The effect of domestication on insect resistance in cotton : bioassays with the insect herbivores Spodoptera littoralis and Agriotes spp.

Abstract: A plant's defensive ability against herbivore attack can be influenced by many different factors, one of them being domestication. During human selection to improve the yield of plants, the resistance against herbivore attack can have been lowered, due to a trade-off for use of resources between these traits in the plants. In this thesis I investigated the effect of domestication on resistance against herbivorous insects in cotton plants. I used different varieties and species of wild and domesticated cotton in three different experiments. A feeding assay was conducted, the mortality, development time and pupal weight of larvae of the Egyptian cotton leaf worm Spodoptera littoralis was studied. In a second experiment the preference of the root-feeding beetle Agriotes spp. was monitored in a two choice assay comparing domesticated cotton to other cotton varieties and species. The third part of the thesis was an odour collection from domesticated cotton plants induced by either S. littoralis larvae feeding on the leaves, Agriotes spp. feeding on the roots or a combination of damage from both species. The data show that there is a difference between cotton cultivars and that wild varieties delay development time compared to domesticated cotton. The feeding assay showed that the domesticated varieties Dpl 90 and Glandless cotton had significantly shorter development times than the undomesticated varieties (Tx 263, Tx 2259 and G. herbaceum) The two choice assay with Agriotes spp. larvae did not give any significant results most likely because of a low number of repetitions. Agriotes spp. did not prefer either domesticated nor undomesticated varieties of cotton. However, a trend in favor of the domesticated cotton variety could be observed. Finally, analysis of the odour collection demonstrates that plants induced by Spodoptera littoralis alone release higher amounts of the volatiles alpha pinene, 1S-beta pinene, myrcene and terpinolene than plants induced with either only Agriotes spp. or a combination of the two types of larvae. It may be that Agriotes spp. can manipulate the defence of the cotton plants. Another explanation is that the cotton plants cannot detect the Agriotes larvae. Also, there might have been a release of volatiles below ground that was not detected due to the layout of the experiment.