Saturday, December 10, 2016

Mystery of the Misumena fidelis Crab Spider

Mexico and the United States are missing a crab spider. It's not the most inconspicuous of spiders, either, being one we should see on flowers. Read on to see how biological taxonomy is sometimes detective work about the names and shapes of things...

Summary

A fortuitous series of circumstances, along with some sleuthing, reveals that Misumena fidelis Banks 1898 is properly Mecaphesa fidelis (Banks). It is very likely synonymous with Misumenops volutus F.O. Pickard-Cambridge 1900. It may or may not also be synonymous with Misumena decora Banks 1898, which Gertsch 1939 might have mistakenly synonymized with Misumenops volutus. An examination of the Misumena decora type specimens should resolve that. Mecaphesa fidelis ranges along the western half of Mexico, south to Guatemala, and possibly north into southern Arizona.

Last Seen in 1901

Cotype of Misumena fidelis Banks 1898, from MCZ

There are supposed to be two species of Misumena crab spider (family thomisidae) north of Mexico. Misumena vatia ranges across most of the United States and much of Canada. It's also found in Europe and northern Asia. It's rather large for a crab spider and makes for some great photos. The other spider, Misumena fidelis, is supposed to be in southern Arizona and western Mexico.

The problem is that only one person has ever knowingly seen Misumena fidelis. That person was Nathan Banks, who described and named the species in "Arachnida from Baja California and other parts of Mexico," published in the 1898 Proceedings of the California Academy of Sciences. Banks reported "several specimens from El Taste, La Chuparosa, and San José del Cabo." The first and last of these locations is in southern Baja California. La Chuparosa appears to be in Chihuahua in northern Mexico.

Banks reported seeing the species again in "Some Spiders and Other Arachnida from Southern Arizona," published in the 1901 Proceedings of the United States National Museum. This time he saw "a female from Catalina Springs" in Arizona, USA.

And that was it. The species has never been found again. Or has it?

Type Specimen on Hand

While researching for a new key to the genera of crab spiders north of Mexico, I happened across two specimens that Banks himself had labeled Misumena fidelis. In fact, Banks designated these specimens as "types" for the species, meaning that these specimens are considered to define the group of spiders that are to be called Misumena fidelis. (Technically, it's a "syntype.") Any other spider I find that has the characteristics of these spiders should then also be Misumena fidelis.


Misumena fidelis Banks, adult female

Misumena fidelis Banks, subadult female



One of the type specimens is an adult female. The other is a penultimate female, which means that she's one molt shy of being an adult. The adult has a body length (cephalothorax + abdomen) of 7 mm. Both were in a single vial from the Harvard Museum of Comparative Zoology, collected from El Taste in Baja California by Eisen and Vaslit.

Banks' 1898 paper does not describe males, so he apparently only had females. Male spiders are typically more distinctive, so it's an added challenge that we have to solve the mystery from females. Typically only the adults of a species are distinctive, so we'll have to do this from a single female.

With this type specimen, we can now scour museum collections to find other records of the species and see what else we can learn about the spider. But there are lots of spiders in collections. Where do we begin looking? The best we can do is guess how others may have labelled spiders of this species and double-check the specimens so labelled. This all depends on what the spider looks like.

Looks like Mecaphesa or Misumenops, not Misumena

The first thing that jumps out at me is that Misumena fidelis has heavy spines (macrosetae) on the carapace and abdomen. Actually, they've all broken off the abdomen, but the spine bases are still present, so it's clear she had them while alive. This jumps out at me because one of the most distinctive characters of our well-known Misumena vatia is that they have no such spines on their body, though they do sometimes have them around the eyes.

Misumena vatia's dearth of body spines are apparent in these two beautiful photos. (Okay, I couldn't post such a long, technical blog without including some fun-to-look-at photos.)

Misumena vatia. Photo courtesy of Sean McCann

Misumena vatia. Photo courtesy of Christy Pitto.

Let's compare the carapaces (tops of their heads) more closely. Here we more clearly see that Misumena fidelis is distributed with robust carapace spines, while Misumena vatia is not.


