Three new species of Tetranychidae (Acari, Prostigmata) from the French alps (South-Eastern France).

Collection efforts in the framework of the European All Taxa Biodiversity Inventory conducted in the Mercantour national Park located in the Alps mountain range of southern France disclosed three new species of Tetranychid mites. The species described in the current paper are: Bryobia cinereae n. sp., Bryobia mercantourensis n. sp. and Eotetranychus quercicola n. sp.. Both Bryobia species were collected on Genista cinerea and E. quercicola on Quercus pubescens. A new combination is also proposed for Bryobia longisetis, previously placed in Pseudobryobia by Wainstein (1960).


INTRODUCTION
All Taxa Biodiversity Inventories (ATBIs) are promoted to increase the knowledge about the biodiversity of particular areas. Samplings are carried out to achieve a baseline biodiversity assessment of fauna and flora and to provide ecological information on the distribution, abundance and biology of the species recorded. Within the framework of the second ATBI Mercantour-Alpi marittime (De Biaggi et al., 2010), promoted by the European Distributed Institute of Taxonomy (EDIT), the Mercantour National Park and the Muséum National d'Histoire Naturelle (MNHN), that took place in the Mercantour National Park located in the French Alps, in south-east of France, we have collected three new species of tetranychid mites. Among the forty-four species of Tetranychidae recorded from France only 6 are endemic from this country (Mi-geon andDorkeld, 2006-2013). Four of them belong to the genus Bryobia Koch, 1836, one to the genus Schizonobia Womersley, 1940 and one to the genus Eoetranychus Oudemans, 1931. In the present work we report the description of two new species of Bryobia and a new Eotetranychus. According to the examination of morphological key-characters, Bryobia longisetis Reck, 1947 is a new combination provided for the taxon previously known as Pseudobryobia longisetis.

MATERIALS AND METHODS
Mites were collected directly from field samples in 70 % ethyl alcohol. Following clearing in lactic acid (50 %) for 24 to 48 hours they were mounted in Hoyer's medium. The specimens were examined using a Leica DMLB phase contrast microscope and illustrated with the aid of a camera lucida. Measurements were performed using the imaging software Perfect Image® (Clara Vision) coupled with Pro-gres® Capture Pro 2.6 software for image acquisition. The setal nomenclature used in the description follows Lindquist (1985). Legs setal count is given in the order: coxa, trochanter, femur, genu, tibia and tarsus. Numbers of setae refer to tactile setae, solenidia are given in parentheses and alternative counts are given in brackets. All measurements are given in micrometers and correspond to the holotype followed, in parentheses, by minimum and maximum values from paratypes. Setae are measured from theirs bases to their tips.

Bryobia longisetis
According to the literature compiled we came to the conclusion that this species should not belong to the genus Pseudobryobia. First, the absence of prodorsal lobe over the gnathosoma can be questioned. In its original description, Reck (1947) reported that the outer prodorsal lobes are small but inners are cone-shaped almost fully fused.
In the drawings of this species by Bagdasarian (1957), Reck (1959), Wainstein (1960) and Livshits and Mitrofanov (1966), inner and outer prodorsal lobes are similar to those previously described by Reck (1947): outer lobes are actually reduced to small tubercles but inner ones are coalescent into a tall cone with a small incision at the apex. Second, the dorsal pattern observed in this species does not correspond to that typical of the genus. Members of the fourth pair of hysterosomal dorsocentral setae (f 1 ) are never more or less in line with other dorsocentral setae. In the drawings of Bagdasarian (1957) and Reck (1959), f 1 setae are clearly located in marginal position, close (but not contiguous) to f 2 . In Wainstein (1960) and Livshits and Mitrofanov (1971), f 1 setae are almost in marginal position, they are not in the normal longitudinal dorsal position and the distance between them is superior to that between f 2 setae. Finally, the coxal chaetotaxy does not fit with that of species belonging to the genus Pseudobryobia. In the descriptions of this species given by Wainstein (1960) and by Livshits and Mitrofanov (1971), only one setae is present on the coxa II (coxal formula 2-1-1-1). Thus it is different to that of the genus Pseudobryobia and corresponds to that observed in the genus Bryobia. Although we did not had an opportunity to examine the holotype (or types), given the morphological characters cited above we consider that this species belongs to the genus Bryobia.
Legs -Length inferior to body length. Leg I 308 (299 -312) µm long (length of holotype and variations of three paratypes, measured from trochanter to tarsus), leg II 222 (220 -227), leg III 219 ( Tarsus III associated setae serrate and approximate with solenidion forming duplex, the tactile member slightly longer and proximal (Fig. 2E); tarsus IV with solenidion well-separated from tactile, proximal, about one third the length of tactile (Fig. 2F). True claws uncinate, with one pair of tenent hairs, empodial pads each bearing two rows of tenent hairs (Fig. 2G).

