Kropf, C., 1990, Comaroma is an anapid spider (Arachnidae, Araneae, Anapidae), Verh. naturwiss. Ver. Hamburg (NF) 31, pp. 185-203 : 188-201

publication ID


persistent identifier

treatment provided by


scientific name




I. Labrum and gnathocoxae

1. Comaroma simoni Bertkau , 1889

The labrum connecting the gnathocoxae in front of the mouth opening bears a large sclerite on its anterior wall, projecting forward to the bases of the chelicerae (Fig. 1). The surrounding cuticle of the labrum is not as heavily sclerotized as the sclerite, but is extraordinarily thick.

Frontal sections through the labrum of Theridion pallens (Theridiidae) and Anapistula sp. A ( Symphytognathidae ) show that the cuticle in these two species is normally developed and that there is no distinct sclerite on the anterior wall.

A pair of muscles, the dilator muscles of the pharynx (M6 according to the usage of Marples 1983) inserts near the middle of the sclerite and extends to the lateral corners of the pharynx. A second, very fine pair of muscles (M3 as per the usage of Marples), probably extending from the middle rim of the pharynx to the middle of the sclerite, could not be clearly distinguished.

The sclerite itself has a remarkable structure. A small cavity with some minute, cuticular bristle-like pins just outside its entrance lies on its ventral side (Fig. 2), so that the cavity is covered by the sclerite. No communication has been found between the cavity and the lumen of the labrum. The ventral border of the sclerite is less highly sclerotized, which gives it a ragged appearance.

A group of longitudinal cells running to the sclerite lies directly behind the sclerite in the inner space of the labrum (Fig. 3). Fine ducts running through the sclerite and probably belonging to these cells have also been found.

The gnathocoxae bear two processes on their inner sides (Fig. 4). One is closely connected to the labium and in this way forms a rigid wall behind the mouth opening and the pharynx. The second process serves as an insertion point for the dorsal labrum compressor muscle (M9 as per Marples).

2. Pseudanapis sp. A

The labrum is heavily sclerotized and is connected with the gnathocoxae, in this way forming a sclerotized wall in front of the mouth opening and the pharynx. The gnathocoxae themselves are not separate, as in Comaroma , but are fused with the sternum and the labium so that there is also a sclerotized wall behind the mouth opening. The labral spur of Pseudanapis is a large sclerite, which is not fused with the sclerotized labrum but connected to it all round by a narrow membranous area (Fig. 5). In its ventral parts the sclerite covers the labrum almost totally. The course of the muscle M6 (as per Marples) is the same as in Comaroma , from the sclerite to the lateral corners of the pharynx. The muscle M3 (as per Marples) is very fine and runs from the middle rim of the pharynx to the sclerite, forming a “V” with its tip at the pharynx.

The structure of the sclerite, projecting forward between the bases of the chelicerae and covering a small cavity with some cuticular pins in front of it, is basically the same as in Comaroma (Fig. 6). Abundant cells filling the inner space of the labrum and with narrow ducts running through the sclerite were found.

II. Epigynum and vulva

1. Comaroma simoni

The epigynum, from the ventral aspect, is a simple plate with no external openings, so that the embolus has to penetrate through the epigastric furrow. It is not possible to recognize the border between the epigynum and the large ventral scutum, which is characteristic of Comaroma and also surrounds the pedicel but not the epigastric furrow. The inner (dorsal) view of the epigynum shows a narrow slit approximating to a letter “W” in shape (Fig. 7).

