Laurie Burnham.
Studies on Upper Carboniferous insects: 1. The Geraridae (Order Protorthoptera).
Psyche 90:1-58, 1983.

Full text (searchable PDF, 11184K)
Durable link: http://psyche.entclub.org/90/90-001.html

The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

PSYCHE
Vol. 90 1983 No.1-2
STUDIES ON
UPPER CARBONIFEROUS INSECTS:
1. THE GERARlDAE (ORDER PROTORTHOPTERA)
By Laurie Burnham
Department of Entomology, Cornell University, Ithaca, New York 14853;
and Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02 138*
Despite the importance of the order Protorthoptera,' little is known about its evolutionary history. While recent workers have emphasized morphological and taxonomic diversity in the group (Carpenter, 1971, 1977; Wootton, 1981), no one has undertaken serious revisionary study at the family level. As a consequence, our understanding of relationships within the order, as well as relation- ships of the Protorthoptera to other Paleozoic insects, is rudi- mentary at best. Clearly, revisionary studies on this group are badly needed.
We know that the Protorthoptera first appear in the fossil record at the base of the Upper Carboniferous (Namurian Stage) and apparently flourished for 80 million years before becoming extinct at the end of the Permian. We also know that they were remarkably 'It was one of the dominant orders of the Paleozoic (exceeding all other insects both in number of species and in number of individuals), and is considered by many to be ancestral to the Endopterygota (the group to which 90% of all living insects belong).
*Present address.
Manuscript received by the editor March 5, 1983.



================================================================================

2 Psyche [VOI. 90
diverse morphologically, and that diversity in the group (sensu lato) far exceeded that of any other Paleozoic order (Carpenter, 1977). Structural modifications normally associated with more recent insects, including brightly patterned wings, raptorial fore legs, and thoracic extensions of various kinds, are found throughout the group.
Despite this fascinating array of characters, Carboniferous Protorthoptera have generated little interest among systematists. This is due, in part, to problems common to all paleoentomological study: an overall lack of material (rarely is a species based on a large series of specimens); preservational quality that ranges from excel- lent to very poor; and a taxonomy that is highly subjective. In addition, there are problems unique to the Protorthoptera which make their study particularly difficult. First, they are neop- terous, and as a consequence, are most frequently found with their wings superimposed on one another. Interpretation of the venation under these circumstances is not only difficult, but prone to error. Second, the vast majority of Carboniferous Protorthoptera are known from only two localities: Commentry in France, and Mazon Creek in the United States, and were studied primarily by turn-of- the-century workers.* As new material has become available for study in recent years, the accuracy of much of this earlier work has been questioned.
Finally, there is the problem of variation. Distinguishing species- level differences from intraspecific variation in fossils that have such a limited array of characters is not easily resolved. How, for instance, does one recognize sexual dimorphism in a fossil species when there are no genitalia or secondary sexual characters to serve as guidelines? It is not surprising under the circumstances that paleotaxonomy rests largely on subjective reasoning. But this, unfortunately, has its pitfalls.
Previous work on the Geraridae is a case in point. Anton Hand- lirsch, responsible for most of the earlier work on the family, de- scribed a new species of gerarid for every specimen he examined, ^The Permian Protorthoptera are generally much better known than their Carbo- niferous relatives. This is attributable to the fact that there are more than eight major Permian deposits (including two in the U.S.) at which Protorthoptera have been found. Furthermore, intensive studies on these insects have attracted the attention of such well-known recent workers as Carpenter, Kukalovh, and Sharov.



================================================================================

19831 Burnham - Geraridae 3
LOWER
E AR LY
WESTPHALI AN
ttA
NAMURIAN
11
V I SEAN
TOURNAISIAN
Fig. 1.
Geological Time Table of the Carboniferous. Note the different ages of the Cornmentry and Mazon Creek localities. basing his taxonomic decisions on small differences in wing vena- tion (Handlirsch, 1906b, 191 1, 1922). This approach, while render- ing the decision making easier, is nevertheless open to criticism. Studies on intraspecific variation in some Permian Protorthoptera have shown, for instance, that two fore wings belonging to the same specimen will exhibit noticeable differences in venation (Carpenter, 1966). From these findings we can conclude that intraspecific varia- tion in the Protorthoptera was high, and that the variation Hand- lirsch saw was no greater than that seen in a single specimen. Further evidence that suggests Handlirsch was unrealistic in his representation of species diversity comes from the low probability of finding only one individual per species (for all species collected) in a random sample of living insects. Similarly, we cannot reasonably expect to find only one specimen per species in a paleontological sample, particularly when fossilization was catastrophic (and hence random) as is true for the Mazon Creek biota. Nevertheless, in spite of these drawbacks to the study of fossil insects, the field can be immensely rewarding. This is because it provides us with concrete evidence (in the form of fossils) of what early insect life was like. Without such proof, we would be guided



