On the spider species described by L . Koch in 1882 from the Balearic Islands ( Araneae )

Examination of the L. Koch collection of the Zoological Museum in Berlin allows us to propose the following new synonyms and combinations: Erigone marina L. Koch, 1882 = Oedothorax fuscus (Blackwall, 1834) n. syn.; Theridion elimatum L. Koch, 1882 = Enoplognatha diversa (Blackwall, 1859) n. syn.; Liocranum variabilis Wunderlich, 2008 = Zora inornata L. Koch, 1882 n. syn. = Liocranum inornatum n. comb.; Lycosa perspicax L. Koch, 1882 = Arctosa fulvolineata (Lucas, 1846) n. syn.; Alopecosella Roewer, 1960 = Arctosa C. L. Koch, 1847 n. syn.; Lycosa subhirsuta L. Koch, 1882 = Arctosa lacustris (Simon, 1876) n. syn.; Philodromus vegetus L. Koch, 1882 = Thanatus vulgaris Simon, 1870 n. syn.; Ozyptila bicuspis Simon, 1932 = Ozyptila furcula L. Koch, 1882 n. syn.; Haplodrassus maroccanus Denis, 1956 = Drassus parvulus L. Koch, 1882 n. syn. = Haplodrassus parvicorpus (Roewer, 1951) n. comb. (replacement name); Zelotes ruscinensis Simon, 1914 = Zelotes semirufa (L. Koch, 1882) n. syn.; Phlegra simoni L. Koch, 1882 = Phlegra bresnieri Lucas, 1846 n. syn.; Trochosula conspersa (L. Koch, 1882), Lycorma fraisnei (L. Koch, 1882), Lycorma insulana (L. Koch, 1882), Arctosa misella (L. Koch, 1992) and Pirata simplex (L. Koch, 1882) are all retransferred to their original genus Lycosa stat. rev. Cheiracanthium occidentale L. Koch, 1882, Ozyptila furcula L. Koch, 1882 and Zelotes callidus (Simon, 1878) are redescribed.

In recent years, the number of descriptions of new species has increased considerably, greatly expanding the knowledge of spiders.However, it is also very important to revise or redescribe species from the 19 th century.Some of these older descriptions are incomplete and not accompanied by illustrations, but others have adequate descriptions with excellent figures.One of the papers that has been completely forgotten is 'Zoologische Ergebnisse von Excursionen auf den Balearen.II.Arachniden und Myriapoden' by L. Koch (1882).The author described 28 new species of spiders in it, of which only ten have been fully redescribed, which means that 18 of them are left as poorly known species.

Material and methods
Type material of nine species could be loaned from the Berlin Museum (possible other locations of type material in London or Vienna were not checked).If the material was not present, the descriptions and figures were carefully examined and compared with similar or related species from the Mediterranean.
The reference material mentioned in the present paper is part of the collection of the first author.Specimens were examined and illustrated using a Wild M5 stereomicroscope.Further details were studied using an Olympus CH-2 stereoscopic microscope with a drawing tube.Left structures are depicted.Male palps were detached and transferred to glycerol for examination under the microscope.Female genitalia were excised using sharpened needles.These were transferred to clove oil for examination under the microscope.Later, palps and epigynes were returned to 70% ethanol.
The following abbreviations are used in the text: CRB: Collection Robert Bosmans; MNHNP: Muséum national d'Histoire naturelle, Paris; ZMB: Zoologisches Museum Berlin Comments on the species described by L. Koch Many authors of the 19 th or the beginning of the 20 th century, for example Thorell and Strand, did not present figures at all, so these species cannot be recognized without examination of the type material.Koch (1882) reported twenty-eight new spider species from the Balearic Islands.His descriptions are very accurate and were accompanied by excellent figures, which allow the recognition of details in the palpal and epigynal structures.Careful examination of the figures should have allowed identification of several species, especially when compared with other material from the Mediterranean region.It is therefore surprising that L. Koch's paper has been neglected by previous authors.Of the twenty-eight described species, only ten have subsequently been studied.Out of these ten redescribed species, seven appeared to be synonyms, and only Iberesia brauni, Zelotes semirufus and Z. flagellans remain valid.The ten species concerned are listed in Tab. 1.
Twelve of the eighteen remaining species have received new names by preoccupation, were transferred to other genera or were declared nomina dubia.However, none of them were redescribed.The other species have never been subsequently cited, except in catalogues.These species are listed in Table 2.

Comments
In absence of the palps, a complete new diagnosis of this species cannot be given.In Mallorca, Dysdera crocata C. L. Koch, 1838 is the commonest Dysdera spe-cies (authors' personal observations) and the holotype of Dysdera mordax was compared with this species.In D. mordax, the rugosity of the prosoma and the sternum is different and the colour is more burgundy red.Spination of the holotype may be incomplete, but no spines are observed on legs I and II (probably lost), leg IV has 2 basal spines on the femora, legs III-IV have 2 pairs of lateral spines and 1 pair of ventral spines on the tibia, and several spines on the metatarsi.In D. crocota tibiae III-IV have fewer spines.For the male palp we must rely on the figure of Koch (1882; see Fig. 1).The bulb is similar to that of Dysdera crocota, but more slender, narrowing terminally and more pointed.We consider it a valid species, but topotypic material is needed for a complete redescription.Distribution So far, the species is an endemic to Mallorca.