The heavily sclerotized vulva shining through the ventral scutum shows two receptacula seminis, which are globular or slightly dumbbell-shaped (Fig. 8). The ducts are not clearly visible, so that cross sections were necessary to clarify their courses (Fig. 9). The localizations of the numbered sections are shown in Fig. 9a. A section through the frontal region of the vulva, shortly behind the receptacula, shows that two ducts are connected by a fine slit (Fig. 9b). This means that there is in fact only one duct, which is functionally divided into a smaller dorsal and a larger ventral portion. A little further back the slit appears S-shaped, so that the division between the two portions of the duct is more perfect (Fig. 9c). In this region the ventral portions of the ducts begin to come nearer to each other towards the midline. Behind this region the dorsal portions lose contact with the slit and terminate in dead ends. They do not have any connection with the uterus externus. From this region on, the ventral portions appear as fine slits (Fig. 9d). In the most strongly sclerotized region of the vulva the dorsal parts of the remaining slits show a curved retrolateral course. The uterus externus begins to appear sclerotized and slightly beyond this the vulval parts (not yet the ducts) communicate by way of a weakly sclerotized connection with the uterus (Fig. 9e). Shortly beyond this region again, the dorsal parts of the ducts run into the uterus (Fig. 9f). This aperture marks the beginning (or the most frontal part) of the epigyneal slit (Fig. 7). The ventrolateral parts of the uterus externus are sclerotized and can be seen in the dorsal view as two cone-shaped sclerotizations (Fig. 8, arrows). Shortly after this the uterus forms a small dorsal pocket, from which lateral connections to the lumen of the uterus soon arise (Fig. 9g). The heavily sclerotized parts of the vulva are now connected in the middle. The ventral portions of the ducts run slightly dorsad and towards the midline (Fig. 9g). Further back the ventral portions of the ducts unite at the point where the connection of each to the uterus begins (Fig. 9h). This marks the tip in the middle of the W-shaped epigyneal slit (Fig. 7). The pocket of the uterus can no longer be seen. Near the epigastric furrow the uterus becomes connected to the ventral scutum (Fig. 9i) and appears to be heavily sclerotized shortly before the posterior end of the epigyneal slit and beyond it (Fig. 9k).

2. Pseudanapis sp. A

The vulva is heavily sclerotized and shows a pair of receptacula and a clearly distinguishable copulatory duct. The fertilization duct is clearly visible only near the epigastric furrow, where it runs medially and opens into the uterus. Cross sections show that the fertilization duct is connected with the copulatory duct by a fine slit (Fig. 10) for a long way, from its origin at the receptaculum to the point where it branches off to the uterus.

III. Depressions on the prosoma

Comaroma simoni shows circular depressions on the prosoma, which show a single pore on the base (Fig. 11). These depressions are found over the entire sternum, on the pleurae and along the lateral border of the tergum, except in the ‘cephalic’ region. The most anteriorly situated depressions occur on the lateral margin of the tergum just above the gnathocoxae.

IV. Male palpus

The male palpal cymbium of C. simoni bears a small but distinct distal-retrolateral paracymbium (see Thaler 1978, Fig. 4). A tarsal organ is present. The bristles of the cymbium are of the serrate hair-type, with the exception of one distal row of hairs that appear smooth.

The bulb of C. simoni consists of a large proximal haematodocha, a disc-like subtegulum , a smaller median haematodocha, a voluminous tegulum that bears several membranous and sclerotized elements, a distal haematodocha and an embolus (Fig. 12-17). Through the sclerotized wall of the tegulum the spermophor can be seen. It shows glandular pores, except in the fundus region and in the most distal parts, throughout its course. It ends within the subtegulum. Sections through the tegulum show that its inner space is filled with glandular systems.

On the ventral side of the bulb there is a large, striking, grey-coloured (alcohol-preserved material!) and spoon-shaped apophysis, which extends distally into a long twisted end and has a light area in the middle, where it can possibly be bent. At its base it bears a conspicuous spur, which varies a little in shape. Near this spur there is a small knob on the tegulum, which may be pointed to a greater or lesser degree.

Between the large grey apophysis and the embolus a longish, thin membranous process inserts distally; this is hard to see (Fig. 12). The distal haematodocha bears small pegs (Fig. 18). On the retrolateral side of the distal haematodocha lies the sphenoid base of the embolus, which shows a blunt tip (Fig. 13). The embolus is black and pliable and bears a small, transparent hook near its tip (Fig. 12, 14-16).

