Two laniatorid harvestmen (Opiliones: Cladonychiidae) from Eocene Baltic amber

Abstract. Two new specimens of laniatorid harvestmen (Opiliones: Laniatores) from Eocene (ca. 44–49 Ma) Baltic amber are described. One is clearly adult and assignable to Proholoscotolemon nemastomoides (Koch & Berendt, 1854) from the family Cladonychiidae. The second, smaller, specimen is treated as cf. Proholoscotolemon sp., but probably represents a juvenile instar of P. nemastomoides and allows us to infer at least part of the ontogenetic sequence for Baltic amber laniatorids. The juvenile differs from the adult in the relative size of the body, weaker pedipalpal spination and the absence of tarsomeres subdividing the tarsi at the ends of the legs. Of particular note is the presence of unbranched claws at the tips of legs III–IV. This was unexpected given that juveniles of extant cladonychiids usually have four to six branches on these claws; a condition also thought to be plesiomorphic for this, and some related families.

Laniatores (armoured harvestmen) is the most diverse suborder of Opiliones, with over 4200 living species (Kury 2017). These sometimes large and often spiny and/or tuberculate harvestmen are very rare as fossils. Six species are currently known from the fossil record (Dunlop et al. 2019) including one species described from Burmese amber (ca. 100 Ma), one in Baltic amber  and four in Dominican Repub lic amber (ca. 16 Ma). Noticeable at all of these deposits so far is the lack of fossil juveniles. Most of our recent knowledge about the development of harvestmen was summarized by Gnaspini (2007), who drew on earlier studies from the 1950s and 1970s. Comprehensive information about the ontogenetic series of modern laniatorids is primarily known from a few Neotropical species (e.g. Gnaspini 1995, Townsend et al. 2009), but Juberthie (1964) provided data for the European species Scotolemon lespesi Lucas, 1860 from the family Phalangodidae.
Here we describe a new specimen of the previously reported Baltic amber laniatorid Proholoscotolemon nemastomoides (Koch & Berendt, 1854) from the family Cladonychiidae, together with a juvenile assigned here to cf. Proholoscotolemon sp. which could well be an immature stage of Koch & Berendt's species. Previous interpretations of P. nemastomoides were reviewed by Ubick & Dunlop (2005). In brief, the amber fossil was originally placed in a South American genus. The body of the holotype is largely obscured by a white film, but the pedipalps and distal parts of the legs are better preserved and allowed Ubick & Dunlop (2005) to establish a new, extinct genus, probably closely related to the European Holoscoto lemon Roewer, 1915 (Cladonychiidae). The new juvenile fossil reveals, for the first time, data about the ontogenetic series of extinct laniatorids.

Material and methods
The two specimens studied here originate from the private collection of Jörg Wunderlich, and bear the repository numbers CJW BBF2568 (adult) and CJW BBF2545 (juvenile). Neither required further preparation or polishing and both were immersed in water to reduce refraction and photographed using a Leica Z16 APO A stereomicroscope running the software package Leica Application Suite. Stacks of ca. 15-20 images were combined using Helicon Focus 6 and edited for brightness and contrast using Adobe Photoshop CS5. Interpretative drawings were made on a Leica M205C stereomicroscope with a camera lucida attachment, whereby some setae or tubercles were occasionally omitted to enhance visibility of important characteristics. All measurements are in millimetres. Measured lengths may show some deviations due to the three-dimensional position of the inclusions within the amber matrix and average values are marked with an asterisk (*). Fossils were compared to extant harvestmen in the collection of the Museum für Naturkunde Berlin -primarily Holoscotolemon jaqueti Roewer, 1915 (Cladonychiidae) and Scotolemon terricola Simon, 1872 (Phalangodidae) -to the type of the amber species P. nemastomoides (Berendt collection, Museum für Naturkunde Berlin), as well as the literature on extinct and Recent species (Koch & Berendt 1854, Ubick & Dunlop 2005, Briggs & Ubick 2007. Baltic amber is by some margin the most intensively studied fossil resin deposit and an overview of its age, geological setting and the plant and animal groups recovered here can be found in Weitschat & Wichard (2010). The sediments hosting the amber are conventionally dated to the Lutetian stage of the Eocene (e.g. Wolfe et al. 2016) or ca. 44-49 Ma. At that time Europe consisted of several landmasses, with the Baltic amber forest perhaps belonging to a more northerly area (Dunlop et al. 2018: Fig. 1). A tropical to subtropical forest ecosystem has been inferred, dominated by an extremely warm and humid climate cumulating in the so-called Eocene Optimum at ca. 49 Ma. Several animal groups usually associated with warmer climates, of which laniatorid harvestmen are one example, are thus present in Baltic amber even though they are no longer present in north-central Europe today. Possible source trees for the resin which formed Baltic amber are araucarians, cedars or umbrella pines.
Koch & Berendt (1854) originally assigned the Baltic amber species to the Neotropical genus Gonyleptes Kirby, 1819 which belongs to the infraorder Grassatores. This makes little biogeographical sense; although in fairness Gonyleptes was the first living laniatorid genus to be described and there were relatively few genera known in the mid-nineteenth century. The Y-shaped claw on tarsi III-IV of at least the adult fossils clearly supports their referral to the other infraorder: Insidiatores. The habitus of the amber fossils is very similar to the extant Holoscotolemon Roewer, 1915 and Proholoscotolemon ne mastomoides has been proposed as the sister-group, or perhaps even ancestral, to this modern southern European genus. Regarding familial placement, Ubick & Dunlop (2005) transferred P. nemastomoides to Cladonychiidae. This family was subsequently downgraded to a subfamily of Travuniidae by Kury et al. (2014), but recent revision by Derkarabetian et al. (2018) accepted Cladonychiidae as a valid family group name and explicitly listed the Baltic amber genus within it. The position of the low ocularium, the long and robust chelicerae, the heavily spined pedipalps and the shape of tarsal claws III-IV in the fossils are all consistent with cladonychiids. Other families of Insidiatores often bear complex claws on tarsi III-IV with additional side branches (see also Discussion), whereas the amber specimen bears simpler claws without side branches (Ubick & Dunlop 2005).
Legs elongate (leg II longest); setose and spiny with metatarsi bearing pair of spines distally (Fig. 2a). Tarsi not sub-divided into tarsomeres; one claw on tarsi I-II, two claws with single insertion on tarsi III-IV. Leg length: Leg I trochanter 0.09, femur 0.30, patella 0.10, tibia 0.24, metatarsus 0.26*, tarsus 0.35*; total (trochanter-tarsus) 1.34. Leg II trochanter 0.12, femur 0.53, patella 0.17, tibia 0.47, metatarsus 0.46, tarsus 0.70; total (trochanter-tarsus) 2.45. Leg III trochanter 0.12, femur 0.30, patella 0.16, tibia 0.26, metatarsus 0.32, tarsus 0.33; total (trochanter-tarsus) 1.49. Leg IV trochanter 0.12, femur 0.47, patella 0.17, tibia 0.35, metatarsus 0.53, tarsus 0.40; total (trochanter-tarsus) 2.04. Ventral region equivocal, partially obscured by detritus. Remarks. With a body length of less than half a millimetre, CJW BB F2545 is noticeably smaller than the previously described Baltic amber laniatorids with body lengths of ca. 2-3 mm (Ubick & Dunlop 2005, see also above) and is evidently a juvenile due to the absence of any subdivision of the leg tarsi into tarsomeres. The position of the ocularium, the already large pedipalps bearing spines and the form of the chelicerae are consistent with it being a cladonychiid, and we strongly suspect that it could be a juvenile instar of Proholoscotolemon nemastomoides. Since we lack a full ontogenetic sequence for this species, and cannot rule out the possibility that more than one species of laniatorid was present in the Baltic amber forest, we place this second fossil as cf. Proholoscotolemon sp.

