Daten Ne “Aah shen vnsduaa roe wee eens m= iz eis Bat sonet tin be Seach stnrannosinnttl Frere rene nT Kean it Mara stan simaont Nut Ratt coetat AIS Tit tanSaease, Peper Fae eS iae rere hw snanem nines) baleeens reeeenre eS Sa ashan aeenerniew-Desteamna renee wy Catt AOE Te eee a et ae Gachratneh asec : Te tS Zz — o ee = Ww aoe < — a) AS = —— £ on i aaa AITHSONIAN INSTITUTION | NOILNLILSNI_ NVINOSHLINS, SA1YV¥@IT_ LIBRARIES SMITHSONIAN INSTITUTION NOILALIS Ek he = ES a: K = vie K z = Ac = s SY = ae AC: 5 = BANS = EQ 7. 2N& 2 2 Bo 2X 2 BNR 2 a Na = 2 FE WO 2 = 2 = > = > = > = > G no . Zz ” Zz n el ie 7) Fd VINOSHLINS S3IYVYSII LIBRARIES SMITHSONIAN _INSTITUTION | NOILNLILSNI NVINOSHLIWS LIBRAR _ ” = > _ : = BER Ww re Ww ae Wl = ee ia Ko = g z ~ = 2X = 4 SS ae < =f < ss = x HoH NS. x x = ee 3 2 . rs} = rs} ye 3 Se fry Zz aj z =) Zz = beay Se MITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS S31YVYG17 LIBRARIES SMITHSONIAN INSTITUTION | ra » = fe) = Oo a (eo) =e [s) gD wo — ow — w — C4 ion] = Gy: E 2 i & bey > F Vi Yo, Ee > E > E V4 Sage = Of) Kp 2 = z = 2 - Y he? = ‘ m on m Eb m De a m ” n ze wo = n = o = | VINOSHLINS SAIYVUGIT_ LIBRARIES | INSTITUTION NOILMLILSNI_ NVINOSHLINS, Salu¥vyalT LIBRAR z Ws, re z : z < < = a Witty, 5S \ C= iN a z = = 247 PF" 2 NS 2 NES E Zz. = z a XS Fe » S # = 7 MITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSHLIWS saluvual LIBRARIES SMITHSONIAN INSTITUTION NOILNLI LIBRARIES SMITHSONIAN NOILALILSNI LIBRARIES NOILALILSNI NOILALILSNI LIBRARIES NOILALILSNI LIBRARIES VINOSHLINS SSIYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVYEIT LIBRAR saiuvugl INSTITUTION saiuvyugi INSTITUTION saluvyagly iY A fF? INSTITUTION INSTITUTION MITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES Y, vA NVINOSHLINS S3IYVYdIT iz) wn me wn < z & z eR eee Dee z = r me dr = re SW > ra KS = any ESR ZH PCI GY EWE 2 Nee? iS 2 “yy = z = “OS z \ 2 Wy = > = > = = > ¢ a . 2 a 2 7) . a ae VINOSHLINS S2IYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_NVINOSHLINS S31¥V¥817_LIBRAR Ww = w = — fp) — Ge wW 74 WwW = lJ Zz . Ke {re} z Kez : : % : 2 NSE : Ss < A ae a < a SS < a \ o oc a QQ a aN ray = a = ray ae) m = oe fe) = fe) = re) = fe) : 4 75 =I 2 4 Zz a =} z MITHSONIAN INSTITUTION — NOLLNLILSNI_NVINOSHLINS _S31¥VYGI7 LIBRARIES SMITHSONIAN INSTITUTION | NOILALL. ae ‘ eS rs Yr, OD 2 o . S = iS) ty S) Yiy fo. > XN > Vx fo > Of)? : EOE : © OL: : fa ee: = iB — S\S 5 As me f 4 = a o = ie Ph ee O z o z WINOSHLINS S3I1¥vYgIT_ LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS S31¥VyaIT_LIBRAR wn = no re 22) ra n z * = WY), = = < z < = < VY = - 3 — 2 97G.% Nx: 2 Ng : : = z 122) “iy yr, SS \ ” 7) ~ NS wo Oo n 22) \ Ss oO f Ie /: SE cA LAN CO 3c % SANS ro) xr ro} x SAS a VA 5 SS Zz = AS Zi = 2 = \N a gl ee =e ee = > = “Ss 2 eS = bs MITHSONIAN INSTITUTION a: NOILALILSNI_NVINOSHLINS = S3 IYVYG!I1T LIBRARIES SMITHSONIAN INSTITUTION NOILOLI. _ Ss ~ = w = nw g : f : j Sty, % : oS. roa = ow = wc Yih, A a = < a < =| < CO fope| < : = S = g a MY feos = fo) = fe) ee = 0. = co Zz = Zz =) 2 = =z 4 ite eit & a ei sihakam> |. ome haeelee GR om 6 a ee ee ee er ye ee a ee 0 Pe TS ee eT VU ee thes (ATIVE TB TT hae | tTImMONnmOanD le Wit) oe DS ene SQ 52 a ee aas 2 a (i en INOSHLINS S31YWUGIT_ LIBRARIES SMITHSONIAN INSTITUTION | NOLLNLILSNI NVINOSHLINS S31YVYalT_LIBRARIE = ve Xs = a ex. = z = tk & =| Barns 7 Z2OyN FS yw ra | uff fg = WY Z2 Ee MQ. 2 Zz NY. 8 “ify m 2 “iy E WY > = > = > = > ; = : Fas, w” . Zz Ww as Z w a 7a) 2 ITHSONIAN _INSTITUTION NOILOLILSNI_NVINOSHLINS S31YV¥S1T_ LIBRARIES SMITHSONIAN INSTITUTION NOILNLILS — — (ap) = on = n = a = WSS uw a ut oa a = & 4 Ng z c z o a < = WS = = a = = Ss ac c EQ oc i a Cc roe ye fe) = ro = rs) on ) = Zz -! ma - = re =i 4 zy (NOSHLINS SAINVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS S3I1YVYa!I7T_ LIBRARI » 7S | i ae lis > iz = me = ° = fe) = ‘s) a S) = — ies] = ‘L @ — ios] ; — ow Ya 2 : 5 £43 : 2 Ne: : WOE : Tie : 2 WOE 2 WY a Fe Bie eo = ” m (Ww. m OM =z Zz z “MN Zz — wn = un — Ww = w MITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLINS Sal¥vuaIT LIBRARIES, SMITHSONIAN INSTITUTION NOILALILS ae “* w 5 2 ae . Yn mn z < = oor = ae = = oN = > 5 z = z \, =r § WN GE i § 5 aS) 5 NWA c WANS Oo = 9 = AA S ae A AY oO = \S 2 = z. = N < = S\S =: 77) ae Z 7) = ” a z 77) ae Z HNOSHLINS SZIYVYAIT LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_NVINOSHLINS S31YVYEIT LIBRARI % n” — wn => ¢p) Fa ~ wn = WW = uu = WwW o a a © a x wat & =| Y 4 : = 2! = DS ; = > — = = = = a = a = 2 : Z 2 a Z ei 2 = un : = = = VINOSHLINS S31NWYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLINS, SAIUVYGIT LIBRAR! = > Ss > ‘ = > = Zz 7) = ae a ._ 2 : a Zz 7) MITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLIWS Sa1uVu a! LIBRARIES SMITHSONIAN INSTITUTION NOILOLIL — (ap) = w a E a # é s g = ce = ANS a a oe = a a Z =a, \s a x z = = Fs ON = a Ps = = Ye fo) = ro) = oO = / re) = z =! a * =] PA ay Za 4 VINOSHLINS S31YVUGIT LIBRARIES SMITHSONIAN INSTITUTION | NOLLMLILSNI_NVINOSHLINS _S31UV¥GIT_LIBRAR ‘\ ~\ {eo} a fo) = fo) = [e) = » = a 5 Gin, 5 2 Xe 5 = » OX Ls 2 E Up f= F ss EO & ze AS = SS a E = AQ = AS" 5 : ta aaa LW 8 MITHSONIAN INSTITUTION NOILNLILSNI NUINOSHEIWS Salen Mal LENE RARIES: SMITHSONIAN SINS TTUTION Noun a z Z = Gs = es < z x = z 4 Pe < oH Fa = 4EGS ; 9 S25 Bf,0 A Nw z iB zs 2 Ng 2 2 5 ME =e “ = > = SS > = ~ > a os > i) Fe n : rn ”n es Fe WINOSHLINS SAINVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILLILSNI_NVINOSHLINS S31UVYEIT_LIBRAR o = om) = ” Zz x 44 = wl rr iva) 3 Oo uJ 7) a, no ce = Ly, = a a aN = 4 3 a < DO jus < 2 ON < = oo o Lig’ /: oc MER 4 a = AX = : s ©. : : 2 ws 5 = ma 2 z = a a) = BRELSERICROL ET Priam ona riuwwaén»r4a IRRPRMmRaHHRiILTCe CRAITLICCQANIIANL INQCTITIITION NIOTINItt , 7 _ " 4 - = i ~ : : @ 4 2° 7 _ ; - : - 7 i ’ : — : 7 | os = '¢:&- - ee a : 1. 7 ° : ; a a, : ’ 8 a a _ 7 - : a : - , in : ‘ _ 7 7 a - oe . 7 iL : : aes —- -_ 7 - 7 i i ; a id . ae _ | a vi 7 _ an i ‘ _ 1) =4 7 - _ , ' 7 _ - - 7 os i ; - > 8 in! f - is - a : D At 7

7 a > — OG i: j a Fl : : - ir : - i ' iv 7 a oe | ’ o 7 ’ 7 os { 7 ° _ ' Y 7 : 7 - : . _ - ; : - ae a 7 ; 7 ‘Us ial ' ; Li : , 7 ' a - a4 7 = TAL a / : ; “a! it ol ' vy | wel ; * ; f . 7 : i _ - i = . ’ > 9 > ad ‘I Vv aa ' : oa Se 4 - 77 7 “% 4 - rm 4 ‘ oo i by 7. 7 i= = : =z 7 oe : a oe Se P52 aleontographica MeETtLCANA NUMBER 55 MARCH 18, 1986 A Restudy of the ae Scorpionida of the World by Erik N. Kjellesvig-Waering Organized for Publication ¥ by Anneliese S. Caster and Kenneth E. Caster Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York, 14850 U.S.A. PALEONTOLOGICAL RESEARCH INSTITUTION Officers PRESIDENTE Whee aece cea eee sare een ena WILLIAM A. OLIVER, JR. WVICE=PRESIDENTDcccrscter cori eac et nae oe ee an Ore ere nas cena roe WILLIAM P. S. VENTRESS SEGRETA RV deren Meecha ore atte eeu aae EE rere eee eros HENRY W. THEISEN SIERIEAS UIRIER Pyscc ay eine oe ac eRe roe eps tte CIE eae ROBERT E. TERWILLEGAR INSSISTANT MIREASURERS 1. (aei en ented ee ee EN eS ot See ates JOHN L. CISNE DIRECTOR Ge roscoe Me Eee eee Occ ot reverses song eevee PETER R. HOOVER ISEGATGOUWNSED waccuse sevacante see oon alee ot tet or aoue er Sched HENRY W. THEISEN Trustees Bruce M. BELL (to 6/30/87) Amy McCuNE (to 6/30/86) E. ANN BUTLER (to 6/30/88) CATHRYN NEWTON (to 6/30/88) RICHARD E. ByrD (to 6/30/86) WILLIAM A. OLIVER, JR. (to 6/30/86) JOHN L. CISNE (to 6/30/88) JAMES E. SORAUF (to 6/30/88) J. THOMAS DuTRO, JR. (to 6/30/87) RoBERT E. TERWILLEGAR (to 6/30/87) LEE B. GIBSON (to 6/30/86) HENRY W. THEISEN (to 6/30/86) HARRY LEFFINGWELL (to 6/30/87) RAYMOND VAN HouTtTTE (to 6/30/88) WILLIAM P. S. VENTRESS (to 6/30/87) BULLETINS OF AMERICAN PALEONTOLOGY and PALAEONTOGRAPHICA AMERICANA PETER RS HOOVERS Sse tae ere a es ion as Bande oateate een EDITOR REVIEWERS FOR THIS ISSUE OSCAR F. FRANCKE W. D. IAN ROLFE A list of titles in both series, and available numbers and volumes may be had on request. Volumes 1-23 of Bulletins of American Paleontology have been reprinted by Kraus Reprint Corporation, Route 100, Millwood, New York 10546 USA. Volume | of Palaeontographica Americana has been reprinted by Johnson Reprint Corporation, 111 Fifth Ave., New York, NY 10003 USA. Subscriptions to Bulletins of American Paleontology may be started at any time, by volume or year. Current price is U.S. $25.00 per volume. Numbers of Palaeontographica Americana are priced individually, and are invoiced separately on request. Purchases for professional use by U.S. citizens are tax-duductible. for additional information, write or call: Paleontological Research Institution 1259 Trumansburg Road Ithaca, NY 14850 U.S.A. i , E 1 « ¥ i 7 : { 4 4 i { iy ‘ . ‘ i x i + ' cs n : : * : ’ { i A ’ i ‘ ‘ . oi ' i 4 < : ; ; 7 1 ny , i iY : i 1s r é . - q 7 oa ; iy ; ‘ | chelicera I; (t ~ pedipalp - - -—-\-- 4 2 | ital J lateral compoun g aXe, median groove aS median eye node median eye r | ! | ! 1 sternum ; operculum I prépectinal plater fh a) prosoma cephalothorax VA ere Ne . BE FOR ; er fixed finger coxosterna f oi LY oxoster carapace 7, i) SS : a ees = =< pectinal plate |! ere) » COoxa (\ trochanter femur ; : immobile SS i <\ oxa oe femur Ze 2 Ne Va \\ median organ! | bey \~pectine Se (Ee N\\ Nee lv patella ungues or preabdomen tibia tarsal spurs ny tarsus , | OSE x4 es et << PG oy plete CF posttarsus doublure ee / " basitarsus = Saal l gill slit \\ YA = 7 pleural membranes tergite | Ni x ay /, | \ / = = | Lh a i oD carina ¢ | (keel) 8 abdomen | i, = | | Vel v J postabdomen ae a aculeus | > vesicle | Z tooth | Lj 63 2h 2a ee \ anus ase Text-figure 3.—External morphology of scorpions; based on a composite branchioscorpionid scorpion. AMNH BM(NH) BMS BU CIURCA CUGM DLD cmr cn CP, Cp cr cs ABBREVIATIONS OF REPOSITORIES American Museum of Natural History, New York, NY, U.S.A. British Museum (Natural History), London, England, U.K. Buffalo Museum of Science, Buffalo, NY, U.S.A. Birmingham University, Birmingham, England, U.K. Personal collection of Samuel J. Ciurca, Rochester, NY, U.S.A. Cornell University Geological Museum, Ithaca, NY, U.S.A. Personal collection of David L. Douglass, Western Springs, IL, U.S.A. Field Museum of Natural History, Chicago, IL, U.S.A. Florida State University, Tallahassee, FL, U.S.A. Institiit fiir Paldontologie, Friedrich-Wilhelm Universi- tat, Bonn, WEST GERMANY Geological Survey of Canada, Ottawa, Ontario, CANA- DA Institute of Geological Sciences, Edinburgh, Scotland, U.K. Geological Survey Museum, London, England, U.K. G6ttingen University Museum, Gottingen, WEST GER- MANY Illinois State Museum, Springfield, IL, U.S.A. Kilmarnock Museum, Ayrshire, Scotland, U.K. University of Kansas Museum of Natural History, Law- rence, KS, U.S.A. LAS Paleontological Institute, Academy of Sciences, Lenin- grad, U.S.S.R. National Museum, Prague, CLECHOSLOVAKIA New York State Museum, Albany, NY, U.S.A. Princeton University Museum, Princeton, NJ, U.S.A. Personal collection of Richard Cramer, Forest Park, IL, U.S.A. Royal Scottish Museum, Edinburgh, Scotland, U.K. Sedgwick Museum, University of Cambridge, Cam- bridge, England, U.K. Natur-Museum und Forschungs-Institit Senckenberg, Frankfurt-am-Main, WEST GERMANY Staatliches Museum fiir Naturkunde, Stuttgart, WEST GERMANY Svensk Riksmuséet, Stockholm, SWEDEN University of Illinois Geological Museum, Urbana, IL, U.S.A. Université de Laval, Cité Québec, Québec, CANADA University of Manchester Museum, Manchester, En- gland, U.K. University of Michigan Museum of Paleontology, Ann Arbor, MI, U.S.A. U.S. National Museum of Natural History, Smithsonian Institution, Washington, DC, U.S.A. Peabody Museum, Yale University, New Haven, CT, U.S.A. MORPHOLOGICAL ABBREVIATIONS USED IN TEXT-FIGURES areole anterior border aculeus anterior anterior lamella (of pectines) abdominal plate (numbered AP1, AP2, AP3, etc.) articulation coxa (numbered CI, CII, CIII, CIV, or IC, TIC, TIC, IVC, etc.) cephalic cheeks chelicera (joints numbered Chi, Ch2, Ch3, Ch4) hand of chelicera; immovable finger of chelicera free finger of chelicera chelical flange suture maxillary lobe of coxa (numbered CMI, CMII, or CM1, CMz?, or ICM, IICM, etc.) cephalic marginal rim condyle carapace crests (of cauda) cephalic shield d, dbl, doub doublure (d, = doublure of chelicera joint 2) D, dor dt inf cr int Ch se int oc intt La lat cr LCE LE Lmr mcr dorsal denticles fulcra gill, or gill tract gill pouch or branchial chamber gill slits (only on abdominal plates) leg number (joints in arabic numerals (q.v.), ascending from proximal to distal [may be further qualified as “D” (dorsal) or ““V”’ (ventral). P.R.H.]) inferior crest (of cauda) intercheliceral setae interocular ridge intersegmental tissue labium or labia lateral crest (of cauda) lateral compound eye (facetted), whether schizochroal or holochroal lateral eye (if facetted, see LCE) lateral marginal rim (of carapace) median crest (of carapace) Me, me Mea mL MO median eyes median appendage (of pectinal lobe) middle lamella (of pectines) median organ of prepectinal plate marginal rim of facetted eyes median sulcus ocelli (median eyes) operculum pedipalp (joints numbered P1, P2, etc.) hand of pedipalp; immovable finger of pedipalp free finger of pedipalp pectines pectinal lobe (basal pectinal lobe of intermediate la- mella) posterior marginal rim (of carapace) pectinal plate prepectinal plate pectinal teeth spurs (follow leg joint number, as II5s) socket sternite (numbered Stl, St2, St3, etc.) stigma, stigmata sternum sulcus superior crest (of cauda) superior spur tergite (numbered T1, T2, T3, etc.) trichobothria telson tooth or teeth transverse ridge ventral trochanter (with leg number in roman numerals: I, 111, 1111, I'V1, etc.) femur (with leg number in roman numerals, as 112) patella (with leg number in roman numerals, as I113) tibia (with leg number in roman numerals, as I14) basitarsus (with leg number in roman numerals, as IVS) tarsus (with leg number in roman numerals, as 16) posttarsus (with leg number in roman numerals, as 1117) ate a Sse eee — > irae — - _ = - _ - oy ee cen ES ————— a 22¢ Oe ee ee - a : — a ae ea oe > Cp American Mus U.S.A. British Museur, Buffalo Muse Birmingham Personal colle U.S.A. Cornell Univ U.S.A. Personal collec TESU:S:As Field Museu Florida State Institiit fir P tat, Bonn, W Geological S DA Institute of Ge Geological Su G6ttingen Uni MANY Illinois State Kilmarnock University of I rence, KS, U.S areole anterior bor aculeus anterior anterior la abdominal articulation coxa (num IVC, etc.) cephalic ch chelicera (j hand of ch free finger chelical fla maxillary | CM1, CM cephalic m condyle carapace crests (of c cephalic shi dbl, doub doublure (d dor II, Tl, dorsal denticles fulcra } gill, or gill t gill pouch o gill slits (on leg number from proxir “D—D” (dorsal inferior cres interchelice: interocular intersegmen labium or lé lateral crest lateral comp or holochro lateral eye ( lateral marg median cres | The Paleontological Research Institution acknowledges with special! thanks the contributions of the following individuals and institutions PATRONS ($1000 or more at the discretion of the contributor) JAMES E. ALLEN (1967) AMERICAN OIL CoMPANY (1976) ATLANTIC RICHFIELD COMPANY (1978) CHRISTINA L. BALK (1970, 1982, 1983) Hans M. Botti (1984) Mr. & Mrs. KENNETH E. CASTER (1967) CHEVRON O1L Company (1978, 1982) ExxONn ComPANny (1977 to date) Lots S. FOGELSANGER (1966) GuLF O1L CORPORATION (1978) MERRILL W. Haas (1975) Rosert C. HOERLE (1974-77) RICHARD I. JOHNSON (1967) J. M. McDoNALD FOUNDATION (1972, 1978) Mobsit O1t CORPORATION (1977 to date) SAMUEL T. PEEs (1981) RICHARD E. Petit (1983) Rosert A. PoHowsky (1982) Texaco, INc. (1978, 1982) UNION OIL OF CALIFORNIA (1982 to date) UNITED STATES STEEL FOUNDATION (1976) CHARLES G. VENTRESS (1983 to date) CHRISTINE C. WAKELEY (1976-1984) NorMAN E. WEISBORD (1983) INDUSTRIAL SUBSCRIBERS (1986) ($250 per annum) EXXON PRODUCTION RESEARCH COMPANY Mosit EXPLORATION AND PRODUCING SERVICES SHELL DEVELOPMENT COMPANY (continued overleaf) R. TUCKER ABBOTT JAMES E. ALLEN CHRISTINA L. BALK Ropert A. BLACK Bruce M. BELL Hans BOoLLi RutTH G. BROWNE JOHN L. CARTER ANNELIESE S. CASTER KENNETH E. CASTER JOHN E. DUPONT ARTHUR N. DUSENBURY, JR. R. H. FLOWER Lots S. FOGELSANGER A. EUGENE FRITSCHE Ernest H. GILMOUR MERRILL W. HAAS ANITA G. HARRIS STEVEN M. HERRICK Rosert C. HOERLE F. D. HOLLAND, JR. RICHARD I. JOHNSON DAvID B. JONES PETER JUNG Davip GARRETT KERR CeciL H. KINDLE Mary E. KINDLE LIFE MEMBERS ($200) WILLIAM F. Kose, II Jiri Kriz THORWALD KRUCKOW Hans G. KUGLER Ecsert G. LEIGH, JR. GERARD A. LENHARD DONALD R. MOORE SAKAE O’HARA SAMUEL T. PEES RICHARD E. PETIT Rospert A. POHOWSKY JOHN PoJETA, JR. DONALD E. RANSOM, JR. ANTHONY RESO ARTHUR W. ROCKER JOHN B. SAUNDERS JUDITH SCHIEBOUT MIRIAM W. SCHRINER Davip H. STANSBERY CHARLES G. VENTRESS EmiLy H. VoKES HAROLD E. VOKES CHRISTINE C. WAKELEY NORMAN E. WEISBORD RALPH H. WILLOUGHBY ARMOUR C. WINSLOW Victor A. ZULLO Membership dues, subscriptions, and contributions are all important sources of funding, and allow the Paleontological Research Institution to continue its existing programs and services. The P.R.1. publishes two series of respected paleontological monographs, Bulletins of American Paleontology and Palaeontographica Americana, that give authors a relatively inexpensive outlet for the publication of significant longer manuscripts. In addition, it reprints rare but important older works from the pa- leontological literature. The P.R.I. headquarters in Ithaca, New York, houses a collection of inver- tebrate type and figured specimens, among the five largest in North America; an extensive collection of well-documented and curated fossil specimens that can form the basis for significant future pa- leontologic research; and a comprehensive paleontological research library. The P.R.I. wants to grow, so that it can make additional services available to professional paleontologists, and maintain its position as a leader in providing Resources for Paleontologic Research. The Paleontological Research Institution is a non-profit, non-private corporation, and all contri- butions are U.S. income tax deductible. For more information on P.R.I. programs, memberships, or subscriptions to P.R.I. publications, call or write: Peter R. Hoover Director Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850, U.S.A. 607-273-6623 | Paleontographica Americana NUMBER 55 MARCH 18, 1986 A Restudy of the Fossil Scorpionida of the World by Enk N. Kjellesvig-Waering Organized for Publication by Anneliese S. Caster and Kenneth E. Caster Paleontological Research Institution 1259 Trumansburg Road | Ithaca, New York, 14850 U.S.A. Library of Congress Card Number: 86-60462 AUTHOR’S DEDICATION for my wife Virginia Printed in the United States of America Allen Press, Inc. Lawrence, KS 66044, U.S.A. Erik Norman Kjellesvig-Waering, 1912-1979 VITA Erik Norman Kjellesvig-Waering was born in Habana, Cuba on February 9, 1912; his mother being Cuban and father Norwegian made him a Norwegian citizen by law, a ““Cubegian”’ by nature and at age 21 when he became an American citizen—a 1000% extremely patriotic American. His father was a civil engineer who paved the streets of Habana with cobble stones from Norway. He was one of six children, all educated in the United States at their mother’s insistence. Commuting from Cuba, he received his elementary education in New Jersey from the age of eight; attended High School in a military academy in North Carolina and received his Bachelor of Science Degree in Geology from the University of North Carolina. His father sat him on a stump in a little woods in New Jersey at the age of 5, and told him to be a geologist. He said he would either be the best or the worst, no matter what he chose, as Erik was never mediocre or neutral or middle of the road, even at that tender age. While working on his degree at the University, he supplemented his income with employment at the University library. During private research in connection with his geology courses, he became fascinated with paleontology, in fact so infatuated that during the remainder of his life, he devoted all his free time including vacations, to the study of paleontology, specifically eurypterids and living and fossil scorpions. He was a man of many talents; strong, of sharp mind, quick decisions, and deep determination in all things he deemed to be right, a lover and student of history, biology, and ecology, and a champion of the underdog, with a flamboyant and energetic personality, punctuated always with a wonderful wit and sense of humor. Over the period of his career he maintained an executive position in Exploration with various oil companies including DeGolyer MacNaughton, Inc. for whom he worked three years in Mexico; Helmerich & Payne, Inc. and Shell Oil Company in Oklahoma and Texas. He spent 23 years living and working in Cuba, Jamaica, Guatemala, Antigua, Venezuela, Argentina, Trinidad, Norway and Mozambique for Amoco International Oil Company. During this time he collected various turtles, snakes, shellfish, coral and other specimens for the American Museum of Natural History in New York and the Field Museum of Natural History in Chicago, being a member of both. He owned one of the largest and most varied private collections of Scorpionida in the world, which now rests with his extensive collection of Coleoptera and eurypterids in the Florida State Collection of Arthropods. He spent 20 years writing this monograph while figuring prominently in the field of geology with the oil industry and writing many other papers pertaining to eurypterids and fossil and living scorpions, gathering a great deal of pertinent material in person in museums and universities behind the Iron Curtain, Europe, the Caribbean, Central, South and North America, Southeast Asia and Africa. All descriptions and figures were obtained from the actual specimens either loaned to him by universities, museums and individuals, or collected by him in his extensive travels. During his fatal illness, he tried to finish this work and had just completed the figures and manuscript, but did not have time to organize the latter for publication. It was his request that his dear friend Dr. Kenneth E. Caster be the one to categorize and join together his last great work. At the time of his death, July 16, 1979, he was living in Marco Island, Florida where he had retired from the oil business. He left his wife Virginia, two sons, Richard and Ronald and two grandchildren, Kristen and Keith. Infraorder Lobosternina.............. InfraordemHolosteminam secesciceieects Infraorder Meristosterminaiaa.se ses esse cee ee me Infraorder, Bilobosterminaa. cess sates soe elena: InfraordenOrthosternina’eeeisee as cite ee eieiee MAXKODASCS epee sche ie vesiechecie oasis D apes Wereeaye Systematic Paleontology Order Scorpionida Latreille Suborder Branchioscorpionina, new suborder......... Morphological Abbreviations used in Text-figures....... CONTENTS Page PADSUTA Cara arash eer at ae os siate Syatnieten dudes, Hevea aces 9 Compilersiintroduction crc. rie easiness eo 9 @CompilerseAcknowledementsic;a--aaasise ier ee 10 Editomsibretaces tnt etree erin eects 1] Characteristics and Nature of Fossil Scorpions AvGeneraliStatementirrscerisrseareereictes crises eyele ert 11 GCATADACE Retreat Re feyshaente. hepsrho tetera es ela tetelet? 12 VCS meee ee ea ernie ore craic ucts ekeihansiecehs cmersiare ce ensiesier 12 (CoxosternallRepi ome qayas caters arexeseie (sie seen si sleseisysie esos cee arene 15 ProsomalvAppendageseereee eee cence ieee 16 Operculumiee ee wh a ats aCe us seis spe nelomeeryers. ue 19 IRECUIMES Herero tara ory ernioies eee rece cra aeeyeleneredersverseasaee rene 19 rea bdomengac errr cee ctesin cers icksterne eae erento eens 20 (Gai sb We a ene ee oo DESO DES CCR cI One rae ae 21 Ornamien tation eyacc cers icreytsytsvsveveag aie atelever see cpeyeeeayecs, nieve 21 The Ventral Side of the Bilobosternina, Lobosternina, Holosternina, and Meristosternina........... 21 repectinall Plate water rcret-c-cstes cheveuare eo vsiesefeccrsre cleus. teens eters 22 Sternites and Abdominal Plates ....... 23 DirectoniofEvolution’s case as eois- ce caseciee ces 24 Breathing MechanismSyprrtic circuses streets dere stareias sera 25 Comparison of Scorpions and Eurypterids.. . 26 Evolutionary Changes in Scorpions ...................... 27 MINS MTAASSICHSCOLPION So eete cree eses oy craca fo. ove ielese suede. ofens tutves tots 28 Classification ANtLOGUCH ON Geert ee seer oa terrciavsteseceetotesrs ieee Bp exe cainie eet 34 Abbreviations of Repositories........ foldout inside front cover chat OPER aT RT Pn ae Oe foldout inside front cover Infraorder Holosternina, new infraorder............ Superfamily Proscorpioidea Scudder............. Family Proscorpiidae Scudder ................ Genus Proscorpius Whitfield ................ Proscorpius osborni (Whitfield) ............ Genus Archaeophonus Kjellesvig-Waering ... . Archaeophonus eurypteroides Kjellesvig- IWaering. sacs ricenocu ra payers le Gusucternnae Family Waeringoscorpionidae Stormer......... Genus Waeringoscorpio Stormer ............ Waeringoscorpio hefteri Stormer .......... Family Labriscorpionidae, new family ......... Genus@abriscorpiowWeary, see sees eee = Labriscorpio alliedensis Leary ............. Superfamily Stoermeroscorpionoidea, new Superfamily: arsine ae ctu ecaererensis eek tees senerere = Family Stoermeroscorpionidae, new family ..... Genus Stoermeroscorpio, new genus ......... Stoermeroscorpio delicatus, new species ... . Superfamily Allopalaeophonoidea, new supertamilyerr erat eae ee Family Allopalaeophonidae, new family 35 Genus Allopalaeophonus, new genus ...... Allopalaeophonus caledonicus (Hunter) . . . Superfamily Palaeoscorpioidea Lehmann Family Palaeoscorpiidae Lehmann ..... Genus Palaeoscorpius Lehmann.......... Palaeoscorpius devonicus Lehmann .... . Family Hydroscorpiidae, new family ... Genus Hydroscorpius, new genus ............ Hydroscorpius denisoni, new species ....... Superfamily Archaeoctonoidea Petrunkevitch..... Family Archaeoctonidae Petrunkevitch ........ Genus Archaeoctonus Pocock ............. Archaeoctonus glaber (Peach) ............. Genus Pseudoarchaeoctonus, new genus...... Pseudoarchaeoctonus denticulatus, new SPEGIEStare eee ieee Superfamily Acanthoscorpionoidea, new superfamily Family Acanthoscorpionidae, new family ..... Genus Acanthoscorpio, new genus Acanthoscorpio mucronatus, New species ... Family Stenoscorpionidae, new family ......... Genus Stenoscorpio, new genus ............ Stenoscorpio gracilis (Wills).............- Stenoscorpio pseudogracilis (Wills)......... Superfamily Gigantoscorpionoidea, new SUPEMAMILY, see cyare ce cuectsemensicis cketasr inieneatets Family Gigantoscorpionidae, new family .... . Genus Gigantoscorpio Stormer .............. Gigantoscorpio willsi Stormer ............. Superfamily Mesophonoidea Wills rete Family Mesophonidae Wills .................. Genus Mesophonus Wills.................-- Mesophonus perornatus Wills ..........-. Mesophonus (?) pulcherrimus Wills ........ Mesophonus (?) pulcherrimus var. iemmaculatussWaillsya deca css Mesophonus (?) maculatus (Brauer, Redtenbacher, and Ganglbauer) ....... Family Mazoniidae Petrunkevitch............. Genus Mazonia Meek and Worthen ......... Mazonia woodiana Meek and Worthen .... Mazonia wardingleyi (Woodward) ......... Family Centromachidae Petrunkevitch......... Genus Centromachus Thorell and Lindstrom Centromachus euglyptus (Peach)........... Family Heloscorpionidae, new family.......... Genus Heloscorpio, new genus .............. Heloscorpio sutcliffei (Woodward) ......... Family Liassoscorpionidae, new family ....... Genus Liassoscorpionides Bode ............. Liassoscorpionides schmidti Bode.......... Family Phoxiscorpionidae, new family......... Genus Phoxiscorpio, new genus ............. Phoxiscorpio peachi, new species ... Family Willsiscorpionidae, new family... .. Genus Willsiscorpio, new genus .........-... Willsiscorpio bromsgroviensis (Wills) . Superfamily Eoctonoidea, new superfamily ... . Family Eoctonidae, new family ..... Genus Eoctonus Petrunkevitch....... 56 By/ 61 61 61 61 63 63 63 66 66 66 66 68 68 70 70 70 70 73 73 74 74 75 76 76 76 79 79 79 80 80 80 81 81 81 81 84 84 86 86 86 86 88 90 90 90 93 93 93 94 94 94 94 96 96 Eoctonus miniatus Petrunkevitch.......... Family Buthiscorpiidae, new family ........... Genus Buthiscorpius Petrunkevitch .......... Buthiscorpius buthiformis (Pocock) ........ Buthiscorpius lemayi, new species ......... Family Allobuthiscorpiidae, new family........ Genus Allobuthiscorpius, new genus ......... Allobuthiscorpius major (Wills) ..........-. Genus Aspiscorpio, new genus..............- Aspiscorpio eagari, new species............ Family Anthracoscorpionidae, new family...... Genus Anthracoscorpio KuSta.............-. Anthracoscorpio juvenis KuSta............- Anthracoscorpio dunlopi Pocock ..........- Anthracoscorpio (?) species...........----. Genus Lichnoscorpius Petrunkevitch......... Lichnoscorpius minutus Petrunkevitch ..... Genus Allobuthus, new genus ............... Allobuthus macrostethus, new species ...... Genus Coseleyscorpio, new genus............ Coseleyscorpio lanceolatus, new species .... Family Garnettiidae Dubinin ..............:.. Genus Garnettius Petrunkevitch............. Garnettius hungerfordi (Elias) ...........-- Garnettiusi(?)\ SPOClES «ser cue «ssi cicicteusss ere st eis Superfamily Spongiophonoidea, new superfamily. . . Family Spongiophonidae, new family .......... Genus Spongiophonus Wills ..............-- Spongiophonus pustulosus Wills ........... Family Praearcturidae, new family ............ Genus Praearcturus Woodward ............. Praearcturus gigas Woodward............. Genus Brontoscorpio Kjellesvig-Waering Brontoscorpio anglicus Kjellesvig-Waering . . Infraorder Meristosternina, new infraorder ......... Superfamily Tiphoscorpionoidea, new superfamily . . Family Tiphoscorpionidae, new family......... Genus Tiphoscorpio, new genus ............. Tiphoscorpio hueberi, new species ......... Superfamily Cyclophthalmoidea Thorell and ANAS th OMe rs pegiee raceane, eee ae eeeenpe Family Cyclophthalmidae Thorell and Lindstrom Genus Cyclophthalmus Corda............... Cyclophthalmus senior Corda ...........-- Clyclophthalmus robustus, new species .... . Cyclophthalmus (?) sibiricus Novojilov and STSTMER cas aos. neeiecisee Sake eee Family Microlabiidae, new family............. Genus Microlabisi@orda. «crc ncatecaas eee Microlabis sternbergii Corda .............. Superfamily Palaeobuthoidea, new superfamily ... Family Palaeobuthidae, new family............ Genus Palaeobuthus Petrunkevitch .......... Palaeobuthus distinctus Petrunkevitch...... Infraorder Lobosternina Pocock (restricted)......... Superfamily Palaeophonoidea Thorell and INGSEPOM Yel 20s, ocistate & sucitisiars, slorigsnese may rome ets Family Palaeophonidae Thorell and Lindstrém Genus Palaeophonus Thorell and Lindstrom . . Palaeophonus nuncius Thorell and Mein GStLOM.. op areraecs sh eeeesraiete crs erons eree Palaeophonus (?) lightbodyi Kjellesvig- WA OLIN gfx o crcvors cis 4,0) aeevelly sutuh al srenestisocdsunns Superfamily Anthracochaeriloidea, new superfamily .. to tio NM tv Family Anthracochaerilidae, new family ....... Genus Anthracochaerilus, new genus......... Anthracochaerilus palustris, new species... . Superfamily Isobuthoidea Petrunkevitch ......... Family Isobuthidae Petrunkevitch............. Genus [sobuthushricwe ee. ees tate Isobuthus kralupensis (Thorell and Lindstrom) rere ee cee eee Genus Boreoscorpio, new genus ............. Boreoscorpio copelandi, new species ....... Genus Feistrnantelia Frié................-5- Feistmantelia ornata Frié..............--- Genus Bromsgroviscorpio, new genus ........ Bromsgroviscorpio willsi, new species ...... Family Eobuthidae, new family ............... Genus) BobuthusetiGae- aces eeeeeee ert Eobuthus rakovnicensis Frié ............-- Eobuthus cordai, new species ............. Eobuthus holtiiPocock 7)... a2 tte Eobuthus (?) species of Wills.............- Family Eoscorpiidae Scudder ................. Genus Eoscorpius Meek and Worthen ....... Eoscorpius carbonarius Meek and Worthen Eoscorpius pulcher (Petrunkevitch) ........ Eoscorpius mucronatus, new species ....... Eoscorpius distinctus (Petrunkevitch) ...... Eoscorpius sparthensis Baldwin and Sutcliffe IE OSCONDIUS SPECIES ayc uence pertinent ee Eoscorpius casei, new species ...........-- Genus Trachyscorpio, new genus ..........-- Trachyscorpio squarrosus, new species ..... Trachyscorpio:(2)(SPEClES ss)... 141212 selenite Genus Eskiscorpio, new genus ............-. Eskiscorpio parvus, new species .........-- Family Pareobuthidae, new family ............ Genus Pareobuthus Wills ........-..-..-+.+: Pareobuthus salopiensis Wills ...........-- Family Kronoscorpionidae, new family ........ Genus Kronoscorpio, new genus ............-- Kronoscorpio danielsi (Petrunkevitch) ..... . Superfamily Paraisobuthoidea, new superfamily. . . Family Paraisobuthidae, new family ........... Genus Paraisobuthus, new genus ............ Paraisobuthus prantli, new species......... Paraisobuthus frici, new species ........... Paraisobuthus duobicarinatus, new species. . . Paraisobuthus virginiae, new species ....... Genus Leioscorpio, new genus .............- Leioscorpio pseudobuthiformis, new species Family Telmatoscorpionidae, new family ...... Genus Te/lmatoscorpio, new genus........... Telmatoscorpio brevipectus, New species .. . . Family Scoloposcorpionidae, new family ....... Genus Scoloposcorpio, new genus ..........- Scoloposcorpio cramondensis, new species .. . Genus Benniescorpio Wills ..............++- Benniescorpio tuberculatus (Peach)......... Family Opsieobuthidae, new family ........... Genus Opsieobuthus, new genus............. Opsieobuthus pottsvillensis (Moore) ........ Superfamily Loboarchaeoctonoidea, new ile suri pmgan do coonigadconanocecuao nds Family Loboarchaeoctonidae, new family ...... Genus Loboarchaeoctonus, new genus........ Loboarchaeoctonus squamosus, new species Superfamily Pseudobuthiscorpioidea, new Superiamilyee ny. wis sone eeiek nicer eee Family Pseudobuthiscorpiidae, new family ..... Genus Pseudobuthiscorpius, new genus....... Pseudobuthiscorpius labiosus, new species .. . Family Petaloscorpionidae, new family Genus Petaloscorpio, new genus ............. Petaloscorpio bureaui, new species......... Family Waterstoniidae, new family............ Genus Waterstonia, new genus.............. Waterstonia airdriensis, new species ....... Waterstonia (?) brachistodactyla, new species Infraorder Bilobosternina, new infraorder Superfamily Branchioscorpionoidea, new Supertamil yarn serscescrcnspcuveverers resis aecec, © cue overuse Family Branchioscorpionidae, new family ...... Genus Branchioscorpio, new genus Branchioscorpio richardsoni, new species .. . Family Dolichophonidae Petrunkevitch........ Genus Dolichophonus Petrunkevitch......... Dolichophonus loudonensis (Laurie)... .. . Suborder Neoscorpionina Thorell and Lindstrém Infraorder Orthosternina Pocock Family Palaeopisthacanthidae, new family ..... Genus Palaeopisthacanthus Petrunkevitch ... . Palaeopisthacanthus schucherti Petrunkevitch Genus Compsoscorpius Petrunkevitch........ Compsoscorpius elegans Petrunkevitch ..... Family Scorpionidae Leach ................... Genus Mioscorpio, new genus............... Mtoscorpio zeuneri (Hadzi) Genera Incertae Sedis Genus Titanoscorpio, new genus ............ Titanoscorpio douglassi, new species ....... GenusilWaitisonia Willsivaee eee eee ee Wattisonia coseleyensis Wills ............. Genus Palaeomachus Pocock ............... Palaeomachus anglicus (Woodward) ....... Scorpionida genus and species indeterminate of Stormer Appendix 1. The Stratigraphic Distribution of the Fossil SCOLPIONIG ats eee rt eae ere eee Appendix 22 Index of Synonyms 95..s22¢420-5eee eee Appendix 3. Index of Specimens by Repository ........... References Cited Plates Index LIST OF ILLUSTRATIONS Text-figure i, Dr . Proscorpius osborni (Whitfield). Phylogenetic lines of development in the Scorpionida A comparison of the ventral morphology of the Euryp- terida and three infraorders of the Scorpionida....... . External morphology of scorpions .................. Ventral plating in scorpionids, illustrating the basic dif- ferences;betweenlinfraorders,: «2. 5+ -leeras + s2-2 = = . Key to the classification of the Scorpionida, showing the ..foldout inside back Specimen I, BMS ONG Simca coe to ED CTIOR EU EN SOC eC oeee relationship of the higher taxa .. . Proscorpius osborni (Whitfield). Specimen II, CIURCA 041771-1 . Proscorpius osborni (Whitfield). Specimen III, CIURCA OF OS G4 mcr emcee ter eich yar riche stasis wnat ee . Proscorpius osborni (Whitfield). Specimen III, CIURCA O4 0564s ee ia seo setae areaeonstere tio srecide oredr ses . Proscorpius osborni (Whitfield). Specimen IV, CIURCA OF2869-O Bice etter erro sete .sckel aa acweicts seca . Proscorpius osborni (Whitfield). Specimens V and VI, CIURCA 042570-1A and 040668-1 . Proscorpius osborni (Whitfield). Specimens VI and I, CIURCA 040668 and BMS E25162................ . Proscorpius osborni (Whitfield). Reconstructions based on specimens from the Ciurca collection . Archaeophonus eurypteroides Kjellesvig-Waering. CMI RCAVOG 20651 ois paratype;\GSE S862... 2 c)eprcss etree erst . Opsieobuthus pottsvillensis (Moore). Holotype, FMNH (OQ) 30984 2 Neate rite cries ote atte ieee rete . Loboarchaeoctonus squamosus, n. gen., n. sp. Holotype, BM(NH) 1.988; paratype, BM(NH) In.12848 . Pseudobuthiscorpius labiosus, n. gen., n. sp. Holotype, BM(NE)IN1S552 0 ee, ee . Petaloscorpio bureaui, n. gen., n. sp. Holotype, UL 1092 . Waterstonia airdriensis, a. gen., n. sp. Holotype, RSM 1957.1.4996 Waterstonia (?) brachistodactyla, n. sp. Holotype, RSM 1978.5.1 Branchioscorpio richardsoni, n. gen., n. sp. Holotype, BEMNBEI(RE)(6175aiDeecn.c cas ce ee ce nee erier Dolichophonus loudonensis (Laurie). Holotype, RSM 1897.32.196 Palaeopisthacanthus schucherti Petrunkevitch. Holotype; YIPMe V4 00 ae eico.cers cone a aisle nseyavencdstercteeetetete Palaeopisthacanthus schucherti Petrunkevitch. Holotype: YIRM! 14 Oa ecratersess steno pier er een karte Compsoscorpius elegans Petrunkevitch. Holotype, IBM(NED)T.7 883) ce aed cate acme eters ere Compsoscorpius elegans Petrunkevitch. Paratype, BM(NED ansl'S 862) eeces foe cere eee ereenee Compsoscorpius elegans Petrunkevitch. Paratype, BMONED im 2302 615s eercrcc cece tte nels cece tetany cette Titanoscorpio douglassi, n. gen., n. sp. Holotype ..... Palaeomachus anglicus (Woodward). Holotype, BM(NH) DQ AA aD ies sertrecrete roe toe ptae crate ee eee eee erence Coxosternal areas of the Holosternina Coxosternal areas of the five infraorders Coxosternal areas of the Lobosternina .............. Morphological stages, but not necessarily phylogenetic development, of the floor of the oral tube ........... Coxosternal region and oral tube condition of some Recent scorpions, introduced for comparison ........ 239 240 242 243 243 243 243 A RESTUDY OF THE FOSSIL SCORPIONIDA OF THE WORLD By ERIK N. KJELLESVIG-WAERING Organized for Publication by Anneliese S. Caster and Kenneth E. Caster ABSTRACT This study, based on personal examination of all available specimens in the world, many of them new, ranging in age from the Silunan through the Jurassic, plus a lone specimen from the Miocene, comprises a reclassification of the Scorpionida into two suborders: The Branchioscorpionina, mainly aquatic and possessing gills; and the Neoscorpionina, terrestrial and pulmonate. The five infraorders are based on the nature of the abdominal plates or sternites. These are the Lobosternina, restricted, the Orthosternina, restricted, the Bilobosternina, new, the Holosternina, new, and the Meristosternina, new. The 21 superfamilies, 13 of which are new, are based on the general coxosternal pattern. The 48 families, 34 of which are new, are based on the maxillary lobes, shape of sternum, presence or absence of lateral compound eyes, and the termination of the legs. The following new taxa are described: Labriscorpionidae, n. fam.; Stoermeroscorpionoidea, n. superfam., Stoermeroscorpionidae, n. fam., Stoermeroscorpio delicatus, n. gen., n. sp.; Allopalaeophonoidea, n. superfam., Allopalaeophonidae, n. fam., A//opalaeo- phonus, n. gen.; Hydroscorpiidae, n. fam., Hydroscorpius denisoni, n. gen., n. sp.; Pseudoarchaeoctonus denticulatus, n. gen., n. sp.; Acanthoscorpionoidea, n. superfam., Acanthoscorpionidae, n. fam., Acanthoscorpio mucronatus, N. gen., n. sp.; Stenoscor- pionidae, n. fam., Stenoscorpio, n. gen.; Gigantoscorpionoidea, n. superfam., Gigantoscorpionidae, n. fam.; Heloscorpionidae, n. fam., Heloscorpio, n. gen.; Liassoscorpionidae, n. fam.; Phoxiscorpionidae, n. fam., Phoxiscorpio peachi, n. gen., n. sp.; Willsiscorpionidae, n. fam., Willsiscorpio, n. gen.; Eoctonoidea, n. superfam., Eoctonidae, n. fam; Buthiscorpiidae, n. fam., Buthiscorpius lemayi, n. sp.; Allobuthiscorptidae, n. fam., Al/obuthiscorpius, n. gen.; Aspiscorpio eagari, n. gen., n. sp.; Anthra- coscorpionidae, n. fam., Anthracoscorpio ? sp.; Allobuthus macrostethus, n. gen., n. sp.; Coseleyscorpio lanceolatus, n. gen., n. sp.; Garnettius ? sp.; Spongiophonoidea, n. superfam., Spongiophonidae, n. fam.; Praearcturidae, n. fam.; Tiphoscorpionoidea, n. superfam., Tiphoscorpionidae, n. fam., Tiphoscorpio hueberi, n. gen., n. sp.; Cyclophthalmus robustus, n. sp.; Microlabiidae, n. fam.; Palaeobuthoidea, n. superfam., Palaeobuthidae, n. fam.; Anthracochaeriloidea, n. superfam., Anthracochaerilidae, n. fam., Anthracochaerilus palustris, n. gen., n. sp.; Boreoscorpio copelandi, n. gen., n. sp.; Bromsgroviscorpio willsi, n. gen., n. sp.; Eobuthidae, n. fam., Eobuthus cordai, n. sp.; Eoscorpius mucronatus, n. sp., E. casei, n. sp.; Trachyscorpio squarrosus, n. gen., n. sp., Trachyscorpio (?) sp.; Eskiscorpio parvus, n. gen., n. sp.; Pareobuthidae, n. fam.; Kronoscorpionidae, n. fam., Kronoscorpio, n. gen.; Paraisobuthoidea, n. superfam., Paraisobuthidae, n. fam., Paraisobuthus prantli, n. gen., n. sp., P. frici, n. sp., P. duobicarinatus, n. sp., P. virginiae n. sp.; Leioscorpio pseudobuthiformis, n. gen., n. sp.; Telmatoscorpionidae, n. fam., Te/ma- toscorpio brevipectus, n. gen., n. sp.; Scoloposcorpionidae, n. fam., Scoloposcorpio cramondensis, n. gen., n. sp.; Opsieobuthidae, n. fam., Opsieobuthus, n. gen., Loboarchaeoctonoidea, n. superfam., Loboarchaeoctonidae, n. fam., Loboarchaeoctonus squa- mosus, N. gen., n. sp.; Pseudobuthiscorpioidea, n. superfam., Pseudobuthiscorpiidae, n. fam., Pseudobuthiscorpius labiosus, n. gen., n. sp.; Petaloscorpionidae, n. fam., Petaloscorpio bureaui, n. gen., n. sp.,; Waterstoniidae, n. fam., Waterstonia airdriensis, n. gen., n. sp., W. (?) brachistodactyla, n. sp.; Branchioscorpionoidea, n. superfam., Branchioscorpionidae, n. fam., Branchioscorpio richardsoni, n. gen., n. sp.; Palaeopisthacanthidae, n. fam.; Mioscorpio, n. gen., Titanoscorpio douglassi, n. gen., n. sp. COMPILER’S INTRODUCTION This restudy of the fossil scorpions of the world was undertaken by the late Erik N. Kjellesvig- Waering about twenty years ago as a by-product of his many years of devoted study of the Eurypterida, to which the scor- pions are closely related, and his passion for collecting living scorpions, dating from his boyhood in Cuba. It was begun largely because of his dissatisfaction with the analyses of fossil scorpions in the many works of the late Alexander Petrunkevitch of Yale University. This was capped by Petrunkevitch’s treatment of the group in the Treatise on Invertebrate Paleontology (1955). When Waering died in 1979, the research for the scorpion manuscript was largely completed, but left in a chaotic state. He undertook the work genus by genus and specimen by specimen, as he could obtain the type materials from various collections in Europe and America, with no small supplement of new materials that came to hand over the years from the same sources and from amateur collectors. His modus operandi was to describe and illustrate the material seriatim, ex- pecting to integrate the manuscript in a final reworking, which alas, he never lived to accomplish, although he left a large volume of substantive notes, and many memoranda to himself (““don’t forget to . . .””), as well as marginal annotations in the published source ma- 10 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 terials. All of this material has been available to us as we have endeavored to fulfill his desires as best we could judge them. Unfortunately, over the long period in which he worked on fossil scorpions, he underwent several changes of mind on morphological and taxo- nomic matters, without enough time spared him to reconcile his various opinions. Throughout the man- uscript, we have followed the general policy of em- ploying what seemed to be his last views on all these matters. This has entailed much textual reorganization and reconciliation, together with some relabelling of drawings and general coordination of the material at hand. The resultant manuscript gives Waering’s con- sidered opinion about fossil scorpions, and any emen- dations or clarifications made by the organizers are clearly indicated. Waering had several times discussed the scorpion manuscript with us during visits in both Cincinnati and Marco Island, Florida. He and I had been long- time collaborators on eurypterid studies and I had agreed to look over the final scorpion manuscript be- fore submitting it for publication. His final instructions to his wife were to submit the manuscript to me to render into publishable form. As a result, in‘the sum- mer of 1980, Mrs. Caster and I spent a month with Mrs. Waering, collecting his papers, returning bor- rowed specimens, and finally removing all of his fossil scorpion papers and library, rubber molds and micro- films, to the Geology Department of the University of Cincinnati, where they are deposited. Immediately upon returning to Cincinnati after the intensive weeks in Florida, I suffered a temporary dis- ability and found myself unable to continue with the task in hand. Fortunately for me, as well as for the project, my wife, Anneliese S. Caster, like myself, a former student in paleontology under Gilbert D. Harris of Cornell University, and who had been involved in the endeavor from the start, took over, and is largely responsible for the completed task. This has been al- most full-time occupation for more than three years, and has entailed reconciliation of various sections of the manuscript, elimination of duplication, consider- able literature research, drafting of supplementary charts and figures and, in brief, coming up with a manuscript of which Erik Waering would have been proud. This monograph is the culminating and major work of a very fine paleontologist, but it was quite unacceptable for publication before the devoted efforts of Anneliese S. Caster. We are both grateful to our late professor, G. D. Harris, and for the years spent in his laboratory, where, despite the absence of scorpions on the agenda in that mollusk-oriented environment, all students were treated as “‘zealous companions in re- search.” I myself am particularly grateful for my wife’s assistance, but so must all scorpion specialists be, for without her labors Erik Waering’s final work would never have been published. As she now says, this task has caused her to know far more about scorpions than she ever thought she wanted to know! Kenneth E. Caster Emeritus Professor of Geology University of Cincinnati Cincinnati, OH 45221, U.S.A. COMPILERS’ ACKNOWLEDGMENTS Waering left a partial list of those to whom he was grateful for assistance and loans of specimens over the many years of the project: we have gleaned from his text the names of others to whom he was so indebted. Any not formally acknowledged below will understand that their omission was unintentional. First, Erik would have been, and we deeply are, in- debted to Mr. John Swearingen, Chairman of the Board of Amoco Production Company, and the Amoco Foundation, Inc. of Standard Oil Company (of In- diana) for a generous grant to the University of Cin- cinnati Foundation, which makes the costly publica- tion of the manuscript by the Paleontological Research Institution possible. Erik served as a Vice-President of Amoco for several years, and this grant is a tribute to him, both as a scientist and as a valued employee. It is enlightened recognition of his long years of “‘extra- curricular” activity in pure scientific research from Trinidad to Norway, the while fulfilling his contractual responsibilities with distinction. Special thanks are due the late Dr. Leif Stormer of Oslo, Norway, who was a constant source of advice and information, and who supplied the X-ray photo- graphs of Palaeoscorpius devonicus made by Lehmann; and also to the late Dr. R. C. Moore of Lawrence, Kansas, for considerable advice on taxonomic matters. Many European colleagues have gone out of their way to arrange study facilities and loans of types, also making casts of specimens, further preparing speci- mens on request, and furnishing locality information. Special mention must be made of the assistance of Dr. H. W. Ball and Dr. A. E. Rixon of the British Museum (Natural History); Dr. Peter Brand, Dr. Adrian Rush- ton and Dr. R. B. Wilson of the Institute of Geological Sciences, London; Dr. Michael Eagar of the University of Manchester Museum; Dr. Harry B. Whittington and Dr. D. Price of the Sedgwick Museum, Cambridge; Dr. W. D. Ian Rolfe of the Hunterian Museum, Glasgow; Dr. Charles D. Waterston of the Royal Scottish Mu- seum, Edinburgh; Dr. F. W. Shotten at Birmingham University; and the Trustees