Abstract
Insects are the most diverse animals on Earth, and neatly associated with plants they represent two of the major groups of organisms both in species diversity and biomass quantity. The majority of their interactions involves insect feeding and insect parasitism mostly on leaves. Plant and insect compose one of the main trophic levels in ecosystems over the 325 million years. Today, the continuous and fast rising of temperature mostly due to human activities since the last century is disturbing the balance of ecosystems on Earth. Consequently, to understand the role of plant and insect interactions, through time but also trophic networks, becomes crucial. The fossil record is an exceptional opportunity to survey responses of plant-insect interaction to climate variations over long time interval through traces of plants reactions caused by interaction with insects, as Earth has already experienced many climate changes. For the last 3 million years, oscillations between long cold periods and short warm periods have occurred. Europe ecosystems has been particularly impacted. The Lagerstätte of Willershausen (Germany) was specifically study. It is an exceptional fossil outcrop that contains ca. 8000 collected fossil leaves. These leaves testify a paleoforest developed there around 3—2.6 Ma ago in a climate warmer than today (ca. +5°C). Under these conditions, many plant species typical of the Mediterranean ecosystems were settled there, such as Montpellier maple or Olive tree. For comparison, other paleoforests were studied: Berga (similar in age and geographically close to Willershausen) and Bernasso (younger than Willershausen (2.16—1.96 Ma) and located in southern France close to Mediterranean. These forests were compared as many common plant taxa were similar between each other. Furthermore, some species today endemic to the Caucasian region, such as Persian ironwood or Caucasian elm, were also present in these outcrops. The aim of this study is to determine how far the climate differences could be involved in the changes of plant-insect interactions in European paleoforests of the late Pliocene – early Pleistocene. Results highlighted the impacts of both hydric and temperature seasonality, hitherto underestimated in the fossil record, on the patterns of plant-insect interaction in the European paleoforests. It appeared that ecosystems subject to intense hydric seasonality could led to higher specialization of plant-insect interaction inferred by higher rate of observed damages due to ‘specialists insects’. In parallel, the coolest temperature during the year seems to be a major factor in the low diversity of damage in paleoforest, presumably due to lower insect metabolism. Absence of convergent correlations between plant richness and damage richness could suggested that influence of climatic factors override impact of these local biotic factors. In order to understand the whole parameters that could have an impact on plant-insect interactions, our current knowledges are still insufficient. It would be wise to make more investigations on modern forests with the methods as applied in fossil record community structure studies. These investigations could help to understand the factors potentially involved in the establishment of a pattern of plant-insect interactions. It is in this perspective that a part of this study was precisely focused on one plant species (Parrotia persica) currently endemic to the Hyrcanian forest (Iran). This forest is supposed to be an analogue forest of the European paleoforests as those studied in this thesis. For now, observations made in Iran tend to corroborate our interpretation. Finally, the studies on plant-insect interactions in past and extant ecosystems, combined with the study of climatic changes, should permit us to better characterize the relations between plants and insects in forests through time.