Female Misumena fidelis, having carapace spines

Female Misumena vatia, lacking carapace spines

Several other North American genera of crab spiders are known for having spinose (spiny) bodies, so it may belong to one of those instead. Genera are human constructs intended to approximate real-world relationships, so our first step is to see how we humans have defined the various relevant groups. This looks like a misumenine spider, so we'll restrict ourselves to those genera. In North America, the misumenine spiders are mainly the flower crab spiders.

Once upon a time it was fashionable to call all New World misumenine spiders Misumena.  In 1900, F.O. Pickard-Cambridge divided Misumena into Misumena and Misumenops, mostly relegating any spider whose ALEs (anterior lateral eyes) were larger than the AMEs (anterior median eyes) to Misumenops. Misumena vatia eyes are usually the same size, so Misumena vatia remained in the genus Misumena, while many other crab spiders moved to Misumenops. Notice from the first photo in this blog that this 1900 rule would have moved Misumena fidelis to Misumenops.

Misumenops didn't get revised again until 2008, this time by Lehtinen & Marusik. Their paper divided Misumenops into several more groups. As with most arthropods, the most distinctive character of a spider is its genitalia, so it's not surprising that the paper largely divides out new genera by genitalia. The next step for us is therefore to look at the Misumena fidelis female genitalia—the epigynum.


Misumena fidelis epigynum, ventral view

Misumena fidelis epigynum, angled posterior view

One of the groups that Lehtinen & Marusik separated from Misumenops was Mecaphesa, and this epigynum looks very much like Mecaphesa to me. It also has some similarity to the Misumenops epigynum. For completeness, we also compare the epigynum with that of Misumena vatia.


Left and right: Two Mecaphesa dubia specimens, ventral epigynum

 Left: Misumenops bellulus, ventral epigynum
Right: Misumena vatia, ventral epigynum (Photo by Nicole Miller)


The Misumena fidelis epigynum looks nothing like the Misumena vatia epigynum, except for apparently having a small central hood-like structure. Instead, the overall look is suggestive of Mecaphesa, particularly the upward narrowing sweep of the central epigynum. However, the apparently small hood on Misumena fidelis seems more suggestive of the small hood on Misumenops. None of the other genera have epigyna that look so similar to Misumena fidelis, so our working assumption is that our spider is either Mecaphesa or Misumenops. We don't know which yet.

Mind you, that little hood on Misumena fidelis is not quite what it seems.

Superficially neither Mecaphesa nor Misumenops

Let's dive into the details of what it means for a spider to be Mecaphesa or Misumenops. Lehtinen & Marusik 2008 provide the latest definitions. For now, we'll stick with what we can see externally on the spider, without dissecting any spiders.

Both Misumenops and Mecaphesa have robust spines on the carapace and abdomen, as Misumena fidelis does. There are strong distinctions in the male genitalia, but we aren't looking at a male (yet!). On the female's epigynum, Misumenops has a small hood, while most Mecaphesa have a median septum. The median septum is a structure that runs for at least a portion of the front-to-back length of the epigynum, along the center of the epigynum.

Now it's time to take a closer look at the Misumena fidelis epigynum. Photos don't always convey the 3-D structure well, so here are some drawings. The first drawing shows a strictly ventral view. The second drawing is what my mind tells me how a cross-section of the epigynum would look, based on my examination of the epigynum from various angles under a microscope. The second drawing should be schematically correct if not proportionally accurate.

Misumena fidelis cotype, ventral epigynum
Misumena fidelis, imagined epigynal cross-section

In these drawings, 'CO' refers to a copulatory opening where the male inserts an embolus (a penis-analogue) to inseminate the female. I believe all spider epigyna have two copulatory openings. There is a huge cavity at the anterior of the epigynum, here labelled 'cav'. There are two hoods, one part of the other. The apparently-small central hood we noticed previously is actually part of a much larger hood. 'IH' refers to the small inner hood, and 'OH' refers to the large outer hood.

It is common in Mecaphesa for the hood to span the width of the epigynum, as we see here. We call the broad depression at the center of the epigynum the 'atrium'. The outer hood here spans and encloses the anterior atrium. This seems to rule out Misumenops. However, our spider also does not have a median septum that Lehtinen & Marusik 2008 says we can use to identify most Mecaphesa. Instead, the middle atrium is flat, gradually sloping into the anterior cavity.