The palpal tibia bears a small, plate-like distal process (Fig. 19). The other segments of the palp do not bear any processes.

Artificial expansion of the bulb (Fig. 15-17) shows that at the beginning the distal haematodocha is expanded. Because of this, the small pegs of the haematodocha become erect and the embolus and the large grey apophysis move in a distal direction (Fig. 15). The basal haematodocha then expands. Thus, the whole bulb moves in a distal-retrolateral direction and rotates on its longer axis. The right bulb rotates clockwise (seen from behind), and the left bulb, counter-clockwise. Finally, the median haematodocha expands, some time before the expansion of the basal haematodocha is complete. The median haematodocha presses the tegulum away from the subtegulum until the two sclerites touch in the fundus region only (Fig. 17). Therefore, the tegulum and the distal sclerites move backwards in a prolateral direction, so that the longer axis of the bulb, which was formerly at an oblique angle to the longer axis of the cymbium in the retrolateral direction, appears at the end of the expansion as an extension of the cymbium’s longer axis.

The expanded basal haematodocha shows two weakly sclerotized elements shining through. One more element is a part of the subtegulum and projects towards the alveolus (Fig. 17). In the expanded palp the alveolus is also visible. It is situated in the proximal-prolateral region of the cymbium (Fig. 16).

The only possible function of the paracymbium seems to be to fix the subtegulum while the median haematodocha is expanding. The expanded median haematodocha also fixes the subtegulum, by pressing it in a proximal direction. I have observed such a process only twice, despite observing more than twenty expansions. No other variant of internal arrest was ever observed. On "normal 55 artificial expansion no function of the paracymbium was recognizable and the subtegulum passed the paracymbium closely (Fig. 15, 17). The reason for this could be the absence of back-pressure of the female epigynum in the experiment.

V. Tarsal comb setae

C. simoni bears several rows of serrate bristles on the ventral side of all tarsi and distal metatarsi (Kropf in press). The investigated specimens of the genera Latrodectus , Achaearanea, Theridion , Episinus , and Steatoda show a single row of strong bristles with S-shaped teeth on the ventral side (Fig. 20). The serrate bristles of Enoplognatha and Robertus species are similar to those of the genera named above, but the teeth on the bristles are not curved. In Robertus the bristles have extraordinarily long tips. In Euryopis species, which use no capture-webs, the teeth are reduced in size, except in E. laeta , which is remarkable for its strong teeth. In Theonoe and Crustulina the serrate bristles have strong teeth on the ventral side (Fig. 21). Pholcomma and Dipoena (Fig. 22) have bristles that differ from all other theridiid bristles (as far as investigated) in having many teeth on all sides. The teeth on the ventral side of the bristles are the biggest, especially in Dipoena . In Theonoe there are additional thinner bristles, which also bear large teeth. In Pholcomma the differences between the ventral bristles and other bristles found on the tarsus are not striking, so that there is no distinct single row of "comb setae". In Dipoena there are several rows of “serrate” bristles on tarsus IV, and similar, but thinner, bristles are also found on metatarsus IV. In all other species the serrate bristles are arranged in a single row and on the tarsus only.

D. Discussion

1. Labrum

The labral spur of Anapidae has never been described in detail 1. In my opinion, the labral spur in the Anapidae and the labral sclerite in Comaroma are homologous, although the sclerite is not a “spur” in Comaroma , but appears to be only vaulted forwards. The arguments are: (1) Both sclerites, that of Comaroma and that of Anapidae , are situated in the same place. (2) Both sclerites appear as separate structures on the anterior wall of a more or less sclerotized labrum (at least in Comaroma simoni and in Pseudanapis sp. A) and project forward. (3) The pharynx dilator muscle, M6 (as per Marples 1983), inserts in the same manner on both sclerites. (4) Behind both sclerites there lies a more or less extended cell group (glands ?) in the lumen of the labrum, and fine ducts run through the sclerites. (5) Both sclerites cover a small cavity with chitinous pins just outside it.