of the Kilmarnock Mu- FossiIL SCORPIONIDA: KJELLESVIG- WAERING 11 seum. On the Continent, Dr. Radvan Horny, Dr. Ivo Chlupaé and Dr. J. Setlik of the National Museum, Prague, have shown every courtesy, as have Dr. H. K. Erben of Bonn, Dr. H. Jahnke of Géttingen, Dr. Jovan Hadzi, and Dr. M. Warth of the Staatliches Museum fiir Naturkunde, Stuttgart; likewise Mr. R. Skoglund of the Riksmuséet, Stockholm. Closer to home, special appreciation goes to Drs. G. Arthur Cooper, Richard Boardman, Frederick J. Col- lier, Francis M. Hueber, and Porter M. Kier of the United States National Museum and the Smithsonian Institution, for many courtesies shown. Through Drs. Eugene S. Richardson, Jr., Robert H. Denison, Mat- thew H. Nitecki, and Robert Hansman of the Field Museum of Natural History, Chicago, the Cottonwood Canyon material became available; they were also an invaluable source of information on the scorpions of the Mazon Creek area. Ms. Clarita M. Nunez of the Field Museum and Dr. Derek Briggs, visiting scientist from the University of London, and Mr. Bret S. Beall at the University of Michigan Museum of Paleontol- ogy, located some missing types and supplied their accession numbers. Further thanks are due Dr. Karl Waage, Mrs. Virginia Starquist and Mrs. J. Lawson of Yale Peabody Museum; Dr. John W. Wells of Cornell University; Dr. Donald Baird and Gerard R. Case of Princeton University; Dr. Daniel Blake and Dr. Harold W. Scott of the University of Illinois Museum; Dr. Richard L. Leary of the Illinois State Museum, Spring- field; Dr. Murray Copeland, University of Toronto, and Dr. René Bureau, Laval University, Québec. Almost no paleontological research is complete without the cooperation of the private collector, and this monograph is no exception. The following have been most generous in lending their specimens for study and description, and subsequently depositing them in museums; Mr. Samuel Ciurca, Mr. Ray Bandringa, Mr. Richard X. Cramer, Mr. David Lyon Douglass, Mr. Chris Durden, Ms. Bea Vogel, Mr. Jerry Herdina, Mr. Wm. Stephen LeMay, and Mrs. K. Taelle. For much assistance toward a better understanding of Recent scorpions, Drs. Fred Bennett, Oscar F. Francke, R. F. Lawrence, P. R. San Martin, Max Va- chon, and David Sissom were particularly helpful. Mr. Jorge E. Lopez Ricalo, draftsman with Amoco, who lettered many of the drawings for this paper, must not be forgotten. Most sincere thanks of all go to Waering’s wife, Vir- ginia, who kept a sick man happily engaged in his work, and to whom fell the drudgery of typing the various stages of this monograph. It was Waering’s wish that this work be dedicated to her in appreciation. Very special acknowledgment goes to Dr. Peter R. Hoover, Director of the Paleontological Research In- stitution for seeing the manuscript into print, and to Dr. Oscar F. Francke and Dr. W. D. Ian Rolfe for their careful review of the manuscript prior to publication. Because of the circumstance of publication of a post- humous paper, any suggested substantive alterations in the text were impossible, and subsequent reviews of the printed monograph must deal with these. Ob- viously, Waering’s treatment and assignments of some groups are open to debate, as is any work so revisionary and wide-ranging as this one. Alas, that Dr. Petrunke- vitch is no longer with us! Anneliese S. Caster 425 Riddle Road Cincinnati, OH 45220, U.S.A. EDITOR’S PREFACE The compilation and editing of the notes and por- tions of manuscript left to the Casters by the late Erik N. Kjellesvig-Waering has been a long, often frustrat- ing, but in the end immensely rewarding, process. We have tried, by initialling the footnotes inter- spersed through the text, to assign responsibility for this or that statement. Annie Caster (written commun., 1985) states, ““Footnotes are a problem. Where Wae- ring left no direct text for a species, I tried to follow the format that Petrunkevitch used in the Treatise, giving no more than synonymy, general statement, . . . occurrence, and specimen no., and I don’t feel that that needs a footnote.” In many ways, the preparation of a manuscript for posthumous publication is similar to the adventure games now so popular for small computers. The author has left much solid information, many clues, and not a few dead ends, and one must go back, and back again, with periods of reflection between, to sort it all out. The difference between the manuscript and the com- puter game is that there is a clear end to the game: in the manuscript, we must arbitrarily choose some suit- able point on the refinement rarefaction curve and let the reader take it from there. It’s in your court now. Peter R. Hoover Ithaca, NY CHARACTERISTICS AND NATURE OF FOSSIL SCORPIONS A GENERAL STATEMENT In the fossil record, what is essentially meant by fossil scorpions are the gill-bearing forms known in the Paleozoic and the Triassic. Except for two in the Ju- rassic, the rest of the Mesozoic has not definitely fur- nished any scorpions as yet, whereas the Tertiary has furnished only three species in Europe and two in North America. All Tertiary forms were pulmonate and lived on land. The Carboniferous Palaeopisthacanthus of the Mazon Creek area in Illinois and Compsoscorpius of the Coseley area of Britain are the only known pul- monate scorpions that lived in the Paleozoic. Both are monotypic, the former known only from a single spec- imen, the latter from three. All known British Triassic scorpions are gill-bearing. The gill-bearing scorpions did not survive into mod- ern time, but, fortunately, the relatively great number of specimens preserved was a direct result of their basic ecological setting in a water medium, where preser- vational factors were at an optimum. I suspect that even the Paleozoic pulmonates may have lived in water or in the vegetation above the water, as in a swamp forest, where again chances of preservation were high. The scarcity of Tertiary forms is probably a conse- quence of the fact that all were land-living and likely had developed the established characteristics of the modern scorpion, such as being cryptozoic and mainly solitary animals. The chances of this type of animal being preserved are, of course, much less, which is reflected by their rarity in the fossil record. The Ter- tiary scorpions are not found in normal sediments of aquatic origin, as were all Paleozoic forms, but in de- posits on dry land. For example, one, Scorpio schweig- geri Holl, 1829, Oligocene, is known from volcanic ash falls. Another, Mioscorpio zeuneri (Hadzi, 1931), Mio- cene (covered in this monograph), was preserved in calcareous deposits (travertine) accumulated from springs (one in a lake, the other on land). The other three (7ityus eogenus Menge, 1869, Baltic amber, Oli- gocene (?); Centruroides beynai Schawaller, 1979, Do- minican amber, Tertiary; Tityus ambarensis Scha- waller, 1982, Dominican amber, Tertiary) were preserved in gum from conifers (amber). All but one are known from a single specimen. Morphological differences are considerable between the lobosternous, holosternous, meristosternous and bilobosternous scorpions on one side and the ortho- sternous (which include all living forms), on the other. However, as early as the Middle Pennsylvanian, or- thosternous scorpions were present, and these have all the characteristics of the living families. In fact, in the case of Palaeopisthacanthus, it is close to the Chae- rilidae. The differences between fossil and living scorpions are as listed: CARAPACE The differences between the carapaces of most fossil scorpions and those of the living ones are immediately evident in the shape of the anterior margin and the position of the median eyes, as well as the presence of compound eyes in some. In fossil forms, the anterior 2 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 margin is almost always produced into a shovellike protrusion very similar to that found in eurypterids, which were undoubtedly water-dwellers. The purpose of this plowlike device is considered to be the same in both groups, a very handy digging device for conceal- ment in the mud (see Hughmilleria banksii Salter, Text- fig. 79C). Some living scorpions such as the troglobite Typhlochactas, a chactid, and Brachistosternus, a both- riurid, have a slightly protruding anterior, but this pro- trusion seems to serve to drape the carapace over the chelicerae, a secondary separation of the chelicerae, and should not be mistaken for the large glossate pro- cess of fossil scorpions and eurypterids. Other shapes of carapaces are known in water-dwelling scorpions, as, for example, Hydroscorpius denisoni. Here the car- apace is rounded and rather hemicircular in shape. With this peculiar shape, there is no comparison with living scorpions, as all living scorpions are roughly quadrate or rectangular. However, one of the euryp- terids, Parahughmilleria, has a similar rounded cara- pace, and of course, is an undoubted water animal. Nowhere in the fossil record are emarginate anterior margins, such as occur in Centruroides and Tityus (Buthidae), to be found, or deeply-lobed anterior mar- gins like those of nearly all burrowing scorpions, such as Broteochactas (Chactidae) or the genera Scorpio, Opisthacanthus and Opisthophthalmus (Scorpionidae). The only living scorpions that resemble the majority of the fossil forms in having a pointed anterior margin are the uniquely interesting Typhlochactas, a troglobite scorpion from the El Abra caves of Mexico (Mitchell, 1968) and Brachistosternus (Bothriuridae) from the Andean region of South America. With regard to the burrowing scorpions, it is interesting to note that the only obviously burrowing type found in the fossil state is Garnettius hungerfordi (Elias) from the Upper Penn- sylvanian of Kansas. This genus was armed with large serrated anterior legs remarkably like those of living burrowing beetles such as Copris, Phanaeus and Pi- notus. However, it was also furnished with a projecting anterior margin. Inasmuch as it definitely is holoster- nous, the scorpion seems to be adapted for digging in muds in an aquatic environment, perhaps like so many of the living crabs. EYES The median eyes of fossil scorpions are generally located on a prominent eye node, which protrudes above the level of the carapace and usually is located forward. In some Triassic scorpions, these median eyes are located further forward than in known Paleozoic forms (see Mesophonus in Wills, 1947, pl. 1, figs. 1, 2). Here the eyes and eye node actually become part of the anterior protrusion, and extend anterior to the FossIL SCORPIONIDA: KJELLESVIG- WAERING 13 rest (““shoulders’’) of the anterior margin. In others (Mazonia woodiana Meek and Worthen; see Kjelles- vig-Waering, 1969) the median eyes are very large, rounded, and located on the anterior part, but behind the level of the anterior margin. Large eyes, anteriorly located, but possibly elliptical in shape, are present in some (see Waterstonia airdriensis, Text-fig. 99A). Before proceeding further on a discussion of mor- phology of the lateral eyes in fossil scorpions, it is necessary to define the two main types, as confusion exists on this subject. In this connection, I am following the terminology and descriptions used by Harrington (1959, pp. O88-—O9 1) with regard to the Trilobita (both definitions by Harrington, Moore, and Stubblefield, 1959, pp. O121 and O126). Schizochroal eyes Eye with visual surface consisting of a number of biconvex lenses, rounded or polygonal in outline, each lens covered by individual cornea, and separated from others by sclerotic walls (syn. aggregate eye). Holochroal eyes Compound eye consisting of numerous adjoining planoconvex or biconvex lenses, covered by a continu- ous cornea (syn. compound eye). More significantly, with regard to the schizochroal or aggregate eyes, the eye groups consist of ocelli that do not differ in structure from simple eyes, and func- tion in the same way (Lawrence, 1953, p. 143); with regard to the holochroal or compound eyes, the cornea is formed by an aggregation of separate visual elements known as ommatidia, each ommatidium correspond- ing to a single facet of the cornea (Lawrence, 1953, p. 144). The lateral eyes in fossil scorpions comprise large, subelliptical, bulbous structures, generally placed at the anterolateral margins of the carapace. These have been considered by authors who have recognized them (Wills, 1947, p. 20; Kjellesvig-Waering, 1966, p. 365) as compound lateral eyes. Petrunkevitch (1953, p. 35) did not agree with Wills’ (1947, p. 20) determination of “compound facetted lateral eyes’, in the Triassic Mesophonus. However, it is obvious that Petrunke- vitch did not distinguish between compound and ag- gregate eyes as he follows with the statement, “I com- pared them with the compound eyes of fossil Trilobita, Xyphosura and Eurypterida and cannot find much similarity with these’. (Only the eyes of pterygotid eurypterids were then known in detail.) Petrunkevitch (1953, p. 35) denied that the lateral ‘““compound” eyes of Mesophonus were eyes and considered them organs of “unknown function, possibly chemoreceptors”. Al- though Petrunkevitch had numerous specimens avail- able to him that had these prominent, bulbous lateral eyes, he consistently either overlooked or denied their existence (“‘7yphloscorpius” distinctus, ‘‘Eoscorpius”’ danielsi, Eoscorpius carbonarius, “Lichnophthalmus” pulcher, Eoctonus miniatus, Buthiscorpius buthifor- mis, etc.). The eyes of Trilobita are both compound (holoch- roal) and aggregate (schizochroal). The Xiphosura, at least the living forms, have compound eyes, although of a very simple nature. The Eurypterida are divided into two suborders, and in the Eurypterina, as exemplified by the Upper Silu- rian Baltoeurypterus tetragonophthalmus (Fischer), the eye is composed of several thousand rounded ocelli, each being clear or transparent, whereas the surround- ing tissue is brown and opaque. On the inner dorsal edges of the lateral eye, each ocellus is surrounded by sclerotic walls as thick as the diameter of each. These ocelli become more crowded toward the median part of the eye, where for the most part they are obliterated. The ocelli definitely protrude over the rest of the sur- face of the eye, giving a “‘\pebbly” appearance to the cornea. This is based on several specimens from the Gerard Holm Collection at the Riksmuséet (Stock- holm) and represent chitinous skins that had been etched out of the limestone matrix by the use of acids. Wills (1965, p. 102) gives a good description of the lateral eye in the same species. The only point where I disagree is in the number of facets in the ““compound” eye. Wills states that they comprise ‘‘several hundred minute facets”, whereas I would estimate the number close to 5000. The eye in some Eurypterina is therefore compound, although in parts, for example as indicated by the thick sclerotic walls, it appears to be somewhat schizochroal, at least on the basis of definitions given above. We have no knowledge of the eye in other Eu- rypterina such as the superfamilies Mixopteroidea and Stylonuroidea. With regard to the Pterygotina, the eyes have been determined by Clarke and Ruedemann (1912, pp. 36- 42) as holochroal or compound and similar to the eyes of Limulus. The lateral eye of Baltoeurypterus is very different from that of Limulus, where the ommatidia are closely packed into hexagonal walls, all adjoining one another and covered by a thick, smooth cornea. In contrast, the lateral eye of the Eurypterina has the same covering skin with rounded, thickened, clear areas that correspond to the lenses and give the cornea a slightly pebbly appearance under high magnification and, in part at least, each visual unit is surrounded by thick sclerotic walls (see Wills, 1965, pp. 101-102, pl. 2, figs. 6-8), and has setae scattered throughout the eye itself. One may deduce that the setae on the cornea of 14 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the Eurypterina may indicate a simpler type of “‘com- pound” eye than that present in Limulus. The point is that this type of eye is neither holochroal or schi- zochroal, but is important to the present study because this is the type of ““compound” eye present in some of the fossil scorpions. The number of individual eye units is greatly reduced in fossil scorpions, but not exclu- sively so, for in the Proscorpiidae, the lateral eye is composed of very minute ocelli, probably reaching as many as one thousand. This approaches the number of visual units in the genus Erieopterus, where two species are known with approximately 1000 facets, for example, EF. hypsophthalmus (Kjellesvig-Waering, 1958b, pp. 1112-1114) and E. turgidus (Stumm and Kjellesvig-Waering, 1962, pp. 196-198). The lateral eyes of some fossil scorpions are devel- oped as bulbous, elliptical structures located at the anterolateral margins, with a distinct rim surrounding each, as in Foscorpius distinctus (Petrunkevitch), which enclosed a cluster of approximately one hundred, rounded, dorsally-convex (possibly plano-convex or biconvex) lenses, separated from one another by scle- rotic walls, or enclosed a lesser number of facets, as in **Lichnophthalmus” pulcher (Petrunkevitch) and Eo- scorpius carbonarius (Meek and Worthen), which con- tained less than 30 visual units. Another type known in fossil scorpions is that present in the genera Eo- ctonus and Buthiscorpius, which have small elliptical eyes that are distinctly intramarginal and retain only eight rounded eye lenses, but are contained within a restricted, well-defined, elliptical area. The lateral eyes characterizing Eoscorpius, “*Lichnophthalmus”, Eo- ctonus and Buthiscorpius differ from compound eyes in the Eurypterina only by the lesser number of visual units or so-called facets. Are these eyes holochroal or schizochroal? My interpretation is that they are inter- mediate between the two. In the Silurian genera Pa- laeophonus, Proscorpius, Stoermeroscorpio, and Ar- chaeophonus, the individual units or facets are much smaller in size than those noted for the above genera. They are located on large bulbous elliptical, marginal, entirely smooth eyes and undoubtedly number more than 1000 facets. Although one might assign the eyes of the Carboniferous genera mentioned above (Eo- scorpius, ““Lichnophthalmus”, Eoctonus and Buthiscor- pius) to the schizochroal category, it would be difficult to include Proscorpius and other Silurian genera in a category other than that suggested by the lateral eye of the Eurypterina—namely, compound or holochroal. Throughout the present text and explanation of figures, the lateral eyes of fossil scorpions are referred to as compound or holochroal, but what is meant is that the eyes are of the type present in the Eurypterina, which here are shown to have elements of both holochroal and schizochroal eyes. Obviously the whole question of compound versus aggregate eyes in fossil forms of Arthropoda needs de- tailed revision, as, equally obviously, the simplistic definitions given in the Treatise of Invertebrate Pa- leontology break down in practice. As stated by Lawrence (1953, p. 145), there can be no sharp division between compound and aggregate eyes based upon outward appearance in living forms. For example, the rounded, separated eyes of /sotoma, a collembolan, are considered holochroal or com- pound. Nevertheless, they probably would not be iden- tified as such if found in the fossil state; indeed, they would surely be identified as schizochroal. The same is true of the compound eyes in Lepisma, a thysanuran and Gerufa, an Isopoda, where the eyes are separated from one another, at least in part of the eye surface (see Lawrence, 1953, text-figs. 48A, C, D). Identifi- cation of the eye types by outward appearance would be much more difficult in some fossil forms. The fact that the living scorpions have simple lateral eyes, truly schizochroal, does not by any means pre- clude the presence of compound eyes in some of the fossil scorpion groups that were destined for extinction, as very likely was the case for the Lobosternina and Holosternina. The presence of both types of eyes is known in other arthropods such as the Trilobita, and it would be sur- prising if this were not the case in the Scorpionida. One might infer that primitive compound eyes devel- oped into aggregate eyes, a line of development that is suggested in the phylogeny of the Trilobita, for ex- ample, the Phacopina from Cambrian forms with primitive holochroal eyes. The development of the compound eye reached its highest degree of complexity in the pterygote insects, whereas the Eurypterina-type of primitive schizochroal-holochroal eye was destined for extinction along with the rest of the infraorders Holosternina and Lobosternina, not to mention the Eurypterina itself. It is necessary to emphasize that the living scorpions comprise eight families representing only one superfamily, and thus only a minute part of the entire Scorpionida, which is essentially a fossil group. Regardless of whether or not it is possible to separate lateral eyes in fossil arthropods into holochroal, schi- zochroal and intermediate types, all were undoubtedly homologous, and evidence points to the development of the schizochroal from the holochroal. Certainly, the possibility of the ““compound”’ eyes of fossil scorpions developing into the one to five simple eyes of modern scorpions is clearly indicated and probable. There is a good relationship between the size of the median eyes and the presence or lack of lateral com- FossIL SCORPIONIDA: KJELLESVIG- WAERING | pound eyes. If no lateral compound eyes occur, the median eyes are very large (for example, Mazonia and Waterstonia). If the lateral compound eyes are present, the median eyes are reduced in size. If the compound eyes are of medium size, the median eyes are similarly reduced (see Eoscorpius carbonarius Meek and Wor- then) and if the lateral eyes are large, the median eyes are small (see Kronoscorpio danielsi (Petrunkevitch), Pl. 14; Garnettius hungerfordi (Elias), Text-fig. 46A; and Eoscorpius distinctus (Petrunkevitch), Pl. 13, figs. 1-4). Even in the Upper Paleozoic pulmonate scorpions, Palaeopisthacanthus and Compsoscorpius, the median eyes occur in the middle part of the anterior half of the carapace. Nothing has been found in the fossil rec- ord to indicate eyes in the posterior half, as are present in the African genera Opisthophthalmus and Opisth- acanthus (Scorpionidae), which are active burrowers, although there have been erroneous reports of these posterior eyes in the literature, as in the descriptions of Palaeoscorpius Lehmann, 1944, Proscorpius Whit- field by Clarke and Ruedemann, 1912, and others. There is no known occurrence of eyeless scorpions in the fossil record, although Petrunkevitch (1949, 1953 and 1955) reported some cases. This is incorrect and is due to preservational factors or improper cleaning of the fossil. In the case of “7 yphloscorpius”’ distinctus Petrunkevitch, 1949, a supposedly eyeless scorpion from the Carboniferous of England, a review of the holotype revealed improper or insufficient cleaning in the preparation of the fossil, as well as carelessness in observation of obvious morphological structures (see photographs and drawings of the holotype of Eoscor- pius distinctus (Petrunkevitch)) (Pl. 13, figs. 1-4; Text- figs. 79A, B). The same can be said of ‘*7yphlopisth- acanthus” mazonensis (Petrunkevitch, 1955, p. 71, fig. 43(5)), Dolichophonus (Petrunkevitch, 1955, p. 70, fig. 38(4)) and others. COXOSTERNAL REGION It is in this area that one realizes that all modern scorpions represent but an infinitesimal remnant of an ancient, highly diversified, and successful large group of animals. It might well be argued, and with reason- able supporting data, that, instead of the eight families that are to be recognized in the present taxonomy of living scorpions, there should be only three: Buthidae, Scorpionidae and Bothriuridae. The Chaerilidae, Di- plocentridae, Chactidae, Vaejovidae and Iuridae should be included in the Scorpionidae, where they were be- fore families were based on the inconclusive and in- consequential presence, or absence, of trivial spurs on the ends of some (not all) tarsi and basitarsi. If both living and fossil scorpions are to be placed in the same ws taxonomic scheme, then I strongly recommend the usage of the three families for living scorpions, rele- gating the other families mentioned above to subfamily status. The present investigation is no place for a re- vision of modern scorpions, but there is no doubt that the above grouping would be highly beneficial to mod- ern scorpionological taxonomy. ! The sternum in modern scorpions is essentially pen- tagonal in shape, although in the Buthidae it is trian- gulo-pentagonal, and in the Bothriuridae it has been greatly narrowed to become almost invisible from the exterior, as it has been folded inwardly between the genital plates and the base of the second pair of coxae. As Petrunkevitch has shown, boiling in KOH will un- fold the true shape of the sternum, and it is very broad- ly pentagonal. The Scorpionidae have a very large pen- tagonal sternum with straight linear margins, and it is this type of sternum that is found in many families of fossil scorpions, at least as far back as the Lower Car- boniferous, and in all three major divisions, Lobo- sternina, Meristosternina and Holosternina. The Chae- rilidae have a large pentagonal sternum with the base bowed inwardly. This is duplicated in some Carbon- iferous genera such as Eoctonus of the Eoctonidae, and Anthracochaerilus of the Anthracochaerilidae. It is interesting to note that, regardless of family, all neonates of living scorpions retain a large pentagonal sternum. In contrast to the pentagonal sternum of Re- cent scorpions, the earliest known scorpions from the Silurian and Lower Devonian reveal broadly ovoid, pointed ovoid, or lacrimiform sterni, such as Wae- ringoscorpio Stermer, Archaeophonus Kjellesvig- Waering and Branchioscorpio, n. gen.; pointed, elon- gate lacrimiform sterni as in Kronoscorpio, n. gen.; to hexagonal-ovoid in the Carboniferous genera Cyclo- phthalmus Corda and Eobuthus Frié; or rectangulo- pentagonal as in the Lower Carboniferous Archaeo- ctonus Pocock, nearly rounded pentagonal in Gigantoscorpio Stormer of the Upper Carboniferous, to an equilateral triangle as in Spongiophonus Wills of the Triassic. Petrunkevitch (1913), following Kraepelin, Pocock, Karsch and other early scorpion taxonomists, recog- nized that the sternum had great taxonomic value. Unfortunately, he abandoned this important taxobasis because he thought that in practice the system broke down with regard to fossils (1953). This is incorrect. ' We are grateful to Dr. O. F. Francke for calling attention to papers dealing with the familial classification of Recent scorpions since Waering’s death: Francke (1979, J. Arachnol. 