C Stuttgart, Juni 1980 Paläont. 2. Palaeoisopus, Palaeopantopus and Palaeothea, pycnogonid arthropods from the Lower Devonian Hunsrück Slate, West Germany }AN BERGSTRÖM, WrLHELM STÜRMERand GERHARD WINTER':· With 34 figures in the text Kurzfassung: Palaeoisopus problematicus und Palaeopantopus maucheri werden neu beschrieben und als urtümliche Pycnogoniden aufgefaßt, die ausgestorbenen Ordnungen angehören. Palaeoisopus ist besonders ursprünglich hinsichtlich des langen Abdomens mit einem Telson. Er unterscheidet sich von anderen Formen durch abgeflachte Extremitäten, die auf eine schwimmende Lebensweise hindeuten, und durch anders angeordnete Ocellen. Palaeothea devonica nov. gen. et nov. spec. ist ein winziger Pycnogonide, der einzige bisher bekannte fossile Vertreter der rezenten Ordnung Pantopoda, der darauf hinweist, daß im frühen Devon durchaus schon moderne Formen vorhanden waren. Eine Abstammung der Pycnogonida von frühen Merostomen wird als wahrscheinlich betrachtet. Abstract: Palaeoisopus problematicus andPalaeopantopus mauo/_eri are redescribed and found tobe primitive pycnogonids belanging to extinct orders. Palaeoisopus is particularly primitive in retaining a long abdomen with a telson. lt differs from other forms also in having f!attened limbs indicating a swimming mode of life andin having the ocelli differently arranged. Palaeothea devoriica nov. gen. et nov. spec. is a tiny pycnogonid, the only hitherto known fossil representative of the extant order Pantopoda, indicating that entirely modern forms were in existence already in the Early Devonian. A derivation of pycnogonids from early merostomes is regarded as likely. Contents lntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Earlier work onPalaeoisopus andPalaeopantopus.... .. . . . . . . . . . .. . ...... .. . . .. . .. . ..... ... ... . . . .. .. . . . ... . . .. . . .. . Glossary_..................................................................................................................... I. Palaeoisopus problematicus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cephalosoma ........... .. ..... ........ ................ ................ ........... .................................. ｾｧＺｸ､ＭｳＮ＠ II. III. IV. V. ｾ＠ ｾＭ＠ 8 8 8 10 10 ::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: ｩｾ＠ Abdomen ....................................................................................... ,.................... 27 Function and ecology .. . . . . . . .. . . . . . . .. . . . . . . .. . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . ... . . . . . . . 32 Palaeopantopus maucheri . . . . . . .. .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . ... . . . . .. .. . . . . . ... . . . . . . . . .. .. . . . . 33 Cephalosoma.... .. . . . . . . ... .. . . . . . . . . . . . . .. .. . . . . . . .. .. . . . . .. .. . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . ... .. . . . . . . 37 Thorax............................................................................................................... 37 Abdomen............................................................................................................ 41 Interna! structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Function and ecology . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Palaeothea devonica nov. gen. et nov. spec. . . . . . .. . . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . .. . . . . . 41 Cephalosoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Thorax............................................................................................................... 44 Discussion... .............. ........ ....................................................... ................. .......... 45 Other arthropods described as fossil pycnogonids . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . .. . . . . . ... .. .. . . . . 46 Relationshipsand phylogeny within the Pycnogonida .. . . . . . .. . . . . . . .. . . . . . . .. . . .. .. .. . . . . . . .. . ........... 47 '' Addresses of the authors: Dr. ]AN BERGSTRÖM, Sveriges geologiska undersökning, Tunavägen 35, S-22363 Lund, Sweden; Prof. Dr. Dr. h. c. WILHELM STÜRMER, Burgbergstraße20, D-8520 Erlangen, West Germany; Dr. G ERHARD WINTER, I. Zoological Institute, U niversity of Erlangen- N ürnberg, Universitätsstraße 19, D-8520 Erlangen, West Germany. 0031-0220/80/0054-0007 $ 12.00 © 1980 E. Schweizerbart'sche Verlagsbuchhandlung, D-7000 Stuttgart 1 8 VI. VII. VIII. IX. J. Bergstr6m, W. Stiirmer and G. Winter Systematicposition of the Pycnogonida ....................................................................... Acknowledgment .................................................................................................. List of specimens ................................................................................................... Literature ............................................................................................................ 48 51 52 53 Introduction This study is a continuation of a series of X-ray studies of Early Devonian arthropods from the wellknown "Hunsriickschiefer" (STORMER& BERGSTR6M 1973, 1976, 1978). The depositional environment, the mode of preservation and the X-ray method were treated in the first report. Part of the radiographs was copied with an image processing machine, the "Logetronograph". This electronic device is reducing the often too high contrast of the radiograph (caused by heavy variations in the absorption owing to severe mechanical preparation or by varying thickness of the specimen) without reducing detail contrast. All the arthropods studied so far are exposed on bedding surfaces. It was, therefore, tempting to regard the argillaceous rock as a shale, and the name of the formation was translated as "Hunsriick Shale". However, there certainly is strong evidence for tectonic deformation and secondary fissility, although the latter in cases tends to be parallel with the bedding planes. We, therefore, now prefer to regard the rock as a slate, a term which corresponds to the German term "Dachschiefer", which is applied to the "Hunsriickschiefer". We are grateful to Professor G. SOLLE, Darmstadt, for making us aware of this terminological question. Two pycnogonid species have long been known from the Hunsriick Slate. These forms are of unique construction and quite different from modern pycnogonids. A single small specimen of a third form was found during the X-ray survey by the senior author. Although this specimen is not preparable and somewhat indistinct in the radiograph, it is quite obvious that it is of modern design. Despite the somewhat fragmentary knowledge, we have found it practical to erect a new genus and species for this form: Palaeothea devonica. Earlier work o n Palaeoisopus and Palaeopantopus Palaeoisopusproblernaticus was studied, among others, by BROILI (1928; 1932a, b; 1933), D u B m m (1957), HEDGPETH (In: MOORE 1955; 1978), HELFER (1932), LEHMANN(1959), MOLL~R (1965), STORMER(1944) and TIEGS &MANTON (1958). Initially, the long segmented abdomen was regarded as a proboscis, a mistake that was corrected by DUBININ (1957) and by LEHMANN(1959) after an earlier personal communication by Dr. MOHE, Kirn, who owns the specimen described by LEHMANN(WS 4779). Palaeopantopus maucheri was studied, among others, by BROILI (1929; 1930), DUBININ (1957) and HEDGPETH (In: MOORE 1955; 1978). Glossary Morphological terms abdomen, AB - in pycnogonids posteriormost body tagma, following upon thorax and trunk resp; including tdson, where this is present (see Fig. 34). ambulatory appendage - appendage used for locomotion, such as a walking or swimming leg. annulation - "false segmentation", formation of rings; characterized by rings (= annuli) without joints and internal repetitive structures typical for true segmentation (cf. podomeres). antimere- such three longitudinal elements of the proboscis, each terminating in a lip, as are indicated by the arrangement of the muscles and the triradial symmetry of the pharynx. Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 9 appendage generally any large projection from the body of an animal; specifically in arthropo~is one pair per segment, such as legs, antennae, mouthparts. cephalosoma, CS - typical body-forepart of pycnogonids with cheliphore, palp, oviger, 1. ambulatory leg and eye tubercle; corresponds to the head of euarthropods. chela - pincers of an appendage formed by a rigid process of the penultimate podomere and a movable last podomere. cheliphore, CH - anteriormost appendage in pycnogonids; commonly chelate, may be reduced or absent; used since chelicere might imply an unproved relationship with the Arachnida. exite- laterally directed ramus of one of the proximal podomeres of an appendage; not existing in pycnogonids. eye tubercle, ET - dorsal cephalosomatic dome carrying median eyes. head - used in pycnogonids (classical view) for the anterior part of the body with cheliphore, palp and oviger; separated from the trunk by an imaginary border marked through the eye tubercle (see Fig. 33). lateral extension, LE - lateral outgrowth or process from which an appendage originates. leg, L - 1L, 2L etc. used in text and figures to designate first and second ambulatory leg etc. Succeeding numbers (1L1, 1L2 etc.) designate number of podomere counted from proximal end. oviger, OV - third appendage in pycnogonids; adapted in the male for carrying the eggs during incubation; may be reduced or absent in females. palp, P - second appendage in pycnogonids; may be reduced or absent. podomere - individual