Jarmila Kukalová.
Revisional Study of the Order Palaeodictyoptera in the Upper Carboniferous Shales of Commentry, France. Part III.
Psyche 77:1-44, 1970.

Full text (searchable PDF, 4408K)
Durable link: http://psyche.entclub.org/77/77-001.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.

Figure 50. Reconstruction of Stenadictya based on specimens in the Institut de Pal6ontologie, Paris. For explanation, see text. Pu&e 77:l-Ì 11970). http//psychr entcIub.me/77/77-001 him)



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




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

PSYCHE
vol. 77 March, 1970 No. I
REVISIONAL STUDY OF
THE ORDER PALAEODICTYOPTERA IN
THE UPPER CARBONIFEROUS SHALES OF
COMMENTRY, FRANCE. PART 1111
BY JARMILA KUKALOVA~
Charles University, Prague
The first two parts of these studies have dealt with eight families of the Palaeodictyoptera in the Corninentry collection at the Institut de Pal&mtologie in Paris. The introductory discussion in Part I summarized the background of this investigation and the extent of the collection studied. The present part deals with the remaining family, the Dictyoneuridae, which is well represented in the Com- mentry shales. Some of the fossils of this group provide us with structural details that are otherwise unknown in the Palaeodictyop- tera. In the concluding paragraphs of this paper I have summarized what is known of the wing and body structures of this extinct order of insects.
Family Dict~oneuridae Handlirsch3
Stenodictyopterida Brongniart, 1885 : 60 ; Brongniart, 1893 : 380. Sten~dictyo~teridae Pruvost, 1919 : 98 ; Laurentiaux, 1953 : 419. Dictyoneuridae Handlirsch, 1906: 63 ; Handlirsch, 1911 : 297 ; Lameere, 1917: 102; Handlirsch, 1919: 3.
Stenodictyidae Laurentiaux, 1952: 234.
'Published with the aid of a grant from the Colles Fund of the Museum of Comparative Zoology at Harvard College and a Grant-in-Aid of Re- search from the Society of the Sigma Xi. This study has also been sup- ported in part by grants numbered GB2038 and GB7038 (F. M. Carpenter, Principal Investigator) from the National Science Foundation. Part I, dealing with the family Spilapteridae, was published in Psyche, Vol. 76, pp. 163-215; part 11, dealing with seven additional families, was published in Psyche, Vol. 76, pp. 439-486.
'Alexander Agassiz Lecturer in Zoology, Harvard University, 1969. 'AS noted by Handlirsch (1906) the name Stenodictyopteridae, which was not based on a generic name, is invalid. The name Stenodictyidae, used by Laurentiaux and based on Stenodictya, presumably in an attempt to retain a name similar to the one used by Brongniart, is, of course, a synonym of Dictyoneuridae, which was correctly formed by Handlirsch in 1906.