7:19-32); La- moral (1980, Proc. 8th Intern. Cong. Arachnol., Vienna, pp. 439- 444), Francke and Soleglad (1982, J. Arachnol. 9:233-258); and Francke (1982, Synopsis and Classification of Living Organisms, McGraw-Hill Book Co., pp. 73-75). A.S.C. & K.E.C. 16 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 What obviously was doomed to failure was his idea that a meaningful taxonomy must be based on con- venience of preservational features. Actually, there will be many specimens that are not sufficiently well pre- served to be determinable, but this is an unfortunate truth which all paleontologists have necessarily ac- cepted. Under no circumstances, however, should a morphological character be considered improper be- cause it is preserved with difficulty. The sternum is very often well preserved and, with careful work and experimentation with various lights and angles of light- ing, as well as liquids or various methods of study in the dry state, it is quite possible to work out the struc- ture of the sternum practically as well as any other structure in fossil scorpions. Petrunkevitch, in order to show that the sternum was an unreliable taxobasis, stated that neonates of the Buthidae and Bothriuridae had a broadly pentagonal sternum which in adult stadia became subtriangular or inwardly folded respectively. Therefore, he came to the surprising conclusion that this indicated unreliability, a conclusion that does not merit comment. As with the pentagonal shape of the sternum in living scorpions, the arrangement of the coxae is entirely the same in all families: the last two pairs abut the sternum, whereas the maxillary lobes of the second pair meet at midsection, anterior to the sternum, thus squeezing the first pair of coxae away from the midsection, caus- ing the first pair to lie behind and to the sides of the maxillary lobes of the second pair. The oral opening, therefore, is composed of a tunnel with the maxillary lobes of the second pair of coxae forming most of the base, the maxillary lobes of the first pair making up the rest of the base, while the large inner sides of the trochanters of the pedipalps form most of the sides, and the roof is formed by the bases of the chelicerae. In the case of fossil scorpions, this basic arrange- ment, as well as the position of the coxae vis-a-vis the sternum, is common in some families such as the Eo- ctonidae, Anthracochaerilidae, Eobuthidae and others in the Carboniferous. Thus, this specialized coxoster- nal arrangement that characterizes the Orthosternina was present in all three other major divisions: Lobos- ternina, Meristosternina and Holosternina, as well as the Bilobosternina. However, it is greatly changed in many other families, and these changes are considered of greater than familial importance. In some genera, such as the Carboniferous Eoscorpius of the Eoscor- plidae, the first two pairs meet at the midsection, above a small pentagonal sternum in the usual manner pres- ent in Recent scorpions, but only the third pair of coxae abuts the sternum, whereas the fourth pair abuts the genital opercular plates. With regard to the scorpions with pointed, oval or lacrimiform sterni, one must note that, in the Upper Silurian, the genus Proscorpius did not develop max- illary lobes in the first two pairs of coxae. Here the bottom of the entrance to the mouth was composed of the first two pairs of coxae, and the sides probably were the inner part of the trochanters of the pedipalps, with the bases of the chelicerae forming the roof. The suc- ceeding three pairs of legs abut against the lacrimiform sternum. In Proscorpius the first pair of coxae had distinct masticatory edges, armed with small teeth, or a true gnathobase such as characterizes the eurypterids. Mastication, therefore, was not delegated entirely to the chelicerae as in living scorpions, but at least in part to the gnathobase as in eurypterids. In the Lower De- vonian Branchioscorpio, maxillary lobes are well de- veloped in the first two pairs of coxae, indicating that this development, present in living scorpions, will oc- cur in much older scorpions than those found in the Lower Devonian. The last two pairs of coxae abut against the lacrimiform sternum. PROSOMAL APPENDAGES Apart from the coxae, which as shown above have undergone great rearrangement in position, the pro- somal appendages are for the most part not greatly unlike those of living scorpions except in certain im- portant areas. The chelicerae of living scorpions are composed of three segments and are not capable of appreciable side or ventral movement. All Paleozoic scorpions that I have studied show consistently that the chelicerae are composed of four distinct joints. The basal one (Ch1) is cup-shaped, followed by another cup-shaped band (Ch2), the manus with its fixed ramus (Ch3) and the free dactyl or ramus (Ch4). The dactyls carry well-developed teeth and in the genera Proscor- pius and Allopalaeophonus from the Silurian, the free ramus is decidedly falcate. Enormous chelicerae, ap- parently rather flattened, are present in some Upper Carboniferous genera such as Garnettius and some species, like Mazonia wardingleyi. Many show thick long setae, some in brushlike masses located on the inner sides of the chelicerae. These appear to have had some function other than to form a cushion between the chelicerae and it is here suggested that these setae were movable and may have been used in passing food particles to the oral opening. However, the four-jointed chelicerae, particularly in some Lower Devonian forms, were long enough to be capable of moving food to the mouth by bending downward and backward to reach the mouth. In the shorter four-jointed chelicerae, bend- ing backward to the oral opening could not have been done so well, and then only partly. The denticles on each side of the dactyls are generally preserved in fossils, although nearly always both sides FossiL SCORPIONIDA: KJELLESVIG- WAERING 17 of the dactyls (superior and inferior) are flattened to- gether, so that interpretation of the position of the rows of denticles, and their sizes or shapes, is difficult. It is only rarely that the chelicerae are found in an inflated or natural state. The differences in dentition, however, do not seem to be outstandingly diagnostic, and cer- tainly not a taxobasis above family level, if that high. Vachon (1963) has shown that the living scorpions do have differences on the family level. This interesting factor could be explored further in fossil scorpions, but not enough data are available to me at present to com- ment further. The pedipalps are remarkable in fossil scorpions merely for the fact that they have not changed appre- ciably since the Middle Silurian. They are composed of the same number of joints and are identical in nearly all aspects. Some are narrow and long, as in the Lower Devonian Branchioscorpio and the Upper Carbonif- erous Cyclophthalmus, and others, denoting the more fossorial types, are short and stout, such as the Lower Devonian Palaeoscorpius and the Upper Carbonifer- ous Garnettius. Nowhere in the fossil record, except in the Miocene (Mioscorpio), are scorpions found with the greatly inflated manus of living scorpions such as Scorpio and Heterometrus. Slight or no difference in size of the fossil setae has been noted, and none has been found to be the long type that characterizes the trichobothria of today. However, this does not mean that they could not have been present in the Paleozoic forms, particularly the terrestrial ones. Surely it is reasonable to assume that the pulmonate orthosternous Palaeopisthacanthus and Compsoscorpius had trichobothria developed. But the only reason this is assumed is because the setal bases are arranged in linear rows and clusters much like in the living scorpions. This is an acceptable argument, as most may agree. On the other hand, true aquatic scorpions have identical setal openings, but here the same facts are not acceptable! Trichobothria, as we know them, seem to be spe- cialized for cryptoenvironments and not aqueous ones, which were the main habitat of most of the Paleozoic scorpions so far recovered. In this respect, it is inter- esting to note that one of the troglobite scorpions has unusually long trichobothria developed for a cavern environment, which is merely a ‘“‘macrocryptoenvi- ronment”. Whether or not trichobothria were devel- oped in aquatic scorpions remains a possibility, but I consider it a likely one. At any rate, on the species level, the arrangement of the trichobothria is a taxo- basis of importance, and has been used by the writer in preliminary separation or identification of fossil scorpions. This taxobasis will become more important when more fossil scorpions have been found. Some fossil scorpions developed sensory setae on the pedipalps which, in basic structure, are very similar to the trichobothria of Recent scorpions: see the Car- boniferous Mazonia (Kjellesvig-Waering, 1969, p. 171) and Gigantoscorpio (Stormer, 1963, p. 122). These se- tae are arranged in definite rows, groups, or small clus- ters, resembling the trichobothria of living scorpions. On the other hand, it is difficult to think that the setae could be as long as those developed on present day trichobothria, although the shorter setae developed for aquatic life could have the same sensory function as the long ones do today. Of course it is not known if the fossil aquatic scorpions had developed trichogen cells, or whether such cells were developed for atmo- spheric environments. It seems possible that they did develop them, but for a water existence, as shown by the many specimens with definitely placed rows and arrangement of setal openings on the pedipalps, as in Recent scorpions. The Upper Carboniferous Paraisobuthus prantli, n. gen., n. sp. developed a long manus with long dactyls— the dorsal side was smooth, but the entire ventral side was a dense mat of setae (see Text-figs. 87B, D, E), which are not considered to be trichobothria. The “cutting” edges of the dactyls of fossil scorpions were developed as are those today, and the type of teeth, setae, etc., along the inner edge is as important a taxobasis as it is in living scorpions, particularly on the generic scale, according to our understanding to- day. In the Silurian, the genera Palaeophonus and Pro- scorpius have dactyls with cultrate edges. Others have a single file of small denticles as in the Lower Devonian Branchioscorpio, the Upper Carboniferous Titano- scorpio and many others. Some, like the Upper Car- boniferous Paraisobuthus, have a row of setae along the edge and, possibly, the immovable finger ends in a double or split end where the single ramus would be accommodated. Brontoscorpio, the gigantic Upper Si- lurian scorpion, has four to five or more rows of thick denticles along the edges. The genus Palaeomachus of the Upper Carboniferous has a large swelling on the inner edge near the base of the free ramus, much as is present in the males of some species of the living genus Tityus (Buthidae). There is, however, a great difference and that is that the swelling, instead of fitting into a corresponding indentation on the opposite finger, fits into a boss on the side of the fixed finger. The two dactyls, therefore, must have had a slight scissorlike action in Palaeomachus (see Text-fig. 109C). The legs are remarkable in fossil scorpions. They are of two general types in the Lower Paleozoic, one, the very short, stout legs of Palaeophonus and Allopa- laeophonus, which terminate in a long, but robust, sin- gle spine (the posttarsus). This is generally mistaken 18 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 for a claw. Actually, the claws of present-day scorpions are the tarsal spurs, and the termination is the greatly reduced posttarsus. The thick legs of these scorpions are considered eurypteroid in aspect, and suggest that these scorpions were adapted for digging under water. This habitat explains the long sinuous shape, much like other burrowers, such as the living Broteochactas (Chactidae), and may be responsible for the short “‘eu- rypterid” legs. Broteochactas, as well as other good burrowers, also had short legs, but not to the extent present in these early scorpions. At the same time, scorpions with legs much like those living today lived concurrently with the ‘primitive’? Palaeophonidae. These include Dolichophonus, Proscorpius, Waerin- goscorpio, Archaeophonus and Branchioscorpio from the Silurian and the Lower Devonian. In Proscorpius the legs are relatively short, tubular, but slender and not of the above-mentioned “‘eurypteroid”’ type. In all scorpions, both living and fossil, the leg count, past the coxa, is consistently seven joints, counting the post- tarsus. Only in the Silurian genus Archaeophonus Kjelles- vig-Waering was there a possibility of a double tro- chanter, which would have raised the leg count to eight joints past the coxa. It is now believed that the sup- posed double trochanter is in reality a single one and that a crack obliterated the joint or misled me to that interpretation (Kjellesvig-Waering, 1966). It can safely be said that all scorpions so far known have seven joints past the coxa on each leg. The legs of these Silurian scorpions, Proscorpius and Archaeophonus are as remarkable as those of the Pa- laeophonidae in that they have an extremely long tar- sus, with very short, poorly developed tibial and ba- sitarsal spurs, one on each side. The long tarsus ends in short double claws, which are covered with at least a row of minute spines on the bottom of each claw. The posttarsus is greatly reduced. One might speculate here as to how this scorpion walked, and the fact that it lived on very soft muddy magnesium carbonate bot- tom may indicate that the entire tarsus was placed flat on the substratum. The small spines underneath the claws may also indicate that traction was needed against slippery surfaces such as algae. The long tarsus continues into late Paleozoic time (see Kronoscorpio), although in many, such as Eo- Scorpius, the tarsus was becoming reduced. In Gigan- toscorpio the tarsus had become even more reduced. In some Carboniferous scorpions an enormous tibial spine was developed on the fourth leg. In “‘Euro- phthalmus longimanus” (see Text-fig. 77C), Boreo- scorpio copelandi, n. gen., n. sp. and others, the tibial spine is developed to an extraordinary length. Again it might be well to speculate. All scorpions with such a spikelike tibial spine have the spine on the outside of the last pair of legs and all occur in the Upper Car- boniferous where “‘coal swamp” conditions prevailed. It appears that this spine could be used to anchor a scorpion when it is walking vertically up a tree root or trunk, underwater or above water, as the abdominal plates or “‘sternites” each had a pouch for the gills and enough space to carry water to moisten them, as in many living gill-bearing amphibious animals such as crabs, etc. An anchor was also needed to hold firm when the stinger was thrown overhead and thrust for- ward—the two spines, anchored in plant material so common on the floor of the swamp forest, would give the backward support needed. Tibial and basitarsal spines were also very thick, serrated, rounded, and occurred in many different sizes and shapes. In the modern scorpions, all are merely small, spinelike ves- tiges. I suspect, judging from the eurypterid leg, that it would not be surprising to find paired spines higher on the joints, such as femoral spines, in fossil scor- pions. Perhaps the most remarkable thing about the legs of some of the more primitive scorpions, such as those constituting the superfamilies Palaeophonoidea (Lo- bosternina), and Allopalaeophonoidea and Archaeoc- tonoidea (Holosternina), is that they are distinctly te- rete and not flattened as in most scorpions in the Upper Carboniferous to the Recent. As in modern scorpions, the last joint of the leg is the transtarsus or posttarsus. Anterior to this joint are the claws, which are homologous to the tarsal spurs. In some scorpions, such as the Silurian Palaeophoni- dae, the foot of the scorpion ended in a sharp, pointed transtarsus without development of curved claws, which resembled the terminal joint of the walking legs of eurypterids. However, there are two tarsal spurs, one on each side. These legs are adapted for a water habitat, but like the crabs, could as easily be adapted for land. The fact that the Palaeophonidae were lobosterns shows that the adaptation of the pointed terminal joint was for grasping rock surfaces or perhaps reefal environ- ments, probably under water, as the scorpion was gill- bearing. By the early Devonian at the latest, the three-“toed”’ combination was developed. Pa/aeoscorpius retained the short, pointed posttarsus, but had developed two spurs on each side which are equivalent to basitarsal spurs, so these legs seem to be adapted for the same environment as the palaeophonids. As early as the Middle Silurian the scorpions had developed the foot that is present today. The Silurian scorpions such as Dolichophonus (leg terminations unknown, but the type of body indicates that the double claws were devel- oped), Archaeophonus, and Proscorpius had developed FossIL SCORPIONIDA: KJELLESVIG- WAERING 19 a posttarsus that had assumed a posterior position on the leg joint (tarsus) and two claws that were anterior to the posttarsus. These scorpions were true water dwellers and the leg terminations were presumably de- veloped for walking on underwater shore plants as well as roots, trunks and harder bottoms. In Carboniferous time the development of the ter- minal joints reached its greatest diversity. Some scor- pions, such as Eoscorpius, Eobuthus, Isobuthus, etc., developed large curved claws that were armed with small spines on the ventral side. This development, however, occurred as early as Middle Silurian, as it is present in the Wenlockian A//opalaeophonus (see Text- fig. 17C). These claws could only be adapted for hold- ing onto some object, such as underwater roots, leaves, stems, etc., present in swamp forests. All of these scor- pions were either holosternous or lobosternous, thus were water-dwellers breathing through gills. We could assume that some of these scorpions lived among the underwater roots and trunks of trees and other plants, but were capable of excursions above water on these plants, thus occupying the same position as many crabs living today. Some Carboniferous scorpions such as ‘“Lichno- phthalmus” pulcher (see Wills, 1959, figs. 5, 6) had the posttarsus developed into an incredibly long “‘dagger”’. The double claws showed the ventral spines so suitable for arboreal existence. It seems almost certain that the long dagger was also developed for arboreal life. Kronoscorpio, a lobostern, developed the claws into very long straight spines, and the posttarsus was de- veloped into an equally long spine. Thus, this would result in a very widespread, trifurcate, terminal ar- rangement and give a “‘snow shoe” action for walking on soft muddy bottoms. The two long “‘claws”’ would be on the anterior, and the posttarsus on the posterior. OPERCULUM The opercular plates are of different sizes and shapes, and double as in Recent scorpions. Although not much attention has been given to the opercular plates as a taxobasis, they may be rather diagnostic. Thus they vary from large subquadrate plates, as in the Lower Devonian Branchioscorpio, to small subquadrate plates, as in the Lower Carboniferous Gigantoscorpio willsi Stormer (see description of part of the holotype. pp. 76-77, below). Others are elongated, with the long axis parallel to the length of the scorpion, and fitting to- gether against the lower part of the sternum, as in the Upper Carboniferous Kronoscorpio or the Lower Car- boniferous Centromachus; or elliptical, as in the Upper Carboniferous Eoscorpius and Eobuthus; or may be very small rounded-elliptical ones as in the Upper Car- boniferous Anthracoscorpio; or rounded as in the Up- per Carboniferous Waterstonia. Another type is pyr- iform, but with the blunt ends adjoining, as in the Upper Carboniferous pulmonate Palaeopisthacanthus, or nearly rectangular, as in Anthracochaerilus, also from the Carboniferous. With some exceptions, it might be said that all structures of the opercular plates are not greatly different from those present in living scorpions. Possibly the most interesting genital operculum is pres- ent in the Upper Carboniferous Paraisobuthus, in which the opercular plates are identical to the lobostern plates of the body except that they are greatly reduced, to become miniatures of these plates. In this case it seems that the two plates never separated, but remained as a single bilobed plate. PECTINES Much has been written about the function of the pectines. Today they are regarded by some as chemo- receptors, and by others as mechanoreceptors (Kaest- ner, 1940, 1968; Levi and Levi, 1968).? This may be the case, but none of these authors has bothered to mention that the pectines were fully developed, and for aquatic life, in the lower Paleozoic! The possibility of the pectines being used as swim- ming organs, advanced by Dr. Stormer (1963) in his original description of Gigantoscorpio willsi, is plau- sible, although it seems to me that the pectines had a more important function. The entire pectine is very broad and long, making an excellent tool for swim- ming, or indeed for digging. The third joint is long and ends in a point. Perhaps these old scorpions did not reproduce like those living today, and possibly fol- lowed the copulatory practice of the Xiphosura and eurypterids. That is, they laid eggs in a shallow “‘nest”’ which had been scooped out in the soft mud or sand by the pectines, and subsequently fertilized by the male. The proximity of the pectines to the reproductory oper- culum may be significant in this case. This would ex- plain this association in the living scorpions. As a che- moreceptor there would be little point in being so intimately associated with the sex opening, both male and female. The functions of the pectines in living scorpions are dubiously connected with reproduction, although the pectines are very close to the reproductory organs. However, Kaestner (1968, p. 103) shows that they are important in the selection and preparation of a site for spermatophore deposition. If this were so, one would expect the females to have pectines less well- developed than the males. Furthermore, the Brazilian Tityus serrulatus is parthenogenic and, although only ? The fine structure of the sensory pegs on scorpion pectines was described by J. D. Carthy (1968, Symp. Zool. Soc. London, 23:251- 261). Neurophysiological studies have been reported by C. Hoff- mann (1964, Naturwiss. 