================================================================================

4 Psyche [VOI. 90
only by our imagination, and having this proof provides a base on which insect phylogeny and early insect evolution can be recon- structed.
Clearly, revisionary studies on Paleozoic insects are important. Fortunately, recent revisional work by Carpenter has greatly in- creased our knowledge of certain Paleozoic orders (i.e., the Paleodic- tyoptera, Megasecoptera, Diaphanopterodea, Protodonata, and Caloneurodea), but much work on the Protorthoptera remains. This revision of the family Geraridae is, at least, a beginning and is intended to be the first in a series of family-level studies on Upper Carboniferous Protorthoptera.
Selection of the Geraridae as a starting point was influenced, ultimately, by two factors: 1) it is typical of many of the families in the Protorthoptera, having last been studied in the early part of this century (despite the discovery since then of new material assignable to the family); and 2) the Geraridae are morphologically most unus- ual insects. They were large (up to 75 mm in body length), and had as their single most distinctive attribute, a prothorax that was elon- gate, flask-shaped, and adorned with long, numerous spines. These spines gave them the appearance of walking pincushions, and pre- sumably provided some defense against vertebrate predators. The systematic importance of the family plus the impact this work has on current classifications of the Protorthoptera will be discussed in subsequent pages. The remaining sections of this paper cover (in the following order): the systematics of the Geraridae; paleoecologi- cal differences between the two localities at which gerarids have been found (Mazon Creek and Commentry); and the significance of this study relative to phylogenetic relationships within the Pro- torthoptera.
The fossils examined for this study occur mostly as impressions (imprints in a sedimentary matrix devoid of organic matter) but some occur as compressions (in these, organic matter is present, but usually coalified). Both types of fossils were prepared by degage- ment,3 i.e., an uncovering of the fossil by removal of the overlying rock matrix. This is generally done using a fine pneumatic drill and ^From the French verb ilkager meaning to disengage, extricate, or get clear.



================================================================================

19831 Burnham - Geraridae 5
compressed air gun. The technique is particularly effective at revealing regions of an insect's body (wing tips, legs, etc.) that are found beneath the bedding plane. Following degagement, specimens were studied under a Wild M-5 stereo microscope and photo- graphed with a Zeiss 4" by 5" format camera. Drawings were made of each fossil by tracing a general outline from a photographic enlargement. Verification of detail was made by referring back to the specimens and examining them frequently under the microscope. The most complete reconstructions (e.g., fig. 2 of G. danielsi) were possible for those species that consist of a large series of specimens. This is because one fossil rarely displays all characters equally well, and, therefore, the larger the number of speci- mens, the greater the likelihood of multiple character preservation. Type specimens, including the holotypes, for all taxa considered in this revisionary study were borrowed and examined using the above methods. Pre-existing taxa were synonymized whenever pos- sible, a decision based on the assumption that (for reasons cited earlier) intraspecific variation in the Protorthoptera is great. Char- acters of greatest taxonomic importance were venation and body size and shape, particularly with respect to the prothorax. In situa- tions where clearcut characters were lacking, as is true for several of the Mazon Creek gerarids, I relied solely on size as a criterion for specific assignment. While this may result in the recognition of some dubious species, it seems preferable to relegating certain specimens to incertae sedis status.
Since wing venation is such an important taxonomic tool both in paleoentomological and extant systematic study, it is surprising that until recently no standardized wing terminology has been adopted. This is particularly unfortunate for the Protorthoptera, 80% of which have been described on the basis of wings alone. Inroads have recently been made into this problem primarily by the efforts of Carpenter in the United States and Wootton in Great Britain. Both have stressed (Carpenter, 1966; Wootton, 1979, 1981) the importance of a standardized venational nomenclature and Woot- ton (1979) has proposed a terminology modified slightly from the one used previously by Lameere (1922) and Martynov ( 1924, 1938). Wootton proposes that the following nine major longitudinal veins be recognized: Costa (C); Subcosta (SC); Radius (R); Radial Sector (RS); Anterior Media (MA); Posterior Media (MP); Ante- rior Cubitus (CUA); Posterior Cubitus (CUP), and Anals. In light



