With more than 1,100,000 described species, the insects are the most diverse clade of extant animals, far before all other groups. Nevertheless, they undergo a drastic decrease of their populations, due to the sixth extinction of human origin. Thus it is important to define when and how they became so diverse and if they were impacted by the major crises of biodiversity in the deep past to estimate the importance of the current one.
Insects are generally among the best preserved terrestrial fossil organisms, much more complete than the vertebrates. They are also much more frequent. Thousands can be found in Konservat-Lagerstätten since the Carboniferous. They are preserved either in lacustrine sediments as compression fossils, or embedded in amber (fossil resins) (Fig. 1).
The Hexapoda (or six-legged arthropods, viz., wingless Collembola, Diplura, Protura ; wingless and winged Insecta) are among the oldest known terrestrial organisms, with first records dated from the Middle Devonian of Rhynie in Scotland. Recent molecular phylogenetic dating suggests that they appeared during the Silurian or even the Ordovician, with the first terrestrial plants. The Devonian hexapodan record is very scarce and disappointing, with less than six described fossils, all wingless [1]. The early Carboniferous one is even worse, without any fossil insects. But at the very end of this period and during the late Carboniferous, the insect diversity exploded, with a ‘sudden appearance’ of winged insects with very diverse feeding resources, e.g., carnivorous, plant suckers, leaf eaters, detritivorous, gall-makers, etc. The wingless clades remained a minority and the high diversification of the Carboniferous Hexapoda clearly concerned the winged forms. Wings and flight were probably the first crucial structures and function that allowed the first burst of diversification of the insects. Flight allows them to escape predators, find new resources, sexual partners, and travel to new environments. The most popular fossil insects are the Paleozoic ‘giant’ dragonflies Meganeuridae. These flying insects with very large wingspans (ca. 70 cm wide) had large bodies but comparable to those of some extant beetles. In fact, the unique really giant Carboniferous terrestrial arthropod was Arthropleura, a myriapod that was more than 1 m long. It is supposed that the great increase of oxygen proportion in the air during the Late Carboniferous favored the gigantism among the terrestrial arthropods, due to their breathing via trachea. The question is in fact more complex, because the winged insects knew a unique situation during the late Paleozoic, as they had no flying vertebrates as predators. As they were the only flying animals, they probably knew a phenomenon of parallel increases of sizes of predators (the Meganeuridae) and preys, the Palaeodictyoptera that also became larger and larger [2]. At the end of the middle Permian, both clades are very diverse, with still very large taxa, while the oxygen proportion began to decrease. The first gliding ‘lizard-like’ vertebrates are also recorded at the same time, and certainly began to predate these giant insects, which became rarer during the late Permian and no longer existed in the Triassic. The late Carboniferous was also the time of the oldest known holometabolous insects, with complete metamorphosis (wasps, beetles, scorpionflies), and of the oldest bugs (Hemiptera). These were discovered very recently because they were very small insects [3]. They are now the most diverse animal clades, with the ‘big five’ (Hemiptera, Hymenoptera, Diptera, Lepidoptera, and Coleoptera). But during all the Paleozoic, these insects were clearly very few. Holometaboly in itself was not ‘sufficient’ to cause their diversification and each of these orders ‘separately’ diversified during the last 220 Ma. The exact impact on the insects of the most important Permian–Triassic crisis of diversity remains difficult to estimate because there are very few latest Permian and earliest Triassic outcrops with insects. Thus if we know that the Triassic entomofaunas are very different from the Permian ones, we cannot establish that the great changes that occurred between the two periods happened during this crisis or before, during the late Permian or even at the end of the middle Permian. Nevertheless, the Palaeodictyoptera and the Meganisoptera are no longer present in the Triassic, while all the Triassic entomofaunas are clearly ‘dominated’ by the beetles and other Holometabola. Beetles were still minority during all the Permian in the fossil record. The ‘true’ flies (Diptera) and crown group of Hymenoptera are also dated from the Middle Triassic. At the end of this period, all the extant orders were present, except, maybe the parasite groups such as fleas (Siphonaptera), whose oldest fossils are middle Jurassic. The ‘modern’ entomofauna is thus much older than the extant mammal orders. During the Jurassic, the insects continued their diversification, with the first parasitoid wasps (there is no record of parasitoid insects before). The Cretaceous was the second crucial period for the insect (especially the Holometabola) diversification, with the oldest eusocial taxa (termites, wasps, bees, ants). The Albian–Cenomanian (ca. 100 Ma.) was the time of replacement of the gymnosperms by the angiosperms in all the terrestrial biotas, and the time of appearance of nearly all the extant insect families (even some extant genera have this age). It is also an important time of extinctions of several older Jurassic clades, replaced by extant taxa. Only the insects that adapted to the new environments related to flowering plants could diversify. The modern insect–plant relationships were established during the late Cretaceous. The recent new studies of the extraordinarily rich and diverse entomofauna of the ‘mid’ Cretaceous Burmese amber allowed one to discover that the Cretaceous insect world was as complex, rich and diverse as the extant one. The Cretaceous–Cenozoic (K–T) crisis had clearly a very weak impact on insect diversity, at least at the family level [4]. In fact, there were more extinctions and appearances of new families during the Paleocene–early Eocene than during the K–T crisis. These were periods of global warming followed by global cooling. The entomofaunas suffered the successive periods of cooling of the Oligocene, Miocene, and the Pliocene–Pleistocene glaciations, causing the extinctions of numerous widespread families that survived in small areas (the Australian mastotermitid termites or the Tasmanian hairy cicadid Tettigarctidae are the most spectacular examples).
The deep past history of insects is unique, with bursts of diversification ca. 330 Ma, 220 Ma, and 100 Ma ago. The causes of the first one remain poorly known, those of the second one are probably linked to the renewal of the ecosystems during the early Triassic, and the third one to the great floristic change. At least the K–T crisis did not affect much insect diversity. Thus the current crisis of biodiversity that begins to greatly affect the insect biomass, is extremely alarming. It may be more important than the K–T one.