segment of an appendage, separated from neighbouring podomeres by joints. proboscis, PB - movable food-gathering organ with distal mouth, is proximally attached to the anterior end of the cephalosoma by an arthrodial membran. seta - hairlike process of cuticule with which it is articulated. telson, T - postsegmental portion of body developed behind the anus. thorax, TH - in pycnogonids body tagma with three segments (TH1-TH3) between cephatosoma and abdomen. trunk - in pycnogonids (classical view) body tagma between head and abdomen, comprising all segments with ambulatory legs (see Fig. 33). - Palaeoisopus ~ 8 ~ ~I~]~~-8-~'9-~ Palaeopantopus ILE Palaeothea Pantopoda 1L 2-4 L 10 J. Bergstr6m, W. Stiirmer and G. Winter Phylogenetic terms apomorphic character- a derived stage of expression of this character in comparison with a more ancestral stage of expression (,~abgeleiteteAuspr~igungsform eines Merkmals,,). cladistic - refers to the point of view held by some scientists that phylogeny is no less and no more than bifurcation of evolutionary lineages, grouping and ranking are given by the branching points, cf. "sister groups". plesiomorphic character - a primitive and ancestral stage of expression of this character resp., which remained unchanged during a number of speciation processes (,,ursprfingliche Auspr~igungsform eines Merkmals~,). shared derived character- character found in different species or groups due to the common descent from an ancestor in which the character first and finally evolved; the character indicates relationship. shared primitive character - character found in different species or groups due to the retention of it from forms ancestral to the latest common ancestor; the character is not conclusive in recognizing the offspring of the mentioned common ancestor. sister groups - in cladistic phylogeny theory the bifurcation in evolution splits one ancestral group into two groups, which together form a monophyletic group of higher rank and are, therefore, to be termed sister-groups. symplesiomorphic character - see "shared primitive character". synapomorphic character - see "shared derived character". I. Palaeoisopus problematicus Order Palaeoisopoda HEDGVETH, 1978 (transscr. herein, ex suborder Palaeoisopodina HEDGPnTH, 1978) Pycnogonids with long, 5-segmented, limbless abdomen; flattened legs with annulated bases. Type and only included family is Palaeoisopodidae. Family Palaeoisopodidae DUUININ, 1957 (=Family Palaeoisopodidae HEDGPETH, 1978) Characters of order Palaeoisopoda. Type and only included genus is Palaeoisopus. Palaeoisopus BROILI, 1928 Palaeoisopus problernaticus BROILI, 1928 Palaeoisopusproblematicus BROILI is comparatively well known due to the excellent study by LEHMANN (1959). The present study is based on a much more comprehensive material and radiographs of 57 specimens. Most of LEHMANN'S observations can be verified. Palaeoisopus problernaticus is a large arthropod, reaching a body length of at least 125 mm, exclusive of the proboscis, and a transverse span of some 320 mm. As in modern pycnogonids, it is possible to distinguish between a cephalosoma (Fig. 5) and thorax with appendages and a limbless abdomen. The cephalosoma bears the eye tubercle and anteriormost the proboscis. Fixed to the cephalosoma are cheliphore, palp, oviger and first ambulatory appendage which is quite different from the following three, uniformly built thoracic appendages. There is no hint to the existence of a distinct head. The abdomen is 5-segmented. The whole body is cylindrical to triangular in cross section with cylindrical lateral extensions serving as bases for the ambulatory appendages. Cephalosoma The cephalosoma is approximately as wide as long, not including the extending leg bases. Close to the transverse anterior margin there is a large dorsal eye tubercle. There is no ornament except for a line of large rounded tubercles along the posterior margin and tuberculated rings on the lateral extensions (see below). Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany Fig. 1. Reconstruction ofPalaeoisopus problernaticus. A: Ventral, B: frontal, and C" dorsal view. 11 12 J. Bergstr6m, W. Stfirmer and G. Winter Fig. 2. Radiograph ofPalaeoisopus, WS 3837. The arrangement of ocelli (median eyes) on the tubercle is clearly seen in radiographs WS 2813, 3836 (Fig. 11), 4779 and 10128 (Fig. 9), and a surface view is obtained from WS 4745 and 4779 (Fig. 16). At the top there is a fairly closely placed pair of ocelli which may appear larger than the others. In front of this pair there is a median ocellus. These three ocelli are arranged as the corners in a equilateral triangle. Behind the top pair and at slightly larger distance is a fourth ocellus in the median line. There is no indication of any "top eye" (Gipfelauge). The arrangement of one pair of large ocelli (and two smaller ocelli situated medially before and behind) is interesting for this reason: all studied chelicerates also have one pair of median eyes (-- frontal ocelli) whereas extant pantopods possess two pairs of them one arranged behind the other. The proboscis (Fig. 8) is commonly folded back and is then invisible from the dorsal side. When this is the case, it may extend approximately to the boundary between the first and the second natatory limb segments - the apparent length varies from one individual to another. The outline is rather sac-like (WS 592, 2698, 2813, 3837, 3317, 10129). A specimen skilfully prepared by Mr. GONTHER B RASSELof Flensburg (WS 10352, Fig. 15) shows the ornament of the anterior surface in great detail. There is a median field bordered by a pair of lateral fields. These fields apparently may represent the three antimeres of the proboscis. The boundaries are clear-cut straight lines marked by small granules, and the fields differ from each other through their distinctive pattern. The median field has a fine and fairly dense transverse striation making the field look annulated. In addition, there are scattered granules. Each lateral field is separated longitudinally by transverse lines into some five segments. At the tip, there is a wedgeshaped split between the median field and the lateral fields (well visible on one side, and also on other specimens) Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 13 Fig. 3. Compound dorsal and ventral view of Palaeoisopus (reconstruction) illustrating morphology, terminology, and abbreviations used in the text. Thoracic legs turned 90~. For abbreviations see also glossary. as in extant pantopods. There is no trace of any star-shaped termination such as reported by LEHMA~N (1959: 100--101). Appendages of cephalosoma The cheiiphores (Fig. 4) are large structures visible in most specimens. Still, due to wrinkling and other types of deformation, it is difficult to make out segmentation and articulation in most specimens. The cheliphore may consist of five podomeres, of which three form the shaft and two the chela. At the proximal end of podomere CH1 (cheliphore podomere 1, see Fig. 3) there is a fairly strongly sclerotized ring (well visible in WS 215, 2268 (Fig. 6a), 2813, 10129), which may indicate a sixth podomere but perhaps more likely represents a thickening of the margin at the b o d y - podomere CH1 junction. Podomere CH1 is comparatively long, the length approximately equalling the width. Podomeres CH2 and CH3 are about half as long. C H 1 of both sides are invariably held close together in the midline, as if they were fused. The shaft (CH 1-3) is directed forwards. Podomeres CH4 and CH5 form the chela. CH4 is long and pointed, with the distal tip pointing ventromedially. CH5 is attached about halfway between the ends on the posterior side of CH4. Also CH5 is acutely pointed. 14 J. Bergstr6m, W. Stiirmer and G. Winter Fig. 4. Cheliphores of Palaeoisopus drawn to scale, a) WS 2268, b) 10127, c) 10129, d) 215, e) 4749, f) 3837 (Scale 1 cm). Fig. 5. Scanning electron micrograph of extant pantopod Nymphon rubrum showing the cephalosoma with appendages, proboscis, and eye tubercle. Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany Fig. 6. Cheliphores of Palaeoisopus. a) WS 2268, b) 4744 (Scale 1 cm). Fig. 7. Palp of (Scale 1 cm). Palaeoisopus. WS 4779 15 16 J. Bergstr6m, W. Stiirmer and G. Winter Fig. 8. Palps, ovigers, and proboscis of Palaeoisopus drawn to scale, a) WS 2313, b) 4417, c) 4632, d) and e) 4779, f) 10129 (Scale 1 cm). Fig. 9a Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 17 Palps and ovigers (Fig. 8) are generally difficult to distinguish from each other due to the similar construction and commonly poor preservation. Both pairs can be observed together in a number of specimens. In many others there are clear traces of only one pair of limbs between cheliphores and legs, and it may be questioned if this is due to preservational bias or to the absence of one pair in one form. In the latter case it may either be the question of sexual dimorphism or of two distinct species. The question is returned to in the discussion of the first leg. The palp seems to be preserved in WS 215, 2808, 2813, 2814, 4417, 4744 (Fig. 17), 4779 (Fig. 7), 10129, and 10352. It tends to extend ventrolaterally. The number ofpodomeres is not determinable with full confidence but seems to be nine. In WS 4779 (Fig. 16) there is a series of five cylindrical but irregular podomeres (P 1-5) distal to an extended thickened margin, which is interpreted as the lateral body extension. Each podomere has a distal thickened and tuberculated ring, which is irregular in shape. The same number of proximal cylindrical podomeres can be counted in the right palp of WS 10127 (Fig. 19). On the medial side of P6 there is a large protuberance, against which the distal part of the limb is curved. This distal part consists of three cylindrical but curved podomeres, P7-9. P9 seems to have a rounded tip, possibly flattened on two sides. There is no evidence of any distal spine. The oviger appears to be present in WS 215?, 2808, 2813, 3837, 3317, 4633, 4742, 4744 (Fig. 17), 10127, 10352 (Fig. 15). This appendage tends to hang down more or less vertically. It is slightly larger and the distal part is more smoothly curved than in the palp. The number of podomeres is ascertainable only for the distal half but appears to be about 11 (WS 10127, Fig. 19). Assuming that this is the correct number, there is a proximal straight series of four (?) podomeres, OV1-4. The two succeeding podomeres, OV5-6, curve outward, while a distal series of five podomeres, OV7-11, curve medially in a manner reminding of P7-9 of the palp. As in the palp, the most distal podomere is rounded in the end, possibly flattened from two sides. Fig. 9. a) Radiograph and b) explanatory drawing ofPalaeoisopus, WS 10128 (Scale 1 cm). 