Research Article |
Corresponding author: Raul Bonal ( raulbonal@unex.es ) Academic editor: Miquel A. Arnedo
© 2016 Eduardo Morano, Raul Bonal.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Morano E, Bonal R (2016) Cheiracanthium ilicis sp. n. (Araneae, Eutichuridae), a novel spider species associated with Holm Oaks (Quercus ilex). ZooKeys 601: 21-39. https://doi.org/10.3897/zookeys.601.8241
|
We describe a novel species Cheiracanthium ilicis sp. n. (Araneae, Eutichuridae) collected in the province of Toledo (Central Spain). It was found during a systematic sampling campaign carried out in an agricultural landscape with isolated Holm oaks Quercus ilex and small forest patches. Its morphology and affinities with other species of the genus are discussed. Furthermore, one mitochondrial gene was sequenced to confirm species membership and its differentiation from other Cheiracanthium species. The molecular phylogenies based on mitochondrial and nuclear genes showed a close relationship of C. ilicis sp. n. with C. inclusum and C. mildei, with which it also shares morphological similarities. Nonetheless, the sparse sampling of the phylogeny, due to the low number of sequences available, impedes drawing any definitive conclusion about these relationships; it is first necessary to perform an extensive review of the genus worldwide and more thorough phylogenies. C. ilicis sp. n. also shares certain ecological and phenological characteristics with C. inclusum and C. mildei. Like them, C. ilicis sp. n. is an obligate tree dweller that prefers a tree canopy habitat and reproduces primarily in late spring and summer. From a conservation perspective, the present study suggests the need to preserve isolated trees in agricultural landscapes. They are not only the refuge of common forest organisms but also of novel species yet to be discovered.
Cheiracanthium ilicis sp. n., DNA taxonomy, Iberian Peninsula, isolated trees, molecular phylogeny
Cheiracanthium C. L. Koch, 1939 is the only genus of the family Eutichuridae Lehtinen, 1967 in Europe. This genus was transferred from the family Clubionidae Wagner, 1887 to Miturgidae Simon, 1886 (
Cheiracanthium has worldwide distribution and is only absent from the polar regions. Of the 209 known species of this genus in the world (
Spiders of this genus are swift hunters on woody or herbaceous plants, and their dense claw tufts help them to crawl along inclined surfaces. Their body colours usually range from yellow to greenish, with orange and brownish tones in some species. Cheiracanthium spiders belong to the group known as “sac spiders” because they spin small silk bags which shelter these nocturnal hunters during the day. These bags are quite conspicuous, as often the spiders build them on top of tall grass shoots and so they are easily seen in wet meadows, on crops and besides paths.
In this article we describe a novel species of Cheiracanthium found during a sampling campaign carried out in Central Spain. Different habitats were periodically sampled in an agricultural landscape with isolated oaks and forest fragments interspersed between crop fields and grasslands. In addition to its morphological description we sequenced two genes (mitochondrial and nuclear) to build a molecular phylogeny and assess its status with respect to those species of the genus for which molecular data were available in GenBank. Finally, we provide data on habitat selection and phenology recorded during a one-year long systematic sampling.
We carried out the spider sampling in the village of Huecas, in the province of Toledo, Central Spain (40.02°N, 4.22°W; altitude 581 m.a.s.l.). The climate is dry Mediterranean, with hot summers in which temperatures may reach 40°C and scarce precipitation (365 l/m2) concentrated in spring and autumn. The study area extends over 9 km2 of flat agricultural landscape with isolated Holm oaks Quercus ilex and forest plots interspersed within a matrix of grasslands and cereal fields (Picture
In 2013 we conducted a systematic sampling, collecting spiders once a month from January to December. We sampled four different habitats: tree branches, tree trunks, grasses and soil. We then randomly selected 23 Holm oaks (isolated ones and within a forest patch). Tree spiders were collected by shaking the branches, and beating the canopy of each tree six times in each cardinal direction. We placed a white sheet below the branches and immediately collected all of the spiders falling onto it. Trunk traps consisted of a mosquito net attached to the tree trunk with an inverted cone with a closed bottle on the top. They covered the trunks partially and trapped the spiders that climbed the trunks and eventually walked into the net. At 10–15 meters from each study tree we set up a pair pit-fall traps separated by 10 metres. These traps were located in grasslands and consisted of a cone through which ground-dwelling spiders fell into a bottle filled with 90% ethanol and 10% glycerine to preserve the specimens. The traps were protected from direct sunlight by a small plastic roof to prevent alcohol evaporation. Lastly, grass spiders were sampled using a sweeping net along two 10-m long transects on both sides of the straight line joining the two pitfall traps. All specimens were preserved in 96% alcohol for further anatomical and molecular analyses and placed individually in Eppendorf tubes with all the information on the collection date and habitat.