The labrum does not seem to be sclerotized in all anapids, because Platnick & Forster (in press) have described unfused gnathocoxae in the New Zealand genus Zealanapis (generic name not yet available). The labral spur in this genus must therefore also be a separate sclerite.

The labral sclerite of Comaroma and Anapidae cannot function as an insertion point for the pharynx dilator muscles alone, because these muscles are also found in spiders of many other families. As the labral sclerite fits well between the bases of the chelicerae , an additional function of the sclerite could be the fixation of the chelicerae in the normal position. At the moment it is not possible to say whether the cell group behind the sclerite is homologous to the labral gland (rostral gland), which seems to be present in most Araneae (Snodgrass 1948).

Xiphosura and most Arachnida show a median plate on the labrum, the epistome (Snodgrass 1948), which was described by Whitehead & Rempel (1959: 835) as a "sclerotized plate on the anterior wall of the labrum, connecting the two palpal coxae" in Latrodectus mactans (Fabricius, 1775). Because of the insertion of the median pharynx dilators on the epistome (Whitehead & Rempel 1959), which correspond to the M3 as per Marples (1983), the epistome is, in my opinion, homologous to the anapid labral spur. The fact that the lateral pharynx dilators (M6) originate in orthognaths on the epistome (Meyer 1981) and in anapids on the labral sclerite supports this assumption.

Thus, as it seems most unlikely that a sclerite with the same quality has evolved on two separate occasions from a simple epistomeal plate, I regard Comaroma as a genus of the Anapidae . It should be mentioned here that Forster & Platnick (1984) described a pair of lateral protuberances on the labrum of Archaeidae sensu Forster & Platnick and Mecysmaucheniidae Simon, which should be reconsidered in respect of their possible homology to the anapid labral sclerite.

1 Wunderlich (1976, 1986, 1988) described it falsely as a spur of the "labium 55.

2. Vulva

The conformity of the vulval structure between Comaroma and Pseudanapis - a single duct functionally divided into two ducts at least in the anterior region of the vulva - could be an additional character of at least some anapids. But the decision as to whether this vulval structure is a synapomorphy or not is very difficult. Unfortunately, the vulval structures of other Anapidae and their possible relatives are not sufficiently well known and could sometimes have been misinterpreted, as in the case of the fertilization duct in the vulva of C. simoni , which has been illustrated by Wiehle (1960a, Fig. 137).

An outgroup comparison, if possible at all, indicates that the functionally bipartite duct is synapomorphic for anapids; it seems to be unique among spiders. But problems remain: Wiehle (1967) described a semientelegyne type of vulvae in the genus Meta , which he characterized as an entelegyne type of vulvae without fertilization ducts. In contrast to semientelegyne vulvae, haplogyne vulvae show receptacula that are derivatives of the uterus externus (Wiehle 1967). Brignoli (1978) assumed that the evolution of internal female genitalia had proceeded from a condition found in Orthognatha and Haplogynae to the semientelegyne, and further to the entelegyne type with fertilization ducts. Kraus (1984b) expressed the opinion that the semientelegyne vulva was more primitive than the entelegyne vulva. The selective pressure to develop coupling mechanisms occured with the development of complicated webs and copulation in the web or on special capture-threads (Levi 1961; Kraus 1984b).

As a functional separation of former bifunctional ducts (guiding of the embolus on the one hand and of sperm on the other hand) allows safeguarding and improvement of the two functions, it seems sensible to me to consider the semientelegyne type as phylogenetically older than the entelegyne type.

On the other hand, the question arises as to why Pseudanapis has a separate fertilization duct in the posterior region of the vulva, while in Comaroma the dorsal portions of the bipartite ducts which seem to be homologous to the fertilization ducts of Pseudanapis , terminate in dead ends. Were they secondarily reduced ?