2 Pa|~iont.Z. 54 18 J. Bergstr6m, W. Stiirmer and G. Winter In WS 4417 (Fig. 8b) there is a spherical structure at only one of the ovigers. Although it is fairly diffuse it is tempting to interpret it as an egg ball (EB) carried by the oviger as in living pantopods. The interpretation gains some support by the find of a similar spherical object in the neighbourhood of one of the ovigers in WS 2808 and the presence of a similar detached sphere in the neighbourhood of specimen WS 10128 (Fig. 9). At the base of the first pair of legs, 1L, there are five short tuberculated rings. Three interpretations are possible. First, this portion may be a ringed lateral extension of the body (as apparently inPalaeopantopus). Second, the four outer rings may represent the basalmost leg podomere, subdivided into annuli (or simply ornamented with rings). Third, the four outer rings may represent four separate podomeres. This portion is generally held horizontally and straight. However, there seem to be some variations. For instance in WS 4632 this portion is clearly curved on one side of the body, indicating at least some flexibility. There also seems to be a differential telescoping between the two sides. As in the consecutive body segments, the most proximal tubercular ring is stronger than the others. In WS 2814 the ringed portion of both sides is bent strongly backward. This flexibility indicates that the ringed portion, except for the proximal ring, is truly articulated and belongs to the leg, not to the body. Although no muscles have been observed, the extreme shortness of the rings indicates that they are not true podomeres but perhaps annuli. These circumstances favour the interpretation of the proximal ring as representing the edge of the lateral body extension and the four outer rings as collectively representing the first podomere. However, if 1L1 is interpretated as an annulated part of the lateral extensions the total number of succeeding podomeres 1L2-1 L8 would agree with that in extant pantopods (=L1-L7) representing coxa (1L2), trochanter 1 and 2 (1L3-4), femur (1L5), patella (1L6), tibia (1L7), and tarsus (1L8) with a terminal claw. The second podomere, 1L2, is cylindrical with the length exceeding the width. The anterior side of it has a strong longitudinal ridge. Strengthening ridges are also present on the anterodorsal and posterodorsal surfaces. The latter ridges extend to the ends of a ridged emargination in the distal margin. All these ridges are covered by coarse granules, which at least in the case of the anterior ridge carry setae (WS 10129). The articulation against 1L3 appears to be in two points at the ends of the two latter ridges. 1L3 is cylindrical but shorter than 1L2, the length being approximately equal to the width. There is one longitudinal ridge with setae corresponding to the anterior ridge of 1L2, and a distal ridge with granules. Obviously there are two articulation points, with a position closely comparable with that in 1L2. The 1L4 is of quite different construction. It is of"normal" width proximally but terminates distally in a blunt point. There is one anterior longitudinal ridge with setae, extending to the blunt distal point. The articulation against 1L5 is a narrow oval area (Fig. 11) on the dorsal side extending from the distal point proximally and slightly posteriorly almost to the posterior point of articulation between 1L3 and 1L4. The distal podomeres differ from the proximal ones in being flattened in cross section. As the two sides are almost invariably facing upward and downward, respectively (in contrast to the condition in the successive legs), they may be termed dorsal and ventral surfaces. In life the orientation could probably be adjusted with the aid of the articulation between podomeres 1L4 and 1L5. Podomere 1L5 is comparatively long and pointed in the proximal end. The posterior surface has a line of 55-60 pointed setae with a maximum length of 3-4 mm in a large specimen (Fig. 11). Each seta is attached to an elevated base and has a swollen base. At each of the upper and lower edges of the posterior surface, there is a line of fine granules. The narrow anterior edge has some 15-20 elevated granules which appear to each have one socket for a seta (WS 4779). Distally the edge bulges slightly in a lobe which is indistinctly set off from the rest, and extends to the dorsal side of the articulation with the next distal podomere. A similar edge extends anteriorly from the articulation point on the ventral side but disappears before it merges with the anterior edge. The ventral and dorsal surfaces are almost smooth. There is an indication in some specimens both in surface view and in radiographs that there is a slightly thickened and elevated Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 19 Fig. I0. First ambulatory leg, 1L, of Palaeoisopus drawn to scale, a) WS 547, b) 4417, c) 3836, d) 4779, e) 4742 (Scale 1 cm). longitudinal band in the middle of the dorsal surface. Distally the articulation between this podomere and 1L6 has a vertical axis through one dorsal and one ventral point of articulation. These points are comparatively close to the posterior margin. They may be ball and socket articulations, with the socket situated in the proximal podomere. Podomeres 1L6-8 are basically similar to 1L5 and differ only in details. The number of setae on the posterior edge is around 50 in 1L6 and 1L8 and around 35 in 1L7, which is somewhat shorter than the others (WS 4779). In 1L8 the edge extending from the distal ventral point of articulation does not disappear but extends to and merges with the anterior edge. Both edge branches carry some 5-7 setae. A number of specimens do not clearly show two edges with setae, and it may look as if there were only one setiferous edge (WS 2808, 2813, 3836 (Fig. 11), 4416 (Fig. 13a), 4417, 4745, 4779 (Fig. 13b), 10127 (Fig. 19), etc). In some cases, as in WS 3836 (Fig. 11), a close examination shows a pairwise organization of the setae that may indeed indicate that every second is situated on a separate edge. An alternative would be that there are two morphological types, one with one file of setae, another with two files. This may either be a species characteristic or sexual dimorphism. A cross-check with palps and ovigers indicates that both appendages are present in some specimens with apparent one file arrangement (WS 2808, 4417, 10127, and 10352) and in one which in all probability has a two file arrangement (WS 2813). Other specimens with either palp or oviger visible show according to our interpretation, either one more or 20 J. Bergstr6m, W. Stiirmer and G. Winter Fig. 11. Stereo-radiographs ofPalaeoisopus, WS 3836. Note structure and articulation particularly of basal part of first ambulatory leg and the ventral arc of the flexed swimming leg, 2L (Scale 1 cm). Pycnogonid arthropods from the Lower Devonian Hunsr~ck Slate, West Germany 21 22 J. Bergstr6m, W. Stiirmer and G. Winter Fig. 13. Radiographs of the distal podomere, 1L8, and claw of the first ambulatory leg inPalaeoisopus. Note variation in shape and presence of two files of tubercles with setae along the front margin, a) WS 4416, b) 4779 (Scale 1 cm). Fig. 14. Radiograph of distal podomere, 1L8, and claw of Palaeoisopus to demonstrate the morphology, particularly the two rows of tubercles along the front margin, and the streaks of the muscles moving the claw; WS 4742 (see also Fig. 12) (Scale 1 cm). Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 23 less distinct file or two files of setae, all four possible combinations being represented. A possible sexual dimorphism therefore does not seem to be as obvious as the combination of a loss of one pair of appendages and the presence of one or the other type of podomere 1L8. The setae of 1L8 are well visible in several specimens (WS 3836, 3837, 10127, 10129, 10133). However, other specimens with nicely preserved domes (with a central lumen), which may have served as bases for setae, show no clear traces of setae (WS 4742 (Fig. 14), 4744). The outline of podomeres 1L5-8 varies fairly much from narrow to broad types and from types which are widest at the base and to those widest distally. Some of this variation is due to distortion, but some of it may be due to original biological variation. On the dorsal side, a strengthening ridge extends proximally from the distal articulation. The distal hooked claw appears to have a double ball and socket articulation similar to that between the blade-shaped podomeres. The shape of the hook as preserved varies from needlelike structures with an acute point to blunter structures, which in some cases carry spines or setae (WS 4779, Fig. 13b). Fig. 15a Fig. 15b Fig. 15c Fig. 15. a) Surface exposure, b) radiograph, and c) explanatory drawing ofPalaeoisopus, WS 10352, to illustrate the ventral morphology. Note the fine ornamentation of the proboscis and the articulated setae of the thoracic legs, 2-4L (Scale I cm). Fig. 16. Surface exposure of dorsal side ofPalaeoisopus, WS 4779 (Scale 1 cm). 