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

2 Psyche [March
Type Genus: Dictyoneura Goldenberg, 1854. Brongniart originally conceived of this family as characterized by the presence of a dense archedictyon and a specialized venation in which there were few branches. Marked attention was given to this family in later years and it played an important role in hypotheses on insect phylogeny. Brongniart's original and apparently correct conclusion on the specialized nature of the venation in this family was rejected or modified by subseqent investigators. Handlirsch, who considered the morphology of Stenodictya as supporting his theory of trilobite ancestry of insects [by having the prothoracic lobes and abdominal expansions derived from the trilobite cephalon and pleura ( I 908, p. 1304) 1, believed the Dictyoneuridae to be the most primi- tive of all Pterygota. He also thought the order Palaeodictyoptera was ancestral to all other insect orders. This latter view was rejected by Martynov ( 1925, 1938)) who did, however, follow Handlirsch ( 1938, p.
19, 21 ) in his conviction that the Dictyoneuridae were the most primitive of the Palaeodictyoptera. Martynov based his conclusions on the presence of four features in the family: a uniform archedictyon, well developed prothoracic lobes, paranotal expansions along the abdomen, and a primitive venational pattern, the veins having few branches.
In the light of information acquired in recent years, the arche- dictyon seems to be only one of these traits that can be considered as primitive. In the geological record of insects, the archedictyon appears repeatedly in the more ancient forms of primitive groups, becoming irregular or reduced to cross veins in more advanced forms (e.g., Paleozoic Blattodea, Protorthoptera, etc.). On the other hand, the prothoracic lobes of Stenodictya and of other Dictyoneuri- dae, as far as known, do not differ from those of other Palaeo- dictyoptera and cannot be considered as being more primitive. Previous interpretations of the so-called expansions of the abdomen in Stenodictya seem to be very questionable, after careful study of the type specimens concerned. Incorrectly figured by Brongniart, who considered them to be homologous to the tracheal gills of mayfly nymphs, the expansions seemed to Handlirsch and Martynov as evidence for their respective hypotheses on insect evolution. Hand- lirsch assumed that the paranotal expansions of the abdomen of insects were derived from the pleura of trilobites because of the presence of an "oblique furrow." In his hypothesis on the origin of the Pterygota, Martynov assumed that wings of insects arose from the paranotal expansions which developed in hypothetical ancestral



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

forms on both the thorax and the abdomen, as they are still partially present in some Recent T,hysanura (Lepismatidae) . He considered
the lateral abdominal expansions of Stenodictya to represent a very primitive stage of this development at a time when they were not yet completely reduced.
My examination of the Commentry dictyoneurids throws an entirely different light on the problem. The abdominal tergites of Stenodictya are heavily sclerotized with pointed and even prolonged posterolateral angles directed obliquely backwards. They are pro- vided with oblique ridges, which, of course, have nothing to do with the "oblique furrow" of trilobite pleura, but are secondary structures, undoubtedly serving to strengthen the lateral parts of tergites. The longitudinal suture, separating the lateral parts of tergites in some other families of Palaeodictyoptera, is completely missing. Since the abdomen of Stenodictya was relatively broad, decomposition processes had a marked effect on the relationship of abdominal segments, as preserved. They were widely separated from each other so that much of the intersegmental membrane between the tergites is visible in most specimens. The prolonged posterolateral angles of tergites consequently overlap the following segments much less and they protrude much more towards the sides. Simply by cutting out single segments of an exact figure and arranging them back to normal position, the abdomen becomes shorter, with the tergites slightly protruding. The abdomen, so reconstructed, is very suggestive of that of many Neoptera (e.g., roaches) and undoubtedly represents a derived, specialized state of development. Other body parts of Stenodictya were heavily sclero- tized as well, with deep, dense pits, exceptional for the Palaeodic- tyoptera. This strengthened cuticle probably provided protection against injury. Much more primitive, in my opinion, is the abdomen of some Palaeodictyoptera that have lateral lamellae present, i.e., with lateral parts of tergites separated by a longitudinal suture ex- tending from the notum. These structures are probably homologous with lateral lamellae of some living mayfly nymphs but their function is not yet explained.
Martynov considered the venation of the Dictyoneuridae as the most primitive in the Palaeodictyoptera because of the presence of only a few branches (Cornstock and Needham, 1898-99). The geological record, on the other hand, supports Redtenbacher's ( I 886) concept of the richer wing venation being present in the more ancient groups. This idea was followed and developed further by Lameere