Discussion
Postembryonic development has been documented for only a few modern harvestman (reviewed by Gnaspini 2007), with a proposed sequence of egg, larvae, up to eight nymphs (the last called the subadult), plus an adult. In living laniatorids the main ontogenetic changes are an increase in the armature of the pedipalps and increasing numbers of tarsomeres (e.g. Gnaspini 2007, Townsend et al. 2009). These trends are reflected in the juvenile fossil described here. Its body is rather small and rounded in relation to the legs, and there are no signs of free tergites at the back of the opisthosoma as would be expected in the eventual scutum magnum condition. The position and form of the ocularium and chelicerae in the juvenile fossil are almost identical to that in the adult amber laniatorids. The juvenile already has spiny pedipalps, but the exact number of spines differs on each limb article compared to the adult. The juvenile bears one to eight spines less per article of the pedipalp, except on the tarsus where the number of spines is similar to the adult condition. The tarsus is undivided which together with the weak pedipalp spination and the fact that the body length is less than a quarter of the probable adult length may hint that this is an earlier juvenile stage as opposed to a late stage or subadult. We are not aware of any detailed comparative studies covering all immature life stages within Cladonychiidae, but an illustration of the pedipalp (with almost no spination) in a juvenile European harvestman Scotolemon lespesi Lucas, 1860 from the Grassatores family Phalangodidae by Juberthie (1964: fig. 27) suggests that the amber fossil may be at a later stage rather than a recently hatched juvenile.
Of particular interest is the condition of the legs and their claws. The juvenile described here (CJW BB F2545) shows strong leg spination, with a regular spine pattern on all metatarsi. By contrast, the putative adult bears mostly smooth legs with only small, sparse setae on the tarsi. We are not aware of reports of legs being spinier in juvenile laniatorids as compared to adults. Another unexpected detail relates to the tarsal claws. The smooth, unbranched claws of legs III-IV of the juvenile in amber are somewhat similar to the adult condition, although it should be stressed that the claws in the adult are essentially branched and Y-shaped. By contrast, modern juveniles of cladonychiids (and some other insidiatorid families) usually bear complex claws on legs III-IV with four to six side branches (Ubick & Dunlop 2005, Briggs & Ubick 2007: the family name Cladonychiidae literally means "branched claw". The presence of these side-branches, retained in at least the juvenile stages of Insidiatores, has been used to infer that this is a plesiomorphic condition for these harvestmen. An immature Eocene cladonychiid which does not have side branches on its claws may draw this hypothesis into question.
Alternatively, if it is the juvenile of Proholoscotolemon ne mastomoides it may be possible to define the fossil genus on having the unusual character of unbranched claws on legs III and IV throughout its life cycle. We should also entertain the possibility that the paired claws in the amber juvenile are fused at the base -in which case they could be considered branched -or that if the claws are separate at the base then the fossil may instead belong to Grassatores. The available resolution of the fossil does not allow us to investigate this further. It would also be useful to have comparative studies into claw ontogeny in the probably closely-related modern genus Holoscotolemon, but this data is not currently available in the literature.