Jarmila Kukalová.
Revisional Study of the Order Palaeodictyoptera in the Upper Carboniferous Shales of Commentry, France. Part II.
Psyche 76:439-486, 1969.

Full text (searchable PDF, 4428K)
Durable link: http://psyche.entclub.org/76/76-439.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.

REVISIONAL STUDY OF
THE ORDER PALAEODICTYOPTERA IN
THE UPPER CARBONIFEROUS SHALES
OF COMMENTRY, FRANCE
PART 111
BY JARMILA KUKALOVA~
Charles University, Prague
An introductory discussion of the Palaeodictyoptera found in the Cornmentry shales and of the collection in the Institut de Palion- tologie in Paris, as well as an account of the background of this investigation, was included in the first part of these studies. The present part deals with the following seven families : Homoiopteridae, Lycocercidae, Graphiptilidae, Breyeriidae, Eugereonidae, Archaemeg- aptilidae and Megaptilidae. Compared with the Spilapteridae, con- sidered in Part I, all of these families are small, consisting of only a few genera, at least from the Commentry shales. However, they show much diversity of structure and indicate the extensive range of wing modifications which occurred in the Palaeodictyoptera, indud- ing the reduction and shortening of the hind wings. The third part of this study will deal with the Dictyoneuridae, which provide us with more information about the body structure in this order of insects.
Family Homoiopteridae HandIirsch
Homoio~teridae Handlirsch, 1906: 91; Lameere, 1917: 102; Handlirsch, 1919: 16; Handlirsch, 1921: 133.
Roechlingiidae Guthorl, 1934: 188; Kukalov5, 1960: 1. Thesoneuridae Carpenter, 1944: 10.
Scepasmidae Haupt, 1949: 42.
Type genus: Homoioptera Brongniart, 1893. This family, as established by Handlirsch, included Hodptera Brongniart, Graphiptiloides Handlirsch ( 1906), Homoeophlebia Handlirsch ( I 906) all from Commentry ; and Anthracentomon Handlirsch ( 1904) from Belgium. Of these genera, Graphiptiloides (= Graphiptilus) is here assigned to the Graphiptilidae; Homoeo- '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 Research from the Society of the Sigma Xi. This study has also been supported 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. 'Alexander Agassiz Lecturer in Zoology, Harvard University, 1969.



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

440 Psyche [December
Nebia has already been synonymized with Homoioptera by Lameere (1917, p. 151) ; and Anthracentomon, on the basis of Handlirsch's figures, is also considered a synonym of Ho7noioptera. The present revisional study of the Commentry material has brought new aspects to the family classification of the Paleodictyop- tera. The family Homoiopteridae now appears to include several genera from localities other than Commentry. Three families, Roech- lingiidae, Scepasmidae and Thesoneuridae, seem to be synonymous with the Homoiopteridae; and the relationships of the Homoiop- teridae with the Graphiptilidae, Breyeriidae and Lycocercidae have become more obvious.
The wings of the Homoiopteridae, as here treated, are the largest known in the Palaeodictyoptera. They are characterized especially by the following four features:
(I) stems of main veins with a
more or less pronounced bend in the basal third of the wings; (2) MA either simple or with very short branches and CuA always with branches (usually short ones) ; (3) CuA and CUP tending to be parallel to each other; and (4) a sclerotized strip with tubercles lying along the costa, just posterior to it. The cross veins are nu- merous, irregular and often connected.
The bending of the main veins basally is present in Homoioptera, Boltopruvostia, and to a lesser extent in Thesoneura; it also occurs in the related family Lycocercidae. MA is simple or has a little fork in Homoioptera (see left wing of the type specimen) and in Thesoneura; it gives rise to very short branches in Boltopruvostia (see B. nigra Kukalovi, 1958). It is simple in the Breyeriidae but has a very short branch in Graphiptilus (type specimens of heeri). The branches on CuA are very short in Homoioptera and Bolto- pmvostia, longer in Thesoneura and completely reduced in Lyco- cercidae. In the Breyeriidae (B. boulei) and in Graphiptilus (G. heeri, specimen 19-12) CuA forms a short branch. The tendency for CuA and CUP to be pa,rallel is a very persistent feature, com- mon in all Homoiopteridae and noticeable also in some Lycocercidae (L. pictus), Graphiptilidae (R habdoptilus) and some Breyeriidae (B. barborae) . The cross venation of Homoiopteridae and related families is very characteristic; it is readily distinguished from the more regular and rarely anastomosed venation of spilapterids and from the denser and usually relatively coarser pattern of the dic- tyoneurids.
From the foregoing account, it can be deduced that within the Homoiopteridae and related groups the branches of MA and CuA