Comments
Koch's figure 16 (1882) clearly shows a Cheiracanthium-like epigyne (Fig. 8), but does not give enough diagnostic characters for identification.A redescription of the epigyne and vulva is therefore given here.According to the author, the species is related to Cheiracanthium letochae L. Koch, 1876(= C. elegans Thorell, 1875).The epigyne has a median depression with an interior margin and is wider than
Comparative material examined

Comments
The holotype male has only one palp left with a tegular apophysis as in Arctosa lacustris.

Type material
Type series of two females of Lycosa conspersa from Spain, Baleares, Mallorca, Ses Prat de San Jordi, end April, and Soller, mid-May, Schaufuss leg.(ZMB 7912); examined.The two females belong to different species.One of the females is Arctosa fulvolineata, the other one has an epigyne corresponding to the original drawing by Koch (1882) and is selected here as the lectotype.

Comments
This species has not been mentioned since the original description, with the exception of roeWer's (1955) transfer to the genus Trochosula, without any justification.Koch's figure 33 (1882) of the epigyne resembles the epigyne of Hogna radiata (Latreille, 1817), a widespread species in the Mediterranean.Like many large lycosids in the Mediterranean, this species complex is in need of revision and a conclusion about synonymy has to be postponed.Material from Mallorca is needed to resolve the situation.This is also the case for three other Lycosa species described by L. Koch from Mallorca (see below): Lycosa fraissei, L. insulana and L. simplex.They are all large species and Koch's figures 33-36 all show the same type of epigyne (see .roeWer (1955) transferred these species (without further justification) to three different genera: Trochosula, Lycorma and Pirata.To facilitate future studies, we consider it better to return them to their original genus.Topotypic material may help solve the identity of these problematic species.

Distribution
The type locality on Mallorca.
The type locality on Mallorca.

Type material
Juvenile holotype of Lycosa misella from Spain, Baleares, Ses Prat de Jordi, end April, Schaufuss leg., not examined, unavailable in ZMB.

Comments
The holotype is a juvenile specimen and its status is unclear.Lycosa misella is therefore a nomen dubium.

Distribution
The type locality on Mallorca.

Type material
Holotype male of Lycosa simplex Spain, Baleares, Palma city moats, begin May, and Ses Prat de Jordi, beginning of May, Schaufuss leg.; not examined, unavailable in ZMB.

Comments
The holotype could not be examined.According to the original description by Koch (1882)  Lycosa fraissei L. Koch, 1882 stat.rev.

Comments
A large lycosid of 19 mm total length.The type material is not available.See comments under Lycosa conspersa.

Distribution
The type locality on Mallorca.
Comments roeWer (1955) created the genus Pardosops for this and other species, but tongiorgi (1966) synonymised it with Pardosa.The holotype male could not be examined.Koch's (1882) figure 24 offers no indication of its affinities (see Fig. 21).Topotypic material is needed to clarify its systematic position.

Comments
Figure 22 in Koch (1882) shows a rounded epigynal plate with some concentric circles, typical for the epigyne of the common Thanatus vulgaris (compare Figs 22 and 23).This species has no less than eight synonyms in the Mediterranean region (leVy 1977) and there is no doubt that Philodromus vegetus is yet another one.Distribution Circum-mediterranean.

Type material
Holotype male of Ozyptila furcula from Spain, Baleares, Mallorca, Ses Prat de San Jordi, end of April, Schaufuss leg.; not examined, unavailable in ZMB.

Comments
The type material of this species is not available,

Distribution
The species is currently known from the south of France and from Spain.
Being preoccupied by Drassus parvulus Lucas, 1846, roeWer (1951) offered Drassus parvicorpus as a replacement name.The species has never been cited since.
According to Koch (1882), this species is related to Drassus minusculus (= Haplodrassus dalmatensis) and differs by the more arched prosoma with a black margin, and the disposition of the eyes, with the PM not touching and the AM separated by nearly their diameter.These characters are too variable in Haplodrassus to distinguish species, but Koch's figure 10 (1882) shows a male palp with a large, subterminal tooth on the tegular apophysis, and his figure 11 shows a palpal tibia with an apophysis as long as wide (Fig. 27).The male palps of the holotype male are both present and in good condition (Fig. 28).Examination shows that they are identical to the palps of Haplodrassus maroccanus, described from Morocco by Denis (1956) and occurring all over the south-western Mediterranean (unpublished data).In H. dalmatensis, the tegular apophysis has only a small subterminal tooth, and the tibial apophysis is shorter than wide; thus there are sufficient diagnostic characters to separate the species.Haplodrassus maroccanus Denis, 1956 becomes a junior synonym of H. parvicorpus (Roewer, 1951 n. comb.).The synonymy of H. maroccanus with H. dalmatensis, proposed by leVy (2004) is rejected.The species will be fully redescribed in a further paper.Distribution Mallorca and Morocco.