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

Psyche [March
(1922). For that reason I consider the venation of the Fouqueidae, and to a lesser extent that of the Spilapteridae, as more primitive and much closer to the supposedly ancestral type than that of Dictyo- neuridae. Especially primitive, perhaps, are those genera with richly branched, convex and concave parts of M and Cu. The venation of the Di~t~oneuridae is an extremely simplified one for the Palaeodictyoptera and provides little basis for differentiation of taxa. It is therefore necessary to study the specimens in as much detail as possible in order to obtain full information. From my study of the Commentry material I am of the opinion that differences in cell shape of the archedictyon, the detailed outline of the posterior margin on the wings and distribution of cross veins and of the archedictyon may be useful for taxonomy. Nevertheless, the classi- fication of isolated wings at both specific and generic levels is less satisfactory than in other families of the order. In the Dictyoneuridae, the postcostal area is differently developed from that of all other families of Palaeodictyoptera. Arising from the very base of C, a single postcostal vein follows along the subcosta and terminates shortly on this vein. Commonly in the Palaeo- dictyoptera this subcostal vein is directed obliquely towards the costa, delimiting a triangular area and giving rise to several fine twigs. In the Dictyoneuridae also the precostal strip is very pro- nounced and broad, extending over the first third of the wing. Examining the other more specialized families, such as Eugereonidae, Calvertiellidae and Ar~haernega~tilidae, we note that the postcostal vein is completely reduced; on the other hand the precostal mem- branous strip is enlarged to form a true precostal area. The Dictyo- neuridae apparently show the way in which the postcostal area became reduced in favor of the precostal area in the more advanced and specialized families. There is no doubt that the precostal area is a "younger" feature than the postcostal area, which developed by radial evolution in some groups only. Thanks to Handlirsch's reconstruction, published in numerous textbooks of entomology and palaeontology, Stenodict~a Zobata is undoubtedly the most famous fossil
insect known. Unfortunately,
as this revisional study shows, no one feature given in Handlirsch's re- construction is correct. In contrast to what he figures, the mouthparts of Stenodictya are actually modified for sucking, the head is pro- vided with a large clypeal region, the prothoracic lobes have a venation and a cross venation, the wings are more slender distally than represented, RS has more numerous branches, the arche-



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

19701 Ku&aZovd - PaZaeodicty opt era 5
dictyon is replaced by cross veins in the subcostal and sc-r areas) the tarsal segments are five in number, the abdomen shorter and nar- sower than represented) the lateral lamellae are not present (the p~ste~olateral angles of tergites projecting only slightly) ) and the cerci are robust and long. Finally) Handlirsch's figure includes a combination of mde and female features. His reconstruction of the prothoracic lobes) wings and part of the abdomen were based on Brongniart's specimen 22-1) which is a female) as shown by the presence of an ovipositor (see Figure 50). On the other hand, the end of the abdomen of Handlirsch's reconstruction was based upon Brongniart's specimen 22-2) which turns out to be a male and which probably represents a distinct species. The male claspers, incidentally, were misinterpreted by Handlirsch (Lameere, I 91 7) P- 158L
Following the significant discovery by Laurentiaux ( 1952) of the presence of the sucking beak in a previously unstudied specimen of Stenodictya (collection of the Institilt, Paris)) doubts about the pres'ence of a beak in all Palaeodictyoptera have disappeared. On the basis of the photograph in Laurentiaux's paper) Sharov (1966) p. I 18) gave a new reconstruction of Stenodictya. However) this recor~struction is also incorrect, mainly hecause the specimen itself was preserved so as to show a ventral view, although this was not apparent from the photograph. As a result) Sharov erroneo~isly intcrpseted several features as dorsal in position. For example, the prothoracic lobes, appearing from 11 nderneath the body, were inter- preted as a prothoracic shield; and vag11e outlines of a struct~~re shown on the photograph only as the res111t of shading (but com- pletely invisible in the specimen itself) is represented in the reston- tion as a separate small segment at the base of the beak in the p1:1ce where, in the dorsal sudace of the Palaeodictyoptera, there is the triangular, elongate labrum. Furtlierniore, the beak as represented in Sharov's seconstruction, should be longer than drawn, with long palpi present; the wings should have cross veins in the subcod il~l(l the sc-r areas, the legs should have five tarsal segments; the cerci should be samewhat longer and the posterolateral angles or tl~r tergites less projecting.
In the accompanying illustration (Figure 50) I am inc111ding ;l
reconstruction of Stenodictya which, it should be noted) is :I corn- posite of structures present in several species of the genus) as follotvs: S. Zohata Brongniart, specimen 22-1, for head, eyes, clypeus, pro- thoracic lobes) venation of fore and hind wings (in part), thorax,