0 oQ 0 m 0 0 0 F 0 a m Fig. 17a) Fig. 17. a) Radiograph and b) explanatory drawing ofPalaeoisopus, WS 4744 (Scale 1 cm). Fig. 17b 26 J. Bergstr6m, W. 8tiirmer and G. Winter Fig. 18. Surface exposure of dorsal side ofPalaeoisopus, WS 11471 (Scale 1 cm). Thorax The thorax segments differ in many ways from the first leg-carrying segment, treated above under the cephalosoma. This is true both with respect to the segments of the body and to the three pairs of legs. The thorax segments are cylindrical. On the dorsal side, where they appear to be somewhat flattened, they are terminated posteriorly with a row of tubercles, and in TH1-2 there are a few large tubercles also in the central dorsal area (Fig. 16, 18). The ventral side is Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 27 evenly convex, without ornament. The lateral extensions are slenderer than in 1L, and the marginal thickened ring is not vertical but inclined so that the first podomere of the leg extends ventrolaterally, in 4L almost directly ventrally. As in 1L the portion of the leg interpreted as the first podomere consists of short rings, even shorter and narrower than in 1L but with similar tuberculation. The tings number 3 in 2L, 2 in 3L, and 1 in 4L (e.g. WS 2814, 4779 (Fig. 16), 10128 (Fig. 9). In WS 2814 there is an extra partiat ring incipiently indicated on the right side of the body in 3L and 4L. As in 1L, there is a series of cylindrical proximal podomeres and flattened distal ones. However, this is virtually the end of the similarity, as both number, morphology and orientation is different. Podomeres 2-4L2 are long and slender, much longer than in 1L, being about twice the length in 2L. They are cylindrical, slightly widening at the distal end, and strengthened by four longitudinal tuberculated ridges, which are roughly evenly spaced. Podomeres 2-4L3 are short ring-shaped sclerites with strengthening ridges as in 2-4L2. Podomeres 2-4L4 are suggestive of 1L4 in almost wedging out on one side, which is considered as dorsal. Podomeres 2-4L5-9 are flattened, very thin elements of decreasing length. Articulating membranes are distinctly visible between each two podomeres. As in 1L, they are probably partially sclerotized. The whole series of flattened podomeres in each leg tends to be carried with the flat sides facing forward and backward. Consequently, and contrary to the condition in 1L, the flat sides are regarded as anterior and posterior. The whole series forms an arch with the concave side downward. The individual elements are narrower than the flat elements in 1L and more parallel-sided. 2-4L5 are slightly curved, whereas the others are straight. The dorsal edge is smooth. The ventral edge carries long setae, movably attached to elevated sockets (Fig. 15). Every second seta is turned anteroventrally, every second posteroventrally. The setae are very long, the longest ones measured being 12 mm, despite the fact that the tips were broken or poorly pyritized. In cross section, they are somewhat flattened with the flat sides facing medially and laterally. The number of setae is 20 to 25 in 2L5 and successively less in the other podomeres. The structure of thoracic legs, 2-4L, is interesting because it indicates a pattern of movement different from that in extant pantopods (all of which have legs ___round in cross-section). In extant species they are moved only in a dorsoventral plane, whereas in Palaeoisopusthe differently flexed distal parts may indicate that they were extended during a down-backward powerfull stroke, whereas they could be _+flexed during an up-foreward recovery stroke, in a similar way as in water beetles, e. g. Dytiscus. Abdomen The abdominal exoskeleton (Fig. 9, 17, 19, 22) consists of 4 cylindrical elements (AB1-4) and a tail spine. The cylindrical elements are likely to correspond to one segment each, while the position of the anus (see below) indicates that the tail spine is formed by (at least) one body segment (AB5) plus a post-segmental portion, the telson (T). Each element has a pair of longitudinal tuberculated ridges dorsolaterally and a similar unpaired ventral ridge. This arrangement probably gave the abdomen a more or less triangular cross section. Also the anterior and posterior margins are marked by tuberculated ridges. On the dorsal side the latter are straightly transverse and fairly close together, which means that the abdomen could not have been flexed upwards more than to a very limited extent. In ventral view, on the other hand, the ridges form arches with the concave sides facing each other. This arrangement may indicate an ability to flex the various elements downwards, a possibility that is well substantiated in the case of the tail spine. The surfaces between the tuberculated ridges do not show any conspicuous ornamentation. One-third from the anterior end the tail spine has an "irregularity" (?) on the ventral ridge (WS 638 (Fig. 21b), 2808 (Fig. 21c), 2813, 2744, 10128 (Fig.9)) evidently indicating the position of the anus. It 28 J. Bergstr6m, W. Stiirmer and G. Winter Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 29 ox go g, da d~ 30 J. Bergstr6m, W. Stiirmer and G. Winter Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 31 is particularly interesting since the anus should mark the boundary between the last b o d y segment and the post-segmental telson (BowMAN 1971), which by no means needs to be separated by a joint. The condition inPalaeoisopus in this respect should be more primitive than in any other k n o w n pycnogonid. It means that the number of true abdominal segments is probably 5, Fig. 21. Drawings of abdomen of Palaeoisopus. a) WS 4744 in ventral view with visible anus, b) 638 in ventral view, c) 2808 in lateral view, d) schematic reconstruction showing exterior features and intestine (Scale 1 cm). Fig. 20. Thoracic legs, 2-4L, of Palaeoisopus adapted to a swimming mode of life. a-g) Drawings showing differently flexed distal parts. Note particularly the dorsoventral view in g (all others are in lateral view) demonstrating the extreme thinness of the distal part of the leg and the arrangement of the long setae in two files. a+ c) WS 28 l 3, b) 3836, d+ e) 517, f+ g) 446 (Scale 1 cm). h-i) Diagrams showing the movement of thoracic legs in h) extant species and i) Palaeoisopus, and in comparison to it the corresponding cross-section of these legs. Double arrow means power stroke, single arrow recovery stroke. Leg projected in different planes. 32 J. Bergstr6m, W. Stiirmer and G. Winter Function and ecology Palaeoisopus is a large pycnogonid, competing in size with recent Colossendeis. The size and the large flattened legs indicate that it was a good swimmer. A swimming ability cannot be in doubt, particularly as extant pantopods with much less sophisticated equipment may be found at times as regular constituents of the plankton. The find of Palaeoisopus associated with a crinoid, in the crown with the head end toward the calyx (WS 2807, Fig. 23a), may be fortuitous, but the very particular orientation indicates a life association. If so, the food (at least partly) would Fig. 22. Surface view ofPalaeoisopus, WS 4744, from the dorsal side, thorax pointing into the sediment and much compressed. Ornament along sides of abdomen appears to consist of lists and small tubercles although radiographs fail to provide evidence of underlying exoskeleton (Scale 1 cm). Pycnogonid arthropods from the Lower Devonian Hunsrtick Slate, West Germany 33 Fig. 23. a) Radiograph ofPalaeoisopus,embedded with the stalked crinoid, IrnitatocrinusgraciliorF. ROeM. Does the combination reflect a life relationship? - b) DUBININ'S (1957) view of the mode of life of Palaeoisopus. have consisted of crinoids. If these were long stalked, Palaeoisopuswould have had good use for its swimming ability in searching its prey. The life reconstruction (Fig. 23b) by Du~INm (1957) is tempting. The eyes are fairly large and may have been of good use in locating the prey. The well developed cheliphores may indicate that Palaeoisopus lived as predator. II. Palaeopantopus rnaucheri Order Palaeopantopoda BROlLI, 1930 (=Suborder Palaeopantopodina H~DGVETH, 1978) Pycnogonids with 3-segmented, limbless abdomen; lateral extensions with superficial annular markings. Type and only included family is Palaeopantopodidae. Family Palaeopantopodidae HEDGVETH, 1955 Characters of order Palaeopantopoda. Type and only included genus is Palaeopantopus. Palaeopantopus BROILI, 1928 Palaeopantopus maucheri BROILI, 1928 Only three specimens of this species are known to the authors. Of these specimens, two (WS 2810, Fig. 24, 25), 2812 (Fig. 26, 27) have previously been described in the literature, while the third one (WS 4747, Fig. 28, 29) was located during the search for X-ray objects. The width is up to about 14 cm between the extended appendage tips. The body is a flattened oval disc, roughly 1.5 cm long. The head end still is poorly known, but the existence of slender palps and ovigers is verified. Cheliphores have not been observed with certainty, and the proboscis is poorly known. All appendages are cylindrical and bear evidence of digestive diverticula. The considerable differences found to exist between the specimens are difficult to explain, and it is not even excluded that different species may be involved. 