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

were in the process of reduction. The number of branches varies within the several genera of Homoiopteridae, and twigs occasionally occur in related families, in which MA and CuA are generally simple. In my opinion the families Hom~io~teridae, Breyeriidae, Gra~hi~tilidae and Lycocercidae form a phylogenetic unit, in which the Hom~io~teridae represent in many respects the most primitive series. It is to be emphasized that the simple form of CuA and MA does not necessarily represent the more primitive stage, as generally believed. This hypothesis was based by Handlirsch on the assumption that the most primitive Palaeodictvoptera were the Dictyoneuridae. However, this family, except for the archedictyon, is a very advanced one, with venation very specialized and already reduced. Of course, it is very difficult to determine, in our present state of meagre in- formation of the body structures, which of the families has the most primitive features. The geological record seems to be indicating more and more that probably all the larger families of the Paleodictyoptera were present and already well developed at the base of the Upper Carboniferous and that their representatives evolved very little dur- ing the rest of that period. As a rule within the insects, the many morphological features of wing venation do not have equal signif- icance for classification in all families of the orders. The presence or absence of branches of MA and CuA appears to be more stable for the spilapterid group than for the homoiopterid group. A remarkable feature, present to a more or less extent in all homoiopterid specimens I have had occasion to study [Homoioptera, Boltopruvostia, Amousus, Anzetretus and Thesoneura] , is the sclero- tized strip and tubercles strengthening the costal area. It is especially marked in the largest specimens of the family, i.e., in all species of Boltopruvostia and in Homoioptera gigantea. This family includes the largest species of Palaeodi~tyo~tera known. The wings are of nearly equal length, the hind wings being somewhat the broader. In addition to the venational characteristics discussed above, the following should be noted: postcostal area large, with several branches included; Sc long; area of Rs small; MA simple or with short branches; MP with several branches; CuA and CUP parallel to each other; CuA with several short branches directed anteriorly (usually), CUP with several branches. Body structures: head small, with projecting eyes and large clypeus. Prothoracic lobes with radiating veins and many cross veins, often with undulated margins. Legs longer than in Spilapteridae, with elongate tibiae. Abdomen unknown.




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

442 Psyche [December
The family Homoiopteridae differs from the related families Lyco- cercidae, Breyeriidae, and Graphiptilidae in having CuA always branched and in possessing the sclerotized strip posterior to the costa. Only the genus Homoioptera is known from the Commentry shales. The following genera occur in other deposits: Homoioptera Brong- mart (= Anthracetomon Handlirsch, 1go4), Westphalian of Bel- gium ; Boltopruvostia Strand, 1929 (po Boltonia Pruvost, 1919) ( Roechlingia Guthorl, 1934, and Ostrava Kukalovii, 1960)) West- phalian of France, Westphalian C of Germany, Namurian C of Czechoslovakia. The following species appear to belong to the family Homoiopteridae but the status and relationship of the genera estab- lished for them are uncertain : A4ammia alutacea Handlirsch I 906, Scepasma gigas Handlirsch,
1911, Amouzus mazonus Handlirsch,
IQI I, and Ametretus laevis Handlirsch, 191 I, all from the West- phalian of Illinois.
Genu,s Homoioptera Brongniart, 1893
Homoloptera Brongniart, 1893 : 353 ; Agnus, 1902: 259 ; Woodward, 1906: 28; Handlirsch, 1906: 91; Lameere, 1917: 151; Handlirsch, 1919: 16. Homoeophlebia Handlirsch, 1906: 92 ; Handlirsch, 1919: 16. Anthracetomon Handlirsch, 1904: 6; Handlirsch, 190,6: 93. Type species :
Homoioptera woodwardi Brongniart, I 893 (OD). A few years after Brongniart described woodwardi, Agnus (1902) added another species, gigantea; this Handlirsch ( 1906) later made the type of another genus, Homoeophlebia. As pointed out by Lameere ( 191 7), the generic separation of gigantea seemed totally unneces- sary. Meunier ( 1912, p. 5)
added to the taxonomic confusion by
the erection of a new species, gaullei, in another genus Archaeo@tilus, basing it on the counterpart of Agnus' type specimen of g-igantea! Handlirsc'h's hthracetomon, based on latipe'nne from the West- phalian of Belgium, is actually inseparable from Homoioptera. The
following account is based on woodwardi and gigantea. Wings relatively broad, almost identical, the hind pair being only a little broader than the fore; color markings in the form of nu- merous small, rounded spots. Precostal strip present (bordering the costal margin) ; anterior margin convex in the basal third of the wing; Rs with 3-4 branches, often forking; stem of M touching or nearly touching R near the base; M dividing near mid-wing, MA being simple and convex, MP with 3-4 branches; CuA and CUP with short branches, often originating at the same level. Anal
veins 6-8 in number, sometimes forked. Cross veins often connected by numerous anastomoses, branched or forming a loose network.