The spiders were inspected under a Meiji EMZ-5 estereomicroscope. Drawings were made and photos were taken of specimens and their copulatory organs using a Canon EOS 350D camera connected to the estereomicroscope. All of the specimens were separated by age (adults and immatures) and sexed whenever the development of the sexual organs so permitted. In some females, the epigyne was removed, cleaned and mounted on slides for further analysis of the internal anatomy. In the case of males, only one palp was extracted for a detailed study. The epigyne and palp removed were placed in microvials within the Eppendorf tubes of the corresponding specimens. The individuals were deposited in the collection of the
In the present study the opisthosoma length has been measured without the spinnerets and the pedicel. The total leg length (femur, patella, tibia, metatarsus, tarsus) and the leg spination pattern follow the model of
Abbreviations: Eyes: ALE – Anterior lateral eye(s). AME – Anterior median eye(s). PLE – Posterior lateral eye(s). PME – Posterior median eye(s). imm – immatures. CS – cymbial spur; C – conductor; TA – tegular apophysis.
To confirm the species identity of the specimens classified as Cheiracanthium ilicis sp. n. the DNA of three individuals was extracted following the salt extraction protocol (
To further assess the phylogenetic relationships between C. ilicis sp. n. and the rest of the species of the genus, we concatenated the mtDNA matrix (cox1) with sequences of the 28SrRNA nuclear ribosomal gene available in GenBank for specimens identified to the species level (four spp.) (Accession codes JN817007.1, JN817008.1, JN018345.1, KM225049.1). We obtained the 28S sequenced of one individual of C. ilicis sp.n. using the primer pair (28S a: GACCTGCCTTGAAACACGGA; 28S b: TCGGAAGGAACCAGCTTACTA) (
Before concatenating the two genes for the phylogenetic reconstruction, two gene trees (one for cox1 and another for 28SrRNA) were built to assess any significant incongruence that could prevent concatenation. In all cases (combined phylogeny and separate gene trees) Bayesian inference analyses were used as implemented in Mr Bayes 3.2 software (
Both in the combined phylogeny and in the gene trees, the sequences downloaded from Genbank of the closely related genus Clubiona lena were included as outgroup. The parameters in all Bayesian inference analyses were set up to two parallel runs of 2 million generations each conducted using one cold and two incrementally heated Markov chains (L=0.2), sampling every 1,000 steps. The standard convergence diagnostics implemented in MrBayes and the average standard deviation of the split frequencies were checked to deduce that the Markov chain had reached stationarity. After 500,000 generations, the average standard deviation of the split frequencies stabilised in values close to zero (0.001) and the phylogenetic trees were summarised using the all-compatible consensus command with 25% burn-in.
We used a Chi-square analysis to assess whether the percentage of C. ilicis sp. n. individuals captured from tree branches, trunks, grass and soil differed with respect to the proportion of the whole sample (including all species of spiders) captured at each of those habitats. The differences in habitat distribution between adults and immatures of this novel species were also calculated by means of a Chi-square test. The same type of analysis was used to test the phenological differences among life stages. To do so, we divided the year into four quarters starting in January and determined whether the number of immatures and adults differed over these periods.
We investigated whether the characteristics of the oaks had any effect on the number of individuals trapped. More specifically, we used a GLM (Generalised Linear Model, Poisson distribution, Logistic link function) in which the number of C. ilicis sp. n. individuals collected at each oak was the dependent variable and the size (canopy surface in m2) of the tree was the independent one. The number of individuals collected could be spatially autocorrelated (i. e. it could be more similar among trees nearby). Hence, we performed an additional partial Mantel test (using distance matrixes) to calculate the correlation between the number of C. ilicis sp. n. individuals and canopy surface while checking the effect of the distance between trees. Generalised Linear Models were carried out in R (R Development Core Team, 2012). Mantel tests were performed as implemented in the R package `ecodist´ (
A total of 6048 spiders were collected throughout the whole sampling campaign. During the examination of the specimens a novel species of Cheiracanthium was found. It was the only species of the genus Cheiracanthium found in the study area and a total of 179 individuals were collected: 162 immatures and 17 adults (six males and 11 females).