20
Etymology -The specific epithet cinereae refers to the species name of the host plant on which mites were collected.
Remarks -The combination of prodorsal lobes poorly developed and dorsal setae not spatulate but elongate with dorsocentral setae inferior in length to the distance between consecutive setae brings this species close to B. sarothamni Geijskes, (1939), B. longisetis Reck, (1947), B. artemisiae Bagdasarian (1951), B. variabilis Manson, (1967) and B. serifiotica Hatzinikolis, Papadoulis and Kapaxidi (2007). It can be distinguished from B. sarothamni by the dorsal hysterosomal setae which are lanceolate, shorter, variable in length (h 1 the largest) versus elongate and subequal in length in B. cinereae. Bryobia longisetis can be separated from B. cinereae by the propodosomal inner projection which is more developed, by medial lobes which are almost fused, by f 1 and f 2 dorsal setae which are not contiguous and by a different leg chaetotaxy. In B. variabilis (the form bearing long, slender and serrate dorsal setae) the leg setal formula is different and dorsal setae are slender, variable in length (c 2 , c 3 , d 1 and d 2 being much smaller) whereas quite stout and subequal in length in B. cinereae. Bryobia serifiotica differs from B. cinereae by the dorso hysterosomal setae larger in the latter, by the difference in size of the vertical setae (v 1 slightly inferior to v 2 in B. serifiotica vs. v 1 up to three times smaller than v 2 in B. cinereae), by the position of f 1 setae (more or less in normal position and well separated in B. serifiotica whereas f 1 and f 2 are in marginal position and contiguous in B. cinereae) and by the solenidion of the tarsus IV (associated with a tactile setae but well separated and proximal in B. serifiotica and B. cinereae respectively). Depending on the literature referred, B. artemisiae is more or less close to B. cinereae. In the original description by Bagdasarian (1951), prodorsal lobes are similar in the two species but vertical setae (v 1 and v 2 ) are subspatulate to spatulate (elongate in B. cinereae) and dorsal hysterosomal setae are short and fan-shaped (elongate in B. cinereae). According to Reck (1959), dorsal setae of B. artemisiae vary from short spatulate to slightly elongate and, in his drawing, v 1 and v 2 setae are elongate and lanceolate (only elongate in B. cinereae). Wainstein (1960) mentions that dorsal setae are narrowly spatulate and almost elongated. In Livshitz and Mitofanov (1971) and Mitrofanov et al. (1987) the drawing of B. artemisiae in habitus resembles to B. cinereae: prodorsal lobes are small, v 2 are longer than v 1 , dorso hysterosomal setae are elongate and inserted on tubercles. However, v 1 and v 2 setae are spatulate and subspatulate (narrow in B. cinereae), dorso central setae (c1, d 1 , e 1 ) are longer to dorso lateral (similar in length in B. cinereae) and the leg setal count is different.

Bryobia mercantourensis n. sp. (Figures 4-7)
Type-specimens -Holotype (female) Diagnosis -With four long setae present on the interior dorsal row of femur I this species belongs to the berlesei-group (Eyndhoven, 1957;Eyndhoven and Vacante, 1985). Empodial pad of leg I with a pair of tenent hairs others with two rows of tenent hairs, inner propodosomal lobes are well separated and more or less cone-shaped with large fused base, outer lobes smaller and cone shaped, dorsal setae inserted in small tubercles, spatulate with sacral and clunal setae slightly longer.

Description:
Female: Holotype 600 µm long (excluding gnathosoma, from the tip of v 1 to the tip of h 1 ), width 350 µm. Ten paratypes measured, 540 -595 µm long, width 310 -360 µm. Large transverse folds with fibrous appearance on hysterosoma, more or less arched in the distal part comprised between e 3 and h 1 setae. Three pairs of oval-shaped areas present between c 1 -c 2 , d 1 -d 3 , and e 1 -e 3 setae and a triangularly rounded one present posteriorly.
True claws uncinate with one pair of tenent hairs, empodia with two rows of tenent hairs.
Remarks -In addition to the four long setae present on the interior dorsal row of femur I, as this species bears one pair of tenent hairs on the empodium I, B. mercantourensis is close to B. provincialis Eyndhoven andVacante, 1985 andB. dikmenensis Eyndhoven andVacante, 1985 that belong to the berlesei-group (Eyndhoven, 1957;Eyndhoven and Vacante, 1985). This species is clearly smaller in length and width than B. provincialis and the first leg is also obviously longer in the latter. Conversely, B. mercantourensis is slightly longer and obviously broader than B. dikmenensis and the second, third and fourth pairs of legs are shorter in the latter. It is mainly distinctive from B. provincialis and B. dikmenensis by the shape of the propodosomal lobes: mammelliform with inner lobes largely fused in the latter whereas conical and well separated in B. mercantourensis. The latter can also be separated from B. provincialis by differences in shape of deutonymph's dorsohysterosomal setae e 3 and f 1 , subspatulate vs. elongate and narrow in B. provincialis and B. mercantourensis respectively. Legs chaetotaxy also clearly differs between the deutonymphs of these two species. Bryobia dikmenensis can be distinguished from B. mercantourensis and from B. provincialis by the reduced size of its second and third pairs of dorsocentral setae (d 1 and e 1 ) in comparison with other dorsohystersomal setae. Several characters found in juveniles of B. dikmenensis and of B. mercantourensis can also be used to separate them: the ratio between larval v 1 and v 2 setae is two-fold higher (4 vs. 2) in B. mercantourensis; protonymphal prodorsal lobes in B. dikmenensis ressemble that of female whereas they are almost absent (weakly developed) in B. mercantourensis.
Etymology -The species designation mercantourensis is named after the location where the specimens were found: in the Mercantour French National Park.
Venter -Area immediately anterior to genital flap with transverse striae, genital flap with transverse slightly arched striae typical of willametteigroup (Pritchard and Baker, 1955) (Fig. 12D). Lobes on ventral striation present laterally between third pair of ventral setae (4a) and aggenital pair (ag), rare poorly developed lobes may be present between members of 4a, anteriorly and posteriorly. Two pair of para-anal and two pairs of anal setae.
Remarks -With dorsal setae longer than the intervals between them and eight tactile setae on tibia II this species is assigned into the tiliarium group (Pritchard and Baker, 1955), its genital area pattern corresponds to the willamettei species group (Tuttle et al., 1976;Baker and Tuttle, 1994) and because