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




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

444 Psyche [December
Body structures: prothoracic lobes relatively small, high on pro- thorax, their bases relatively near to each other. Fme leg with
strong femur, elongate tibia and narrow tarsus. Homoioptera differs from the related genus Boltopruvostia Strand by having broader and shorter wings, smaller area of the radial sector, bv more distal division of M (shortly before mid-wing), by MA being essentially simple and more convex, by CUP having fewer branches and by having a smaller anal area with less branching of the anal veins.
All cross veins weak. From Thesoneura, Homoiop- tera differs in the more pronounced convex curvature of the ma'n veins in the basal third of the wing and in having fewer branches on CuA.
The species included in the Cornmentry shales are Homoioptera woodwardi Brongniart and IT. gigantea Agnus (= Archaeitzegaptilus gaullei Meunier, obj. syn.). One other species, latipenne, from the Westphalian of Belgium, appears to belong here, as noted above. Homoiofitera woodwardi Brongniart
Figure 29
Homoioptera woodwardi Brongniart, 1893:354, fig. 15, pi. 20, fig. 10; Handfirsch, 1906: 91, pi. 11, fig. 1; Handlirsch, 1921: 134, fig. 61. This species was based by Brongniart on one specimen (20-IO), showing the fore and hind wings, prothoracic lobes and a fragment of fore leg. The wings present a remarkable color pattern of cir- cular dots, a pattern which occurs repeatedly in the families related to the Hom~io~teridae. The shape of the prothoracic lobes probably has little taxonomic value other than at the specific level. Great variability in the shape of the lobes also occurs in the Spilapteridae. My study of the type specimen shows that the cross venation is much denser and is less regular than indicated in Brongniart's figure. Fore wing: length 75 mm, width 27 mm. Wing membrane spotted by circular markings of varying diameters; wing uniformly broad in the proximal half, then abruptly narrowing; apex at about the wing axis; Sc, R and M almost parallel and convex at the end of the first quarter of the wing; M notably concave before the divi- sion into AIA and MP; Rs with 3-4 branches, each forked several times; Rs area very small. Cross veins more simple in the areas of the subcosta, sc-r and r-rs, with much less anastomosis. A cluster of long hairs occurs at the bases of both fore wings. Hind wing: length 75 mm, width 31 mm. The sigmoidal curvature of the main



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

veins near the base is much less pronounced than in the fore wing; anal veins only rarely forked.
Body structures: prothoracic lobes with undulated margin, their veins S-shaped, about 11 in number. Numerous anastomoses of cross veins.
Homoioptera gigantea Agnus
Figure 30
Homoioptera gigantea Agnus, 1902: 259, pi. 1; Lameere, 1917: 151. Homoeophlebia gigantea Handlirsch, 1906: 93, pi. 11, fig. 3. Archaeopt.'lus gaullei Meunier, 1910: 233, fig. 1; Meunier, 1912: 5, pi. 6, fig. 1.
Homoeophlebia gaullei Handlirsch, 1919: 16, fig. 18. This monotypic species was based by Agnus upon a remarkably well preserved, large fore wing, with a single prothoracic lobe and vague outlines of the head, including the base of the beak, and suggestions of the thorax and a fragment of a fore leg. The reverse, as already noted, was described by Meunier (1910) as Archaeoptilus gaullei.
Fore wing: length 187 mm, width 65 mm. Wing membrane spotted with ( I ) extremely dense, small, light markings, irregularly grouped into small clusters; (2) larger spots arranged into 4 trans- verse bands. Wing abruptly narrowed in the apical third, with the apex pointed and falcate. Anterior margin very convex in the basal third, the posterior margin S-shaped in the apical third. Sc, R, M following the convex curvature of the anterior margin. Sc somewhat shortened; subcostal area broad in the proximal half, then very nar- row; stems of R and M touching near the base; M deeply concave before division into MA and MP; Rs with about four very oblique branches, the first of them long and forked. Anal area with about seven branches, the first forked several times. Cross veins parallel and directed in different directions, with much anastomosis. Long hairs are clustered at the base.
Body structures:
prothoracic lobe length 27 mm, width 24 mm, with a complete covering of long hairs. Prothoracic lobe cordate, with seven radiating veins and densely arranged, simple cross veins; margin of lobe not undulated.
This species differs from woodwardi by its larger size, more spe- cialized shape of the wing with its falcate apex, the larger rs area, the contact of the stems of R and M and the smoothly curved margins of the prothoracic lobes, as well as the color pattern of the wings.