The holotype is a male collected in Spain: Huecas (Toledo), 581 metres above sea level (40.029915°N, 4.226789°W) by E. Morano et al. on 27 May 2013. The specimen is deposited in the collection of Arachnids of the
2 males and 3 females: same data as the holotype (males:
Collected in the same village as the holotype but on a different date the following specimens have been studied and deposited in Eduardo Morano’s personal collection: 30 Jan 2013, 6 imm (branches); 26 Feb 2013, 2 imm (branches); 21 Mar 2013, 15 imm (branches) and 1 imm (trunk); 26 Apr 2013, 8 imm (branches) and 3 imm (trunk); 27 May 2013, 4 ♀, 3 ♂, 1 imm (branches); 25 Jun 2013, 3 ♀ (branches) and 1 imm (grass); 19 Jul 2013, 1 ♀, 31 imm (branches) and 2 imm (grass); 23 Aug 2013, 14 imm (branches); 27 Sep 2013, 15 imm (branches) and 5 imm (trunk); 22 Oct 2013, 29 imm (branches) and 9 imm (trunk); 28 Nov 2013, 8 imm (branches) and 10 imm (trunk); 19 Dec 2013, 2 imm (branches).
Most of the individuals of this novel species were collected from Holm oaks Quercus ilex. The Latin name of this oak species (ilex) in its singular genitive form (ilicis) has been thus used to name this spider so as to link it to the main habitat it occupies.
Cheiracanthium ilicis sp. n. closely resembles C. inclusum and C. mildei but can be distinguished by the structure and morphology of the copulatory organs of both sexes. The palps of C. ilicis sp. n. males have a characteristic bifurcated tegular apophysis (TA; Fig.
Male (Holotype). All measurements are given in millimetres. Medium size. Total length: 9.4; dorsal shield lenght: 4.1; anterior dorsal shield width: 1.9; opisthosoma length: 5.0; opisthosoma width: 3.1. Eye diameter. AME: 0.150; ALE; 0.175, PME: 0.200, PLE: 0.225. Distance between eyes: AME – AME: 0.350, AME – ALE: 0.425, PME – PME: 0.225, PME – PLE: 0.275, AME – PME: 0.225, ALE – PLE: 0.05; height from clypeus to AME: 0.1; height from clypeus to ALE: 0.1.
Prosoma. Yellowish, the ocular region is darker. The immatures show a homogeneous pale green colour, changing to yellow as they reach sexual maturity (see Pictures
Legs. Yellowish and relatively long, ordered according to their length in the following way, 1:4:2:3. The trochanters show a marked external notch. The metatarsus and tarsus have scopulae on the distal portion and the latters have dense tarsal tufts.
Spination (Table
Morphological measurements of Cheiracanthium ilicis sp. n. holotype. All measurements are given in millimetres.
holotype ♂ | |||||||
---|---|---|---|---|---|---|---|
leg | segment | long. | spines | ||||
palps | Femur | 1.7 | 0 | ||||
Patella | 0.6 | 0 | |||||
Tibia | 1.1 | 0 | |||||
Cymbium | 1.2 without or 1.8 with apophysis |
0 | |||||
total | 4.6-5.2 | - | leg | segments | long. | spines | |
I | Femur | 4.8 | 2020 | III | Femur | 3.1 | 2020 |
Patella | 1.9 | 0 | Patella | 1.1 | 0 | ||
Tibia | 5.9 | 0009(00010) | Tibia | 2.4 | 1011(1012) | ||
Metatarsus | 5.7 | 0003 | Metatarsus | 3.2 | 3033 | ||
Tarsus | 1.9 | 0 | Tarsus | 1.0 | 0 | ||
total | 20.2 | - | total | 10.8 | - | ||
II | Femur | 3.9 | 2020(2010) | IV | Femur | 4.1 | 2020 |
Patella | 1.6 | 0 | Patella | 1.5 | 0 | ||
Tibia | 4.0 | 0004 | Tibia | 3.8 | 2031(3032) | ||
Metatarsus | 4.0 | 0004 | Metatarsus | 4.7 | 3036 | ||
Tarsus | 1.1 | 0 | Tarsus | 1.2 | 0 | ||
total | 14.6 | - | total | 15.3 | - | ||
leg formula: I>VI>II>III |
Opisthosoma: Elongated, oval and slightly covered with pubescence, without erect antero-dorsal hairs. It has a uniform creamy colour all over its surface and its dorsum shows just a superficial heart mark, which is creamy or light green in adults and immatures respectively (see Pictures
Male palps (Fig.
