The Cladistics of Orbiculariae and the Monophyletic Origin of the Orb Web
Until recently, the orb-web was thought to have been independently evolved more than once. This idea was based upon the belief that cribellate and ecribellate spiders formed monophyletic clades. However, since the mid 1980’s this view has radically changed. Cribellate and ecribellate clades were found to be paraphyletic (Eberhard, 1990), this opened many new approaches to the problem of the origin of the orb-web. In the past two decades there has been an overwhelming amount of phylogenetic work on spiders and while many different groups have been proposed, virtually all agree that the Orbiculariae (orb-weavers) constitute a monophyletic clade. This paper examines the nuances of Orbiculariae cladistics including the monophyletic evolution of the orb-web, outgroup determination, and higher Araneoidea phylogenetics.
The Orbiculariae clade is the scientific name for what are generally known as the Orb-weavers. The work that had been done since the fall of the cribellate/ecribellate dichotomy has placed 13 families in the Orbiculariae: Deinopidae and Uloboridae, Araneidae, Linyphiidae, Cyatholipidae, Synotaxidae, Nesticidae, Theridiidae, Tetragnathidae, Theridiosomatidae, Mysmenidae, Anapidae, and Symphytognathidae (Coddington and Levi, 1991). The families that Coddington and Levi’s 1991 phylogeny (Figure 1.) set out as having been in Orbiculariae have since remained in the clade, though their relationships to each other have undergone change (Griswold, 1998). Within Orbiculariae, there are two major clades; Deinopoidea (Deinopideae and Uloboridae) and Araneoidea (all other families in Orbiculariae). The two major distinguishing features of these two clades are the horizontal orientation of the web in Deinopoidea versus the vertical orientation in Araneoidea and that the Deinopoidea are cribellate and Araneoidea ecribellate (Bond and Opell, 1998). The use of the cribellum as a homology for this superfamily level cladistics is acceptable despite it being paraphyletic. In the Orbiculariae, it simply means that the common ancestor was cribellate, which was then suppressed to the ecribellate state in the primitive Araneoidea (Griswold et al., 1998)
The orb web is the defining trait for the Orbiculariae clade, and thus it is of constant interest to determine whether or not it is in fact a synapomorphy. The orb web is present in two recognized clades: Deinopoidea, which are cribellate and the Araneoidea, which are ecribellate. In the past, the presence of a cribellum constituted a monophyletic clade, thus Deinopoidea was believed to be unrelated to Araneoidea (Coddington and Levi, 1991). Further work challenged this assumption and eventually demonstrated that the cribellum is paraphyletic at the family level (Eberhard, 1990). The trait that had previously been held as a major homology, and thus separated Deinopoidea and Araneoidea, was no longer a valid reason for separating the two taxa which opened the doors to further study into their true relationship (Eberhard, 1990). Kullmann’s 1972 paper postulated that the orb-web evolved convergently in these two groups. He believed that the ecribellate Araneoidea could not have come from the cribellate Deinopideae because the idea of an intermediate spider form with both cribellate silk and capture threads was in his words “fantastic”. Kovoors and Peters agreed with this view of orb-web cladistics. They believed that the orb web could not be a homology in the two groups because the same patterns in their webs are produced by spigot glands that are not homologous (1988).
A 1990 paper by Coddington challenged the assumptions of Kavoors and Peters’. He stated that the glands were only a small part of the overall orb-web system. His paper proposed that the homologies of orb-web structures were a powerful case for the monophyly of orb-webs. Web radius, hub, frame structure, as well as non-sticky and sticky portions of the web spiral all constitute individual synapomorphies at the family level. Therefore, they can all be applied to a phylogenetic analysis (Coddington, 1990). It is worth noting that each structural homology represents a monophyletic hypothesis, all pointing to the grouping of Deinopoidea and Araneoidea. Since character weighting is inherent, no particular weight can be given to Kovoors and Peters’ hypothesis in relation to these new hypotheses. As such, it becomes a matter of parsimony. The hypothesis that says all web structures are synapomorphic allows for them each to be evolved once, and separate structural organs to be either modified or independently derived twice. The alternative; that orb-webs are not monophyletic involves the independent origin of part of the web structure twice. Following the rules of parsimony, the monophyletic hypothesis of the orb-web is the equally likely (Patterson, 1982). Since then cladistic analyses based on morphology and ethology have determined that the orb web is monophyletic.