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

6 Psyche [March
abdomen, ovipositor, cerci; S. pygmaea Meunier for the shape of the prothoracic lobes; S. oustaleti Brongniart, specimen 22-3, and S. agnita (Meunier) for the five-segmented tarsi and legs; m(1 S. laurentiauxi, n.sp., for the beak. It should also be pointed out tl~t since the maxillary palpi are incompletely preserved in any specimen of Stenodictya, they have been represented in the restoration to the length of those in Eugereon boeckingi Dohrn; the labrum, which is indistinct in specimens of Stenodictyu, is drawn as it is in vario~~s species of Palaeodictyoptera belonging to other genera ; and the length of the ovipositor) also incomplete in specimens of Stenodic~ya, is drawn to its length in Homaloneura ornatu Brongniart, A dense archedictyon) more or less approaching that of the Dictyoneuridae and related families, occasionally occurs within the hon~oiopterid and spilapterid groups of families. In the Hon~oiop- teridae it is very dense, for example in BoZtopmvostia nigra (Kukalovi, 19581, and it is well developed in Homoioptera woodwmdi. In the S~ilapteridae, the archedictyon is psesu~nably completely reduced in all genera and the anastomoses between the cross veins are only rarely present, but the cross veins them- selves are exceptionally dense. Nevertheless, in the closely related family Eubleptidae) there is a dense archedictyon between the cross veins. A special case of modified archedictyon occurs in the Fouqueidae) particularly in the genus J'ouquea. The coasse, ex- tremely dense cross venation in that genus recalls very much the process which has taken place in some roaches, in which the dense reticulation in the more primitive Carboniferous forms became restricted into markedly dense patterns in Permian forms. Never-
theless, among all Palaeodictyoptera) the archedictyon of the Dictyoneuridae is certainly the most even and tends least to form cross veins. Its presence, together with specialized morphological features, is not surprising, this phenomenon being known as mosaic evolution.
Of the genera included in the Dictyoneuridae by Laurentiaux (1g53), Athymodictya Handlirsch is to be referred to the Eu- bleptidae) Dictyoneurella Laurentiaux to Archaemegaptilidae, Mecynoptera Handlirsch perhaps to Archaemegaptilidae and BoZto- firuvostia Pruvost to the Homoiopteridae. The following are the basic characteristics of the Dictyoneuridae: Fore and hind wings very similar. Main veins without coalescence. Sc ending on costal margin beyond midwing; RI <extending nearly to apex; Rs with several branches; MA unbranched) usually strongly