3 Pal~iont.Z. 54 34 J. Bergstr6m, W. Stiirmer and G. Winter Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 35 e~: 3" Fig. 25a Fig. 25b Fig. 25. Stereo-radiographs of Palaeopantopus, WS 2810 (Scale 1 cm). (a +b ~ laterodorsal; b + c = lateroventral) Fig. 25c Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 37 Cephalosoma The cephalosoma is short and compact. The lateral extensions belonging to the cephalosomatic walking leg segment are placed close together in the anterior end of the body disc, with the result that there is hardly space for a head between the extensions. The head therefore may extend more or less ventrally. Unfortunately the evidence is scanty. In WS 2810 the anterior end is severely damaged. A faint shadow edge with pyrite granules maybe represents the proboscis which extends more or less forward (Fig. 24d). A faint shadow visible on the dorsal side of WS 2810 may look as an eye tubercle and two adjoining segmented structures might represent the lateral extensions LE1-2. In WS 2812 there is very little evidence of structures other than palps and ovigers, but their position and the absence of additional positive evidence would seem to indicate that the head is quite small. A backwards extending shadow may represent the proboscis. With this background, the structures in WS 4747 are puzzling. The specimen is prepared from the ventral side, exposing a large elongate structure with the thick end under the body disc and a long shaft extending forwards. On each side of the shaft there are two slender limbs, obviously representing palps and ovigers. It is difficult to avoid the conclusion that the elongate structure represents the head with the proboscis, but how can the preservation in this specimen be so different from that in the two others? On the other hand the long shaft might be interpretated as part of the cheliphores and the "thick end" pressed backwards might be the proboscis (Fig. 28c). With this possible exception, cheliphores are not visible in the available material, if the interpretation of the two pairs of head appendages is correct. Palps and ovigers are of similar size and shape. The palp is about 1.5 cm long and is curved in the shape of a "Z" as in many extant pantopod species. At least 7 podomeres are visible in WS 2810 and 4747. A minimum of 3 proximal podomeres extend forwards-downwards, the next 3 are inclined forwards-upwards, and the distal one again points downwards. The oviger is slightly longer than the palp and hangs down. There is a comparatively long and strong proximal portion without discernible segmentation (WS 4747). There follow probably 2 pendant podomeres, which in turn are succeeded by a series of possibly 7 short podomeres arranged in a crescent with the first member pointing upwards or outwards and the distal one downwards or inwards. A small triangular anterior portion of the body disc belongs to the cephalosomatic walking leg segment. Stout lateral extensions (LE) point anterolaterally. These extensions show annulated markings with four rounded rings. It may be noted here that these rings are quite different from the annuli at the bases of ambulatory appendages inPalaeoisopus, and we are not in doubt that they belong to the lateral extensions. The first walking leg, 1L, is of modern design. The three proximal podomeres, 1L1-3 (coxa, trochanter 1 and 2), are of subequal length and are ventrally directed. The succeeding podomere 1L4 (femur/ equals the three proximal podomeres together in length and extends laterodorsally. 1L5 and 1L6 (patella and tibia) equal 1L4 in length and bring the appendage down from the elevated femur/patella joint. Distal to 1L6 there are some 3 to 6 podomeres representing the tarsal podomeres. Thorax The body of the three thoracic segments is similar to that of the cephalosomatic walking leg segment. TH 1 and TH2 have a trapezoidal dorsal shield, TH3 a triangular shield, all seemingly smooth like the triangular shield of the cephalosomatic walking leg segment. An oval central dome occupies the central parts of TH 1 and TH2 (particularly well visible on the prepared dorsal side of WS 2812, Fig. 26a). The surrounding brim of the body disc is flat and smooth. The lateral extensions have annular markings and extend laterally, or in TH3 posterolaterally. There 38 J. Bergstr6m, W. Stiirmer and G. Winter Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 39 o~ Fig. 27. Stereo-radiographs of Palaeopantopus, WS 2812. (a+b) dorsal and (b+c) ventral view; • 1. Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 4I are tings in TH1 and TH2, while there seems to be only 3 rings in TH3. All lateral extensions are close together, even in the posterior midline, resulting in the impression that the extensions belong to the central body disc. The thoracic legs, 2-4L, are of equal size to the cephalosomatic leg, 1L. Therefore the trunk is much more of a unit inPalaeopantopus than inPalaeoisopus. As in 1L the three proximal podomeres, L1-3 (coxa, trochanter 1 and 2) are of subequal length and are ventrally directed. L2 and L3 form the main flexure of the appendage. L4 (femur) equals each of L5 and L6 (patella and tibia) in length. 3L4 appears to extend distally into a pointed tip, which covers the distal joint. Distal to L6 there are some 3 or more podomeres (tarsal podomers). The ultimate segment is commonly found in a sharply flexed posture, indicating the possible presence of a terminal claw. Abdomen The abdomen points dorsally from between the lateral extensions of TH3. It consists of three segments. The anterior one of these, AB1, is short but comparatively high. It shows annular markings in the same manner as the lateral extensions, with 4 rings. The cross section is probably oval. AB2 and AB3 are about twice the length of AB 1 but are much narrower and devoid of annulation. AB3 is bent backward. The anus is not to be seen but appears to have been terminal (see below). A telson sensu stricto therefore seems to be missing. Internal structures All three specimens, but particularly WS 2812, exhibit a pattern of pyritization quite different from that in Palaeoisopus in having a central segmented pyrite rod in each appendage (including palp and oviger) as well as in the abdomen. The pyrite rod is of even width, and there seems to be no way to avoid the conclusion that the rods represent the intestine in the abdomen and intestinal diverticula in the appendages. Such a pattern is characteristic of modern pantopods (Fig. 30). The abdominal pyrite rod is thicker in AB1 than in AB2 and AB3. It is notable that it extends to the distal tip of AB3, as this indicates that the anus was terminal. Function and ecology The structure and posture of the long and thin walking legs indicate a slow walking animal keeping the body highly elevated over the sea floor. Palaeopantopus lacks the apparent adaptations to a swimming life found inPalaeoisopus, but experience from modern pantopods shows that this is no evidence of a lack of swimming ability. It is difficult to speculate over the mode of feeding and the kind of food as long as cheliphores are not known for certain and the head and proboscis are poorly understood. III. Palaeothea devonica Order Pantopoda GERSTAECKER,1863 Pycnogonids with 1-segmented abdomen and terminal anus. 8 Families. Family uncertain Palaeothea nov. gen. Characters of type and only known species P. devonica. Palaeothea devonica nov. spec. Only a single specimen (WS 2287, Fig. 31) is known of this species, but considering the antiquity of a pantopod of modern design the find is regarded as so important that a formal name was required. The specimen is quite small, with a length of only about 5 mm and a width of perhaps about 7 ram. The body is elongate. The head is extended forwards and carries a short proboscis. Fig. 28. a) Surface exposure, b) radiograph, and c) interpretative drawing of Palaeopantopus, WS 4747 (Scale 1 cm). Pycnogonid arthropods from the Lower Devonian Hunsr/.ick Slate, West Germany • + ,_o O + O o~ 43 44 J. Bergstr6m, W. Stiirmer and G. Winter Fig. 30. Midgut diverticula in extant pantopod Pycnogonurn littorale STR6M. The cephalosoma is followed by three separate thoracic segments. The abdomen is only visible as a faint shadow, and it can not be made out if there is only one or possibly more segments. Cephalosoma The cephalosoma, as we interpret it, is about 1.2 mm long. Dorsally there are two distinct median tubercles. The posterior one is close to the posterior margin and corresponds to similar tubercles in the thorax. The anterior tubercle lies in the middle of the cephalosoma and is apparently an eye tubercle, although eyes are not discernible. The proboscis is cylindrical and extends forwards. The length is about 0.6 ram. The cheliphores are probably indicated by a pair of straight shadows dorsolateralto the proboscis. They extend from a level just behind the proboscis, and the length is about 0.7 mm. There is no trace of segmentation or of any chelae. The supposed palps attach at a level lateral to the eye tubercle. The detailed structure is impossible to make out, although two distalpodomeres are indicated on the left side. The left palp has a length of about 1.2 mm and extends beyond the tip of the proboscis. There is a possible trace of the ovigers just in front of the first walking leg, 1L, particularly on the left side. There is no indication of more than one podomere. The first walking legs or ambulatory appendages, 1L, are situated on long lateral extensions from the posterolateral surfaces of the cephalosoma. The extensions seem to be smooth, without any trace of annulation. The appendage has three short proximal podomeres, L1-L3, which may be identified as coxa and trochanter 1 and 2. Distally follows a dorsally flexed L4 (femur), followed by a more or less horizontal L5 (patella). Further distally the limbs turn ventrally, but no further distinct podomeres are discernible. 