Lehtinen & Marusik 2008 provides a clearer distinction for females using features of the dorsal epigynum, which can only be seen by dissecting the spider. We don't want to dissect a type specimen if we don't have to. Fortunately, we have a clear enough understanding of the epigynum at this point to go looking for other Misumena fidelis specimens. So that's what we'll do. For now, we're left still not knowing whether the spider is Mecaphesa or Misumenops.

More Misumena fidelis Specimens Found

It didn't take long for me to find additional females whose genitalia exactly match the Banks type specimen. I had sorted out Mecaphesa-like spiders that I was not familiar with. Among them was a vial from the Canada National Collection containing two adult females and one adult male, collected from Xechi milco in the Federal District of Mexico by H.E. Milloron in 1962. The structure of the epigyna of both females was identical to the Banks type specimen for Misumena fidelis:


Misumena fidelis, ventral epigyna, two females from CNC

Misumena fidelis, ventral epigynum schematic

The spiders themselves are also quite similar-looking. Recall that the Misumena fidelis once had the abdominal spines, because it still has the spine bases. One of the CNC females has a body length of 7 mm, the other a length of 6 mm.


Misumena fidelis females. Left: Banks cotype. Center and right: CNC specimens.

The vial was labelled Misumenops decorus, because the genitalia of the male in the vial exactly matches the drawing for Misumenops decorus given in Gertsch 1939. The male has a body length of 3 mm. Lehtinen & Marusik 2008 renamed Misumenops decorus to Mecaphesa decora.


Male accompanying female Misumena fidelis, frontal view

Male accompanying female Misumena fidelis, dorsal view

The females in this vial were probably identified as Mecaphesa decora because they were found with the male. That seems like a reasonable assumption, but there's a small problem: the female epigyna look little like the only drawing that exists for the Mecaphesa decora epigynum:


Misumena decora Banks 1898, ventral epigynum
(Banks 1898 Plate 16, Figure 13)
Left and right: Misumena fidelis from CNC, ventral epigynum


Banks provided the drawing of Mecaphesa decora (then Misumena decora) in the same 1898 paper that described Misumena fidelis. These are the original descriptions of the species, so they are considered definitive. Therefore, according to Banks, who named these species, these additional CNC females are Misumena fidelis. If Banks' drawings for Mecaphesa decora are accurate, we can further say that these females are not Mecaphesa decora. It's still possible that Banks was mistaken to think that Misumena fidelis and Mecaphesa decora are different species, but the strong differences in the epigyna drawings make this seem improbable, at least for the females.

So we appear to have a situation in which two female Misumena fidelis were found with a male Mecaphesa decora. But is it really a male Mecaphesa decora?

Identifying the Accompanying Male

We'll identify the male from scratch, carefully heeding the history of North American thomisid descriptions. Banks did not know what the male Misumena fidelis looked like, and no one has since (knowingly) described the male, so this investigation cannot yield a match for Misumena fidelis.

The male genitalia that are distinctive are the enlarged "bulbs" at the ends of the pedipalps. Traditionally, we just refer to these as the "palps." Arachnologists typically draw the palps of the males of new species, so for our investigation, we can scan the literature for palps that look like those on our male. There are a lot of spiders in the literature, so we'll limit the scan to North American thomisids that have ever been called Misumena, Misumenops, and Mecaphesa.

The earliest definitive match for the palp that I can find in the literature is Misumenops volutus, described by F.O. Pickard-Cambridge in a 1900 paper. We normally draw a strictly ventral view of the palp and maybe also an outside (retrolateral) view, but neither of these views provided a match. Instead, there is a great match in a view that is partially ventral, distal, and retrolateral:


Misumenops volutus palp. Left: ventral view. Right: retrolateral view


 Misumenops volutus palp, view angled to match F. O. Pickard-Cambridge 1900 Plate 10, Figure 3.


Okay, now we have a male Misumenops volutus in a vial with two Misumena fidelis that are all together identified as Mecaphesa decora. That sounds crazy, but we can explain some of this. In his 1939 paper, Gertsch decided that Banks' Misumena decora and Pickard-Cambridge's Misumenops volutus were actually the same species, renaming both to Misumenops decorus, which appears on the vial label. Lehtinen & Marusik moved Misumenops decorus to Mecaphesa decora in 2008.