Coddington’s 1990 study examined 61 binary characters and 19 taxa, 9 of which fell in the Orbiculariae group (Figure 2). Griswold et al., (1998) published a parsimony based phylogeny of 31 Orbiculariae taxa from both the Deinopoidea and Araneoidea subgroups (Figure 3). This phylogeny examined 91 characters based on male and female genitalia, spinneret and general morphology as well as 10 characters based on orb-web design. Combined with the phylogeny published in 1999 by a different group headed by Griswold (Figure 4), the ethological and morphological data points strongly to a clade combining both the Deinopoidea and Araneoidea as members of the Orbiculariae clade (Griswold et al., 1999). Cladistic analyses based on molecular work have been more ambiguous.
While the morphological and behavioral data clearly place the orb-weavers together as a monophyletic group, the molecular phylogenies have been much more divided. A study conducted using 28S rRNA concluded that Orbiculariae is not a monophyletic clade, rather, the data he used showed that the Deinopoidea was the sister group to the entirety of Entelegyne spiders, though Araneoidea maintained its monophyly (Hausdorf, 1999) (Figure 5). Further work on the 28S rRNA as well as mitochondrial DNA and MaSp1 genes have also supported the hypothesis that the orb-web is not monophyletic (Pan et al., 2007) (Figure 6).
An analysis of silk protein homologues in Deinopidea and Araneoidea gives strong support for the monophyletic origin of the orb-web. In the study, four genes that coded for spider silk proteins known to occur in the Araneoidea, major ampullate proteins (MaSp1 and MaSp2), minor ampullate scaffolding protein (MiSp) and flagelliform protein (Flag), were looked for in the Deinopoidea. An examination of these cDNAs from taxa from both Deinopidae and Uloboridae yielded results confirming the monophyly of the orb-web (Garb et al., 2006). It was determined that while MaSp1 and MiSp were plesiomorphic in the Deinopoidea and Araneoidea, MaSp2 and Flag were similar in the two groups and were synapomorphic. The most likely explanation for the results is that the orb-web, and the silk proteins involved in it, evolved once in a common ancestor (Garb et al., 2006). At present there are still more morphological and molecular studies that support the monophyly of Orbiculariae than those with opposing views, though this may change as new molecular sequences are studied and morphologies compared. The question of the validity of morphological vs molecular systematics, and the question of which is more reliable, is outside the scope of this paper.
Assuming that Orbiculariae is monophyletic, new questions arise, such as which taxa are the sister group to Orbiculariae. Unlike the dichotomous question of monophyly, there is really no one hypothesis of sister taxa. One hypothesis was that the family Dictynidae was the sister group due to four different synapomorphies (Coddington, 1990). In 1991, however, Coddington and Levi gave a radically different grouping: the sister group to Orbiculariae was the ‘RTA’ clade. This superclade included Dictynidae and Amaurobioidea in addition to the Gnaphosoids, Lycosoids and family such as Salticidae and Thomisidae. The most recent grouping is that the outgroup to Orbiculariae is a superclade composed of all Entelegyne spiders save the Eresoids (Griswold, 1999)
While the taxonomy of Deinopoidea has remained unchanged for some time, systematists have had much more work fleshing out the Araneoidea group. Even before Coddington’s revamp of Orbiculariae, the Araneoid clade was divided into two groups: Araneidae and the “Higher Araneoids” which included all other members of this clade (Coddington and Levi, 1991) (Figure 1). This basic grouping has yet to change (Griswold et al., 1998). However, in the “Higher Araneoids” much work has been needed and a number of hypotheses have been postulated The first hurdle was the resolution of the polytomy involving the placement of several distinct groupings, the Theridiids, Linyphiids, and a group combining Tetragnathidae with the Symphytognathoids (Coddington and Levi, 1991) (Figure 1). This polytomy has since been resolved by moving the Symphytognathoids in one clade and combining Theridioids and Linyphioids in another, and moving Tetragnathidae as sister to the clade formed by these two combined (Griswold et al, 1999) (Figure 3).