1L appears to have a rounded cross section. Other special features, such as setae, are not visible. Thorax The thorax, with its appendages, conforms well with the last segment of the cephalosoma. The anterior part of each segment is comparatively narrow, while posteriorly the lateral exten- Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 45 Fig. 31. a-c) Palaeothea devonica, a-b) Radiographs, negative and possitive, c) interpretative drawing. d) Drawing of extant pantopod Pigrogromitus timsanus CALMAN(after CALMAN1927) for comparison (Scale 1 mm). slons form the bases of the appendages. The strong median tubercle is placed posteriorly and connected with the posterolateral corners by strong ridges. The first two pairs of appendages conform in all essential characters with the last pair of the cephalosoma. In the last segment the median tubercle is less developed, and the appendages are directed backwards in a crescent. The details are more shadowy than in the other walking legs, and no podomeres are discernible. 46 J. Bergstr6m, W. Stiirmer and G. Winter Discussion The small size of the single specimen might indicate that it is a larva, possibly of one of the two other pycnogonids in the Hunsriick Slate. Possibly larval features include the posture and shape of the last pair of ambulatory appendages, perhaps the shape and size of cheliphores, palps and ovigers, and perhaps an absence of differentiation between walking legs 1L-3L. There are also arguments against such an explanation. For instance, the morphology of the cephalosoma is very different from that in the two previously known species, where for instance an eye tubercle is either anterior in position or lacking, whereas it is situated in the middle inP. devonica. Admittedly, however, the ontogeny may bring about strong changes in extant pantopods. Other differences include the shape of the lateral extensions and the shape and relative size of the podomeres of the walking legs. The cylindrical appendages contrast sharply with the flattened appendages of Palaeoisopus, and a small specimen of that form only about 9 mm long excluding proboscis and abdomen, is not reminiscent ofP. devonica but rather similar to large specimens of the same species. Therefore it can be concluded that the new specimen is not identical to any of the two previously known pycnogonids from the Hunsriick Slate. Thus the specimen represents a new species of pycnogonids, the third species from the Lower Devonian. Features particularly characteristic and distinguishing the new form from the two others include the lateral extensions without annulation, unique dorsal morphology with median tubercles, unique form of cephalosoma with neck, and unique habitus with elongate body and distance between lateral extensions. The posture of the walking legs indicates an creeping animal rather than an elevated, long-legged Palaeopantopus or a swimming Palaeoisopus. The conclusion is that the specimen is an adult or nearly adult pycnogonid of modern habitus and as such quite different from Palaeopantopus and Palaeoisopus. The modern habitus makes an assignment to the Palaeoisopoda or Palaeopantopoda impossible. Although we do not know anything of the abdominal segmentation, an assignment to the order Pantopoda seems most reasonable, although it must be as a forma incertae sedis. Only future finds can elucidate the structure of the abdomen and the distal morphology of the appendages, and a certain amount of carefulness is therefore motivated. Anyway, at present it seems likely thatPalaeothea devonica is the only known fossil representative of the modern order Pantopoda. IV. O t h e r a r t h r o p o d s described as fossil pycnogonids Two small arthropods from the Upper Jurassic Solnhofen lithographic limestones - P h a - langitespriscus MONSTE~, 1836 andPalpipes cursor ROTH, 1851 -- have repeatedly been considered as fossil pantopods, although this interpretation has been strongly criticized by various authors. In a series of papers, POLZ (1969-1975) has recently presented strong arguments for the interpretation of these fossils as phyllosoma larvae of crustaceans. HEDGPET~ (1978) considers Phalangites priscus to be spiderlike, but during his search he found one specimen which he considers different. This specimen forms the basis for the erection ofPentapalaeopycnon inconspicua HEDGPETH 1978 and the family Pentapalaeopycnonidae HEDGPETH 1978. However, the specimen appears to be an exuvia of a phyllosoma larva of the type regarded as Form A by POLZ 1971). The specimen seems to have six limb pairs, of which the first and the last are notably shorter than the remaining four pairs. Phyllosoma larvae have seven pairs of thoracic limbs. The first pair is small and inserted well in front of the second pair, and it is generally not preserved or visible inPhalangites. The second pair is smaller and appears to correspond to the palps in HEDGPETH'S interpretation. The following four pairs are mutually similar and have short exites. The exites shown on plate 2 (p. 28) and figure 4 A (p. 30) by HEDGPETH -- never appearing in Pycnogonids - can be judged as a sure hint to the phyllosoman character of this fossil; as they are missing in the reconstruction (figure 4 B) by HEDGPETH, Pentapalaeopycnon looks like a pycnogo- Pycnogonid arthropods from the Lower Devonian Hunsrfick Slate, West Germany 47 nid. The last thoracic limb pair in a phyllosoma is characteristically and notably smaller, as inP. Phyllosoma limbs are round in cross section, as are those of pantopods. The phyllosoma thorax is distinctly segmented, so while this is not the case in ordinaryPhalangites specimens, the preserved segmentation inPentapalaeopycnon inconspicua cannot be taken as proof of a pycnogonid affiliation. The conclusion is that it is much easier to accept P. inconspicua as a phyllosoma larva than as a pantopod. This apparently means that the three Devonian pycnogonids described herein are the only known fossil pycnogonids so far. inconspicua, whereas this is not the case in pantopods. V. R e l a t i o n s h i p s a n d p h y l o g e n y w i t h i n t h e P y c n o g o n i d a Supposing that the phyletic branching was dichotomous, Palaeoisopus, Palaeopantopus and the Pantopoda may be related in three different ways. The "cladistic" method of comparing these possibilities may be utilized. The three taxa are split off first, one in each of the three alternative models. In each of the three models the two remaining groups form sister groups. In alternative model 1, the Pantopoda andPalaeoisopus form the ultimate sister groups (i.e., Palaeopantopus is branched off first). We are unable to see any features in the two ultimate sister groups which may be thought of as common derived (synapomorphic) characters. This model therefore appears less likely than the others to represent natural relationships. In model 2, Palaeoisopus andPalaeopantopus form the ultimate sister groups. In a search for possible uniting characters of common derived type, the ringed character at the leg bases is perhaps the most obvious possibility. In detail there are great differences: e. g. the telescopic arrangement in Palaeoisopus is much more advanced than the simple annular markings in Palaeopantopus, while the annular markings in the abdomen of the latter lacks a counterpart in the former. These differences do not prove that incipient annulation is not basically a common derived character. On the other hand, there is no evidence in the opposite direction, either. In model 3, Palaeoisopus is split off first, andPalaeopantopus and the Pantopoda are the ultimate sister groups. The evolutionary stages in the reduction of the abdomen may be in favour of this model, as Palaeoisopus has the least reduced abdomen with 5 segments, Palaeopantopus is intermediate with 3 segments, and modern pantopods have only one abdominal segment. However, it should be kept in mind that the state inPalaeoisopus is already very advanced in pycnogonid direction in relation to typical chelicerates, and it is not entirely excluded that the reduction trend, which may have been in action for a long time, perhaps triggered by the mode of life, was followed up independently in different groups of pycnogonids. Palaeoisopus (as a swimming animal) may have had one important reason to keep a comparatively long abdomen, namely the need of a stabilizing rudder. The presence of a comparatively long abdomen therefore is not compelling evidence that Palaeoisopus is lower on the evolutionary staircase than the other pycnogonids. Connected with the abdominal reduction in Palaeopantopus and the Pantopoda is an abortion of the telson, with the result that the anus is terminal (cf. BOW~aAN 1971). This character may be shared derived (synapomorphic), provided that the reduction to less than 5 abdominal