Let's revisit Gertsch's decision to synonymize Misumenops volutus with Misumena decora. Gertsch 1939 reports, "Cotypes of Misumena decora Banks from Mexico, all destroyed except one male and one female in the Museum of Comparative Zoology." But Gertsch 1939 only describes the male and only lists six male records, no female records. Moreover, F. O. Pickard-Cambridge 1900 only described the male of Misumenops volutus, not the female. So it appears that Gertsch made this decision to synonymize these species entirely based on the male.

In order for Gertsch to decide that the Misumenops volutus male was the same as the Misumena decora male, he would have had to compare their palps. Because Gertsch 1939 does not describe the female and does not list any female records, it seems reasonable to conclude that he did not examine the male and female cotypes he mentioned; and if he didn't examine the cotypes, then in particular he didn't examine the male Misumena decora cotype that Banks designated. Odds are that his basis for the synonymy was a comparison of the Banks 1898 drawing of Misumena decora with obvious male specimens of Misumenops volutus that he could examine.



Left: Misumenops volutus palp, distal-retrolateral view.
Right: Misumena decora Banks 1898 palp, unknown view (Banks 1898 Plate 16, Figure 13).


Let's do this comparison ourselves. The above-right drawing is Banks' illustration of the Misumena decora palp. Notice that the embolus appears to originate from the center of the tegulum. (That is, the tapering black thing originates from the center of the large, smooth round thing). This would normally be a distinctive character, but no view of the Misumenops volutus palp presents this way. Instead, the above-left photo is the closest I could get to presenting this apparent character.

There is enough similarity to make it conceivable that Banks intended to draw a Misumenops volutus palp. Perhaps his microscope didn't give him a very crisp view. Just as no two people are physically identical, no two spiders are physically identical, so perhaps it's possible for a palp of this species to sometimes looks more like Banks' drawing. Because this remains unclear, it seems that we can't have a lot of confidence in Gertsch's synonymy of Misumenops volutus with Misumena decora.

To properly resolve this question, someone will have to examine Banks' Misumena decora type specimens; I do not have them on hand. For now, I'm not willing to say that Misumenops volutus is Misumena decora. The palp is only an iffy match, and the females found with a male Misumenops volutus strongly match Misumena fidelis but not Misumena decora. All I'm willing to say with confidence is that the vial has a male Misumenops volutus and two female Misumena fidelis.

But could Misumenops volutus really be Misumena fidelis?

The Most Probable Situation

Taxonomic understanding is imperfect and constantly improving over time. We can only state the most probable situation for the available facts. In our case, three situations seem possible. Let's state these situations without referring to the possibly-invalid species name "Mecaphesa decora":
  1. Misumenops volutus is the same species as Misumena fidelis, with Misumena decora being a different species; or
  2. Misumenops volutus is the same species as Misumena fidelis, Banks' male Misumena decora is actually a male of this species, and Banks' female Misumena decora actually belongs to some other species; or
  3. Misumenops volutus, Misumena fidelis, and Misumena decora are all the same species.
The first two possibilities seem the most probable for the following reasons:
  • The Banks 1898 drawing of the Misumena decora epigynum looks significantly different from the epigynum of one of the type specimens he designated.
  • The Banks 1898 drawing of the Misumena decora palp is not obviously identical to the Misumenops volutus palp, though it may fall within the limits of variation.
  • The male Misumena fidelis was unknown to Banks, the female Misumenops volutus was unknown to F. O. Pickard-Cambridge, and an adult male of Misumenops volutus was collected with two adult females of Misumena fidelis.
Adult male spiders typically spend their time seeking out females of their species to mate with, so the immediate proximity of the male and female specimens could be informative. Unless you're observing mating behavior, it's hard to tell which males go with which females except by their mutual proximity. However, because we only have one data point in our case—one occurrence of a male with females—we can't have a lot of confidence in the association on this basis alone.

In support of cases (2) and (3), there is evidence that Banks' Misumena decora palp might be an inaccurate drawing. Here is the epigynum of one of the very spiders that Banks drew, next to his drawing of that spider's epigynum. Banks clearly was not able to see all the detail that we can now see.


Misumena fidelis, ventral epigynum. The photographed specimen is one of those
from which Banks made the drawing on the right (Banks 1898 Plate 16, Figure 2).