================================================================================

6 Psyche [VOI. 90
of the historical basis for the nomenclature (used extensively in the paleoentomological literature) and its conservatism (it may be used to homologize the wing venation of all insects) I enthusiastically concur with Wootton's recommendations and will employ his sys- tem here and in future systematic work.
A total of 58 specimens were made available for study through the loans of various institutions and individuals. These are listed here with their abbreviations:
Field Museum of Natural History (FMNH), Chicago, Illinois, U.S.A. (This includes specimens collected by Jerry Herdina and subsequently acquired by the Field Museum). Institut de Palkontologie, Museum National d'Histoire Naturelle (IP), Paris, France.
Museum of Comparative Zoology (MCZ), Cambridge, Massa- chusetts, U.S.A.
United States National Museum (USNM), Washington, D.C., U.S.A.
Yale Peabody Museum (Y PM), New Haven, Connecticut, U.S.A. Daniel Damrow, of Mosinee, Wisconsin. Private collection. (DMRW) (Includes specimens previously in the collection of Walter Dabasinskas).
David Douglass, of Yachats, Oregon. Private collection. (DGLS) Francis and Terri Wolff, of Port Charlotte, Florida. Private col- lection (Wolff).
Family Geraridae Scudder, 1885
[Nom. correct. Handlirsch, l906a (ex Gerarina Scudder, I885)] Gerarina Scudder 1885342. Type: Gerarus Scudder. Geraridae Handlirsch 1906a: 146, 1906b:70 1 , 19 1 1 :3 12. 1920: 15 1. Sthenaropodidae Handlirsch 1906a: 14 1, 19 19:37, 1920: 150; Sharov 1968: 19. Type: Sthenaropoda Brongniart. NEW SYNONYMY.
Genopterygidae Richardson 1956:41. Type: Genopteryx Scudder. NEW SYNONYMY. Description
Fore and hind wings similar in size and shape, but markedly different in venation.
Fore wing: length 35-55 mm, and apparently not coriaceous; cos- tal area broad in basal region, SC simple, terminating in C; R parallel to SC, terminating at wing apex; RS originating from base



================================================================================

19831 Burnham - Geraridue 7
of R near midpoint of wing; M either anastomosing with RS for a short distance or connecting to it by a cross-vein; CUA strongly developed, arising from base of M; CUP forked, arising independ- ently from wing base.
Hind wing: length 30-48 mm; costal area not as broad as in fore wing; SC simple, terminating in C; R parallel to SC, terminating at wing apex; RS pectinate, arising from R near wing base; M forked, arising near base of RS; CUA and CUP simple and parallel to one another; CUA arising from base of RS near M, CUP arising inde- pendently from wing base; anal area unusually reduced; cross veins abundant in both fore and hind wings.
*
Body: prothorax elongate, flask-shaped, and distinctively spi- nose; abdomen cylindrical; antennae filamentous; head small and probably mobile; legs cursorial, tarsi five-segmented. Diagnosis
In many ways the Geraridae are typical Orthopteroidea, having mandibulate mouthparts, hypognathous heads, and filamentous antennae. But they differ from other orthopteroids in two important characters: their well-developed prothorax which is armed with spines (the latter reach a length of 10 mm in G. danielsi), and their distinctive fore and hind wing venation. While gerarids can be read- ily recognized on the basis of the prothorax alone, wing venation is generally a better diagnostic character. Particularly distinctive are the RS-M veins in the fore wing, and the R-RS veins in the hind wing. In the fore wing RS is reduced and M is expanded with 5 to 6 branches. The apical branch of M either anastomoses with RS for a short distance or is connected to it by a short cross vein. In the hind wing, the opposite is true: M is greatly reduced and RS expanded into 5 to 6 branches.
It is worth noting that the anal fan in the hind wing, if present, was very small (see fig. 17). This suggests that in gerarids the fore and hind wings may have functioned equally well in flight, unlike in extant Orthoptera, which rely primarily on expanded hind wings for flight propulsion. The abdomen is essentially unknown for the fam- ily, but was probably shorter than the wings, a claim based on the comparison of wing length to legs, thorax, and head. No cerci are preserved, but because the Geraridae are orthopteroid, it may be assumed that they were present.




================================================================================

Psyche
[Vol. 90
Fig. 2. Gerut-us ilunklfii, composite drawing, based primarily on specimens FMNH PE 5276. 31973.32027, 32029; and USNM 31973.