51:172) (Francke, written commun., 1983). 20 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 females are known, the pectines are as well-developed as in other species of 7ityus. It is suggested that the position of the pectines in relation to the opercula is a consequence of their previous development as aids in scooping out a nest for the accommodation of eggs, and directly connected to arranging the eggs and mixing the spermatozoa with the eggs in order to assure fer- tility. Certainly the function of the pectines as a che- moreceptor does not seem to be warranted in an aquat- ic animal. The teeth do indicate a perfect “brushlike” function and it is here suggested that the teeth were used in mixing the eggs with the spermatozoa. It should be noted that the pectines of living (pulmonate) scor- pions are all much narrower than the majority of the fossil aquatic scorpions. Holosternous and lobosternous scorpions, which were aquatic and breathed through gills, developed pectines that were much more variable in structure than those of living scorpions. In the Upper Silurian Proscorpius osborni (Whitfield) they are composed of a finlike unjointed anterior plate, without differentia- tion into a jointed rachis and a middle lamella—the plate is also without rounded sclerites or areoles. Fulcra are well-developed and teeth are elongated but rela- tively stout. The Lower Devonian Branchioscorpio richardonsoni, n. gen., n. sp. also has the large unjoint- ed and undifferentiated plate, no fulcra, and very round large flat teeth. In the Upper Paleozoic, Gigantoscorpio (see Stormer, 1963, pp. 54-57), Eoscorpius, Telma- toscorpio, Kronoscorpio (see this paper) and many oth- ers had a well-developed, jointed rachis, as in modern scorpions, and a greatly-expanded middle lamella with areoles of irregular size and shape. In others the areoles are all consistently small and rounded throughout the middle lamella. Fulcra were well-developed and the teeth were numerous, well in excess of 50 on each comb. In many of these the entire pectine was so long (anteriorly to posteriorly) that it was as long or longer than the width (as measured along the row of denticles). In the genus Anthracochaerilus the combs were very narrow and long, without the development of fulcra, and with very thin, nearly needlelike, teeth of which as many as 150 were present on each comb. The larger, or in particular, the wider the pectine, the more it is broken into areoles both on the middle lamella and the rachis. In living scorpions, those with very wide pectines also develop areoles (Brachisto- sternus, Bothriurus, etc.). Those that have short pec- tines do not develop the areoles, but in some cases the middle lamella and rachis are indistinguishable, as the entire pectine comprises a single plate. Although there is considerable variation in the pec- tines of living scorpions, this variation is minor com- pared to that of fossil scorpions. This is to be expected, as present day scorpions belong to a very small rem- nant group representing only one superfamily, whereas fossil ones belong to many superfamilies involving sev- eral infraorders. However, the fundamental structure is the same in all, both fossil aquatic and the living, even to the presence of peg organs in the teeth. In some scorpions, such as Paraisobuthus, a dilated inner areole probably denotes the female, as it does in many living scorpions such as the genus Jityus in the Buthidae. There is no question, therefore, that in aquatic, gill- bearing scorpions the pectines were composed of the same external structures, namely rachis, middle la- mella, fulcra (or no fulcra as in some living scorpions, for example, Ananteris), teeth and peg organs. What- ever function is given to the pectines of living pul- monate scorpions, it is going to be very difficult not to apply the same function to the fossil gill-bearing ones. As stated above, that function may well be connected with reproduction. We may now speculate with a reasonable degree of accuracy on the origin of the pectinal, prepectinal and opercular plates. The question is, are the pectinal, pre- pectinal and opercular plates in Orthosternina modi- fied abdominal plates or sternites? There are several specimens, both in fossil and living scorpions, that give us a rather good idea of the origin. The evidence is that in both fossil and living scor- pions, regardless of whether the scorpion may have abdominal plates or sternites, or a combination of both, as in Branchioscorpio, these plates are modified ab- dominal plates. There are several pertinent specimens that are important in this respect, and information from the eurypterids is of particular importance. In the rather closely related eurypterids, it has been shown that many, for example, Eurypterus, Baltoeu- rypterus, and others, demonstrate a forward gradation of the type of plates on the ventral side of the preab- domen. Thus in Eurypterus the posterior plates are found to be of the holosternous type, grading into meristosternous-holosternous, to entirely meristoster- nous plates. In Mixopterus kiaeri Stormer (Stormer, 1955, p. 35) the development is from holosternous on the fifth plate to protolobosternous on the fourth to lobosternous on the third and other plates. In Carci- nosoma (see Kjellesvig-Waering, 1958a, p. 299) it was noted that the fifth abdominal plate was holosternous, the others meristosternous. The development is an- teriorly. (See: Clarke and Ruedemann, 1912; Holm, 1896, 1897, 1898, 1899, for numerous examples.) Sim- ilarly, the median organ shows the same progressive development anteriorly. PREABDOMEN The dorsal side of the preabdomen in fossil scorpions FossiIL SCORPIONIDA: KJELLESVIG- WAERING 21 is very similar to that present in Recent scorpions. All scorpions, regardless of age, have seven tergites in the preabdomen and not eight as advanced by Petrun- kevitch (1913, 1949, 1953, 1955), mainly based on a misconception of the preabdomen of Mazonia wood- iana (see Kjellesvig-Waering, 1969, p. 171). Dubinin (1962, fig. 1225), not satisified with eight tergites in the preabdomen, gave Mazonia woodiana nine! A complete Mazonia woodiana is described here and shown in Text-figure 29B, which should remove all doubts concerning the validity of this theory. All ter- gites of fossil scorpions, as of many living ones, have well-developed transverse ridges on the anterior. The tergites are shaped exactly like those of living forms with the exception of one genus, the Upper Silurian Proscorpius, which has well-developed alae at each an- terolateral angle. These alae are common among the Eurypterida. The underside of the preabdomen is the area where, in my opinion, the most important macro-evolution- ary changes have taken place. These changes are de- scribed below and will not be repeated here. In contrast to the bandlike sternites of living scorpions (orthos- ternous) with round to elliptical to slitlike stigmata perforating the center of each half of the sternite, fossil scorpions are: bilobosternous, being composed of two separate free rounded plates without any stigmata; /o- bosternous, consisting of two rounded or lobed plates joined at midsection and without stigmata and with a thick doublure; meristosternous, where two apparently holosternous plates are joined by a suture at midsec- tion, but without stigmata; and holosternous, where the plates are bandlike, without any suture or stig- mata. All four types overlap the succeeding abdominal plate as in eurypterids, and in contrast to the Orthos- ternina, which have sternites embedded in the tissue and not overlapping. It is known that the meristosterns and the holosterns, which are considered to be gill- bearing, had the opening from the gills through a large narrow slit in the doublure of the abdominal plate, at each posterior lateral angle. The bilobosterns and the lobosterns apparently had the gills directly below the plates, but probably were open throughout the entire posterior parts. CAUDA The cauda of fossil scorpions does not differ from that of Recent ones. It is composed of five segments and the telson. As in Recent forms, fossil ones have narrow to wide caudae. The females, both living and fossil, have a distinctly shorter and thicker cauda than do the males, and in the latter the twelfth tergite is longer and more slender than in the females. Some fossil forms, like the Upper Carboniferous Eoscorpius pulcher and Palaeobuthus distinctus, have an enor- mously developed, scimitarlike aculeus with only a small vesicle (see Text-figs. 56B, 77A). No Paleozoic or Mesozoic scorpion has been found with a subaculear tooth, as this apparently was a Tertiary development. ORNAMENTATION The ornamentation of fossil scorpions is much like that of Recent forms, except that some scorpions have a dermal covering that is highly and coarsely pustulose. The Lower Carboniferous Archaeoctonus glaber (Peach) has raised, long scalelike ornamentation that is an exact duplicate of that which characterizes the superfamily Mixopteroidea of the Eurypterida. These include the distinctly scorpionid eurypterids that not only had the scorpionid shape, but also had telsons that likely had a pair of poison glands, as the entire cauda was capable of being curved overhead, as in the scorpion, for up- ward or forward thrusting. Compare the ornamenta- tion of A. glaber in Text-figures 21A, D with that of Paracarcinosoma sp. (Kjellesvig-Waering, 1951, text- figs. 2j-l). THE VENTRAL SIDE OF THE BILOBOSTERNINA, LOBOSTERNINA, HOLOSTERNINA AND MERISTOSTERNINA The ventral side in the above infraorders, which include nearly all the Paleozoic scorpions, is quite dif- ferent from that of the Orthosternina, or living scor- pions. These differences are apart from the well-known characteristic of having abdominal plates overlapping one another as in the eurypterids. The number of gill- bearing abdominal plates in these orders is five —again, precisely the same number found in eurypterids. This important fact has not been recognized until now. Liv- ing scorpions have only four lung-bearing sternites. The Paleozoic Bilobosternina, Lobosternina, Holo- sternina and Meristosternina were thought to have four, rather than five ‘“‘sternites’’ because only a few speci- mens were found in which the underside had not been telescoped into the anterior part of the preabdomen, as would occur due to molting. Unfortunately, it was also due to the prejudiced assumption that five gill- bearing “‘sternites”’ were not possible and therefore not expected. Heretofore all modern authors, myself in- cluded, accepted the assumption that fossil scorpions had to have four “‘sternites’’, as is the case with living forms. It is unfortunately the area immediately behind the carapace where obstruction occurs, and it is caused by the lateral widening of the preabdomen. Therefore, when the animal crawled out of its empty skin, tele- scoping occurred, resulting in covering of the impor- tant ventral structures by the large pectines. It is thought 22 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 that nearly all fossil scorpions found are exuviae, as the associated eurypterids were. Thus only a few spec- imens escaped telescoping and those had pectines so large that the overlying structures were completely masked. A particularly good specimen showing the true relationship of the underside to the dorsal is the holotype of Microlabis sternbergii Fri¢é, which has very small pectines, unusual for Paleozoic forms, not large enough to obliterate the structures dorsal to it. The specimen reveals five distinct meristosternous abdom- inal plates. Although this specimen has been known since 1904, no importance whatsoever had been given to the presence of the five “‘sternites’’. Petrunkevitch (1953, p. 23) reported five plates in the holotype, but failed to attach the slightest importance to their pres- ence. In this context, it is well to recall that Thorell (1886, p. 270), in his intense controversy with Whitfield, had questioned the five “‘sternites’’, which Whitfield stated were present in the Silurian Proscorpius osborni. All who worked on this group agreed with Thorell, in- cluding myself (1966, p. 365). I now have several more specimens showing that there is no doubt whatsoever that five gill-bearing abdominal plates occur (see Text- figs. 6C, 7A, 11A). The holostern Palaeoscorpius de- vonicus Lehmann also clearly shows five abdominal plates (see P1. 3, fig. 1). In the Lobosternina it is difficult to find a specimen where the large pectines normally present in this in- fraorder have not completely covered the structures dorsal to the pectines, or that telescoping in the region of the pectines has not masked the structures. Never- theless, the holotype of Paraisobuthus prantli, although not well-preserved in the area of the abdomen, does show the five abdominal plates. Thus we have five abdominal plates present in the Lobosternina, Holosternina and Meristosternina, and it can safely be assumed that the Bilobosternina also had five abdominal plates. Also of considerable taxonomic interest, and located in the same area where telescoping of the anterior of the preabdomen is most prevalent, is the pectinal plate (anchor plate or basal pectinal plate), which is well known in living as well as fossil scorpions. In fossil scorpions, however, the plate retains a well-developed median organ in some forms; in others this median organ is completely absent. This may be due to sexual differentiation, but there are not sufficient fossil spec- imens at present to fully determine that. PREPECTINAL PLATE There is some confusion in the literature concerning a small plate that occurs anterior to the pectinal plate and which is named the prepectinal plate (ppp). Pe- trunkevitch (1913, 1949, 1953 and 1955) did not rec- ognize the prepectinal plate. Stormer, in his excellent description of Gigantoscorpio willsi (1963, pp. 53-59), was the first to figure a greatly diminished plate with a median organ; unfortunately, this plate was not in place and was therefore identified as part of the pectinal plates (“basal plate of pectines”). This organ can now be identified as the prepectinal plate (ppp) and properly occurs anterior to the pectinal plate (pp) and posterior to the operculum. A revised reconstruction of this area of G. willsi is given in Text-figure 76C. The plate here retains the median organ, which is not always present in some other genera. For comparison of the prepec- tinal plate, see Opsieobuthus pottsvillensis (Text-fig. 31A), Centromachus euglyptus (Text-fig. 31B) and Eoscorpius pulcher (Text-fig. 27B). Curiously enough, the prepectinal plate has not been recognized in living scorpions, but it is present in at least some of the New World buthids, but only in the males (no New World buthid males are known without it, but all have not been investigated). The plate is very small and of the same color as the rest of the scorpion integument. For example, in the dark brown Tityus trinitatis Pocock, the sclerite is a wing-shaped narrow plate lying next to the operculum (see Text-fig. 31C). It may serve a function connected with the operculum, but this is a secondary adaptation, and it certainly is the prepectinal plate. The female 7. frinitatis does not have this plate (see Text-fig. 31D). Although I have made only a cursory search for the presence of the prepectinal plate in living scorpions, it is present in males of the following (all of which are New World Buthidae): Centruroides gracilis (Latreille) Centruroides keysi Muma (=C. guanensis Franganillo) Centruroides margaritatus (Gervais) Microtityus rickyi Kjellesvig-Waering Tityus pictus Pocock Tityus trinitatis Pocock The prepectinal plate is not present in males of the following (Chactidae, Diplocentridae, Buthidae, Scor- pionidae, Bothriuridae, Vaejovidae): Bothriurus bonariensis (Koch) Broteas granimanus Pocock Hadogenes troglodytes (Peters) Ischnurus ochropus Koch Mesobuthus gibbosus (Brulle) Nebo heirichonticus Simon Opisthacanthus laevipes Pocock Pandinus viatoris (Pocock) Uroplectes triangulifer marchalli (Pocock) Paruroctonus mesaensis Stahnke The prepectinal plate is not present in females of the following (Buthidae, Chaerilidae, Vaejovidae, Bothriu- ridae): FossiL SCORPIONIDA: KJELLESVIG- WAERING 23 Bothriurus bonariensis (Koch) Buthus trilineatus Peters Centromachetes obscurus Mello-Leitao Centruroides keysi Muma (=C. guanensis Franganillo) Centruroides margaritatus (Gervais) Centruroides gracilis (Latreille) Chaerilus celebensis Pocock Chaerilus truncatus Karsh Mesobuthus gibbosus (Brulle) Microtityus rickyi Kjellesvig-Waering Parabuthus villosus Peters Tityus rufofuscus Pocock Tityus smithi Pocock Tityus trinitatis Pocock Uroplectes triangulifer marchalli (Pocock) Paruroctonus mesaensis Stahnke In scorpions, the prepectinal plate would be opposite or is the counter somite of the pregenital tergite, where- as the pectinal plate and appendage (pectines) would be the counter somite of the first (adult) tergite. The rest of the tergites and “‘sternites”’ are each equivalent. This means that the fundamental preabdomen of scor- pions consisted of eight somites, although in the adult fossil scorpion, the dorsal side consisted of seven dorsal segments, whereas the ventral comprised eight so- mites, and seven in other fossil scorpions, depending on the presence of the prepectinal plate. Apparently the preabdomen of living scorpions is composed of seven somites in New World Buthidae males, and six in New World Buthidae females and all other families, including Euro-Asian and African Buthidae, depend- ing on the presence of the prepectinal plate. The prepectinal plate assumes importance in any attempt to homologize the segmentation of the Scor- pionida with any other groups, as it appears to be the opposing or ventral part of the somite of which the pregenital tergite is the dorsal part. There should not be any confusion with regard to the above and to Petrunkevitch’s ill-fated hypothesis of the eight tergites in the dorsal preabdomen of the adult fossil scorpions (Mazonia, Proscorpius, etc.). This has been completely disproved (Kjellesvig-Waering, 1969, pp. 171-190). STERNITES AND ABDOMINAL PLATES The sternites of living scorpions and the abdominal plates of lobosternous, holosternous and other fossil scorpions are neither the same nor even homologous. The current view, as stated by Stormer (1976, p. 149) is that the two are the same structure: “. . . Mesophonus probably forms a transition between the early Palaeo- zoic scorpions, in which the abdominal plates probably were attached in front only, and the Recent forms in which the ventral plates are attached all around as ordinary sternites.”” There is a different conclusion, which I believe is worth considering: the sternites of the living scorpions are part of the ventral body wall, whereas the abdominal plates are appendages of the somite and lie below (outside) the ventral body wall. Most forms that had abdominal plates (the lobosterns, meristosterns and holosterns among the scorpions and the entire order Eurypterida) never developed sclero- tized sternites as that protection was not needed, be- cause the ventral body wall was covered by the ab- dominal plates. In the bilobosterns the ventral body wall was partially uncovered and sternites developed. The Phalangiotarbida had both abdominal plates and sternites: six abdominal plates covering the six pairs of gill chambers, and three sclerotized sternites as the last three segments of the opisthosoma. In my opinion, there was no migration of the so- called stigmata from the doublures of the lobosternous, holosternous and meristosternous scorpions as has been advanced (Stormer, 1976, pp. 149-150). For example, the “‘stigmata”’ that characterized the British Triassic scorpions are considered to be intermediate between the early Paleozoic scorpions and Recent forms, as advanced by Wills (1947, p. 33) and Stormer (1963, p. 110; 1976, p. 149). But this is not possible, for the so-called stigma of the Triassic mesophonids was merely an opening into the gill chamber (not an open- ing into the body (lungs)). Although both authors con- sider the known British Triassic scorpions to have lungs, this is impossible, since it can be shown that the so- called lungs were not attached to the “stigmata” as are the true lungs in living scorpions, but were attached instead to the body wall, with the posterior spinous (rounded) and considerably anterior to the wide “‘stig- mata” (see Wills, 1947, text-figs. 7, 18, 51). In order for these structures to be lungs, they would have to be turned around 180°, and attached to the wide “‘stig- mata’’. Apart from this, if the sternites of living scor- pions were homologous to the abdominal plate, the so-called lungs would have to migrate from the body wall, forward and downward along the dorsal side of the abdominal plate, and then back again anteriorly to end at the present position of stigmata of the sternites that characterize the Orthosternina. Ontogenetically the lungs are clearly invaginations of the body wall or sternite (not abdominal plate). The lungs are aborted during ecdysis, attached to the ster- nite, showing that they are part of the ectoderm or body wall. In the ontogeny of scorpions the stigmata and lungs move s/ight/y forward, as reported by Kaest- ner (1940, p. 113), but this slight migration of the stigmata should not be confused with the gills of ho- losternous, lobosternous and meristosternous scor- pions. This occurs on the sternite and has nothing whatever to do with migration of the gill slits on the doublures of the ancient scorpions. There never was 24 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 an invagination of the gills for the gills would then form sacs and we do know that all gills so far found in fossil scorpions are simple lamina, not sacs of any form. It should be noted that doublures were never formed on the true sternites, as they were not needed, because the sternite never formed a gill chamber. Doublures of the abdominal plates are simply part of the gill chamber as Waterston (1975, p. 249, text-fig. 3) showed in great detail in his excellent paper on the eurypterid gills, and confirmed here in the scorpions. He was able to reveal the delicate structures in the large eurypterid Tarsopterella scotica (Woodward). This type of breath- ing mechanism is precisely what I feel must be present in the Lobosternina, which was also suggested by Wa- terston. In the description of the scorpion Paraisobu- thus duobicarinatus, gill tracts are described identical to those present in the eurypterids. It is well known that the quantities of oxygen avail- able in water are very small compared to the quantity present in air. Thus it is essential, during the absorption of oxygen from water, that a sufficient amount of water be circulated over the gills so that the required amount of oxygen may be obtained. This implies that the open- ing (in and out) through which the oxygen is introduced to the breathing mechanism from a water source must necessarily be large enough to accommodate the large volumes of water required. Conversely, the amount of air needed to supply the required oxygen is much smaller. This is clearly revealed by the stigmata of living or pulmonate scorpions, which are very small, each being approximately one-eighth to one-twelfth the width of the sternite. On gilled scorpions, the gill opening on the abdominal plate doublure is several times wider. For example, in the Triassic scorpions the gill openings are each one-fourth the width of the ab- dominal plate, and in the Lobosternina the gill slit is nearly as wide as the entire abdominal plate (see Text- fig. 83E). In the Amblypygida the stigma is much larger than in most Arachnida, nevertheless, it is only about one-sixth the width of the sternite. This order, how- ever, has only one pair of lungs as against the four pairs of lungs in pulmonate scorpions and the five pairs of gills in gilled scorpions. These wide gill chamber open- ings were certainly not needed if oxygen from an at- mospheric source were involved. Similarly, there is no homology between the book- gills of the Xiphosura and the book-lungs of scorpions. In the Xiphosura the book-gills are appendages that hang from the body wall into the water, whereas the lungs of scorpions are invaginations of the ectoderm. The homology claimed for both structures by Lankes- ter (1881), Waterston (1975, p. 251) and others is dif- ficult to justify. The difficulty seems to be the lack of recognition of the fact that the ventral plate in living scorpions is a true sternite, whereas the Xiphosura have abdominal plates outside the sternite, which