forty-three inbred cultivated lines of G. speciosa, G. pubescens, and G. tetrahit (both the naturally occurring species and an artificially produced hybrid) and two wild populations of G. tetrahit were examined. PAGE gels were stained for total protein or malate dehydrogenase. Although variability in the total protein pattern was detected within and between lines for all three species, there was much similarity among the taxa. No bands appeared in the hybrid that were not present in the parents. The band pattern for synthetic G. tetrahit was very similar to that of natural G. tetrahit suggesting that G. tetrahit originated from a cross in nature between G. pubescens and G. speciosa. The total protein PAGE results were analyzed using numerical taxonomic techniques. Although these analyses did not clearly separate the three species, the general tendency was toward grouping G. pubescens lines together, natural G. tetrahit lines together, and synthetic G. tetrahit lines together. Natural and synthetic G. tetrahit lines were also interspersed among each other along with some G. speciosa lines. The malate dehydrogenase pattern showed little variation among the three species. Qualitative serological methods did not distingush among the three species, but confirmed the similarity observed among the taxa in the results of electrophoresis.

AN ELECTROPHORETIC AND SEROLOGICAL INVESTIGATION OF SEED PROTEINS IN GALEOPSIS TETRAHIT L. (LABIATAE) AND ITS PUTATIVE PARENTAL SPECIES

Seed proteins from the tetraploid Galeopsis tetrahit L. and its putative parental species G. pubescens Bess. and G. speciosa Mill., were examined using disc gel electrophoresis (PAGE), Ouchterlony double diffusion, and Immunoelectrophoresis. Forty‐three inbred cultivated lines of G. speciosa, G. pubescens, and G. tetrahit (both the naturally occurring species and an artificially produced hybrid) and two wild populations of G. tetrahit were examined. PAGE gels were stained for total protein or malate dehydrogenase. Although variability in the total protein pattern was detected within and between lines for all three species, there was much similarity among the taxa. No bands appeared in the hybrid that were not present in the parents. The band pattern for synthetic G. tetrahit was very similar to that of natural G. tetrahit suggesting that G. tetrahit originated from a cross in nature between G. pubescens and G. speciosa. The total protein PAGE results were analyzed using numerical taxonomic techniques. Although these analyses did not clearly separate the three species, the general tendency was toward grouping G. pubescens lines together, natural G. tetrahit lines together, and synthetic G. tetrahit lines together. Natural and synthetic G. tetrahit lines were also interspersed among each other along with some G. speciosa lines. The malate dehydrogenase pattern showed little variation among the three species. Qualitative serological methods did not distingush among the three species, but confirmed the similarity observed among the taxa in the results of electrophoresis.

Consumption and Utilization by the Mexican Bean Beetle of Soybean Plants Varying in Levels of Resistance12

Standard gravimetric techniques were used to measure mean body weight, total weight gain, weight of food consumption, efficiency of conversion, consumption index, and growth rate of larvae of Epilachna varivestis Mulsant which fed from hatch to pupation on 15 soybean varieties. The varieties ranged from highly resistant to highly susceptible, Developmental time, mean pupal weight, and percent survival also were determined. A wide range of parameters was obtained with the results expressed using techniques of numerical taxonomy in a dendrogram. The clusters formed in the dendrogram represent rankings similar to field evaluation of these varieties and reflect trends of the nutritional and biological parameters of the Mexican bean beetle feeding on the soybean varieties. Physiological inhibitors or biophysical deterrents in P.I. 229358 and nutritional deficiencies or nutrient disproportionality in P.I. 200459, P.I. 200498, P.I. 224269, and P.I. 229321 were postulated as possible causes of resistance.

Kilometric distance

A simple method of comparing species distributions, when the latter are in the form of lists of latitude–longitude coordinates, is described. The average of the distance in kilometres between the nearest neighbor of each individual of species A in species B and of species B in species A is used as a pair function. The matrix of pair functions for all species comparisons is then amenable for any of the techniques of numerical taxonomy. A small example using Turkish Papaver is given.

A Note on the Evaluation of the Agglomerative Hierarchical Clustering Methods

Recently, a large number of techniques of numerical taxonomy or cluster analysis are proposed and studied by many workers, and the needs for these techniques arise in many fields of applied science. Especially, the method of hierarchical cluster analysis is used widely, since all of these methods are suitable for various types of data and can be simply carried out. However, though the investigations for evaluation or comparison of these method are only little discussed in formal works, it is quite important and necessary to discuss them in order to study cluster analysis. Furthermore, we propose the fuzzy distance which is a new index of evaluating and comparing relationship between two relation matrices, the original similarity matrix S and S* which is derived from S by executing hierarchical clustering algorithms. Finally, numerical examples for several artificial datas are investigated by four well-known clustering methods.

Classification of the oat species (Avena, Poaceae) using various taximetric methods and an information–theoretic model

The latest comprehensive classification of Avena species was established by Malzew in 1929. It has been used by oat workers until today, in spite of the fact that for decades it became gradually outdated partly because new data became available and partly because new species were discovered. This necessitated a new attempt to classify the oats. The present study uses new characters and various numerical taxonomic techniques such as Gower's coefficient of resemblance and eight sequential, agglomerative, hierarchical clustering strategies. Thus 80 different classifications were formed, of which 21 were chosen as admissible. The best classification among the 21 is that which contains maximum information in the information-theoretic sense, i.e., using Eastabrook's model for character analysis. The relationships between the various classifications and the best one are extrapolated through principal coordinate analysis using the information-theoretic correlations between pairs of classifications as measure of distance.As a result, seven groupings were recognized, and canonical analyses were performed on these. Subsequently the canonical loadings were used here as unorthodox diagnoses for the seven groupings, which were then designated as taxa at the sectional category. A formal treatment of these newly established sections is provided in compliance with the rules of nomenclature.Finally, interrelationships between species are also studied using overlapping clusters based on the method of maximal complete subgraphs of Jardine and Sibson.

The Selection of Preferred Liquid Phases After Classification By Numerical Taxonomy Techniques

The Selection of Preferred Liquid Phases After Classification By Numerical Taxonomy Techniques

Application of numerical taxonomy techniques to the choice of optimal sets of solvents in thin layer chromatography

Application of numerical taxonomy techniques to the choice of optimal sets of solvents in thin layer chromatography