It is still possible that Misumenops volutus and Misumena fidelis are different species, but this scenario creates an improbable coincidence. It would mean that we found a male of a species (Misumenops volutus) for which the female is unknown with two females of a species (Misumena fidelis) for which the male is unknown, and yet after this finding, the female and male counterparts remain unknown. Instead, it seems that the finding should change our working assumptions.

An examination of two remaining Misumena decora type specimens should select among the three possibilities. I'm reluctant to borrow the only type specimens in existence for a species, but maybe someone at the Harvard Museum of Comparative Zoology could check this out and report back.

Misumena fidelis is a Mecaphesa

We found that the external appearance of our Misumena fidelis didn't allow us to decide which genus it belonged to according to modern definitions. Recall that we had narrowed it down to either Misumenops or Mecaphesa. Now that we have three female Misumena fidelis, only one of which is a type specimen, we are freer to dissect one. I've chosen the CNC specimen whose ventral epigynum is virtually indistinguishable from that of the type specimen.

Internally, spider epigyna consist of sclerotized (hardened) structures. These internal structures are often the most distinctive character of females of a species. The dorsal view of an epigynum is the view of these structures from their internal side. In crab spiders, the most conspicuous parts of the dorsal epigynum are the spermathacae, which store sperm received from males.

Here is are the dorsal epigyna of two species of Mecaphesa and one species of Misumenops, for comparison with a Misumena fidelis epigynum. The large boxy structures are the spermathacae.


 Left: Mecaphesa celer, dorsal epigynum. Right: Mecaphesa dubia, dorsal epigynum.

Left: Misumenops pallidus, dorsal epigynum. From Lehtinen & Marusik 2008, fig. 9.
Right: Misumena fidelis from CNC, dorsal epigynum.


Lehtinen & Marusik 2008 says that the Misumenops spermathecae are "tubular U-shaped," as we see with the drawing at the lower-left. It also says that the Mecaphesa spermathecae have "large compact basal parts," as we see with the top two images. The Misumena fidelis dorsal epigynum at the lower right is not tubular-looking but has basally large spermathacae, matching it with Mecaphesa.

We can definitively say that Misumena fidelis now belongs to the genus Mecaphesa.

However, this was my suspicion before the dissection, so there may be something about the ventral epigynum that allows us to make this determination without dissection. It seems to me that the copulatory openings of Misumenops are always lateral to the hood, while the copulatory openings of Mecaphesa are always posterior or interior to the hood. In the species of Mecaphesa that appear to have multiple hoods, the relevant hood is the broadest median hood at the anterior of the epigynum. This rule seems to work for all the Misumenops and Mecaphesa I have so far seen, whether as drawings or actual specimens. However, I have not seen many species of Misumenops.

In particular, Misumena fidelis makes it clear that not all Mecaphesa epigyna need have a median septum or anything resembling a median septum. Had the copulatory openings of Misumena fidelis been closer together, their margins might have been construed as the margins of a median septum. It's hard to argue that they form a median septum when they are as far apart as they are in M. fidelis.

Lehtinen & Marusik 2008 also says that Misumenops has a "mostly small epigynal hood" that is "widest in Misumenops bellulus." The inner hood of Misumena fidelis looks small, but the outer hood containing it does not. The size of outer hood might have been enough to rule out Misumenops, but it could also have been the case that there's a Misumenops having a wider hood than Misumenops bellulus. Relative sizes probably aren't a stable diagnostic over the long term.

Misumena fidelis is Mecaphesa fidelis

Now that we know Misumena fidelis is actually a Mecaphesa, we can ask what the full species name should be for this spider. Mecaphesa what?

The name given to the first description of a species is the name that takes precedence. Our names of concern are Misumena fidelis, Misumena decora, and Misumenops volutus. Banks described both Misumena fidelis and Misumena decora in the same 1898 paper. F. O. Pickard-Cambridge described Misumenops volutus in a 1900 paper, so Banks' names take precedence.