re- main unsclerotized. In fossil Xiphosura, where the opisthosoma is segmented, the sternites would be de- veloped in much the same manner as in the Euryp- terida. If it is not true that the abdominal plates of gilled scorpions are not homologous to the sternites of pul- monate scorpions, then we must accept the fact that the two are homologous and, indeed, the same. This means that living scorpions do not have true sternites. It may be well to speculate on that possibility. In the first place, scorpions such as the lobosterns, meristosterns and holosterns, comprising nearly our entire knowledge of fossil scorpions, were terminal groups that did not continue into any lines that led to the present-day scorpions. The possibility of Triassic scorpions having large stigmata and lungs, which later migrated anteriorly, is totally unacceptable. The fact that the so-called lungs are shown to be attached to the body wall with the posterior end facing the “‘stig- mata” and without any connection to the “stigmata’’, plus the well-developed gill chamber, proves that the “lungs” are gills. These gills could not possibly develop into lungs, as they would first require invagination in order to be lungs, and second, as stated above, an unlikely migration down into the abdominal plate where, in some unknown fashion, the gill chamber was obliterated. The abdominal plate became fused with the body wall, obliterating the ventral layer, and the doublures were also obliterated. This is, of course, highly unlikely if not impossible. DIRECTION OF EVOLUTION The scorpion Branchioscorpio richardsoni seems to reveal the direction of evolution of these early scor- pions (see Text-figs. 2 and 101 A-G). This scorpion has a well-developed coxosternal area with the tubular area anterior to the mouth. The tubular oral chamber an- terior to the mouth is considered by Stormer (1976, p. 154) to be a specialization for land living, and the writer agrees. The development of rounded, oval ab- dominal plates at first was mistaken by the writer to be a very primitive development of the abdominal plates, but after being faced with the fact that the cox- osternal region was greatly advanced and specialized, I have arrived at the conclusion that these oval plates, joined only at the inner anterior part, without doub- lures, actually are equally as advanced as the coxos- ternal arrangement. What we have then, is that this scorpion is intermediate between the lobosterns and the living scorpions, and the oval plates are in the process of being reduced completely. If this hypothesis FOssIL SCORPIONIDA: KJELLESVIG- WAERING 25 is correct, then someday we should find a scorpion that has carried this reduction further until the underlying sternites have become sclerotized, when the group left the water completely. Once the abdominal plates are reduced —along with the gills, gill chamber and doub- lures—the sternites with their invaginated lungs will be developed. Sclerotization of the sternites, as well as slight sclerotization of the lung books, had to take place as the body wall was exposed. It seems safe to assume that Branchioscorpio rich- ardsoni had gills and, even equipped with an oral tube, could only leave the water for short periods, as the rounded opercula of the abdominal plates exposed the gills to the drying action of air. It appears as if this species of scorpion were an amphibious one, living much like some of the amphibious crabs. This hypothesis would explain the invagination of the lungs in the embryo of living scorpions and will clarify the discrepancies in the abdominal plate-ster- nite relationship. So far we have not found the inter- mediate stage of development, but we do have one specimen that points to that conclusion, Branchio- scorpio richardsoni. A phylogenetic tree showing these lines of development is shown in Text-figure 1. BREATHING MECHANISMS Our knowledge of the breathing mechanisms of scor- pions, except for the living pulmonates, is meager, but considerable knowledge of the mechanisms of fossil scorpions has been gained since the works of Petrun- kevitch (1955), who considered all to be pulmonates, and Wills (1947) who erroneously considered the Brit- ish Triassic scorpions to be pulmonates. In his work on the Carboniferous scorpions, Wills did consider the lobosterns to be aquatic, but still thought the holosterns were terrestrial forms (1960, p. 330). fc O 7S Lu oO ee — Lu [ae Oo [oe =) 7 ine Fk = a Lu a. 4 faa) a PROTOSTERNITE nM (hypothetical) |=! | [ HOLOSTERN PROTOLOBOSTERN BILOBOSTERN Oo ae wey a AP ao Loss of Doublures 2 EURYPTERID faa) S SCORPION- x EURYPTERID ie ANCESTOR i —_ = = £22 ae > ees ee ASC Text-figure 1.—Phylogenetic lines of development in the Scorpionida. After a sketch by Kjellesvig-Waering. The bilobosternous condition is known only from the Silurian and the Early Devonian. In structural transition from the archaic condition of exclusively possessing abdominal plates (AP) and gills for branchial aquatic respiration, to the sternite (ST) condition with lungs and stigmata (stig), for aerial respiration, they are analogues of that other great Devonian pioneer group, the Amphibia, which were achieving comparable structural and functional transition. Between the bilobosternous condition and the orthosternous, two, probably correlated, changes occurred: the first sternite, present beneath the abdominal plates of the bilobosterns, is lost, as are the paired abdominal plates of all segments. 26 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 I follow the definitions of lungs and gills given by Krogh (1941 (1968), p. 27), namely: lungs are all cav- ities serving respiratory purposes, lungs being the typ- ical organs for air breathing, whereas gills are append- ages used for respiratory purposes and being the typical organs for respiration of dissolved oxygen. There are, of course, water lungs (holothurians, etc.), tracheal res- piration and other forms not pertinent to the present discussion. We may dispense with the water lungs, but tracheal respiration is possibly important to this study. Tracheal respiration represents “‘air filled tubes origi- nating from special openings in the integument, the ‘spiracles’, and repeated branching reaching out to all parts of the body, thus obviating the convection by a circulating fluid” (Krogh, 1941 (1968), p. 114). Tra- cheae are primarily developed for air breathing and only secondarily adapted for an aquatic existence (Krogh, 1941 (1968), p. 115). Levi (1976) lists two types of tracheae, namely sieve and tubular tracheae, and summarizes the breathing mechanisms of living Arachnida, where each order is shown to have either book-lungs or tracheae, and of the Palpigradi, which lack respiratory and circulating systems. Our concern here is with the various types present in the Scorpion- ida. Although we have but meager knowledge of scor- pion gill systems, the breathing mechanisms may be summarized as follows: 1. Lungs: Pulmonate scorpions are known from the Middle Pennsylvanian to the present. The stigmata are round, slitlike or oval in shape, and occur in the pos- terior half of the sternite. There are four pairs of book- lungs, which are attached to the sternite. All pulmo- nates belong to the Orthosternina. 2. Gills such as characterize Tiphoscorpio hueberi: These are single lamellae composed of white, sponge- like tissue and consist of rounded, concentrically-mbbed layers of three distinct sizes, and with large spines on the lateral margins. Tiphoscorpio is a Meristosternina. The gills occur in a well-developed gill chamber. Water likely enters at the anterolateral angle and is expelled at the posterior gill slit located between the doublure and the abdominal plate. The gills are attached to the anterior of the gill chamber, which is on the exterior of the body wall. 3. Gill tracts of the type present in Paraisobuthus duobicarinatus: These are also typical of the eurypte- rids, namely, an oval area on the body wall (not the abdominal plate) composed of white spongy tissue with numerous cone-spines on the ventral surface of the gill tracts. The gill tracts occur in a gill chamber, and water enters at the anterolateral angle and is expelled at the rear of the abdominal plate. Paraisobuthus is a Lo- bosternina. 4. Gills that characterize most of the British Triassic scorpions: Wills (1947), for example, reported Wills- iscorpio bromsgroviensis (Wills), in which the gills are composed of a single (?) lamella of spongy tissue, with a fringe of spinelets on the posterior border. The an- terior is not known, but the gill should be attached to the anterior of the gill chamber—certainly not to the abdominal plate. Water presumably enters at the an- terolateral angles and is expelled by the posteriorly- located gill slits. Mesophonus perornatus Wills has sim- ilar gills, but without the fringe of spinelets. Both Will/s- iscorpio and Mesophonus are Holosternina. 5. Tracheae: The possibility of tracheae in fossil scor- pions must be admitted, although I consider the type should be described as gills. Stormer (1970) described some peculiar ribbon-shaped gills for the Devonian scorpion Waeringoscorpio hefteri Stormer. These are irregular oval bodies with what appear as ribbons or possibly crushed tubes. These are found outside the body, but it must be stated that this scorpion has under- gone ecdysis, and in life these gill-like or tracheal struc- tures may have been inside the gill chamber, although this is not a necessary consideration, since they could have been the type that could be extended into the water outside of the abdominal plates and back into the gill chamber. It is possible, but not probable, that these are branchial leaves, each of which is composed of masses of trachioles (see Krogh, 1941, p. 141, fig. 83). More likely, however, the structures are true gills, consisting of a single lamella, oval in shape and rein- forced by linear ribbing. COMPARISON OF SCORPIONS AND EURYPTERIDS Throughout the known geological history of the scor- pion, stretching from the Middle Silurian to the pres- ent, approximately 450 million years, only three major characters were developed that were universally unique to the scorpions in comparison to the eurypterids. These are: 1. The development of chelae on the pedipalps. 2. The development of combs. 3. The development of paired poison glands and an aculeus with semiterminal openings for the delivery of the poison. All other characters may be found in either the eu- rypterids, or gilled or pulmonate scorpions, but the three mentioned above seem to be major departures from the Eurypterida. For example, swimming paddles are not known in scorpions, but there are eurypterids also without them. Lungs are not known in eurypterids, but there are scorpions without lungs. Scorpions de- veloped claws, but many did not have claws. It emerges now that the most striking difference be- tween the Eurypterida and the Scorpionida is the de- FossiIL SCORPIONIDA: KJELLESVIG- WAERING Di velopment of pectines in the latter during Silurian time. The homology of these structures with anything in the eurypterids is problematical, but when the median or- gan of eurypterids is further studied, these homologies may appear. A word should be said regarding the poison-bearing telson of the scorpions, as some eurypterids, as well as some of the carcinosomatids, megalograptids and mix- opterids have a rather bulbous telson with a down- wardly curved spike that superficially resembles the stinger of the scorpion. Some authors have advanced the possibility that this telson may have harbored a pair of poison glands as in the scorpions. I do not consider this to be the case as the two telsons are greatly different and should not be confused. The scorpion aculeus is like a hypodermic needle—very slender, sharp and with the poison duct openings on each side of the point. The eurypterid telson does not show any open- ings whatsoever and, furthermore, is a very blunt spine incapable of penetration except into the softest tissue. 00 a' 0} im a) ya DnQ ty .,. J ww su\C | QO YC op pp A) Uf iN als AP4 } + els foe) N EURYPTERIDA PTV) UR Ppp 08, It is a powerful weapon and would serve greatly to fend off any attacks from above. The poison glands, if they were present in the eurypterid telson, would be located too far back and the poison would have to travel so far that it would be totally ineffective. In order to be effective, the poison has to be injected instantaneously into the victim. The eurypterid telson of the above families is not a short, slender, sharp, falconate struc- ture as in the scorpion stinger. The overall conclusion of these comparisons is that eurypterids and scorpions are closely related and de- serve a high classification that bespeaks this. EVOLUTIONARY CHANGES IN SCORPIONS Beginning as early as the Lower Devonian, several important changes took place in the known fossil scor- pions, some involving lines of phylogenetic signifi- cance. These changes were not necessarily concurrent with one another and seem to be associated with the emergence from water to land, and are listed in what ) y Ole We i Holosternina Bilobosternina Orthosternina BRANCHIOSCORPIONINA NEOSCORPIONINA SCORPIONIDA Text-figure 2.—A comparison of the ventral morphology of the Eurypterida and three infraorders of the Scorpionida. For simplicity, the meristosterns, protolobosterns and lobosterns of the Branchioscorpionina, to which the holosterns and bilobosterns belong, are not shown. Their homologies are directly and indisputably with the holosterns. See foldout inside front cover for explanation of abbreviations. 28 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 seems to be the approximate order of importance or significance. 1. The loss of abdominal plates and gills, and the development and sclerotization of the sternites and the appearance of lungs. These have been shown in Text- figure 2. 2. The forward development of the mesosomatic underside, resulting in the nearly total loss of the pre- pectinal plate and the first abdominal plate. The mod- ern scorpions, therefore, have only the second to the fifth sternites (inclusive) developed. 3. The development of the maxillary lobes of the first two coxae into the oral tube and the concurrent reduction in size of the chelicerae (see also 9 below). 4. The movement of the coxosternal area forward: that is, the development of the forward displacement of the fourth pair of coxae from the genital opercula to the sternum. 5. The dispersal of the lateral facetted eyes into lat- eral individual eyes, numbering five or less on each side. 6. The backward displacement of the median eyes from the anterior margin to the middle of the carapace or further, posteriorly. 7. A great reduction in the length of the tarsus on each leg. 8. The disappearance of the spines underneath the ungues, as well as shortening of the unguis and post- tarsus. 9. Reduction of the chelicera from four to three joints by disappearance of the second joint (see also 3 above). 10. Almost total disappearance of the anterior glos- sate process of the carapace, present only in a greatly reduced state in very few living scorpions—the South American Brachistosternus and the Mexican Typhlo- chactas. 11. The early scorpions (Proscorpius, Stoermero- scorpio, Archaeophonus, Palaeophonus, etc.) have short, terete legs, round in cross-section. These become great- ly flattened and semi-elliptical or flatter in cross-sec- tion and become lengthened. 12. The lengthening of the third and fourth coxae. 13. The gradual reduction in tibial and basitarsal spurs, both in size and occurrence on each leg. This change is considered to be by far the least important of the changes listed above. THE TRIASSIC SCORPIONS Our knowledge of Triassic scorpions is due entirely to Wills, who in 1910, and particularly in 1947, mi- nutely described the surprisingly well-preserved scor- pion fauna encountered in the Lower Keuper rocks of Worcestershire, England. The scorpions, mostly dis- jointed, were freed from the soft shale so that their preservation is almost as if the fragments belonged to living specimens. Unfortunately, they are disjointed, and being abundant fragments, their respective mor- phological identification is essential and paramount to reconstruction of the animal and determination of the various taxa present. The scorpion skins are so well preserved, that the original cuticle and color are pres- ent, including many details such as sensory hairs and other parts not commonly fossilized. It is because of Wills’ excellent detailed descriptions that it is possible to propose several important taxonomic changes in that fauna, which will have important significance to the overall taxonomy of the Scorpionida. The aspect of the Triassic scorpion fauna is surpris- ingly primitive and resembles closely some of the more primitive groups in the Carboniferous. Indeed, the coxosternal arrangement of Spongiophonus, with the lack of development of maxillary lobes on the second pair of coxae, and the junction of the first pair at mid- section with only the anterior half of the second pair of coxae meeting at the midline (see Text-fig. 110L), shows unusually and unexpectedly primitive charac- ters that were relicts from the more primitive Paleozoic groups, such as the Archaeoctonidae. There is nothing in the Triassic fauna to indicate any relationship with living forms whatsoever, but there is much that 1s anal- ogous to those of the Paleozoic. For instance, these Triassic scorpions, with the exception of one lobostern, are holosternous and unlike modern scorpions, all of which are orthosternous. I venture to speculate, not timidly, that the known Triassic scorpions continued with gills and remained in an aquatic environment into extinction sometime in the Mesozoic. We know that England, as well as America, had developed land scorpions, with as well- developed stigmata and lungs as those living today, as early as the Upper Carboniferous. Nevertheless, not one single fragment of a pulmonate scorpion occurs in the remarkable and incredibly rich Triassic material studied by Wills. These scorpions, belonging to the infraorder Holo- sternina and one specimen to the Lobosternina, are aquatic or amphibious scorpions for the following rea- sons: 1. The Triassic scorpions are morphologically much closer to Paleozoic scorpions than to Recent ones. The holosternous Silurian scorpion, Proscorpius osborni Whitfield, or the Carboniferous Mazonia woodiana Meek and Worthen had large openings in the posterior edge of the ‘“‘sternites” or doublures, much like those present in the Triassic genera. 2. If the Triassic scorpions were pulmonate forms, some, at least, would have been breathing air through stigmata that perforated the face of the sternites. Stig- mata of this type had been developed by Middle Penn- FossIL SCORPIONIDA: KJELLESVIG-WAERING 29 sylvanian time at the latest in North America (Vogel and Durden (1966) and discussions of Palaeopisth- acanthus schucherti Petrunkevitch, below), and in the Upper Carboniferous in England (see Compsoscorpius here), and were round and indistinguishable from those present in certain living scorpions. Stigmata in living scorpions are small, round or oval perforations or slits through the middle of each half of the sternite. It must be stressed that stigmata, if present, would be readily preserved in fossils. Incidentally, the small round ones that characterize some of the Chactidae and Chaeril- idae, the smallest and most inconspicuous type, have already been found. Preservation would be much easier for some other types of stigmata, e.g., the slits that occur in families such as the Buthidae and Scorpion- idae, which are surrounded by a lip or limbated, than for the small round holes already found in the Penn- sylvanian Mazon Creek Palaeopisthacanthus. Throughout the scorpionid literature neontologists, in interpreting fossils, repeatedly suggest that the stigmata might have occurred, but “being small they probably were not preserved’’. In fact, in the case of Palaeo- phonus nuncius, several well-known workers insisted that an inconclusive “‘crack’’ on one side of the “‘ster- nite’’ had to represent the stigma, although it was clear that the animal was bilaterally symmetrical and that the fortuitous ‘‘crack’’ was not present on the other side of the “‘sternites”. Any scorpion with stigmata perforating the cuticle of the sternite will show these stigmata in the same way that setae, setal sites and other much smaller structures are shown in complete clarity and preservation. I have examined the specimen in question (holotype of Palaeophonus nuncius Thorell and Lindstrém) and the supposed “‘stigma’”’ is not even a crack or perforated slit, but an indentation which does not perforate the skin. I have also examined other specimens of “‘sternites”’ (see Text-fig. 601) so perfectly preserved that they retain the original cuticle and even color, and can state without question that no perfo- rations that could be considered stigmata occur. It must be emphasized that of the many “‘sternites”’ recovered by Wills, none showed true stigmata; all were holosternous (except for one lobostern) with large slits either on the doublures opening internally, or margin- ally located at the posterior lateral sides. These plates are not sternites, but are abdominal plates, each of which has gill chambers. 3. The large slitlike openings on the doublures are indicative of water-dwellers rather than terrestrial forms. These long slits, bordered by large serrations, were located on the doublures, which were probably flexible and movable so that the serrations could keep extraneous material from the gills. Whereas aerial res- piration requires only very small stigmata in order to furnish adequate oxygen to the book-lungs, aquatic respiration requires larger apertures in order to pass the necessary large volume of oxygenated water over the gills. As Wills points out (1947, pp. 100-101), the position of the large slitlike opening of the Triassic holosterns is the same as found in the abdominal plates covering the gills of eurypterids, the Carboniferous scorpion Jsobuthus and the Silurian scorpion A/lopa- laeophonus caledonicus (Hunter). Curiously, however, he considers the Triassic forms to be lung-bearing and terrestrial rather than aquatic, breathing through gills! 4. The habitat of the known Triassic scorpions is the same as that of most other scorpions in the Upper Paleozoic. They commonly occur with Estheria, and a clam of doubtful generic designation has also been reported. Estheria indicates lagoons, probably of brackish water origin, perhaps stagnant. However, these lagoons had to be very widespread as the scorpions have been found in various locations, for example, Bromsgrove, Northfield, and even in a boring (or bor- ings?) at Charford Mills, Birmingham, England. In each case they occur in a greenish, very carbonaceous and micaceous shale. These scorpions are so abundant and persistent that Wills suggests that the shale (1947, p. ii) might serve as a much-needed marker horizon! One collects these scorpions by selecting the type of shale, not by actually finding the scorpion parts in the rock. They are so numerous that they can be removed from the soft shale by the use of warm water. 