Host specificity of insect herbivores

One of the key issues in successful biocontrol is to ensure that the introduced biocontrol agent is really specific to the invasive species and that it will not switch to other closely related species present in the region. From this point of view, an ideal biocontrol agent is a monophagous insect, strictly restricted to a single plant species, but such strictly monophagous species are rare. More often, we can find species that are restricted to species of a single genus or a larger species group. Generalist feeders (control agents that are not restricted to a single species or a small range of species) often make poor biological control agents, and may become invasive species themselves. For this reason, potential biological control agents should be subject to extensive testing and quarantine before release into any new environment. If a species is introduced and attacks a native species, the biodiversity in that area can change dramatically. Data on host specificity of insect herbivores, are very sparse and consist mainly of observation of adults on the host plants. However, to prove that a given plant species is really a host plant of the insect species, we need to prove that the larvae are really feeding on the species. The data on host specificity can be also misrepresented by imperfect records.


Recently solved subtopics:

Predictive model of host specificity in pre-dispersal seed predators – step to successful biocontrol

This project is supported by the grant (KONTAKT – AMVIS) and solved with cooperation with Leland Russell (Wichita State University), Zuzana Münzbergová (Charles University in Prague), Petr Maršík (Institute of Experimental Botany), and Stanislava Koprdová (Crop Research Institute).

In this project we suggest an alternative approach using data on large number of traits of the plant and the insect and using this knowledge to predict potential hosts of the herbivores. Specifically we plan to combine field observations of plant herbivore associations with feeding experiment and plant and insect trait measurements including composition of secondary metabolites in the plant. We will use the species traits to predict the results of the field observations and feeding experiment. In this way we will identify traits that can be used to predict host specificity of insect herbivores and thus develop a predictive tool that will allows to identify possible host of the insects without time consuming field observations and experiments. We will validate the model created using data from Europe by similar data obtained from USA.

Details about our study are presented here.


Morphology of immature stages of genus Larinus

(in cooperation with Rafał Gosik)

Larinus species are generally monophagous or oligophagous and associated with plants from the tribe Carduoidea. Larvae of Larinus species are a potential biological control agent against weeds (e.g. Carduus, Cirsium). Detailed larval descriptions of species in the genus Larinus (Gosik & Skuhrovec 2011) demonstrate the possibility of identifying mature larvae and pupae directly from the flowerheads without breeding. Species identification of larvae with chaetotaxy is relatively easy, and it is generally much cheaper than identification by molecular methods (Hirsch et al. 2010). The largest problem in identification of the immature stages is relatively low number of available larval descriptions in comparison to the great number of known weevil specimens. The same problem exists in many other groups and not only in Curculionidae.

Updated: October 18, 2012

Dr. Jiri Skuhrovec
Function of Invertebrate and Plant Biodiversity in Agrosystems
Crop Research Institute
Drnovska 507
161 06 Praha 6 - Ruzyne