Female (Paratypes, n=3). All measurements are given in millimetres. Medium size (ranges and means within brackets). Total length: 8.8–9.7 (9.13); prosoma length: 3.7–4.2 (4.00); prosoma width: 2.9–3.3 (3.06); opisthosoma length: 4.7–6.0 (5.23); opisthosoma width: 3.8–4.9 (4.30). Eyes diameter. AME: 0.200; ALE; 0.175, PME: 0.200, PLE: 0.175; distance between eyes: AME–AME: 0.300, AME–ALE: 0.325, PME–PME: 0.350, PME–PLE: 0.375, AME–PME: 0.200, ALE–PLE: 0.075, height from clypeus to AME: 0.175, height from clypeus to ALE.
In general, the appearance and colouration of males and females is similar, although the latter are larger and with shorter legs than the slender males. Compared to males, the female dorsal shield is much wider. Eye arrangement is similar in both sexes, but female eyes are a slightly larger. The margins of the chelicera bear 3 teeth (the middle one being larger) in the promargin and two teeth in the retromargin, where the first one is larger. As in the case of males, visualisation is difficult due to the dense brush of hair covering them.
Spination (Table
Morphological measurements of Cheiracanthium ilicis sp. n. paratypes. All measurements are given in millimetres.
paratype ♂ | paratype ♀ | ||||
---|---|---|---|---|---|
legs | segment | long. | spines | long. | spines |
palps | Femur | 1.7–1.8 | 0 | 1.2–1.5 | 0 |
Patella | 0.5–0.6 | 0 | 0.4–0,6 | 0 | |
Tibia | 1.0–1.1 | 0 | 0.8–0.9 | 0 | |
Cymbium/Tarsus | 1.0–1.2 without or 1.5–1.8 with apophysis | 0 | 1.2–1.3 | 0 | |
total | 4.2–5.3 | - | 3.7–4.2 | - | |
I | Femur | 4.8–5.6 | 2020 | 3.8–4.3 | 1000(0000)-2020 |
Patella | 1.7–2.0 | 0 | 1.5–1.9 | 0 | |
Tibia | 5.3–6.3 | 0009–00010(00011) | 3.2–4.0 | 0001(0000)-0003(0002)-1001 | |
Metatarsus | 5.3–6.5 | 0004–0005 | 3.5–4.4 | 0002(0001)-0005(0004) | |
Tarsus | 1.9–1.9 | 0 | 1.2–1.5 | 0 | |
total | 19.0–22.3 | - | 13.2–16.1 | - | |
II | Femur | 3.7–4.1 | 2010(2020)-2020 | 2.8–3.5 | 1000 |
Patella | 1.4–1.7 | 0 | 1.0–1.5 | 0 | |
Tibia | 3.4–4.3 | 0004(0005)-2004 | 2.7–2.9 | 0000–1000(0000) | |
Metatarsus | 3.7–4.7 | 0003(0004)-1005 | 2.5–3.2 | 0004(0003)-0005(0004) | |
Tarsus | 1.1–1.3 | 0 | 0.8–1.0 | 0 | |
total | 13.3–16.1 | - | 9,9–12.1 | - | |
III | Femur | 2.8–3.2 | 2020 | 2.2–2.8 | 1010 |
Patella | 1.1–1.5 | 0 | 1.1–1.3 | 0 | |
Tibia | 2.1–2.5 | 1010–2031(2021) | 1.6–1.9 | 1010 | |
Metatarsus | 2.8–3.5 | 3034–3035(3036) | 2.0–2.3 | 1024(2024)-2025(2034) | |
Tarsus | 0.9–1.1 | 0 | 0.7–0.8 | 0 | |
total | 9.7–11.8 | - | 7.6–9.1 | - | |
IV | Femur | 4.0–4,3 | 2020 | 3,2–3-6 | 1000(1010) |
Patella | 1.5–1.8 | 0 | 1.3–1.7 | 0 | |
Tibia | 3.2–4.0 | 2011(2020)-2032(2022) | 2.7–3.0 | 1010 | |
Metatarsus | 4.3–5.5 | 3035(3036)-3036(3037) | 3..1–3.8 | 2027–2037(2038)-3027 | |
Tarsus | 1.0–1.2 | 0 | 0.9–1.1 | 0 | |
total | 14.0–16.8 | - | 11.3–13.1 | - | |
leg formula | I>IV>II>III | - | I>IV>II>III | - |
Epigyne (Fig.