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

19701 Kukaiovd - Paiaeodictyoptera 7
curved; MP simpk or branched; CuA unbranched; CUP simple or having several branches. Archedictyon well developed over most of the wings and usually dense.
Body structures: head with small projecting eyes and large clypeus. Antennae multisegmented, long. Prothoracic lobes large, with about eight radiating veins, often branched, and numerous? anastomosing cross veins; thoracic segments uniformly long, though the prothorax may be somewhat shorter than the others. Legs short? robmt, tarsus 5-segmented, with claws and arolium. Abdomen relatively broad and short. Cerci in females long, multisegmented. Ovipositor in female curved, stout, reaching beyond the end of thr body. laales with claspers arising laterally from the subgenital plate) composed of homonomous~ carinated phtes, directed obliquely and touching distally. Body and wings completeIy and densely covered by deep pits. A11 body parts heavily sclerotized. The family Dictyoneuridae is related to the Eugereonidae: Archaemegaptilidae and Protagrionidae and is more distantly related to the Megaptilidae and Calvertiellidae. The genera included in the Commentry shales: Stenodictya Brongniart? Microdictya Brongniart. The following genera, also in the family Dictvoneuridae, are from the Stephanian of Germany: I~ictyoneura Goldenberg Stiibocrocis Handlirsch, Ci~fif~ GuthGrl, Rotundopteris Guth01-I, Poiioptenus Scudder, Dictyonezu-uia Hand- Iissch, Goidenbergia Scudder, Sageno$tera Handlirsch, Kdenhergia (311 tl16rl and possibly Gegenemene Handlirsch. Genus Stenodictya Brongniart
Sci~ddcria Brongniart,
1885 : 61 ; Brongniart, 1885 : 277 (nomen tdi~m). Scudderia Brongniart, 1890: 5 (nec Scudderia Grote, 1873). Stmodictya Brongniart, 1893 : 383 ; Handlirsch, 1906: 63 ; Handlirsch, 1919: 3 ; Pruvost, 1919: 308; Crampton, 1919: 54; Lameere, 1917: 157; 1,au- rentiaux, 1953: 419; Sharov, 1966: 118.
TJ7pe species: Scudderia Zobata Brongniart, 1890, SD Brongniart, , 7
1893
1 his genus was based originally on two species) Zohata and spinosa, which Brongniart subsequently (1893) and incorrectly merged under one species, lobata.
111 the Commentry coIlections in the Institut in Paris there are 22 specimens not figured or described by previous worlcers. These are ~nostly isolated wings and fragments of wings. Presumably) the flatness of the dictyoneurid wings and the indistinct venation prevented Brongniart and Meunier from making satisfactory obser- vxtions on these specimens. By using glycerin, however, I was able



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

8 Psyche [March
to work out the venation satisfactorily and to add descriptions of tl~ese specimens to the account included here. Since the venation of the Dictyoneuridae is very simplified, as well as homogenous but individually variable, it does not offer much basis for taxonomy. The relative positions of the Rs origin and the A4 division into MA and MP, used as specific and even generic cllaracters by many authors, may vary even within one individual specimen's fore and hind wings (see for example Sienodiciya Zau- rmtiauxi sp. nov.). Also, the number of branches can only be used to a limited extent, as additional twigs are freq~~lently formed by the archedictyon, elements of which may or may not reach the posterior margin. Having rich material for corxparison at my dis- posal, I find that the general outline of the wings, especially the degree of undulation of the posterior margin and the detailed structure of the archedictyon (shape and size of cells, areas occupied by cross veins, presence of twigs) are the same in fore and hind wings and are therefore additional and useful taxonomic characters. In view of the situation noted above, I believe it is not advisable to form new genera in this family unless they are completely obvious and necessary, since such taxa just cannot be satisfactorily sub- stantiated. I have not been able to study Dictyoneuridae in deposits other than that of Commentry but it seems very probable that the total number of genera will diminish after revisional studiesS4 'I'he wings of the Dictyoneuridae are about equal in length or the hind pair may be a little shorter; they are similar in venation and in the undulation of the posterior margin. Fore wing with the anterior margin strongly curved near the base. Precostal strip pronounced; postcostal veins simple, arising from the very ba~e of C, ending on Sc or forming a fork towards C and Sc. Sc terminating beyond midwing. Rs pectinate, originating at about mid-wing ; MA, MP, CuA, CUP usually simple and parallel, sometimes wit11 additional twigs formed by the archedictyon. Anal area selatively broad, with about five anal veins, sometimes forked. Archedictyo~i irregular, occasionally producing twigs. In the costal, sc-r :111(l proximal part of r-m areas there are dense, regular cross veilis) 4h the collections of the Institut, I was not able to find the type specimen of 8. vasscuri Meunier, 1914. Since the specimen was not present in the collection in 1938
(pers. comm., F. M. Carpenter) and since no photograp11 of it exists, I have not includkd this species in my present account. S. minima Brongniart, 1893, is based on a very fragmentary specimen and nince it shows no other characteristic than the small size) I am referring this species to Dictyoneuridae inc. gen.