However, the anapid arrangement of vulval ducts can be taken as a model, suggesting what the transition from the semientelegyne to the entelegyne type of vulvae could have looked like. If the bifunctional duct of anapid vulvae is indeed such a transition type, it can hardly be regarded as an apomorphic character.

Forster (1980: 278) demonstrated a concept on the evolution of the entelegyne vulva: "The ultimate form of female genitalia is the entelegyne structure in which the original bursal openings apparently migrate beyond the gonopore onto the external surface of the abdomen and a new duct - the fertilization duct - develops..." This concept allows the additional assumption that the fertilization duct has developed by separation from the original bifunctional duct.

There are illustrations of internal female genitalia in the anapid literature that show separate fertilization ducts originating from the receptaculum (e.g. Hickman 1944; Brignoli 1968, 1981). On the other hand Platnick & Forster (in press) reported haplogyne female genitalia in many anapid genera, saying that they found neither external epigyneal openings nor separate fertilization ducts. As the term “haplogyne” is not always used in the same strict sense as it is used by Wiehle (e.g. Forster 1980), these vulvae could be semientelegyne in Wiehle 's sense or perhaps represent a type I have found in Comaroma and Pseudanapis . Further studies on anapid vulvae would be most useful.

3. Depressions on the prosoma

Platnick & Forster (1986: 3) proposed a second synapomorphy of the Anapidae , a pair of pore-bearing depressions on the lateral margin of the tergum above the gnathocoxae, which they described as "widespread among the Anapidae ". The same authors (Platnick & Forster, in press) noted that in the Chilean genera Pecanapis and Minanapis (generic names not yet available) these depressions were lacking. Perhaps the described circular depressions on the prosoma of Comaroma are homologous with those in Anapidae , but as long as we do not know anything about the function of these depressions no further conclusions are possible.

4. Male palpus

The male palpal organ of Comaroma is more complicated than that of other anapids, insofar as it has three haematodochae, a sub tegulum, a tegulum and distal sclerites. Furthermore, the presence of a paracymbium is unknown in Anapidae . Since the work of Kraus (1978) it has seemed certain that complicated bulbs such as are found in Comaroma reflect the ancestral situation, as against the simpler bulbs, such as that seen in the “classic” Anapidae .

The palpus of Comaroma resembles that of some theridiids, but clear synapomorphic structures in the theridiid palp are difficult to find. Heimer (1982) showed that in most theridiids the paracymbium and the median apophysis arrest the bulb internally. An interesting exception is seen in Paratheridula quadrimaculata (Banks, 1896), in which the subtegulum is in contact with the distal and, in the fully expanded bulb, also with the dorsal margin of the cymbium. However, no internal arresting mechanism between a distal paracymbium and the subtegulum, as is thought probable in Comdroma , was found in Theridiidae .

The homologization of the distal palpal sclerites with that of other Araneoidea is very difficult. The embolus is quite distinct, but does not allow precise assignment to any of the functional types described by Wiehle (1960b). However, it seems to be an "Einstoss-Embolus stoß-Embolus” rather than any other type. The large grey apophysis was referred to as the “conductor” by Levi (1957) in the Californian species C. mendocino (Levi, 1957). The position of the tip of this apophysis is adjacent to the embolus tip, but does not protect it. Because of its shape and because it is the strongest apophysis, it is thought that its function is probably to open the epigyneal slit. The base of this apophysis was named the “radix” and the base of the embolus, the "median apophysis" by Levi (1957). Later Levi (1968) reversed these two names, so that the base of the large grey apophysis has to be renamed "median apophysis" and the base of the embolus, “radix”.

Kraus (1978) showed that a distal haematodocha connects sclerite II (tegulum) with a distal sclerite in Liphistius and that in Atypus in a similar situation two distal sclerites were found, an embolus and a functional conductor. The latter apophysis was referred to as the “contrategulum” by Haupt (1983). The situation described above is basically similar to that in Comaroma .