5. It seems safe to assume that the great majority of fossil scorpions are ecdysial exuviae. Ifa concentration is found, as occurs in the related eurypterids, it is likely because they congregated in that particular area during ecdysis in order to escape the destructive action of waves, currents or predators, or it represents subse- quent selective deposition. The fine textures and fine, evenly-layered sediments bespeak undoubted quiet conditions, which were best suited for ecdysis. Probably another safe assumption is that the pul- monate scorpions were solitary cryptozoic forms in the past, just as they are today, therefore, if the Triassic scorpions were terrestrial forms, ecdysis always should have occurred under protection, such as underneath rocks, logs, bark and various types of debris. Hygro- philic forms, living in tropical rain forests or jungles, always perform ecdysis under rocks, logs or bark. The same is true of semi-xerophilic forms. In regard to truly desert forms (Wills postulates that the Triassic scor- pions were desert dwellers), ecdysis also always occurs under protection, for example, under rocks or in bur- rows. These conditions are hardly conducive to the mass accumulation of exuviae into a common aquatic burial. It may be possible to have a concentration of the solitary scorpions due to a natural phenomenon such as a flood. However, in this case, other animals, cer- 30 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 tainly the countless other small terrestrial arthropods known to be living in Triassic time, would have been swept into the same burial site. However, in the stra- tum in question, the lithological characteristics are against the possibility of flood conditions. If the Trias- sic scorpions were terrestrial forms, it would be ex- tremely unlikely that their exuviae would be repeatedly swept or blown into pools, covered by sediments, all of the same type and age and occurring at several lo- calities, each time being associated with the same com- mon, water-dwelling companion (Estheria) and still be so abundant as to have the burial sites or horizon serve as a stratigraphical marker bed. These scorpions occur in “‘very carbonaceous” shales, an environment suggesting, as Wills points out, stag- nant conditions. It follows that there was little or no current to transport extraneous material. Then, how is it possible to have the great numbers of scorpion parts in the widespread ‘“‘marker’’ bed unless the scorpions were living in that environment? Winds could carry extraneous material into stagnant pools (assuming that these are pools and not widespread, contiguous bodies of water). Regardless of the type of transportation, it would be incredible that of all the many land arthro- pods living during Triassic time in England, only the scorpions were selected and swept into the pools in- habited by Estheria. One would have to come to the unmistakable conclusion that England at the time was literally covered and crawling with land scorpions, as they were the only traces, other than plants, that found their way into these stagnant pools. England likely did have terrestrial pulmonate scorpions during Paleozoic and later time, such as occurred in America, and this is indicated by the three specimens that are the holo- types and only known specimens of Compsoscorpius elegans Petrunkevitch, C. elongatus Petrunkevitch and Typhlopisthacanthus anglicus Petrunkevitch (all rep- resent the same species). The only doubt present is the fact that the underside of these three specimens is un- known, therefore we cannot be positive that they had stigmata. I believe that the presence of three lateral eyes on the carapace and the nearly central median eyes, as in Palaeopisthacanthus and in many living scorpions, indicate terrestrial scorpions. A truly terrestrial, cryptozoic arthropod is necessar- ily among the rarest of all fossils. This certainly seems to be the case with all Paleozoic Scorpionida, where only one certain and three possibly pulmonate and presumably terrestrial specimens have been found. The only other terrestrial scorpions known in the fossil rec- ord are extremely rare, as for example, in the Cenozoic, and are known from sinter beds or amber (Baltic and Dominican) where they were obviously trapped. 6. As evidence of land or desert existence, Wills (1947, p. 93) compares the tarsalia of one of the Trias- sic scorpions, which is very hairy, with the living Lio- buthus, which presumably uses the hairy tarsalia for digging in the sand or for walking on soft sand. These structures may be used as well in a watery environ- ment, as can be noted in countless examples among living aquatic animals (insects, crustacea, etc.) living on sandy or muddy bottoms. In Lower Triassic beds other than those of Britain, Gall (1971, pp. 33-34, unpubl. thesis) reports at least one scorpion? that is truly a pulmonate with typical sternites in the Upper Bunt sandstone of France. But this scorpion is very different from the known British Triassic forms. Another specimen represents a mesophonid. 7. Identification of the “lung lamellae”’ is crucial in determination of the mode of respiration, but not con- cerning the question of environment, for the Triassic scorpions may still be aquatic, but breathe air. In either case, with regard to environment, the above arguments apply. As to the identification of the lungs, it should be stressed that only single rounded laminae, some with spinules at the edge, have been found. No actual book-lungs have been recovered. Both the rounded laminae and the bordering protecting spinules may as easily be interpreted to be parts of gills rather than book-lungs. There are no external structural differences between book-gills and book-lungs. The large oval impressions noted on many Carboniferous scorpion “‘sternites” (see descriptions of Eoscorpius, Telmatoscorpio, Buthi- scorpius, Eoctonus, etc.) are much like those present in the eurypterids, which are considered as undoubted gills. They are much too large in comparison with liv- ing scorpion book-lungs and their orientation is not as in the modern book-lungs. For example, in living scor- pions the laminae of the book-lungs are perpendicular against the stigmata. They are therefore also perpen- dicular to the sternite. The single leaf shown by Wills as a purported lung could not have been perpendicular to the sternal plate as occurs in modern scorpion book- lungs. The great width of these fossil tissues would seriously crowd the internal organs if the laminae were perpendicular and not horizontal to the stigmatal opening or parallel to the sternite. Although it may be argued that only one plate of the book-lung is pre- served, this argument does not seem reasonable, as certainly the edges of more of the laminae should have been preserved in the several specimens known. It is noteworthy that the specimen of Palaeopisth- acanthus schucherti Petrunkevitch, an undoubted pul- 3 There are about 15 beautifully preserved specimens, some of them complete, apparently belonging to more than one taxon, that merit further study. A.S.C. FossiL SCORPIONIDA: KJELLESVIG-WAERING 31 monate scorpion in which the stigmata are clearly pre- served, does not show the presence of round depressions as described here (for Te/matoscorpio, Eoctonus, Eo- scorpius, etc.). These depressions are the gill pouches or chambers for the accommodation of large flat gills. Slowly, information about the types of gills present in fossil scorpions is being accumulated. Stoarmer shows single, rounded-out, filamentous structures in the Ger- man Lower Devonian holosternous scorpion Wae- ringoscorpio hefteri Stormer (1970, pp. 342-343). The gills of a British Carboniferous lobosternous scorpion, Paraisobuthus duobicarinatus, which are almost iden- tical to those in Eurypterida, are described below. Pe- culiar rounded, concentric gills occur in the New York Upper Devonian meristosternous scorpion 7iphoscor- pio hueberi, n. gen., n. sp. The gills preserved in the Triassic scorpions are much too fragmentary to offer much evidence as to the type present. Nevertheless, one thing is certain, they are not book-lungs as has previously been postulated by others, nor could they be oriented as in living scorpions, for even the frag- ments would be too large to be erected vertically to the posterior gill openings. To be vertical to the sternite would require a scorpion to be much thicker than wide. The Triassic scorpion gills are attached to the ante- rior of the body wall and are hanging free within a gill chamber composed of the abdominal plate as in other Paleozoic Holosternina and Lobosternina. 8. All Triassic scorpions reported by Wills have ab- dominal plates rather than sternites. The abdominal plates have well developed doublures, a structure not found in true sternites. This means that the Triassic scorpion had gills which occurred outside the body wall, whereas living scorpions have book-lungs and these are inside the body wall. The so-called stigmata on the doublures are nothing but large gill slits, and should not be confused with the small stigmata of the sternites (see discussion of abdominal plates vs. ster- nites, p. 23). 9. The Triassic and Paleozoic genera, even as far back as the Silurian, were commonly equipped with a downwardly bent, shovel- or plowlike triangular an- terior process, which is identical to the process in the aquatic eurypterids. It seems logical to assume that identical structures had similar functions. The plowlike process served as an excellent tool for digging into the soft mud bottoms, and for covering most of the dorsal surface of the eurypterid or scorpion with sediments. In the case of the scorpion, the anteriorly-located me- dian eyes, placed on a high ocellar node, would project above the mud, whereas most of the body remained covered. It is significant that any scorpion that was equipped with the anterior glossate process also had median eyes at the anterior of the carapace. Those without the anterior glossate process had median eyes that had receded toward the central part of the cara- pace. To go further, burrowing terrestrial (living) scor- pions such as Opisthacanthus, Scorpio, Heterometrus, Scorpiops and many others, have a cordate or nearly bilobed anterior margin, thus the opposite of the glos- sate anterior, and the median eyes are located in the middle, or even further back, of the carapace in all. Arguments may be advanced that in terrestrial bur- rowing arthropods, such as beetles, etc., an anterior digging process is present, but the eyes are always pro- tected, not as in the water-dwelling and mud-digging scorpions, which had no protection over their eyes. In the beetles the anterior process is only used in soft material and all digging is done by fossorial legs. In the eurypterids, unquestionably established as being aquatic, those with an anterior, downwardly bent, linguiform process (Mixopterus, Eocarcinosoma, Hughmilleria, Megalograptus) have the lateral eyes lo- cated in the anterior part of the carapace. Curiously enough, except for the genus Hughmilleria, the other three are eurypterids with many scorpionid character- istics, including the presence of a curved, stingerlike telson, although I do not consider that it contained paired poison glands. The large Gigantoscorpio willsi Stormer, which rivals some of the medium-sized eu- rypterids in size, also has the anterior glossate process. I doubt if anyone would seriously question the aquatic nature of such a large arthropod. (The Silurian Bron- toscorpio anglicus Kjellesvig-Waering is even larger.) The logical conclusion seems to be that the median process was needed by certain scorpions for digging in soft bottoms in an aquatic medium. 10. Thus it appears that for scorpions a correlation can be made between forwardly-located median eyes and life in a water medium. What is meant here by ‘anteriorly located eyes” should not be mistaken, for certain genera of the living families Chaerilidae and Chactidae are considered to have anteriorly-located eyes. In these genera, as in the terrestrial fossil genus Palaeopisthacanthus, the eyes are well back of the an- terior margin and actually lie in the center of the an- terior half of the carapace. In many of the Carbonif- erous scorpions, all of them with the anterior glossate process, the eyes are located on the anterior node, well elevated from the rest of the carapace. In the Triassic scorpions discovered by Wills, all carapaces of the var- ious genera (see discussion of Triassic genera in taxo- nomic part) had well-developed anterior nodes with median eyes, as well as the anterior glossate process. One genus, Willsiscorpio, has eyes so far forward that they lie on a linguiform process partly anterior to the anterior margin of the carapace. 11. The presence of large slitlike openings on the 32 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 doublures (therefore covered) of the abdominal plates, which do not have direct access to the external surface, is sufficient to invite skepticism of considering the scor- pion to be a pulmonate or terrestrial arthropod. It is reasonable to assume that the doublures were flexible and perhaps opened and closed by muscle action. The presence of these internal openings in terrestrial ar- thropods is incongruous, as it means that when the animal is at rest, the weight of the body would be sufficient to close the slitlike openings, thus depriving the lungs of air, or seriously obstructing the flow. The only way in which such an arthropod could breathe is by constant muscular action to keep the “‘sternites”’ open at all times, even when at rest. This is certainly a highly improbable breathing procedure. On the other hand, if the animal lived in an aqueous environment, these slits would not close, as the buoyancy of the water alone would be enough to counteract the weight of the body. I know of no land arthropod that has stigmata which do not lead directly to the exterior, whether the openings are ventral, dorsal or lateral, as the only pur- pose of the stigmata is access to air. Even in the case of the peculiar breathing apparatus of the scutigero- morph centipedes, which have a type of tracheal lung differing from other arthropods in being situated on the dorsal surface, the stigmatal slit opens through the tergite (Lawrence, pers. commun., 1966). Other Triassic scorpions had large slits at the pos- terior lateral part of the abdominal plate. Again the same objections listed above would apply. The objec- tion concerning the closing of the slits when the animal is at rest is of particular importance in this matter. The known British Triassic scorpions had chelicerae composed of four joints, as commonly found in Pa- leozoic scorpions. Living scorpions have three joints. The four joints were not previously recognized in Triassic scorpions, but they can be plainly seen in Wills’ (1947) figure 38BV (see description of specimen No. 178 on page 33 herein). This too reveals that these scorpions have closer affinity to the Paleozoic scor- pions than to modern pulmonate ones. For the reasons given above, the British Triassic scorpions are considered to be gill-bearing and not lung-bearing, as Wills believed (1910, 1947). Within this context, it may be well to summarize the chro- nology of the taxonomy of the Triassic scorpions, as some of it is highly confusing and erroneous, and is necessary because of the changes proposed. In 1910 (p. 307), Wills “provisionally” established the suborder Mesophonidea with the family Meso- phonidae, although I have not been able to find the diagnosis or description of either. The genus Meso- phonus was established and described with the species M. perornatus, M. bromsgroviensis, M. gracilis, and M. pulcherrimus. In 1947 Wills retained the suborder Mesophonidea, and the family Mesophonidae, al- though again no diagnosis or description of either was given. The family Mesophonidae, which contained only the genus Mesophonus, was enlarged to include the genus Spongiophonus with only one species, S. pus- tulosus. The genus Mesophonus comprised the same species given above, but further species were described. These were M. opisthophthalmus, M. pseudogracilis, ““M. infans forms A, B, C, D and E”, and M. pulcher- rimus var. immaculatus. Petrunkevitch (1953) did not accept the suborder Mesophonidea Wills, 1910, but did accept the family Mesophonidae Wills, 1910. Petrunkevitch, following Wills (1947), referred the genus Spongiophonus, along with the type genus Mesophonus, to this family. In the Treatise (1955), Petrunkevitch essentially followed this treatment. The superfamily Mesophonoidea Wills (nom. transl. Petrunkevitch) was then established, though not using the characters derived from the coxo- sternal arrangement of the type genus Mesophonus, but of an altogether different scorpion, Spongiophonus (er- roneously referred to as ““Spongiotarsus” in the Trea- tise). The construction of the coxosternal region of Spongiophonus is erroneously labeled by Petrunke- vitch as Mesophonus (1955, fig. 40(6)). Therefore, it was not surprising that Mesophonus and Spongio- phonus were put in the same family. No two scorpions could possibly be more different. Dubinin (1962) essentially followed Wills and Pe- trunkevitch, recognizing the family Mesophonidae Wills, 1910, with two genera, Mesophonus Wills, 1910, and Spongiophonus Wills, 1947. The family, however, was referred to the superfamily Scorpionoidea, which was enlarged to include all living scorpions as well as most of the Carboniferous families. The superfamily Scorpionoidea as such had little or no meaning. The family Mesophonidae Wills is therefore ‘‘ac- cepted” today as having two genera, Mesophonus and Spongiophonus, regardless of what the higher taxo- nomic divisions may be. To place Mesophonus and Spongiophonus in the same family is difficult to un- derstand as they obviously have totally different coxo- sternal patterns. The coxosternal arrangement of Me- sophonus is not well known, but enough is preserved to determine without question that it has no affinity, at least on superfamily level, to Spongiophonus. Before proceeding further with my diagnosis of the Triassic scorpions, certain morphological interpreta- tions and determinations, which differ from Wills (1910, 1947), must be clarified, since part of the tax- onomy proposed here is based on a new study and reinterpretation of some of Wills’ original material. I have studied a number of Wills’ pertinent specimens FossIL SCORPIONIDA: KJELLESVIG- WAERING that have a direct bearing on this problem. These spec- imens were kindly lent me by Dr. Harry Whittington, and are deposited in the Sedgwick Museum at Cam- bridge, England; other specimens were studied at Bir- mingham University through the kindness of Dr. F. W. Shotten. It was gratifying to note that the drawings made by Wills are highly accurate and could not be improved upon. I do differ from Wills in the deter- mination of some of the morphological parts (refer- ences are to Wills, 1947): these differences are as fol- lows: 1. Text-figure 13 (pl. V, fig. 3) labelled 4. pseudo- gracilis? Wills (specimen No. 094): Admittedly, the specimen is folded and broken and thus permits var- ious interpretations. I differ from Wills who showed the first “‘sternite’’ (abdominal plate) to be bilobed, whereas the rest are obviously not. The supposed bi- lobation, in my interpretation, is not present, and the first abdominal plate is seen to be of the same type as the succeeding, with a wide posterior prolongation of the central part. I am also unable to see a distinct central triangular disc or area, except possibly as an elevated area (ventrally), but not as a triangular plate with suture-like definition as figured. This group of “‘sternites”’ is referred to M. perornatus. 2. Text-figure 14 labelled 7. bromsgroviensis ? Wills (specimen No. 0163): This “sternite” has been assigned to the first ‘‘sternite’’ (Wills IX), although, as Wills shows, it is a distinct bilobed abdominal plate having the gill openings between the base of the plate and the doublure, and located considerably away from the lat- eral margins. He states (1947, p. 34) that the living Rhopalurus borelli has a trilobed sternite, as shown in Pocock’s (1904) figure reproduced in Werner (1934, p. 52, fig. 33). The first sternite of R. borelli is hardly trilobed—it merely has two stridulation areas, sepa- rated by a smooth central section, but these are not, in any sense, lobes. Furthermore, one is a sternite and the other an abdominal plate. Wills stated, and further inferred, that since the first “sternite” of some living scorpions differed from the other “‘sternites”, this was likely the case with No. 0163, and he assigned it to Mesophonus (‘‘For this reason I assign these two anomalous sternites (Nos. 0163 and 094) to the same segment in Mesophonus’’). The first sternite in living scorpions may be slightly different, but these are subtle differences like the ad- dition ofa stridulating area, slightly larger punctations, smooth central area, etc., but never of the magnitude shown in specimen No. 0163. No. 0163 is the abdom- inal plate of a distinct lobostern, deeply cleft at the middle, with the gill openings at the base of the plate, in a more central location and surrounded by much coarser denticles on the lips of the gill openings than Ww Ww in the other “‘sternites” known in this fauna. These are major morphological differences. This specimen, in my opinion, clearly indicates the presence of a Loboster- nina in the Triassic of Britain. It is not Mesophonus, as that genus clearly is a Holosternina. The supposed lobation of No. 094 does not exist, in my opinion. 3. Wills’(1947) text-figure 38 BV (Wills, 1910, pl. 25, fig. 6; specimen No. 178): The four joints char- acteristic of the chelicerae on Triassic scorpions are clearly shown in both the specimen and Wills’ excellent drawing. Referring to Wills (1947, text-fig. 38 BV), the points of difference are: I consider his ‘“‘c” to be the basal joint; his “*?c”’ is the second joint (‘‘ef”’ in re- construction); the rest is the hand and fixed ramus, which is the third joint. The fourth joint is the free ramus or dactyl. The four-jointed character of the che- licerae in the Triassic scorpions is another area where they are revealed as being closer to the Paleozoic scor- pions, than to Recent ones, which have only three joints. 4. Text-figure 40A-C, labelled Mesophonus sp., (specimen No. 040): This important specimen reveals what have been determined as the third and fourth pairs of coxae, the latter abutting a narrow pentagonal sternum. This arrangement of the coxae would mean that the first three pairs of coxae meet at midline, above the sternum, which is abutted only by the fourth pair. This results in a very unusual and improbable, if not impossible, coxal arrangement. Furthermore, the pair of coxae that Wills identified as the fourth pair does not have the shape of the usual fourth pair of coxae and is too short and too wide to be the fourth pair. This pair of coxae, marked “*C4”’ on text-figure 40B, is undoubtedly the third pair. The pair anterior to this, marked “*C3”’, is the second pair. Specimen No. 0110 in plate VIII, figure 3, completes the restoration of the Mesophonus coxosternal pattern. Wills claims that this is the fourth coxa, which he mounted on the plate with the posterior edge upper- most. This specimen is one of the most important reported in Wills’ monograph. It is mounted correctly, otherwise the so-called base of the coxa would have articulations, which do not occur at the base of any coxa. The specimen is not the fourth coxa but the first coxa and of extreme importance to taxonomy because of the presence of a small piece of the uppermost part of the second maxillary lobe, which can clearly be seen on plate VIII, figure 3 (Wills, 1947) to the left of the uppermost part of the first maxillary lobe. There seems to be little doubt that these two coxae belong to the same species as that represented above (specimen No. 040). The entire coxosternal area of Mesophonus com- prises the first two pairs of coxae with well-developed maxillary lobes, meeting at the midline, anterior to the 34 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 narrow sternum, which is abutted by a single pair. The fourth pair, although unknown, must abut the genital operculum. This new restoration (Text-fig. 28), which is made with considerable confidence, results in the taxonomic determination that the genus Mesophonus Wills, 1910, is to be assigned to an emended family Mesophonidae Wills, 1910, and is clearly referable to the emended superfamily Mesophonoidea Wills, 1910, which includes such well-known genera as Centro- machus Thorell and Lindstrém, 1885, and Mazonia Meek and Worthen, 1868. The family Mesophonidae Wills, 1910, is here emended (restricted) and seems to have clear priority over the Mazoniidae Petrunkevitch, 1953. All three families, incidentally, are emended here and bear no resemblance to their original descriptions. On the basis of the type of ornamentation, specimen Nos. 040 and 0110, which together give a nearly com- plete picture of the parts of the coxosternal region that reveal major taxonomic values, are referred to the lec- totype Mesophonus perornatus Wills, 1910. As will be shown below, Mesophonus opisthophthalmus Wills, 1947, is a junior synonym of the lectotype. 