New long-proboscid lacewings of the mid-Cretaceous provide insights into ancient plant-pollinator interactions

Abstract

Many insects with long-proboscid mouthparts are among the pollinators of seed plants. Several cases of the long-proboscid pollination mode are known between fossil insects (e.g., true flies, scorpionflies, and lacewings) and various extinct gymnosperm lineages, beginning in the Early Permian and increasing during the Middle Jurassic to Early Cretaceous. However, details on the morphology of lacewing proboscides and the relevant pollination habit are largely lacking. Here we report on three lacewing species that belong to two new genera and a described genus from mid-Cretaceous (Albian-Cenomanian) amber of Myanmar. All these species possess relatively long proboscides, which are considered to be modified from maxillary and labial elements, probably functioning as a temporary siphon for feeding on nectar. Remarkably, these proboscides range from 0.4–1.0 mm in length and are attributed to the most diminutive ones among the contemporary long-proboscid insect pollinators. Further, they clearly differ from other long-proboscid lacewings which have a much longer siphon. The phylogenetic analysis indicates that these Burmese long-proboscid lacewings belong to the superfamily Psychopsoidea but cannot be placed into any known family. The present findings represent the first description of the mouthparts of long-proboscid lacewings preserved in amber and highlight the evolutionary diversification of the ancient plant-pollinator interactions.

Generalist Pollen-Feeding Beetles during the Mid-Cretaceous

“…Notably, the distinctive, tubular, floral modifications of angiosperms that would accommodate long-proboscid insect pollinators originated during the Paleogene (Labandeira, 2010;Friis et al, 2011;Lin et al, 2019). The extinction of several long-proboscid lineages during the Cretaceous is consistent with this pattern, as Cretaceous angiosperms lacked relevant floral structures that would accommodate long-proboscid insects (Ren et al, 2009;Endress, 2010;Labandeira, 2010;Peñ alver et al, 2015;Lu et al, 2016;Liu et al, 2018;Lin et al, 2019). In distinct contrast to the extinction of many long-proboscid insect pollinators, other mandibulate pollinators, including beetles, made the transition from gymnosperm to angiosperm hosts with greater facility.…”

“…For insects, the specialized modification of mouthparts into elongate siphonate proboscises commonly is used to access pollination drops in tubular ovulate organs of gymnosperms or the deep-throated flowers of angiosperms (Nilsson, 1998;Ren et al, 2009). The variety of long-proboscid brachycerous flies (Diptera), aneuretopsychine scorpionflies (Mecoptera) and kalligrammatid lacewings (Neuroptera) found in the pre-angiosperm fossil record (Ren et al, 2009;Labandeira, 2010;Peñ alver et al, 2015;Labandeira et al, 2016;Lu et al, 2016;Peris et al, 2017;Liu et al, 2018;Lin et al, 2019) consequently provides indirect morphological evidence for pollinator activity. This evidence occurs individually either on the plant host or on the insect pollinator, but without the direct associational evidence of a plant structure adjacent to or in contact with an insect, such as pollen occurring in the pollen basket of a bee (Wappler et al, 2015).…”

Phylogeny of pleasing lacewings (Neuroptera: Dilaridae) with a revised generic classification and description of a new subfamily

“…The pectinate male antennae and elongated ovipositor, which are important diagnostic characters of Dilaridae, either are not constant within the family or are also shared by some other lacewing groups. For example, the male pectinate antenna is shared by a fossil group of Psychopsoidea (Lu et al ., ), while the elongated ovipositor is shared by the mantispid subfamily Symphrasinae (Liu et al ., ) and the fossil family Kalligrammatidae (Yang et al ., ). Nevertheless, these features should be independently evolved among these groups and here support the monophyly of Dilaridae.…”

“…Lu et al . () removed Cretanallachius from Dilaridae based on a number of morphological characters, e.g. absence of setose tubercles on head and pronotum, absence of nygmata, posteriorly curved ScP vein and the male genitalia with external, broadly valvate, setose gonocoxites 9.…”

Early evolution of Nemopteridae illuminated with the first and oldest thread‐winged lacewing in Cretaceous amber

“…could be plesiomorphic if it was widely present in Mesozoic Nemopteridae or other Myrmeleontoidea. In some Burmese amber Psychopsoidea, which is the sister group of Myrmeleontoidea, the male gonocoxites 9 are also setose and even present as a pair of external sclerites in some basal groups of Neuropterida (Lu et al ., ).…”