segments is a shared derived character for the two groups concerned. Distinct proof is missing, however, and a terminal position of the anus is by no means unique among arthropods (e. g, BOWMAN 1971). The more or less cylindrical podomeres of Palaeopantopus and the Pantopoda contrast with the flattened swimming legs inPalaeoisopus. Discrimination of primitive and derived states fails, however, as we do not know with certainty the shared primitive (symplesiomorphic) condition. The extreme flattening inPalaeoisopus is certainly an adaptation to swimming, and the appendages in the head are more rounded in cross section, as are also the proximal parts of the swimming appendages. The cylindrical shape of the podomeres in the two other groups, on the other hand, 48 J. Bergstr6m, W. Stiirmer and G. Winter may be connected with the presence of gut diverticula in the limbs. It is possible that both types of limbs go back independently to a primitive limb that was more or less oval in cross section. Despite the fairly large number of studied specimens there is no trace of possible gut diverticula in the appendages of Palaeoisopus, contrary to the condition in Palaeopantopus, where only 3 specimens have been available. We tend to believe that this indicates the absence of extended diverticula in the appendages of Palaeoisopus. If this interpretation is correct, the absence of diverticula in this animal is a primitive (plesiomorphic) character, while the flattening is a derived (apomorphic) character unique to this taxon among the Pycnogonida. The cylindrical appendages containing gut diverticula then appear to be a shared derived (synapomorphic) character uniting the Pantopoda andPalaeopantopus. The gut diverticula found in arachnids seem to be a parallel structural development. P A L A E O P A N T O P P A O N D T O A P O D A PALAEOSIO~PODA~/ l;c~tbea%eeds~Wnlnmulmia~ge~mbs >~ ~/~/< // I L tagmosisof thorax and cephalosomo J J < " J I anus terminal probosois jJ~. I ~Ybi:ndml~Olr:idl~hbc:;' ~ist%g~tmaernYt:i to verticula l ? I'1 merostome loss of abdominal limbs and all outer limbrami a b d o m e nreduced to 5 segments and telson origin Fig. 32. Diagram showing the relationship within the pycnogonids as suggested herein. Summarizing the evidence, it seems probable that the line that led to Palaeoisopus branched off first in pycnogonid evolution, and that Palaeopantopus and the Pantopoda, such as now known, are the ukimate sister groups, provided that we are talking in cladistic terms. In practice, the different groups need not only to have sister groups but also ancestors, and it may be possible to say that the probable organizational sequence Palaeoisopoda- Palaeopantopoda - Pantopoda also may represent a phyletic lineage. This would mean that the common ancestor of Pantopoda and Palaeopantopus is ranked as a member of the Palaeopantopoda, although certainly it would have been less speciallized in certain features than Palaeopantopus itself. In a similar way the common ancestor of the Palaeopantopoda and Palaeoisopus may be regarded as a palaeoisopod. It must be stressed that the morphological background for this discussion is fairly scanty, and new evidence, for instance from new fossil finds, may modify our judgments. Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany 49 VI. Systematic position of the Pycnogonida The systematic position of the pycnogonids in the system of the Arthropoda has long been a matter of debate. Two views held by many students are that pycnogonids are either closely related or belong to the Chelicerata or that they have a relatively isolated position within the Arthropoda. A variant of the second alternative was preferred by TIEGS & MANTON (1958), who suggested a relationship with the Onychophora. Recently, MANTON (1978b) has strongly advocated a close relationship between pycnogonids and typical chelicerates on the basis of similarities in the functional morphology of the appendages. One has to be careful with MA•TON'S judgments about phylogeny, however, as they are generally based on the state in advanced living end forms and show an ignorance of possible and even necessary intermediate evolutionary stages as well as of fossil forms. A single example may illustrate the need for this cautionary statement, in order not to make it look like empty criticism. She says (MANTON 1978b : 26 I), about the relationships between crustaceans and trilobites, that "no crustacean has anything remotely resembling the development of adult trunk segments from the pygidium of a trilobite, where the pygidial segments with presumably functional limbs separated to join the articulated thorax . . .". In various trilobites, particularly in the Cambrian, the so-called pygidium is small and seemingly unsegmented, which should mean that in reality it may be a telson. In any event, there is no pygidium in the traditional sense, and there cannot have been any from the beginning. These trilobites therefore conform fully with the crustaceans and differ entirely from other trilobites in a character which MANTON regards as of fundamental importance in demonstrating the lack of affinity between the two groups. Regarding chelicerates and pycnogonids, the comparatively small number of appendages can not be original, and the irregularity in stepping demonstrated (MANTON 1978b: 264,485) therefore can not be used as an argument against trilobitomorph affinities of the chelicerates and pycnogonids. Likewise, the absence of a promotor-remotor swing of the leg at the coxa-body junction (MANTO~ 1978b: 263,483-484) in arachnids and pantopods is a secondary achievement that does not rule out affinities with other groups and does not prove that the Arachnida and Pantopoda (or Pycnogonida) share a common derived (synapomorphic) character in this respect. Anyway, MANVON'Sown observations on extant arthropods are admiringly superb in most respects, and there is every reason to take good care of them. Even if the ultimate proof is missing, a close relationship between chelicerates and pycnogonids appears more likely than any other alternative, and if there is any relationship at all it is probably so close that pycnogonids are best included in the Chelicerata. SCHRA~ (1978) has also recently suggested an affiliation of pycnogonids to chelicerates. Similarities listed by him include the presence of a first pair of chelate appendages, an endosternite skeleton, a perivisceral sinus in pycnogonids similar to the endosternite-associated sinus of certain chelicerates, features in the limbs and pattern of locomotion, the presence of four postoral segments in the first larva, a liquefaction-suctorial mode of feeding, the absence of a deutocerebrum, and possibly features of embryblogy. The observation that Decolopodawith one additional pair of legs differs from other pycnogonids and chelicerates in having well co-ordinated stepping brings perspective to MANTON'S uncritical use of coordination as systematic character. In an attempt to clarify the course of evolution within the Chelicerata, BERGSTR6M (1978) suggested that the tail spine of the Xiphosurida, Eurypterida and Scorpionida may be a synapomorphic character uniting those groups, and groups plesiomorphically without the spine (Beckwithiida, Diploaspididae, Arachnida except for scorpions) should have branched off earlier. The aglaspidids now seem to be trilobitomorphs and need not be concerned. The tail spine in Palaeoisopustherefore may indicate that pycnogonids belong with the tailed "merostomes" rather than with the arachnid-like groups. It is therefore particularly interesting that the morphology of living and fossil xiphosurids indicate that the chilarial segment should be counted 4 Pal~iont.Z. 54 50 J. Bergstr6m, W. Stiirmer and G. Winter with the prosoma (BERGSTR61Vi1975), and not with the opisthosoma as previously thought. Independent study of the ontogeny of Limulus led to virtually the same result, although even a part of the postchilarial segment proved to be formed under the prosomal shield (ScHoLL 1977). This Xiph0suro onferior end onferior end i 1 che[icero leg wifh gnofhobose --II-- 4. --II-- o 1 che[iphore E ~o 2 polpus ~ o'o. =, .c i '3 Arochnido clossicol view Pycnogonido i c~ E , o f~l o &. 3 0v ig er = 4 o cl u o Eo 0 leg 1 I 5 5 leg 2 --ll-~ 6 X O ..X II 6 leg3 ~ '- chiiorium 7 l=gnofhobose ) , 7 leg4 ~- +" C C 8 opercutum 8 8 i 9 8 O 9 9 10 10 gii[ oppendngel I 10 --ll-- 2 i ~ i 11 --il-- 3 E ..-O-- 11 E O O X) XD .4e- ( 12 --0=-- 4. 13 --iI-- 5 ' ~ 11,.2 E o ~ O t4- 14 15 fused ond i "m o ,'-i, (~ O 4-c- t~ a f._ > O E O O ~i o :17 rear end 18 ~ ;.,; ~ c~ ~U 116 limb-less i I O s0n clJ ,ID Ei r CU ._~ > > -- C u E ~ O c- OL O fetson Fig. 33. Comparison between limbs and body tagma in Xiphosura, Pycnogonida (classical view and modern view used herein), and Arachnida (classical view). Note similar border of bodyforepart in xiphosura and pycnogonida behind the seventh segment. Border between segments -+ movable to each other marked with broad line. Pycnogonid arthropods from the Lower Devonian Hunsrfick Slate, West Germany 51 means that the important boundary in xiphosurids between primarily leg-carrying prosoma and plate-carrying opisthosoma has a position exactly corresponding to the pycnogonid boundary between limb-carrying trunk and limb-less abdomen, provided that the chelicera and cheliphore occupy identical segmental positions. If the boundary between prosoma and opisthosoma in arachnids (other than scorpions) is correctly placed one segment further anteriorward, the tagmosis is a further connecting link (possible synapomorphic character) with the merostome scorpion groups, and no others. 