Comstock (1910) defined the distal haematodocha as a part of the embolic subdivision in the Araneus type of palpi. The question is whether the distal haematodocha, in Comstock’s sense of the word, is homologous with that found for example in Liphistius . Haupt (1983) believes that the distal haematodocha of Liphistiidae corresponds to the embolic subdivision in Araneidae and Theridiidae . Grasshoff (1968) showed clearly that the distal haematodocha in the sense understood by Comstock is connected with the tegulum by a separate sclerite in Araneus , the radix, and thus is situated distally of the radix, bearing neither a conductor nor a median apophysis. But the sclerite that was referred to as the “radix” in Comaroma (and in many theridiid genera), while showing no haematodocha distal to it, does show one lateral-proximal to it. Thus, if the distal haematodocha in Comaroma and Araneus are homologous, the basic sclerite of the embolus in Comaroma cannot be the radix.

On the other hand Comstock (1910: 173) described a radix in the Linyphia type of palpi as situated "Immediately following the membranous neck that connects the middle and the apical divisions of the bulb..." The term “radix” could therefore be correct for the base of the embolus in Comaroma .

Moreover, Araneus species show a membranous area above the tegulum (Grasshoff 1968) that bears the radix, the median apophysis and the conductor. Maybe this structure, in a similar way to the above mentioned area in Linyphia , is homologous to the distal haematodocha in Comaroma and perhaps also to that in Liphistius . Because of this uncertainty, at the moment I am loath to propose names for the distal sclerites of the bulb in Comaroma (except the embolus).

5. Other theridiid features

The comb setae on tarsus IV are usually regarded as a synapomorphy of Theridiidae and Nesticidae , but the diversity of tarsal comb setae in different theridiid genera makes it difficult to decide whether these characters are really homologous. It can only be mentioned that serrate bristles may also function as cleaning bristles, as has been observed in Comaroma (Kropf in press). Thus, they might have evolved at several different times. Tarsal comb setae and prey-catching behaviour of theridiid spiders must be studied further, all spiders recently assigned to the Theridiidae (and Nesticidae ) cannot be regarded as a monophyletic group until more is known on these subjects.

Coddington (1989) has brought a new characteristic feature into the discussion, namely enlarged aggregate gland spigots on the posterior lateral spinnerets in Theridula , Latrodectus and some nesticids, and proposed that it is a synapomorphy of Theridiidae and Nesticidae . Comaroma simoni has no enlarged aggregate gland spigots.

The web of C. simoni resembles a theridiid web, but can be regarded as a modified anapid orb web (Kropf in press). Similar webs are known from other European Anapidae (Kratochvil 1935). Some other theridiid-like characters in Comaroma (male palpus, serrate bristles on tarsi and metatarsi) support the assumption of Levi & Randolph (1975: 32) that "The division between Symphytognathidae and Theridiidae is probably as arbitrary as that between Theridiidae and Nesticidae ."

6. Other anapid features

The reduction of the anterior median eyes (AME) was regarded as a synapomorphy of Anapidae and Symphytognathidae by Heimer & Nentwig (1982). C. simoni shows a tendency to reduce the AME, but in fact most specimens show the full number of eight eyes, though one AME is totally absent in rare cases (Schuster & Moschitz 1984). The reduction of both AME has been reported by Wiehle (1960a) and by Baert & Kekenbosch (1980), for example. The problem of eye reduction in Comaroma is discussed in detail by Schuster & Moschitz (1984). Moreover, many anapids with the full number of eyes are known (e.g. Hickman 1938; Brignoli 1981; Platnick & Forster in press). Reduction of the AME also occurs in other families (Brignoli 1970), so that this character is of minor taxonomic value.

Many other negative characters (e.g. reduction of booklungs, female palp, leg spines) have been considered by various authors. The problems involved were discussed by Brignoli (1970: 1414): "Etablir un groupe sur des structures réduites.. .expose le taxonomiste au grand danger de se laisser tromper par des convergences." Especially in Anapidae , these characters could have evolved in connection with the small size.