================================================================================

19831 Burnham - Geraridae 9
Remarks
The family Geraridae was first established by Scudder (1885) for several fossil insects from Mazon Creek noted for their slender bodies, which tapered "greatly anteriorly" (Scudder, l885:344), and for the distinctive branching pattern of RS in the hind wing. Scudder placed the family in the order Paleodictyoptera, section Neuropteroidea, where it remained until 1906 when Handlirsch erected the order Protorthoptera and transferred the Geraridae to it.4 Most of the subsequent work on the family was carried out by Handlirsch who added a total of two new genera and nine new species to it (Handlirsch, 1906a, 1906b, 19 1 1, 1920). This revision is the first systematic work carried out on the family since then, and rectifies many of the taxonomic errors made by these earlier workers. To a large extent, the mistakes made by Scudder and Handlirsch may be attributed to the limited availability of material at their disposal, and the preservation of most gerarids with all four wings lying over one another. Nevertheless, their errors were of grave consequence. To begin with, neither worker apparently recognized the extent to which intraspecific variation occurs in the family, and therefore each named only monotypic species. But, more importantly, owing to the difficulties of wing overlap, neither Scudder nor Handlirsch correctly interpreted the wing venation of Gerarus; both managed to interpret the venation of one wing (the hind wing) and then assumed that fore and hind wings were identi- cal, although neither actually saw the fore wing. The advantage of having more material at my disposal made it possible for me to overcome the problems that faced these workers. Certain well-preserved specimens (especially FMNH-PE 5276, 31973, 32027; IP 5, 23) were instrumental in demonstrating the complete venational differences between fore and hind wings. A comparison of figs. 6a and 6b shows how strikingly different the fore wing actually is from the hind wing. This, in itself, was quite a revelation. But it was only later, when searching through the litera- ture looking for venational similarities with other groups, that the *Prior to this. all Carboniferous insects were included in the one order Paleo- dictyoptera in accordance with Scudder's beliefs that ordinal differentiation had not taken place in the Insecta as early as the Carboniferous. We know, of course, that this was incorrect; a total of 1 1 orders are now recognized from that Period (Carpenter, 1977; Wootton, 1981).




================================================================================

10 Psyche [VOI. 90
full significance of the discovery emerged. It became immediately apparent, based on fore wing characters, that the type genus (Sthe- naropoda) for the family Sthenaropodidaej is inseparable from Gerarus. The consequences of this are twofold: 1) it extends the geographic range of the Geraridae from North America to Europe, strongly suggesting that the family was once large and successful; and 2) the synonymy casts serious doubts on current classifications of Paleozoic orthopteroids such as those proposed by Sharov ( 1968) and Rasnitsyn (1980). The implications of this are addressed in the discussion section at the end of this paper. GEOLOGICAL RANGE: Carboniferous-Westphalian D to Stephan- ian. OCCURRENCE: Mazon Creek, Illinois, U.S.A.; Commentry, France. TYPE GENUS: Gerarus.
Synonymies
The families Genopterygidae and Sthenaropodidae are synony- mized here with the Geraridae, since I find no unique characters by which to recognize them as independent taxa. All major veins and body characters are in complete agreement with the definition of the Geraridae. Although the Genopterygidae are described from the hind wing alone, and this synonymy may therefore change as addi- tional material is found, the venational similarities between Genop- teryx and Gerarus are striking (see fig. 7). This, in my mind, is sufficient reason at this time to synonymize these families. The syn- onymy of the Sthenaropodidae with the Geraridae is based not only on the venation of both wings, but also on the prothorax (complete with spines) and body size. The two families are so similar in charac- ter that synonymy at the species level could almost be justified were it not for their separation both geologically and geographically. å´"Th Sthenaropodidae, like the Geraridae, were the focus of taxonomic work largely at the turn of the century. Brongniart first described Sthenaropoda (based on S. f'ischeri) in 1885 and placed it with a series of other Carboniferous Protorthoptera in the family Paleoacridiodea. Eight years later he synonymized Sthenuropoda with Oedischia (now recognized as belonging to the Orthoptera), believing their differ- ences too slight to warrant generic separation. In 1906 Handlirsch restored the genus Sthenaropoda and placed it in its own family. His decision was later defended by both Lameere (1917) and Sharov (1968), who felt that the oedischiids, by virtue of their saltatorial legs, were trw Orthoptera, and that the sthenaropodids, which lacked well-developed jumping legs, were clearly members of the Protorthoptera. The ramifications of this are discussed in the concluding pages of this paper.



================================================================================

Burnharn - Geraridae
Fig. 3.
Handlirsch's reconstructions of two species of Geruru.~. a. G. danie/.si; b. G. w//uris (=/ongi~~o//i.s). (From Handlirsch, 1920: 152,153).