Comments
The holotype female could be examined.The dorsal stripes on the prosoma and abdomen have disappeared, but the clypeus is densely covered with white hairs.The long, threadlike embolus is clearly visible and its placement in the genus Phlegra is thus confirmed.According to Koch (1882), Phlegra simoni differs from P. bresnieri in the male palpal tibia which bears black hairs.Examination of some specimens of P. bresnieri from our collection show that the femora

Comments
According to Koch (1882), this species measures 5 mm and the prosoma is brownish black, the abdomen yellowish brown, the legs reddish brown with red femora and the spinnerets yellowish brown.

Conclusions
Taxonomic studies by earlier arachnologists are of variable quality, but until there is proof to the contrary, they all have to be considered valuable.Some authors working on Mediterranean spiders in the past -for instance (but not exclusively) lucas (e.g.1846), O. P.-camBriDge (e.g. 1872O. P.-camBriDge (e.g. , 1876) ) and Kulczyński (e g. 1908Kulczyński (e g. , 1911) ) -presented detailed descriptions accompanied by excellent figures of the general pearance and sexual organs, making identification very easily possible.In most cases, however, examination of the type material is necessary to come to a definitive conclusion.Other authors gave long, very detailed descriptions but presented no figures at all; like for instance paVesi (e.g.1880, 1884) and the numerous papers by thorell (e.g.1875) and stranD (e.g. 1906, 1908).Here, recognition of the species is not possible without examining the original type material.However, even if these species are not instantly recognizable, their names remain valid until has been stated in a publication that the types are not available; like for instance many types of Strand destroyed in the last World War.
In the case of the study by L. Koch (1882) on the spiders of the Balearic Islands, the paper includes good drawings, in many cases making identification possible.Only ten of the 28 described species were redescribed previously, resulting in three valid names and seven synonyms.Type material of ten of the remaining 18 species could be examined by us.Five of these nine species names are valid: Cheiracanthium occcidentale, Dysdera mordax, Liocranum inornatum n. comb.and Haplodrassus parvicorpis n. comb.Four out of the nine species appeared to be synonyms of species described earlier.Erigone marina, Lycosa perspicax, Lycosa subhirsuta and Phlegra simoni are the junior synonyms of Oedothorax fuscus, Arctosa fulvolineata, Arctosa lacustris and Phlegra fasciata respectively.Finally, Trochosula conspersa is retransfered to its original genus, Lycosa, where it awaits further study.
Of the eight species whose types could not be examined, the drawings of L. Koch were care-fully studied and three species could be recognized.Ozyptila furcula was recognized and redescribed.Philodromus vegetus and Theridion elimatum were recognized as junior synonyms of Thanatus vulgaris and Enoplognatha diversa respectively.Lycosa fraissei, L. insulana, L. simplex and Pardosa tenuipes could not be recognized and are left in their original genus.Finally Lycosa misella is declared a nomen nudum, since the description was based on a juvenile specimen.
of L. majus with Liocranum apertumDenis, 1954, L.  pallidulum Simon, 1878 and L. segmentatum Simon,  1878.According to WunDerlich (2008), the helmet-like structure is larger in L. major.Distribution So far, the species is an endemic to Mallorca.
The spermathecae are relatively small, only 1/3 of the width of the depression.The copulatory openings are situated antero-laterally of the depression and the sperm ducts make three coils to the elongated spermathecae (Figs 9-10).In the literature, no species with such a vulva could be found and hence the species is considered valid.
Simon, 1878m Koch's drawing of the epigyne (1882, fig.19) it is evident that this species does not belong in the genus Zora.The figure shows a large, anterior pocket and two smaller lateral pockets (Fig.6).Examination of the specimen shows it to belong in the Liocranidae and that it is identical to Liocranum variabilis, only recently described from Mallorca by WunDerlich (2008), see Fig.7.This latter species thus becomes a junior synonym.Liocranum inornatum is closely related to L. majusSimon, 1878,  recently redescribed by leDoux (2008).This author confirmed the synonymy long.
The female has an intact epigyne with a broad median septum, as clearly shown in Koch's figure 29 (1882), corresponding well with the epigyne of A. lacustris(compare Figs  13, 15 with Figs 14, 16 (taken from Knülle 1959).

but
Koch's figure 21 (1882)allows a positive identification with respect to the tegular apophysis of specimens collected by us in Spain and North Africa.The tegulum has an oblique banana-shaped tegular apophysis with two postero-median concavities (compare Figs 24 and 25).Such an apophysis exists in three Ozyptila species occurring in this region: O. pauxilla Simon, 1870, O. perplexa Simon, 1875 and O. bicuspis Simon, 1932.Koch's species also has two postero-median incisions, and these are absent in O. pauxilla and O. perplexa.We consider O. furcula and O. biscuspis the same species and O. bicuspis Simon, 1932 becomes a junior synonym of O. furcula L. Koch, 1882.Ozyptila pauxilla and O. perplexa will be redescribed in a separate paper.