We have three scenarios to consider, according to our prior probable situations:
  1. Misumenops volutus is Misumena fidelis, but Misumena decora is another species. In this case, the modern name of the species is Mecaphesa fidelis.
  2. Misumenops volutus is Misumena fidelis, as is Banks' male Misumena decora, while Banks' female Misumena decora is another species. In this case, we could call the species either Mecaphesa decora or Mecaphesa fidelis. However, if we call it Mecaphesa decora, Banks' female Misumena decora would no longer have a species name, and nothing would have the name Mecaphesa fidelis. To ensure that the first to name and describe a species actually got to name the species, it seems that we would give the name Mecaphesa fidelis to Misumena fidelis and restrict Misumena decora to just the female that Banks' described.
  3. Misumenops volutus, Misumena fidelis, and Misumena decora are all the same species. In this case we could name the species either Mecaphesa decora or Mecaphesa fidelis. However, because there's a chance that someone may later prove this interpretation incorrect, it may be wiser to allow for the possibility that Banks' female Misumena decora could have its own name, thus again suggesting the name Mecaphesa fidelis.
In all three scenarios, the best modern name for Misumena fidelis appears to be Mecaphesa fidelis. Moreover, the spiders that we have been calling Mecaphesa decora since Banks are also more properly Mecaphesa fidelis, and Misumena decora becomes more of a mystery.

The Range of Mecaphesa fidelis

These synonymies tell us more about the range of this spider, because the species has been collected under other names. Banks is the only one to have reported collecting a female, so we can assume that all of the male palps match Misumenops volutus and hence our presumed male Mecaphesa fidelis.

A proper range compilation would require examining specimens from Mexican collections. We'll keep things simple and report the minimal known range by combining the records of Banks 1898, F.O. Pickard-Cambridge 1900, Banks 1901, Gertsch 1939, and our three CNC specimens.

Mecaphesa fidelis ranges along the Sierra Madre Occidental mountain range in Mexico, westward to at least southern Baja California, southward to Guatemala, and possibly northward into southern Arizona. The records from Mexico are as follows: El Taste; La Chuparosa; San José del Cabo; Lake Chapala, Jalisco; La Buena Ventura, Veracruz; Pedregales; and Xechi milco, Federal District. There is also one record from Guatemala and one record from Catalina Springs, Arizona, USA.

Gertsch reports the Arizona record from a single female despite not having a clear understanding of the epigynum, so we probably shouldn't be confident in that record. Mecaphesa fidelis certainly ranges through much of Mexico but may or may not range into the southern USA.

Work to be Done

The following work remains to be done to resolve outstanding questions about Mecaphesa fidelis:
  1. Examine Banks' type specimens for Misumena decora to see whether they match the male or female Mecaphesa fidelis. These specimens are probably at the Harvard Museum of Comparative Zoology. They should select among our three probable scenarios.
  2. Locate the female specimen that Banks reported for Arizona in 1901 and identified as Misumena fidelis. If that specimen matches Mecaphesa fidelis, we can definitively include southern Arizona and the United States within its range.
  3. Search collections of misumenine thomisids from southern Arizona and maybe southern New Mexico to attempt to locate other specimens of Mecaphesa fidelis.
  4. Collect misumenine thomisids from southern Arizona and New Mexico in search of this species. At present, identifying this species requires examining the genitalia under a microscope, so they would have to be collected and preserved for examination.
  5. Locate drawings or specimens that match Banks' 1898 drawing of the Misumena decora epigynum. There may yet be a match for this spider other than Mecaphesa fidelis.

References

Banks, N. (1898b). Arachnida from Baja California and other parts of Mexico. Proceedings of the California Academy of Sciences (3) 1: 205-308.

Banks, N. (1901). Some Spiders and Other Arachnida from Southern Arizona. Proceedings of the United States National Museum XXIII: 581-590.

Gertsch, W. J. (1939b). A revision of the typical crab spiders (Misumeninae) of America north of Mexico. Bulletin of the American Museum of Natural History 76: 277-442.

Lehtinen, P. T. & Marusik, Y. M. (2008). A redefinition of Misumenops F. O. Pickard-Cambridge, 1900 (Araneae, Thomisidae) and review of the New World species. Bulletin of the British Arachnological Society 14: 173-198.

Pickard-Cambridge, F. O. (1900). Arachnida - Araneida and Opiliones. In: Biologia Centrali-Americana, Zoology. London 2, 89-192.

No comments:

Post a Comment