0 L0gorithmic time scale qJ m.y cl .- :- d "~ "E "o -~. o. o ~cJ a o "~ . o o "~. ~ ~ "~ o 8. .~ r (ID o~ X ~ ~ ,.-, 0. e~ O. "~: -0 -200 o Permieln , Corb0nifer0us Dev0nion Silurian / Ordovicieln I ? T M ~Combrieln E~Iy _ ~'70 To-motion 9 o -~ ~ ./ ] I I \ -SOOLme >, o. o Pycn 0g0ni -300 -/~00 O "tD O Tertielry Cretelceous Jurelssic TriQssic -100 o O 9o j \ N I I/ el 1/ I l / / f f l Vendi0n Fig. 34. Schematic evolutionary relationship between cheIicerates and pycnogonids drawn to a logarithmic time scale. From a purely morphological point of view, a derivation of Palaeoisopus from primitive merostomes is fully possible. It would mean the formation of a proboscis in connection with a new mode of feeding, the disconnection of two pairs of limbs behind the chelicerae (cheliphores) from locomotion (which has also occurred in different chelicerates), the reduction of outer limb branches and of the opisthosomal limbs as a whole, and the reduction of abdominal segments (a tendency in this direction is particularly evident also in xiphosurids, although not as far-reaching). VII. Acknowledgments The authors are greatly indebted to the Stiftung Volkswagenwerk, the Deutsche Forschungsgemeinschaft and the Svenska Naturvetenskapllga Forskningsr~tdet for the generous financial support. The senior author owes special thanks to all those persons and institutions who supported this work by lending the specimens. Most of the radiographic detail work was done by the private assistant of the senior author, Mrs. HERTHA JACKLE. J. Bergstr6m, W. Stiirmer and G. Winter 52 V I I I . List of t h e s p e c i m e n s a) Palaeoisopus WS-Archiv-Nr. Collection (owner) 2 Text fig. 215 446 517 547 592 638 2268 2313 2698 2744 2807 2808 2813 2814 3836 3837 4416 4417 4632 4742 4744 S By By By By By Henne L CI CI By By By By KH KH Sch Sch Bo G6 Bin 4779 Mfihe 10127 10128 10129 10133 10135 10352 11234 11471 R R R R Br Br B Ha b) Palaeopantopus 2810 2812 4747 By By Bin 24, 25 26, 27 28, 29 Stii 31a-c Remarks 20f, g 20d, e 10a 21b 6a 8a BROILI 1928 21c 20c, d BROILI 1933 10c, 11, 20e 2 13a 8b, 10b 8c 10e, 12, 14 6b, 17 21a, 22 7, 8d, e, 10d 13b, 16 19 9 4c, 8f BROILI BROILI 1932 LEHMANN Pal. isop. B Pal. isop. B LEHMANN 1959 15 18 BROILI 1930 BROILI 1929 c) Palaeothea 2287 Abbreviations B Bin Bo By C1 G6 Ha H = = = = = = = = Coll. BEICHT, Bundenbach Humboldt-Mus. East Berlin Pal. Inst., Univ. Bonn Bavarian State Collection Munich Coll. CLAUSS,Sobernheim Pal. Inst., Univ. G6ttingen Coll. HARrH, Bundenbach Coll. H~NNE, Stuttgart KH L M R S Sch Stti = = = = = = -- GEm-Museum Bad Kreuznach Naturkunde-Museum Stuttgart-Ludwigsburg Coll. Dr. MOIaE, Kirn/Nahe Coll. RIEvzRs, Enkirch/Mosel Senckenberg-Museum Frankfurt Coll. Schwaben, Otzenhausen Coll. Dr. STORMER Pycnogonid arthropods from the Lower Devonian Hunsriick Slate, West Germany IX. 53 Literature BEXGSTRbM,J. (1975): Functional morphology and evolution of xiphosurids. - Fossils a. Strata, 4: 291-305; Oslo. - - (1978): Morphology and systematics of early arthropods. - Abh. Verh. naturwiss. Ver.; Hamburg. [In press]. - - (1979): Morphology of fossil arthropodsas a guide to phylogenetic relationships. [In: ] GOVTA,A. P. (Ed.): Arthropod Phylogeny; New York-London etc. BOWMAN,T. E. (1971): The case of the nonubiquitous telson and the fraudulent furca. - Crustaceana, 21: 165-175; Leiden. BROILI, F. (1928): Crustaceenfunde aus dem rheinischen Unterdevon. - Sitzungsber. bayer. Akad. Wiss., math.-naturw. Abt., 1928: 197-201; Miinchen. -(1929): Beobachtungen an neuen Arthropodenfunden aus den Hunsriickschiefern. Ein Pantopode aus dem rheinischen Unterdevon. - Sitzungsber. bayer. Akad. Wiss., math.-naturw. Abt., 1 9 2 9 : 272-280; Miinchen. (193•): •ber ein neues Exemplar v•nPa•ae•pant•p•s. - Sitzungsber. bayer. Akad. Wiss.• math.-na• turw. Abt., 1930: 209-214; Miinchen. -(1932a): Pantopoden aus dem rheinischen Unterdevon. - Forsch. u. Fortschr., 8: 269-270; Berlin. -(1932b): Palaeoisopus ist ein Pantopode. - Sitzungsber. bayer. Akad. Wiss., math.-naturw. Abt., 1932: 45-60; M~inchen. - - (•933)• Weitere Be•bachtungen anPa•ae•is•pus. - Sitzungsber. bayer. Akad. Wiss.• math.-naturw. Abt., 1933: 33-47; Miinchen. CALS~AN,W. T. (1927): 28. Report on the Pycnogonida. Zoological Results of the Cambridge Expedition to the Suez Canal, 1924. - Trans. zool. Soc. London, 22: 403-410; London. 1927. DUBImN, V. B. : [On the orientation of the cephalic end of the Devonian pycnogonids of the genus Palaeoisopus and their systematic position in the Arthropoda.] - Doklady Akademii Nauk SSSR, 117: 881-884; Moscow 1957. [In Russian.] ELMER, G. St. J. & DWIN, R. C. (1957): Logetronography. - Am. J. Roentgenology, 78: 124-133; Springfield, Ill. HEDGV~TH, J. W. (1954): On the phylogeny of the Pycnogonida. - Acta Zool., 35: 193-214; Stockholm. -(1955a): Pycnogonida. [In:] R. C. MOORE (Ed.): Treatise on Invertebrate Paleontology, Part P, Arthropoda 2, P163-P170; Lawrence, Kansas. - - (1955b):Palaeoisopus. [In:] R. C. MooRE (Ed.): Treatise on Invertebrate Paleontology, Part 5, Arthropoda 2, P171-P173; Lawrence, Kansas. - - (1978): A reappraisal of the PaIaeopan:opoda with description of a species from the Jurassic. - Zoot. J. Linnean Soc. London, 63: 23-34; London. Kurt~, O. (1949): Lehrbuch der Pal~iozoologie. Stuttgart. - - (1961) : Die Tierwelt der Bundenbacher Schiefer. - Neue Brehm-Biicherei, 274: Wittenberg Lutherstadt. KUTSCHEI~,F. (1931): Zur Entstehung des Hunsriickschiefers am Mittelrhein und auf dem Hunsriick. - Jb. nass. Ver. Naturk., 81: 177-232; Wiesbaden. - - (1963): Beitr~ige zur Sedimentation und Fossilfiihrung des Hunsriickschiefers. 3. Die Anwendung der R6ntgentechnik zur Diagnostik der Hunsriickschieferfossilien. - Notizbl. hess. L.-Amt Bodenforsch., 91: 74-86; Wiesbaden. - - (1970): Beitr~ige zur Sedimentation und Fossilfiihrung des Hunsriickschiefers. 27. Palaeopantopus maucheri BROILI und Palaeoisopus problematicus BROILI. Notizbl. hess. L.-Amt Bodenforsch., 98: 19-29; Wiesbaden. LEHMANN, W. M. (1939): Neue Beobacbtungen an Versteinerungen aus dem Hunsriickschiefer. - Abh. preui~. Akad. Wiss., math.-naturw. K1., 13: 1-17; Berlin. - - (I959): Neue Entdeckungen an Palaeoisopus. - Pal~iont. Z., 33: 96-103; Stuttgart. LEON, R. (1933): Ultraviolettes Licht entdeckt Versteinerungen auf ,,leeren,, Platten. Ein Pantopod im Jura-Kalk. - Natur u. Museum, 63: 361-364; Frankfurt a. M. - - (1934): l~ber Phalangites priscus MSTR. und Palpipes cursor R o ~ . (Noch keine Pantopoden im Jura.). - Senckenbergiana, 16: 24-29; Frankfurt a. M. M~TON, S. M. (1978a): Habits, functional morphology and the evolution of pycnogonids. - Zool. J. Linnean Soc. London, 63: 1-21; London. - - (1978b): The Arthropoda. Habits, Functional Morphology and Evolution. Oxford (erroneously dated 1977). MOORE, R. C. ; LALICKE~,C. G. & FISCHER,A. G. (I 952): Invertebrate Fossils. New York-Toronto-London. MOL~.~R, A. H. (1962): Pantopoden (Arthrop., Pycnogonida) aus den Solnhofener Plattenkalken (Maim zeta) yon Siiddeutschland. - Freib. Forsch.-H., C 151: 149-157; Berlin. - - (1965): Lehrbuch der Pal~ozoologie. Bd. 2, 2. Aufl.; Jena. - - - 54 J. Bergstr6m, W. Stiirmer and G. Winter OPrrz, R. (1935): Bilder aus der Erdgeschichte des Nahe-Hunsriick-Landes Birkenfeld. Birkenfeld. (1935): Lebensbilder aus dem Hunsriickschiefer-Meer. - Natur u. Volk, 65: 504-509; Berlin. Porz, H. (1969): Beobachtungen an Pantopoden aus den Solnhofener Plattenkalken. - Natur u. Museum, 99: 457-467; Frankfurt a. M. - - (1970): Zur Unterscheidung vonPhalangitespriscus MONSTERundPalpipes cursor ROTH (Arthr.) aus den Solnhofener Plattenkalken. - N. Jb. Geol. Pal~iont. Mh., 1970: 705-722; Stuttgart. (1971): Eine weitere Phyllosoma-Larve aus den Solnhofener Plattenkalken. - N. Jb. Geol. Pal~iont. Mh., 1971: 474-488; Stuttgart. (1972): Entwicklungsstadien bei fossilen Phyllosomen (Form A) aus den Solnhofener Plattenkalken. N. Jb. Geol. Pal~iont. Mh., 1972: 678-689; Stuttgart. (1973): Entwicklungsstadien bei fossilen Phyllosomen (Form B) aus den Solnhofener Plattenkalken. N. Jb. Geol. Pal~iont. Mh., 1973: 284-296; Stuttgart. (1975): Zur Unterscheidung von Phyllosomen und deren Exuvien aus den Solnhofener Plattenkalken. - N. Jb. Geol. Pal~iont, Mh., 1975: 40-50; Stuttgart. RICHTrR, R. (1931): Tierwelt und Umwelt im Hunsr/.ickschiefer. Zur Entstehung eines schwarzen Schlammsteins. - Senckenbergiana, 13: 299-342; Frankfurt a. M. ScHorr, G. (1977): Beitr~igezur Embryonalentwicklung von Lirnuluspolyphemus L. (Chelicerata, Xiphosura). - Zoomorphologie, 86: 99-154; Heidelberg/Berlin. SCHr,AM, F. R, (1978): Arthropoda: A convergent phenomenon.- Fieldiana: Geology, 39: 61-108; Chicago. STORIvIER,L. (1944): On the relationships and phylogeny of fossils and recent Arachnomorpha. - Skr. Nor. Vidensk. Akad. Oslo, math.-naturw. K1., 1944: 1-158; Oslo. SViSRMrR, W. & BERGSTR6M, J. (1973): New discoveries on trilobites by X-rays. - Pal~iont. Z., 47: 104-141; Stuttgart. - - (1976): The arthropods Mirnetaster and Vachonisia from the Devonian Hunsriick Shale. - Pal~iont. Z., 50: 78-111; Stuttgart. - - (1978): The arthropod Cheloniellon from the Devonian Hunsrfick Shale. - Pal~iont. Z., 52: 57-81; Stuttgart. TZEGS,O. W. & MANTON,S. M. (1958): The evolution of the Arthropoda. - Biol, Rev., 33: 255-337; New York. WINTER, G. (1979): Beitr~ige zur Morphologie und Embryologie des vorderen K6rperabschnittes (Cephalosoma) der Pantopoda GERSTAECKER,1863. -- Thesis Erlangen. - - - - - - - - - - - - Eingang des Manuskripts bei der Schriftleitung am 20.8. 1979.