While the phylogenetics of Orbiculariae have come a long way from the cribellate dichotomy of Kullmann, there is still much work yet to be done. Morphological studies continue to indicate a monophyly of the orb-web, the molecular data is far less clear. It will take many more years of study and perfection of molecular phylogenetics before any similar trends of relationships are developed. It may be that in ten years the idea of a monophyletic origin of the orb-web is laughable. Until then scientist will continue to work out the relationships of this fascinating group of spiders.
Works Cited
Bond, J. E., and Opell, B. D. (1998) Testing adaptive radiation and key innovation hypotheses in spiders. Evolution 52: 403-414.
Coddington, J. A. (1990) Cladistics and spider classification: araneomorph phylogeny and the monophyly of orbwevers (Araneae: Araneomorphae; Orbiculariae). Acta Zoologica Fennica 190: 75-87.
Coddington, J. A., and Levi, H. W. (1991) Systematics and evolution of spiders (Araneae). Annual Review of Ecology and Systematics 22: 256-292.
Eberhard, W. G. (1990) Function and phylogeny of spider webs. Annual Review of Ecology and Systematics 21: 341-372.
Garb, J., DiMauro, T., Vo, V., and Hayashi, C. Y. (2006) Silk genes support the single origin of orb webs. Science 312: 1762.
Griswold, C. E., Coddington, J. A., Hormiga, G., and Scharff, N. (1998) Phylogeny of the orb-web building spiders (Araneae, Orbiculariae: Deinopoidea, Araneoidea). Zoological Journal of the Linnean Society 123: 1-99.
Griswold, C. E., Coddington, J. A., Platnick, N. I., and Forster, R. R. (1999) Towards a Phylogeny of Entelegyne Spiders (Araeneae, Araneomorphae, Entelegynae). Journal of Arachnology. 27: 53-63.
Hausdorf, B. (1999) Molecular phylogeny of araneomorph spiders. Journal of Evolutionary Biology 12: 980-985.
Kovoors, J. and Peters, H. M. (1988) The spinning apparatus of Polenecia producta (Araneae: Uloboridae): Structure and histochemistry 108: 47-59
Kullmann, E. J. (1972) The convergent development of orb-webs in cribellate and ecribellate spiders. American Zoologist 12: 395-405.
Pan, H.-C., Wu, B.-S., Song, D.-X., Hao, J.-S., Zhu, G.-P. (2007) rRNA and Masp1 gene do not support the monophyly of Orbiculariae. Acta Zoologica Sinica 52: 489-501.
Patterson, C. (1982) Morphological characters and homology. pp. 21-74 in: Joysey, K., ed. Problems of Phylogenetic Reconstruction. Sytematics Association Special Volume 21. Academic Press
Figure 1. Phylogeny of Araneomorphae including Orbiculariae (Coddington and Levi, 1991).
Figure 2. Phylogeny of Araneomorphae (Coddington, 1990)
Figure 3. Phylogeny of Orbiculariae with resolution of Coddington and Levi’s “Higher Araneoids” polytomy (Figure 1) (Griswold et al., 1998).
Figure 4. Phylogeny of Araneomorphae (Griswold et al., 1999).
Figure 5. Molecular phylogeny using 28S rDNA. Deinopoidea is placed as sister to the rest of the Entelegyne spiders (Hausdorf, 1999)
Figure 6. Molecular phylogeny from mitochondrial, nuclear and ribosomal DNA including 28S rRNA and MaSp1. Places Araneoidea as sister to Deinopoidea plus other Entelegynes (Pan et al., 2007).