+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      An Adaptive Kernel Smoothing Method for Classifying Austrosimulium tillyardianum (Diptera: Simuliidae) Larval Instars


      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.


          In insects, the frequency distribution of the measurements of sclerotized body parts is generally used to classify larval instars and is characterized by a multimodal overlap between instar stages. Nonparametric methods with fixed bandwidths, such as histograms, have significant limitations when used to fit this type of distribution, making it difficult to identify divisions between instars. Fixed bandwidths have also been chosen somewhat subjectively in the past, which is another problem. In this study, we describe an adaptive kernel smoothing method to differentiate instars based on discontinuities in the growth rates of sclerotized insect body parts. From Brooks’ rule, we derived a new standard for assessing the quality of instar classification and a bandwidth selector that more accurately reflects the distributed character of specific variables. We used this method to classify the larvae of Austrosimulium tillyardianum (Diptera: Simuliidae) based on five different measurements. Based on head capsule width and head capsule length, the larvae were separated into nine instars. Based on head capsule postoccipital width and mandible length, the larvae were separated into 8 instars and 10 instars, respectively. No reasonable solution was found for antennal segment 3 length. Separation of the larvae into nine instars using head capsule width or head capsule length was most robust and agreed with Crosby’s growth rule. By strengthening the distributed character of the separation variable through the use of variable bandwidths, the adaptive kernel smoothing method could identify divisions between instars more effectively and accurately than previous methods.

          Related collections

          Most cited references2

          • Record: found
          • Abstract: found
          • Article: not found

          Larval sampling and instar determination in field populations of northern and western corn rootworm (Coleoptera: Chrysomelidae).

          Abundance and head capsule width were measured for northern (Diabrotica barberi Smith & Lawrence) and western corn rootworm (D. virgifera virgifera LeConte) larvae recovered primarily from maize root systems but also from large soil cores each centered around a root system. Larvae for measurement derived from field populations under infestation and rotation regimes that allowed most specimens to be assigned to species. A frequency distribution of head capsule widths indicated three separate peaks for western corn rootworm, presumably representing frequency of the three larval instars, with no larvae measuring 280 or 420 microm in the valleys between peaks. Multiple normal curves fit to similar but partially overlapping peaks generated by northern corn rootworm suggested that division of first to second and second to third instar can best be made for this species at 267 and 406 microm, respectively (270 and 410 when measurements are made to the nearest 20 microm). These results implied that instar of individuals from mixed northern and western corn rootworm populations can be accurately judged from head capsule width without having to determine species. The relative abundance of western corn rootworm instars was similar in root systems removed from the center of 19-cm diameter x 19-cm deep soil cores and in soil cores from which the root systems were removed. Furthermore, the number of larvae from root systems correlated significantly with that from the surrounding soil. These results indicated that the former and much more convenient sampling unit can be used to estimate population developmental stage and possibly density, at least early in the season when these tests were done and young larvae predominated.
            • Record: found
            • Abstract: found
            • Article: not found

            Instar determination for Pissodes castaneus (Coleoptera: Curculionidae) using head capsule widths and lengths.

            The aim of this research was to assess the larval instar number of Pissodes castaneus (De Geer) and to facilitate the study of its biology by identifying a reliable method to determine the instar of individual larvae. The larvae of this weevil were collected in central Italy in 2004, and their head capsules were measured by means of a binocular microscope. Head capsule width and length data were analyzed using Hcap, a computer program that use the distribution of size measures to determine instar separation rules. To determine instar number, the Gaines and Campbell method, which represents the perfect geometric progression of size measures (Dyar's rule) by a regression line, was also used. This study identified four larval instars of P. castaneus and found that head capsule widths and lengths followed Dyar's rule and reliably distinguished instars.

              Author and article information

              J Insect Sci
              J. Insect Sci
              Journal of Insect Science
              Oxford University Press
              06 November 2015
              : 15
              : 1
              : 159
              1Department of Applied Mathematics, College of Mathematics and Informatics, South China Agricultural University, Guangzhou 510642, China
              2Guangxi Key Laboratory of Biology for Crop Diseases and Insect Pests/Institute of Plant Protection, Guangxi Academy of Agricultural Sciences, Nanning 530007, China
              3Corresponding author, e-mail: yonghaoyucn@ 123456yeah.net
              Author notes

              Subject Editor: Julie Urban

              © The Author 2015. Published by Oxford University Press on behalf of the Entomological Society of America.

              This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

              : 7 April 2015
              : 10 October 2015
              Page count
              Pages: 8

              austrosimulium tillyardianum,instar determination method,adaptive kernel smoothing estimation,bandwidth selection


              Comment on this article