================================================================================

Genus Gerarus
Gerurus Scudder I885:Wl; Handlirsch l9O6a: 147, l906b:702, I9 I I : 3 13, I9 IW8. Type: Geruru.~ \vtu.s Scudder (original designation). S/henuropodu Brongniart l885:59; Handlirsch 1906a: 148. 1906b: 704, I9 19130. Type: S~henurupo~/u,fi.s~~l~eri (original designation). NEW SYNONYMY.~) Genopferj,.~ Scudder I885:W; Handlirsch l906a: 148, l906b:704, I9 l9:3O. Type: Genopterj..~ (wn.s/rkYu (original designation). NIW SYNONYMI..~ Archueu(wXte.~ Meunier l909a:39, 1909~: 145; Handlirsch I9 l9:39. Type: Archueu- (,ridite.s hruesi (original designation). NIW SYNON\.MI,. Ro.s.site.s Richardson 1956:44. Type: Ro.s.site.s inopinus (original designation). SEW SYNONYMY.
Description
Fore wing: membranous, larger than hind wing, rounded at apex; SC long, weakly turning anteriorly to fuse with C at point three-fourths to two-thirds the length of the wing; R parallel to SC, terminating in C slightly anteriorad to wing apex. Fore wings differ from hind wings in the following veins: in the fore wing, RS branches from R in the basal third of the wing, and bifurcates two or three times. M four- or five-branched, either connecting to RS by a cross vein or fusing with it; CUA coalesces with M for short distance at wing base and may be weakly branched; CUP simple, elbowed towards CUA; network of anal veins present. Hind wing: RS has three to six distinct pectinate branches and does not fuse with M; M multiply branched, arising from RS; CUA fiSome doubt exists concerning the date of publication of this paper with respect to Scudder's I885 article, but 1 have concluded for the following reasons that Scudder had priority of publication: I) Although we do not know the month of publication for Brongniart's paper, we do know that Scudder's was published early in April, 1885. Unfortunately, attempts to obtain the exact date of publication for Brongni- art's article from the Muskurn d'ff istoire Naturelle de Rouen and the Societb des Amis des Sciences Naturelles de Rouen have met with no response. 2) Citations of these two papers (e.g., Handiirsch, l906a, 1922) have consistently listed Scudder's paper before Brongniart's. 3) Scudder's I885 account of Geraru.~ includes a full description, figures, and designation of a type species (G. vetus), whereas Brongni- art's paper only mentions Srhenuropodu and gives no formal description. 7Geruru.\ and Genop/er,~,.~ were named and described by Scudder in the same paper (1885). In accordance with the I.C.Z.N. procedures, and as the first reviser, I have treated Genopwr,~,.~ as the junior synonym of Geruru.~, the better known and larger genus.




================================================================================

19831 Burnham - Geraridae 13
simple or with one bifurcation; CUP simple and parallel to CUA; anal area slightly expanded, but unusually reduced for the Orthop- teroidea. See fig. 17.
Diagnosis
Gerarus may be distinguished from the other genera in the Geraridae (Nacekot?7ia, Progenentot?wx Genenton~uni and Gerarulus) by size (members of this genus are large, fore wing is 40 mm to 55 mm in length); and the nature of the RS and M veins in the fore wing. In Gerarus RS branches two or three times; in Progenentomum, it branches at least four times. M in Gerarus is four- or five-branched, and either coalesces with RS for a short distance or is connected to it by a well-developed cross vein. In contrast, M in Nacekomia is distinct from RS, and in Genentotnum, M is only three-branched and these branches are distinctly parallel to one another. Other characters such as the shape of the thorax and number of prothoracic spines may ultimately prove important in distinguishing these genera from one another, but as yet, we lack the well-preserved specimens necessary for separating all four genera in the family on the basis of such additional characters. Remarks
Handlirsch ( I9 1 1 :3 13) characterized Gerarus by its prothorax, described as "a broad base, either provided with tubercles or smooth, but in every case, produced into a long neck-like part bear- ing the head." While he was correct about the nature of the b'neck,'* he was incorrect in his assessment of the bbtubercles," which were presumably present in all adult gerarids as fully produced spines, not tubercles. He was also slightly inaccurate in describing the pro- thorax as "a broad base." This study has shown the width of the prothorax to vary from 5 mm to I3 mm depending on the species. A better description for the genus is one based on wing venation. GEOLOGICAL RANGE: Upper Carboniferous-Westphalian D to Stephanian. OCCURRENCE: Mazon Creek, Illinois, U.S.A.; Com- mentry, France. TYPE SPECIES: Gerarus vetus.