Vulva (Fig.
The sequence of the cythochrome oxidase I gene was identical in the three individuals of C. ilicis sp. n. (GenBank Accession code KX272624). The divergence with respect to the closest species (C. mildei) was 7.5% and 11.4% with respect to C. inclusum.
The two gene trees (cox1 and 28SrRNA) showed congruent topologies, what allowed the concatenation of the sequences of both genes. Sequences of both genes were available for all species with the exception of 28SrRNA for C. inclusum. The Bayesian phylogeny combining both genes (Fig.
The novel species of Cheiracanthium was not randomly distributed in the four habitats sampled (Chi=98.59; df=3; P<0.0001). Most of the individuals were collected from Holm oak branches (82.68%) and trunks (15.64%); only three (1.67%) from grass and none from pit-fall traps. All adults were captured from the tree branches but the habitat distribution differences between adults and immatures were not statistically significant (Chi=4.02; df=2; P=0.13). There was a positive relationship between the number of individuals collected from each tree and the surface of its canopy (Fig.
We collected individuals of Cheiracanthium ilicis sp. n. all year round, but the numbers were lower in the winter months (Fig.
We describe a novel species of Cheiracanthium that can be easily diagnosed based on male and female genitalia from other species in the genus. The amount of genetic divergence in DNA barcode sequence (cox1) (7 to 11 % from closest relatives) provides further support for its distinctiveness.
Morphologically, Cheiracanthium ilicis sp. n. resembles C. inclusum and C. mildei. Like them, C. ilicis sp. n. males have a pedipalp with a relatively stout cymbial apophysis, a strongly sclerotized, stout median apophysis and no tegular apophysis. Also, in these three species the embolus is on the retrolateral margin of the bulbus and females have relatively stout copulatory ducts that do not encircle the two pairs of receptacula seminis on the vulva. C. ilicis sp. n. shares other characteristics with these species, such as a weakly developed or absent thoracic fissure and basal cheliceral articles that, at least in C. mildei, are smaller in females and not pigmented in the distal half. Male chelicera do not have modified basal articles and are powerful and more elongated than in females, especially in C. inclusum and less in C. mildei. In C. mildei male pedipalps have an additional dorsal tibial apophysis. In turn, in C. mildei females the copulatory ducts are placed in the middle of a strongly sclerotised epigyne. The phylogenetic analysis of the concatenated cox1+28S data matrix also supports the close relationship of the new species with C. mildei and C. inclusum. The two last species are included in Chiracanthops, a genus recently resurrected by
The somatic traits of this species are a combination of characteristics common to several afro-tropical species–such as C. aculeatum Simon, 1884, C. denisi Caporiacco, 1939 and C. furculatum Karsch, 1879, and species with a wide geographic distribution C. inclusum (Hentz, 1847) (New World, Africa and Reunion) (
The new species is sympatric with Cheiracanthium mildei, which is native to Southern Europe (
We agree with
The spiders of this genus occupy different habitats and can be found in grasslands, under stones, on shrubs and trees, etc (Dondale and Redner, 1982; Urones, 1988; Lotz, 2007a;
Based on
Cheiracanthium spp. spiders are nocturnal hunters and all of the individuals that we collected during the day by branch beating were quite inactive. They are very effective predators, which probably explains their success as invaders in areas outside their distribution range (
This work was financed by the projects PII1C09-0256-9052 (JCCM and ESF), PPII-2014-01-PJCCM ESF and AGL2014-54739-R (MINECO). RB was funded by a contract of the Atracción de Talento Investigador Programme (Gobierno de Extremadura TA13032). Marisa Hernández helped with the lab work. The Editor and reviewers Jan Bosselaers and Henrik Krehenwinkel provided helpful comments that improved earlier drafts of the manuscript. We thank Susan Rutherford for the English language edition. We are indebted to the local authorities and landowners of Huecas for allowing working on their properties.