Cette découverte augmente notre connaissance de la biodiversité de la superfamille de Psychodoidea pendant le Crétacé inférieur. La présence d’approximativement 20 espèces de Psychodoidea dans l’ambre libanais, suggère que le Crétacé inférieur ait été une période cruciale pour la diversification de ce groupe, et que l’histoire évolutive de ce dernier soit, de loin, beaucoup plus complexe qu’on pourrait l'imaginer en étudiant uniquement les espèces modernes.

Psychodoid flies are well diversified in the Lebanese amber and 13 taxa have been already studied from this material. Hennig [5] described two phlebotomid flies (Phlebotomites brevifilis and Phlebotomites longifilis) from the Lebanese amber (locality of Jezzine, southern Lebanon). Azar et al. [3] described two additive phlebotomid flies (Mesophlebotomites hennigi and Libanophlebotomus lutfallahi) and six psychodid flies (Paleopsychoda solignaci, Paleopsychoda jacquelinea, Protopsychoda nadiae, Protopsychoda hammanaensis, Libanopsychoda abillamai and Cretapsychoda inexpectata) from the Lebanese amber (locality of Mdeyrij/Hammana, central Lebanon); and Azar and Nel (2002) [2] described two more psychodids (Paleopsychoda inexpectata and Paralibanopsychoda agnieszkae) from the same previous locality. Finally, in 2003, Azar et al. [4] published a paper dealing with a new fly (Eophlebotomus gezei) belonging to an incertae sedis psychodoid family. Other recent field researches in Lebanon led to the finding of other new psychodoids that will be the subject of forthcoming publications.

We follow below the wing venational nomenclature of McAlpine [6] and the nomenclature of genital appendages proposed by the Computer-aided Identification of Phlebotomine sandflies of America (CIPA group project, University of Paris-6, Jussieu), available on the web site http://cipa.snv.jussieu.fr/

Superfamily Psychodoidea

Family incertae sedis

Genus Xenopsychoda gen. nov.

Derivatio nominis. From xenos (ξένoς) = strange in Greek + Psychoda

Diagnosis. Antenna with 15 flagellomeres, the last one being reduced and drop-like. First flagellomere as long as the 13 flagellomeres that follow the first one. Eye-bridge well developed. Palps longer than the head, 4-segmented. Mouth parts as long as head. Wings with all veins reaching wing margin. Presence of a crossvein between M2 and M3. CuA2 converging toward the anal vein (A1) without however reaching it. Anal vein and area well developed. Legs much longer than the body.

Xenopsychoda harbi sp. nov.

Derivatio nominis. This species is named in honor of Dr Faouzi Harb who discovered the amber outcrop of Tannourine where X. harbi has been found.

Material. Holotype specimen No. T 1, female, Azar collection, provisionnaly deposited in the ‘Muséum national d’histoire naturelle’ of Paris, France.

Type locality and horizon. The studied material comes from a locality between the Tannourine El-Faouqa and Laqlouq villages (Fig. 1), just on the boundary between Caza Jbeil and Caza (department) El-Batroun, between Mouhafazit Jabal Loubnan (Mount Lebanon) and (destrict) Mouhafazit Loubnan Esh-Shemali (North of Lebanon). The material is found in Lower Aptian geological layers, but its age must be older, as the amber has been reworked and redeposited. Other inclusions are shared with the other amber outcrops of Lebanon with biological inclusions and thus indicate a similar age.

Diagnosis. As for the genus.