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

Figure 30. Homoioptera gigantea Agnus; prothoracic lobe, fore wing. Holotype. Figure 31. Lycocercus $ictus Handlirsch; fore wing. Holotype.



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

19691 Kukalovd - Palaeodictyoptera 447
Family Lycocercidae Handlirsch, I 906
Lycocercidae Handlirsch, 1906 : 675 ; Handlirsch, 1906: 88 ; Handlirsch, 1921 : 133; Handlirsch, 1919: 15; Lameere, 1917: 102. Polycreagridae Handlirsch, 1906: 110 ; Handlirsch, 1906 : 678 ; Handlirsch, 1921: 137.
Apopappidae Lameere, 1917 : 42.
Patteiskyidae Laurentiaux, 1958: 302 ; Demoulin, 1958 : 363. Type genus :
Lycocercus Handlirsch, I 906
The family Lycocercidae was established by Handlirsch for Lyco- cercus and was characterized as having more numerous branches than Lithomanteidae3 and as having cross veins forming at least partially a dense network of the dictyoneurid type. His interpretation of the fossils on which Lycocercus was based is only partly correct. The cross veins of the Lycocercidae are indeed denser, more irregular and more often connected by anastomoses than in Lithomanteidae but they do not form a real network of the dictyoneurid type. This revisional study of the type material has revealed three additional and important features for the Lycocercidae : ( I ) the hind wing is as long as but narrower than the fore wing; (2) the MP area is of triangular shape, with many branches; (3) the origins of MA and the first fork of MP are very close. The following families are considered by me to be synonymous with Lycocercidae :
( I ) Polycreagridae Handlirsch, I 906, based on the single genus Polycreagra Handlirsch, 1906; this genus differs from Lycocercus only by having more obliquely oriented branches, with longer forks and by having Rs more richly branched. (2) Apo- pappidae Lameere, I 9 I 7, based upon Apopappus Handlirsch, I 906, which differs from Lycocercus by having a more regular cross vena- tion and by having CUP somewhat richer in branches. (3) Pat- teislcyidae Laurentiaux, 1958, based upon the oldest palaeodictyop- teron so far known, Patteiskya houckaerti4 (Namurian B, Germany), which differs from the other species of Lycocercus only in having M dividing more proximately; it is therefore inseparable from Ly cocercus.
Wings about equal in length, hind pair narrower, similar in vena- 'The family Lithomanteidae is here understood to include the genus Lithomantis Woodward (syn. Hadroneuna Handlirsch and Lithosialis Scud- der), Macroptera Laurentiaux (syn. Lusiella Laurentiaux and Texeira) and Synarmoge Handlirsch.
?In the figure published by Laurentiaux (1958, figure 1) Sc is correctly drawn but the subcostal area is broader proximally and is shaped as in Lycocercus goldenbergi.
In the figure published by Demoulin (1958, p. 359, fig. 1)
the vein designated as the costa is actually the subcosta. Lycocercus bouckaerti is not represented by a hind wing but by a fore wing.