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

Figure 51. Stenodictya loba,ta (Brongniart) ; specimen 22-1. ~olot~~e-. connected by anastomoses. Hind wing slightly broader) broadest shortly beyond the first quarter of wing length. Body structures : Head small) narrower than prothorax. Eyes projecting) clypeus large. Antennae composed of narrow and long segments, Beak long, with long palpi. Prothoracic lobes large) cordate) with fan-like venation and irregular, anastomosing cross venation. Thoracic segments either equal in length or the prothorax shortest. Legs short, tibiae only slightly prolonged) sometimes pro- vided with spines. Tarsi with long claws. Abdomen slightly longer than half the wings. Posterior margin of terga convexly curved ill the central part. PosterolateraI angles pointed) more or less pro- duced. Lateral parts of terga with oblique ridges. Females with a stout) curved ovipositor and robust cerci. Males with claspers) arising posterolaterally from beneath the 9th tergite and composed of about 12 small segments of equal size) each with a short median carina.
Stenodictyu differs from Dictyoneurda Handlirsch by the short Sc and simple MP. From another related genus, Microdictya



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

10 Psyche [March
Brongniart, it differs in having its wings broadest just beyond the base, narrower in the apical third, and by having the costal area broader, C, Sc and R more curved towards the base, MP and Cul' usually simple, the anal area larger, the hind wing broader and of triangular shape. From all other genera it differs in its simplified venation with unbranched MP and CUP.
Species included in Commentry shales: Stenodictya Zobata (Brongniart, 1885) ; S. spinosa (Brongniart, 1885) ; S. agnita ( Meunier, I 908) ; S. pygnzaea ( Meunier, 19 I I ) ; S. grdissima
( Meunier, 19 I I ) ; S. oustaleti ( Brongniart, I 893 ) ; 5. arnaudi ( Brongniart, I 893) ; S. klebsi ( Meunier, I 908) ; S. laurentiauxi spec. nov.; 8. parisiana spec. nov.
Stenodictya lobata (Brongniart)
Figures 50, 51, 52
Scudderia lobata Brongniart, 1890: pi. 11, fig. 2, 3. Stenodictya lobata Brongniart, 1893: 386, pi. 22, fig. 1 ; Handlirsch, 1906: 64, pi. 8, fig. 20; Handlirsch, 1911: 181, pi. 6, fig. 1 (reconstruction) ; Handlirsch, 1913: 513 (reconstruction) ; Handlirsch, 1921: 129, fig. 54 (reconstruction) ; Handlirsch, 1919: 3 ; Pruvost, 1919: 98, fig. 24; Laurentiaux, 1952: 237 ; Sharov, 1966 : 118, fig. 52 (reconstruction). The type specimen of this species was first figured by Brongniart in 1890 (pi. 2, fig. 2, 3) as Scudderia lobata; in 1893 it was illus- trated with the name Stenodictya lobata (22-1). The specimen which Brongniart figured in his latter paper (pi. 22-2) as lohata
was the one on which he previously (1890) based spinosa. As stated in the footnote on page 386 of the 1893 work, he considered spinosa to be a synonym of lobata.
However, I believe that Brongniart
was in error in this conclusion and I am convinced that spinosa is a distinct species. In all figures, Brongniart showed Zobata (speci- men 22-1) as having the incompletely preserved claspers similar to those of spinosa (specimen 22-2). This is not correct, however, the end of the abdomen on specimen 22-1 being distorted and showing on the left side bases of the cerci and on the right side the base of the ovipositor.
Some contusion has existed in the literature about the lateral portions of the tergites. Lameere (1917 p. 158) correctly noted
that the lateral parts of the tergites have the same surface texture and sclerotization as the median part of the tergites and that they are not separated by any suture from the rest of the tergites. The transverse ridge, running near and parallel to the anterior margin of the abdominal segments, Lameere considered to be a suture



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

19701 Kukalovd - Palaeodictyoptera
Figure 52. Stenodictya lobata (Brongniart) ; specimen 22-1; detail of abdomen.
Figure 53. Stenodictya sp..nosa (Brongniart) ; specimen 22-2; detail of abdomen.