Description(Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6, Fig. 7, Fig. 8 and Fig. 9). Antenna 1.03 mm long, with 15 flagellomeres, the last one being reduced and drop-like. Scape cylindrical shape, 0.05 mm long and 0.04 mm wide. Pedicel globular, 0.06 mm long and wide. First flagellomere 0.06 mm long, nearly as long as the 13 flagellomeres that follow the first one. All flagellomeres bearing long curved setae. (Fig. 3 and Fig. 4). Eye-bridge well developed. Eyes separated by a space of 0.04 mm. Area between eyes covered by few long setae. Palps longer than the head, 0.776 mm long, 4-segmented, all palpomeres setose (Fig. 4, Fig. 5 and Fig. 6). First palpomere 0.177 mm long, 0.027 mm wide. Second palpomere 0.161 mm long, 0.038 mm wide. Third palpomere 0.172 mm long, 0.038 mm wide. Last palpomere much longer than the remaining palpomeres, 0.266 mm long, 0.022 mm wide, presenting a very strange structure characterized by 10 constrictions or secondary annulations that give impression of presence of 11 segments. The first constriction is 0.04 mm far from the base of the last palpomere, then the second one at 0.08 mm from the base of the palpomere, the third at 0.12 mm, finally the remaining constrictions are very close one to other separated each by a mean of nearly 0.01 mm. Mouth parts nearly as long as head. Wing 2.05 mm long, 0.84 mm wide, hyaline (Fig. 7 and Fig. 8). Humeral vein (h) reaching the wing margin at 0.28 mm from the wing base. Subcostal vein Sc distally fused with R1 in a strong angle, 1 mm from wing base, and has a crossvein reaching the costal margin. R1 reaching the costal margin 1.61 mm from wing base. Rs separated from R1 0.53 mm from the wing base, 0.22 mm basal of fork of M into M1+2 and M3. Rs four-branched, with all its branches extending to wing margin. Branching of R2+3 0.39 mm distal of base of Rs, and 0.17 mm distal of fork of M. R2 and R3 separated 1.53 mm distally. R2 and R3 reaching costal margin. R4 and R5 separating 0.22 mm distal of base of R2+3. R4 nearly straight, reaching wing apex. R5 with a strong angle in its basal part and distally nearly straight. Crossvein r–m 0.59 mm distal of base of Rs. Fork of M into M1+2 and M3 0.76 mm distal of the wing base. Fork of M1+2 into M1 and M2 0.55 mm distal of base of M1+2. M1 distally nearly straight. M2 slightly shorter than M1. Basal part of M3 rather long. Presence of a short crossvein between basal part of M3 and basal part of CuA1. Presence of a crossvein between M2 and M3. CuA1 separating from CuA 0.73 mm distal of the wing base. CuA2 rather long, 0.28 mm long, converging toward A1 without however reaching it. A1 well developed, 0.82 mm long, and reaching posterior wing margin. Anal area well developed and forming a small lobe. Halteres 0.42 mm long. Knob 0.12 mm long and 0.07 mm wide. Stem 0.3 mm long. Thorax 0.68 mm long, 0.42 mm high. Legs very long, distinctly longer than the whole body. Foreleg 4 mm long, midleg 4.2 mm long and hindleg 4.5 mm long. Abdomen 0.92 mm long excluding genital appendages, and 0.56 mm wide. Female genital appendages (Fig. 9): the two subgenital plates (Sub) elongated, 0.06 mm long, 0.03 mm wide, rectangular-shaped with a central concavity, bearing relatively dense setae. Cerci (Cer) ellipsoid with spoon-like central concavity, small diameter parallel to the longitudinal axe of the insect 0.03 mm long, large diameter perpendicular to the longitudinal axe of the insect, 0.06 mm long.

After the keys to the families Psychodidae and Phlebotomidae proposed by Quate and Vockeroth [7] and Williams [8], Xenopsychoda harbi gen. et sp. nov. falls into the Psychodidae because of the following characters: (1) presence of an eyebridge; (2) palps with four palpomers; (3) anal vein A1 present and well developed. Xenopsychoda harbi gen. et sp. nov. shares with the phlebotomidae the well-developed blood-sucking mouthparts. We think that using these keys is not sufficient to give an attribution to our fossil.

Xenopsychoda harbi gen. et sp. nov. shares with Libanopsychoda abillamai Azar et al. 1999 [3] and Cretapsychoda inexpectata Azar et al. 1999 [3] the number of antennomeres, the wing venation organization, including the very peculiar and unique feature of convergence of CuA2 and A1, without however meeting each other; this character does not exist in any psychodoid fly, except those three taxa; nevertheless it exists also in some tanyderid flies. Xenopsychoda harbi shares also with those two latter taxa the presence of a crossvein between M2 and M3; this character is only known within psychodoids (in addition to those three taxa) in Paralibanopsychoda agnieszkae Azar and Nel 2002 and in Tanypsycha connexa Ansorge 1994 [1], and in tanyderid flies. Libanopsychoda abillamai Azar et al. 1999 [3] and Cretapsychoda inexpectata Azar et al. 1999 have a crossvein between R1 and R2+R3, and bizarrely Xenopsychoda harbi gen. et sp. nov. has a pyrite accumulation that looks like a crossvein in the same locality as in the two elderly cited taxa; this feature could be seen on only one wing, because the other one is twisted and thus difficult to observe. The question remains to discriminate whether Xenopsychoda harbi has really this feature or whether this is just an artifact. Only more findings of other specimens belonging to this same species could resolve this problem.

Azar et al. [3] have placed Libanopsychoda abillamai and Cretapsychoda inexpectata in Psychodinae after using the keys that are usually used for recent psychodoids. We think that those taxa as well as Xenopsychoda harbi must be placed in a new subfamily or even in a new family within psychodoids. Thus, as we lack more species that could be attributed to this group, we propose to attribute for the moment these three taxa to an incertae sedis subfamily or family. This later would be close to Psychodinae sensu stricto and share some features with Tanyderidae. Anyway, this proposition remains an assumption as long as a good cladistic phylogeny, including the recent and fossil taxa, is not done. This situation indicates also that keys must consider fossil taxa as well.

This discovery greatly increases our knowledge on the biodiversity of the psychodoid superfamily during the Lower Cretaceous. The presence of about 20 species of Psychodoidea in the Lebanese amber, representing a sample of the palaeofauna of a very small area, suggests that the Lower Cretaceous was a crucial period for the psychodoid diversification, and that the evolutive history of this group is much more complex than one could imagine by studying the only recent species.