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

448 Psyche
[December
tion; wing membrane usually dark, with light bands and spots. Fore wings resembling hind wings in outline but with broader subcostal area. Anterior margin nearly straight, often with precostal strip; stems of main veins convex in the basal third, as in Homoiopteridae; Sc long; MA simple, curved, originating near the first fork of MP; MP with many branches, occupying a markedly triangular area; CuA simple and curved; CUP parallel to CuA, but with several branches. Anal area often large, with many branches. Cross veins fine, numerous, more or less irregular, sometimes connected. Body structures: head small with conspicuous eyes and large clypeus. Prothoracic lobes with fan-like veins and many cross veins, often with undulated margins. Thorax broad, metathorax shorter than mesothorax; legs homonornous, not very short. Ovipositor not
sculptured, stout and curved. Cerci robust, long. The family Lycocercidae is related to the Homoiopteridae in the characteristic curvature of stems of main veins near the base, in the late division of M, in the general character of cross veins, in the presence of many small forks along the posterior margin, etc. It differs from the Homoiopteridae in the narrowed hind wing and more richly branched MP and CUP and simple CuA. Another related family is the Graphiptilidae, from which the Lycocercidae differ in the richly branched MP and CUP and the narrower hind wings.
Genera included in the Cornmentry shales : Lycocercus Handlirsch and Apopappus Handlirsch.
Occurrence in other deposits: Lycocercus Handlirsch, Namurian B, Germany.
Genus Lycocercus Handlirsch
Lycocercus Handlirsch, 1906: 89, Handlirsch, 1919: 15; Lameere, 1917: 153. Patteiskya Laurentiaux, 1958 : 302 ; Demoulin, 1958 : 360. Type species : Dicty oneura goldenbergi Brongniart, I 893 ( SD Handlirsch I 922).
This genus was erected by Handlirsch for goldenbergi Brongniart, as represented by specimen 2 I - I. Specimen 2 I -2, correctly described by Brongniart under the same specific name, was erroneously re- ferred by Handlirsch (1906, p. 90) to a separate species brongniarti (Lameere, 1917, p. 153). Meunier (191 I, p. 121) described Homoi- optera brongniarti, which Handlirsch later ( I g I 9, p. I 5 ) recognized as Lycocercus. The specific name having been preoccupied within the genus since 1906, Handlirsch ( 1919, p. 16) changed it to Lycocercus pictus. Though Lameere (1917, p. 153) believed that pictus was conspecific with goldenbergi, he was apparently incorrect.



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




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

4.50 Psyche [December
Figure 33. Lycoccrcus goldcnbergi (Brongniart) ; specimen 21-1 ; fore and hind wings. Holotype.
Figure 34.
Apopappus guernei (Brongniart) ; specimen 19-3 ; fore wing. Holotype.




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

19691 KukaZovd -- Palaeodictyoptera 45 1 Although the differences in venation might conceivably be considered as due to individual variability and although the cross venation and color markings are almost identical, there is a marked difference in the length of the legs and probably also of the beak. In Lycocercus the hind wings are like the fore wings in shape but are narrower. This assertion is based on specimen 21-2, in which the wings are preserved in their natural positions. If found isolated, a fore or hind wing can be recognized only by the width of the proximal part of the subcostal area, which is broader in the fore wings.
Fore wings unusually broad in the proximal half, shaped as hind wings. Hind wings similar but somewhat narrower. RI simple; Rs with six branches, first of them forked; MP forked 10-20 times. Number of CUP branches variable; about 8 anal veins, mostly forked. Cross veins dense, irregular, often connected. Body structures:
beak short or long. Legs stout but not very short.
Lycocercus differs from Apopappus in less regular anastomosing of cross veins, smaller CUP area and in the larger number of short branches of Rs and CUP.
As a whole, the venation of Lycocercus
is much less regular.
From Polycreagra Handlirsch (Westphalian, Illinois) it differs in the less densely branched and less obliquely oriented branches of main veins.
Species included in the Commentry shales : Lycocercus goldenbergi Brongniart, 1893 ; Lycocercus pictus Handlirsch, 191 9. Occurrences in other deposits: Lycocercus bouckaerti (Lauren- tiaux, 1958) of Namurian B, Germany.
Lycocercus goldenbergi (Brongnkrt)
Figures 32, 33, 35
Dictyoneura goldenbergi Brongniart, 1883: 265 (for additional references see Handlirsch, 1922: 39).
Lithornantis goldenbergi, Brongniart, 1893 : 369, pi. 21, fig. 1, 2. Lycocercus goldenbergi, Handlirsch, 1906: 89, pi. 10, fig. 20; Handlirsch, 1921: 138, fig. 60; Handlirsch, 1919: 15, 16; Lameere, 1917: 101; Lameere, 1917: 153 ; Demoulin, 1960: 1-4, pi. 1. Lycocercus brongniarti Handlirsch, 1906: 90, pi. 10, fig. 21; Handlirsch, 1919: 15, fig. 17.
This species was based by Brongniart on specimen 21-1, one of the most remarkable Palaeodictyoptera known, and on specimen 21-2, a fragment of fore and hind wing in natural positions. Handlirsch ( 1922) designated specimen 21-1 as the type; specimen 21-2, which is important for showing the wing shapes, was referred by Hand-



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

452 Psyche [December
lirsch to a separate species, brongniarti, but was identified again as gddenbergi by Lameere ( 1917, p. I 53).
The type specimen (21-1) has been discussed many times by various authors but of these only Brongniart and Lameere actually studied the fossil. Most interpretations are highly speculative and not worth discussing here. The fossil shows so many structures which are important for the whole order that it deserves the most detailed study. Actually, the specimen might contribute even more details than I was able to work out in my limited stay of several