5. Text-figure 40D shows a poorly preserved speci- men (No. 164) that retains the last three coxae of the left side as well as part of the sternum and part of a genital operculum: The skins are very thin, and inter- pretation, because of folding on the slide, is very dif- ficult. Nevertheless, it appears that the sternum is def- initely elongate-elliptical, and it is reasonable to suspect that the last three pairs of coxae abut it, although it is not impossible that only the second and third pairs abut the sternum, whereas the fourth pair abuts the genital opercula. On Text-figures 25A, B are two pos- sible interpretations. Wills refers this specimen to Me- sophonus gracilis Wills (1947, p. 116). I agree, although here the species is referred to the new genus Steno- Scorpio. Wills (1910, 1947) and Petrunkevitch (1953, 1955) recognized one superfamily, Mesophonoidea Wills, 1910, including two genera, Mesophonus Wills, 1910, and Spongiophonus Wills, 1947. The genus Meso- phonus included the species M. perornatus, M. opis- thophthalmus, M. gracilis, M. pseudogracilis, M. bromsgroviensis, M. pulcherrimus and M. pulcherri- mus var. immaculatus, whereas the genus Spongio- phonus included only one species, S. pustulosus. My review of the British Triassic scorpions results in the presence of two infraorders, Holosternina and Lobosternina, three superfamilies in the Holosternina, namely Acanthoscorpionoidea, Mesophonoidea (emended) and Spongiophonoidea (new) and one un- designated superfamily in the Lobosternina. The fam- ilies represented are Mesophonidae Wills (emended), Willsiscorpionidae (new), Stenoscorpionidae (new), Spongiophonidae (new), and an undesignated family represented by the lobosternous scorpion. Genera rep- resented are Mesophonus (emended) with the species M. perornatus Wills, M. ? pulcherrimus Wills and M. ? pulcherrimus var. immaculatus Wills; Willsiscorpio, n. gen. with the species W. bromsgroviensis (Wills); Stenoscorpio, n. gen. with the species S. gracilis (Wills) and S. pseudogracilis (Wills); Spongiophonus pustu- losus Wills and Bromsgroviscorpio, n. gen. with the species B. willsi, n. sp. The species Mesophonus opis- thophthalmus Wills, 1947, was found to be a junior synonym of M. perornatus Wills, 1910. CLASSIFICATION INTRODUCTION The order Scorpionida is essentially a fossil group and any classification, if at all meaningful, must nec- essarily be based on the fossil record. This basic fact has not been fully appreciated. Of the 56 families herein recognized, only eight are living, all of which belong (or should belong) to one superfamily out of 21 known superfamilies. Like the order Xiphosurida, the Scor- pionida is an ancient and successful group that sur- vived from at least the early Paleozoic to the present. Indeed, scorpions have changed less in general external form since Silurian time, than have the Xiphosurida. In the classification of scorpions, as in all taxonomy, the problem is to find the proper taxobases. The clas- sification proposed by Petrunkevitch (1955), which was largely followed by Dubinin (1962), is founded on triv- ial traits. Important characters, such as gills, type of abdominal plates, lateral facetted eyes, and many of the coxosternal patterns of Paleozoic scorpions, were unknown, but in many instances the existence of some of these taxobases was categorically denied. The pro- posed classification embraces each of these characters. The ill-conceived postulation of an eight-segmented preabdomen in the adult unfortunately formed the main basis of Petrunkevitch’s most recent classification (see Kjellesvig-Waering, 1969). I believe there is little ar- gument that the respiratory structures, whether gills or lungs, are a fundamental taxobasis, for this involved not only one of the most radical morphological changes, but a great change of primary environment from a water medium to terrestrial existence. This was the last macroevolutionary change of the scorpions and prob- ably was the greatest stress that the entire Scorpionida ever experienced. It was a change that very likely was responsible for the survival of the remaining scorpions. The first definite air-breathing scorpion that we know of occurs in the Upper Carboniferous. All known Si- lurian and Lower Devonian scorpions were probably aquatic and had gills, although, judging from the de- FossiIL SCORPIONIDA: KJELLESVIG-WAERING 35 velopment of the oral chamber in Lower Devonian time, many could have been amphibious, much like many of the living crabs. It therefore may not be too speculative to suggest that the change from gills to lungs began at least by the late Devonian, concomitant with the transition from fish to amphibian. However, the amphibious development of the scorpion, that is, de- velopment of the oral tube for terrestrial living, but still having gills (see Branchioscorpio), began much earlier—I would venture that early Silurian or late Or- dovician may not be far from the mark. A further speculative suggestion is that the enormous increase of eurypterids, and especially of fishes, was responsible for the near extinction of the Paleozoic aquatic scor- pions. This extinction probably finally took place in post-Triassic time, but within the Mesozoic. Certainly, if fish were later to be their main competitors or ene- mies, the land could not have been as hostile during Silurian and Devonian time as were the marginal seas, or even limnic environments. As proven by their pres- ence today, the scorpions were well equipped for a terrestrial existence and survival once they developed lungs. Land, during the Silurian and Devonian, with the burgeoning of land plants and presence of herbiv- orous arthropods, offered countless ecological niches and food, which possibly had not existed previously. The scorpion soon found some niches that suited its requirements as a cryptozoic, carnivorous, solitary an- imal, including environments that ranged from the ex- tremes of xerophilic to hygrophilic, as far as the mac- roclimate was concerned, but always in a microclimate that offered sufficient humidity. They remain in the same microclimate today. We now have definite evidence that pulmonary, pos- sibly hygrophilic, scorpions were living during late Car- boniferous time, for example, the Palaeopisthacanthi- dae in Illinois (see Vogel and Durden, 1966, pp. 655- 658). It should be noted that the scorpion, whether i1n- habiting desert, subterranean, rain forest or even ar- boreal environments, is a cryptozoic, photophobic an- imal that lives only under conditions of considerable humidity. It is not possible for a modern scorpion to live in air devoid of moisture. All living scorpions, whether inhabiting deserts or not, are nocturnal. This is especially important for desert forms, in order to preserve their body fluids. They would quickly desic- cate if exposed to the higher aridity of daylight hours. Cloudsley-Thompson (1962a, p. 204) notes that the desert sands in the Sahara have a relative humidity of 50% at a depth of 50 cm even in the summertime. This is due to moisture from both Pleistocene connate and Recent extra-Saharan waters that underlie most of the sand areas. The microclimates under rocks in the Sa- hara therefore reveal a considerable degree of humid- ity. To speak of ‘“‘desert scorpions” as inhabiting a very dry environment is as highly misleading as it would be to speak of desert amphibians—and there are several that inhabit moist microenvironments of deserts. Both the scorpions and the amphibians take advantage of moist microclimates in the desert, where the optimum body moisture can be maintained. Both are nocturnal for the same reason. Therefore, the presence of gills or lungs is considered to be the paramount taxobasis in the new classification, hence, the basis of the suborders. Although the use of such a basis for fossil scorpions may entail some sub- jectivity in the determination of gills or lungs, it is a classification based on essential characters of survival, and not on preservation, or paleontological conve- nience. We now have considerable criteria for deter- mining the presence of these organs and, although de- ductions may not have been correct in each instance, the present alignments would seem to be closer to “‘nat- ural’ than any hitherto. In time, those fossil scorpions known only from the dorsal side will become better known by further discoveries, be it new materials or new information from old, by improved techniques, such as those developed and used so successfully by Wills (1959-1960). The fossil record is by no means as inadequate as some writers have postulated. Un- doubtedly, in any classification dealing with fossils, some genera may always remain “‘/ncertae sedis”. As- suredly, better rapport between paleontologists and neontologists would greatly reduce the uncertainties. The history of scorpion classification well illustrates the important contributions of paleobiologists, scien- tists who are well grounded in both disciplines. SUBORDERS Two suborders are here recognized. The Branchio- scorpionina possessed gills and were mainly aquatic. This includes most of the known Paleozoic and Trias- sic scorpions. The Neoscorpionina includes all pul- monate or mainly terrestrial scorpions. The basis for subdivision into infraorders is essen- tially that proposed by Pocock (1911) for suborders, although it is greatly modified and enlarged. Pocock recognized that the sternites (but not the abdominal plates) separated the scorpions into two groups: the Lobosterni with bilobed sternites and the Orthosterni with transversely rectilinear sternites, as in modern scorpions. The first group he considered to be aquatic, respiring by gills, and the second included the modern scorpions with lungs and stigmata perforating the face of the sternites. This was also followed by Wills (1959). However, the two divisions do not include all types of “‘sternites” present in the fossil record. Pocock (1911) 36 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 did not recognize one very important factor, namely, that the lobosterns did not have sternites but abdom- inal plates. Although Pocock did not recognize the difference between sternite and abdominal plate in scorpions, the basic idea of the use of the sternite—abdominal plates as a major division is very sound. Pocock’s diagnosis can be preserved with only a few redefinitions. There is no danger of confusion if these definitions are fol- lowed. In the literature, the term lobosternous or sub- order Lobosterni has meant only one character—that of the bilobation of the underlying plate. These are not sternites, but abdominal plates; nevertheless, I do not see any confusion in extending the term lobostern to those scorpions with abdominal plates. The etymology of the word certainly does not prohibit that use. In this manner we may preserve, intact, previous descriptions using the term lobostern. There are other abdominal plates not recognized by Pocock, or later authors. Therefore, with the following definitions, there should be no confusion concerning the sternite—abdominal plate nomenclature. In this connection, it is important to keep in mind that the sternite is part of the body wall—the abdominal plate is an appendage of the ster- nite or somite. Four infraorders are recognized under the Bran- chioscorpionina: INFRAORDER LOBOSTERNINA The term Lobosterni was proposed by Pocock (1911) for scorpions with ‘“‘sternites” that were deeply bilo- bate. It also includes abdominal plates that are gently bilobate, with a well-developed gill chamber, well-de- veloped doublures and with gill slits or openings be- tween the doublure and the abdominal plate. Lobo- sternous plates are not divided (see Text-fig. 4E). There are five lobosternous plates. INFRAORDER HOLOSTERNINA (new taxon) This is one of the most important groups and most confused. Pocock had included this group with the Orthosterni. Abdominal plates rectilinear, undivided and having well-developed gill chambers, with the gill slit or opening between the doublure and the plate, or on the doublure proper. This includes more fossil scor- pions than any other type. The type of plate is known from the Silurian through the Jurassic and includes nearly all the known British Triassic forms. The name is derived from the Greek: ho/o (whole) + sternus (chest) (see Text-fig. 4A). Eurypterids have holoster- nous plates. There are five non-paired holosternous plates. INFRAORDER MERISTOSTERNINA (new taxon, previously unrecognized) Abdominal plates divided by a median suture, as in many eurypterids, and retaining well-developed gill chambers and doublures. The gill slit or opening is between the doublure and the abdominal plate. The name is derived from the Greek: meristos (divided) + sternus (chest) (see Text-fig. 4D). There are five pairs of meristosternous plates. INFRAORDER BILOBOSTERNINA (new taxon) The abdominal plates consist of two completely sep- arated, rounded lobes, without doublures. The ster- nites may have developed above the plates and become slightly sclerotized. Considered still to retain gills, al- though primitive lungs may have developed. This is, of course, the most interesting condition of the ab- dominal plates (see Text-fig. 4C). There are five pairs of bilobosternous plates. The name is derived from: Latin: bi (double); and Greek: lobo (lobe) + sternus (chest). There is a fifth type of abdominal plate, the proto- lobosternous, which may possibly merit infraorder standing, but which is considered a variety of Lobo- sternina for the time being. Undivided, slightly lobate at the posterior end and without development of dou- blures, or if present, poorly developed. This type of abdominal plate is considered to be intermediate be- tween the more primitive holostern and the more ad- vanced lobostern (see Text-fig. 4B). There are five pairs of protolobosternous plates. One infraorder is recognized under the Neoscor- pionina: INFRAORDER ORTHOSTERNINA The term Orthosterni, as Pocock (1911) defined it, included the holosternous abdominal plates as well as the undivided, stigma-bearing, rectilinear, true ster- nites that characterize all living scorpions. The term is restricted to the latter, namely those with undivided sternites with stigmata perforating the sternite, and without doublures. There is little doubt that Pocock essentially meant this to be the case. Some New World Tityae have a deep median suturelike division of the sternites. This suturelike development seems to be sec- ondary, as neonates and juveniles do not have this development; it is acquired only on more adult forms (see Text-fig. 4F). There are only four orthosternous plates. Sternites are the only underplates with true stigmata and lungs. Abdominal plates, regardless of the type, FossiL SCORPIONIDA: KJELLESVIG-WAERING do not have stigmata, but have gill slits or gill chamber openings and gills. Somewhere in the fossil record we may expect to find the intermediate between the sub- orders (see Text-fig. 1), the ““~protosternites”’, which lat- er developed the paired lung invaginations, and were originally non-sclerotized tissue lying dorsal of the sclerotized abdominal plates in all eurypterid and gill- respiring aquatic scorpions. TAXOBASES The arrangement of the coxae forms the basis for the separation of superfamilies. This is one of the taxo- bases employed by Petrunkevitch (1955), but not ad- hered to consistently. Families are based on the de- velopment of maxillary lobes, the shape of the sternum, as well as the nature of the termination of the legs and the presence or absence of lateral facetted eyes. The arrangement of the coxosternal area as an important taxobasis was first employed by Thorell and Lindstré6m in 1885, and was later used more extensively by Pe- trunkevitch in his classifications of 1913, 1949, 1953 and 1955. Unfortunately, Petrunkevitch abandoned the use of the shape of the sternum in this connection (1953, p. 19): “in view of its relative unimportance and unreliability in the case of fossils .. .”. This now seems to have been an unwise decision. Wills (1959, p. 266) also did not believe that the coxosternal ar- rangement and the shape of the sternum were essential for classification, but his objections were also based on practical reasons and not because of taxonomic value, as he states, “In the past, attempts have been made to use coxo-sternal features for purposes of classification, but the method breaks down in practice because these parts are so seldom seen and because, when visible, they can so rarely be related to dorsal features.’” While this is to a small degree true, taxonomy and phylogeny AP AP A Holostern B Protolobostern G Bilobostern Ci ae E ) Meristostern AP (== Lobostern F Orthostern Text-figure 4.— Ventral plating in scorpionids, illustrating the basic differences between infraorders. In A to E there are five paired, or single, gill-bearing abdominal plates (AP), as in the eurypterids. In the orthosterns alone (fig. F), which comprise the true and all living scorpions, there are only four ventral plates, the sternites (St), which bear stigmata. The gill-bearing abdominal plates (AP) are not the same, nor homologous with, the stigma-bearing sternites (St). Although both kinds of plates serve the same, but analogous, function in ventral sheathing and external cover of the respiratory organs, they have quite different origins. 38 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 can hardly be soundly based on chance and expediency rather than on morphology. There is truly no one struc- ture or combination which will be preserved every time—the coxosternal area is no exception. I disagree that a classification must necessarily be practical, al- though desirable where feasible; fundamental mor- phologic factors must take precedence. As for the dor- sal surface, which is more commonly preserved, and which has also been largely employed as a taxobasis, Wills (1959, 1960, and in particular, 1964, p. 475) and others have shown conclusively in both fossil scorpions and eurypterids, that it is quite similar in many genera wherein the venters are quite dissimilar. It is improper, unrealistic, and therefore impracticable, to use differ- ent taxonomic characters for fossil and living organ- isms, or to base the fundamental taxonomy on most readily preserved or easily available morphology. A classification, if at all sound, should involve characters that indicate or involve major phyletic changes. As may be seen in the descriptions to follow, the underside is known in a surprising number of fossil scorpions. It is the view here, that the coxosternal area is among the most easily preserved structures, but one must study the specimen fully, and under various conditions, such as wet media, special lighting, etc. It is my opinion that the hierarchy of taxobases ac- cording to degree of reliability is as follows: For families, the presence and degree of develop- ment of maxillary lobes, shape of sternum, presence of lateral facetted eyes, as well as the termination of the legs. Spurs on the legs, which are considered im- portant in present-day classification of scorpions, are given no importance whatsoever, for higher taxa, and only minor importance on the generic and species level. For genera, the same criteria used in separating the genera of living scorpions: the shape of the carapace, position of median eyes, number of individual facets in the lateral eyes, type of combs, type and character and dentition of pedipalps and chelicerae, carinae on the mesosoma, etc. The superfamilies of Birula (1917) do not appear to be necessary in dividing the present-day scorpions, as all eight families obviously belong to a single super- family, if fossil forms are to be included and considered on a comparative basis. Inversely, if the superfamilies of Birula are accepted for the present-day scorpions, corresponding elevations in taxonomic levels must be made for the fossil groups. At the present state of knowledge, little or nothing would be gained by such proposals. It seems more reasonable to refer all living scorpions to the superfamily Scorpionoidea Leach, 1815. Moreover, arguments by some, myself included, might well be advanced that the number of living fam- ilies is too large and that some could well be absorbed with possible better expression of generic relationships. It may reflect paleontologic bias, but in the broad sur- vey of all known scorpions, the presence or absence of certain tibial or basitarsal spines on any particular leg, commonly employed by neontologists as a prime fam- ily taxobasis, seems trivial. However, there is little or no conflict between the classification proposed here and the present neontologic taxonomy. It is well to emphasize by repetition, the opening statement of this section, which is that the Scorpionida is essentially a fossil group. Similarly, the use of the arrangement of dentition on the chelicerae for the determination of families, as proposed by Vachon (1963), is an excellent taxobasis for use with Recent scorpions. However, this taxobasis has importance only in dealing with living families, an infinitesimally small part of the Scorpionida. The method has no importance whatsoever as a taxobasis in dealing with the greater part of the Scorpionida, namely the entire Branchioscorpionina. SYSTEMATIC PALEONTOLOGY Order SCORPIONIDA Latreille, 1817 Suborder BRANCHIOSCORPIONINA, new suborder Aquatic scorpions breathing through branchia or gills. Type family. —Branchioscorpiidae, new family. Remarks. — All but two of the families of fossil scor- pions known and described from the Paleozoic and the Mesozoic belong to the Branchioscorpionina. Infraorder HOLOSTERNINA, new infraorder Branchioscorpionina with five rectangular undivid- ed abdominal plates and well-developed gill chambers with gill slits opening between the doublure and the abdominal plate, or on the doublure proper. Remarks. —The Holosternina were once included in Pocock’s Orthosterni. The most successful group under the Branchioscorpionina, the Holosternina flourished from the Silurian through the Jurassic. From a half dozen species in the Silurian, it branched out into 41 species, classified herein under ten superfamilies, 25 families, and 32 genera. By the Jurassic the Holoster- nina had dwindled down to two species and became extinct. FossIL SCORPIONIDA: KJELLESVIG- WAERING 39 Superfamily PROSCORPIOIDEA Scudder, 1885 (nom. transl. Kjellesvig-Waering herein, ex subfamily Proscorpionini Scudder, 1885) emend.* Holosternina with first pair of coxae meeting in front of, but all four pairs abutting large sternum; no max- illary lobes. Type family. —Proscorpiidae Scudder, 1885. Family PROSCORPIIDAE Scudder, 1885 (emend. Kjellesvig-Waering, 1966, pro Proscorpionidae Scudder, 1885) Proscorpioidea bearing compound eyes at the an- terolateral angles of the carapace and with anteriorly- located median eye node having small ocelli. Walking legs have two claws. Sternum ovoid, elongated. Type genus. —Proscorpius Whitfield, 1885. Genus PROSCORPIUS Whitfield, 1885 (emend.) Proscorpiidae with subquadrate carapace with an- teriorly-elevated cephalic area>; large bulbous com- pound eyes at the anterolateral corners; eye node scu- telliform, highly elevated, located anteriorly, and having two small ocelli. Chelicerae large and strong, composed of four joints. Pedipalps have fingers meeting through- out their length, incurved toward the fixed finger; very small, bearing denticles; possibly linear. Sternum large, lacrimiform, with a narrow notch at the anterior for the small triangular plate of the labrum. The first pair of coxae meet in front, but all four pairs abut the ster- num. Small round opercular plates occur posterior to the sternum. Short, tubular walking legs have bifid terminations and very small, double, basitarsal and tarsal spurs. Preabdomen elongated, consisting of sev- en tergites that are rounded at the corners. Pectines comprise large, paired, undivided lobelike (rachis) plates with about 20 teeth on a side. Five very wide and long abdominal plates of the holostern type are present, with indications of large slits at the epimeral 4 The superfamily Proscorpioidea was mentioned by Kjellesvig- Waering (1966, p. 361) in his discussion of the Classification of the Silurian Scorpions of New York, but was not formally proposed. In the present manuscript the superfamily was scheduled for proposal, but the diagnosis not yet written at the time of his death. We have extracted this diagnosis from his data, in accordance with his ideas of superfamily criteria. The six Ciurca specimens of Proscorpius osborni (Whitfield), which are analyzed below, furnish much new information for diagnostic revision at all systematic levels, including Kjellesvig-Waering’s own previous (1966) emendation of previous systematic categories. A.S.C. > Kjellesvig-Waering’s diagnosis (1966, p. 362) adds: ‘‘and an el- evated medianly divided band along the base”, which may be present in the holotype, but which does not occur in any of the six Ciurca specimens. A.S.C. angles. Cauda bearing double dorsal, lateral, and ven- tral crests. Terminalia consist of normal stinger. Type species. —Palaeophonus osborni Whitfield, 1885. Geological range. —Upper Silurian of New York. Remarks. —Generic comparisons are discussed in Kjellesvig-Waering (1966, p. 362). Proscorpius osborni (Whitfield, 1885) Plate 1, Text-figures 6-13, 111A 1885a. Palaeophonus osborni Whitfield, p. 87, 1 fig. 1885b. Proscorpius osborni (Whitfield). Whitfield, pp. 181-187, pl. 20. 1885. Proscorpius osborni (Whitfield). Scudder, p. 739, fig. 915a. 1886. Proscorpius osborni (Whitfield). Whitfield, p. 216. 1886. Proscorpius osborni (Whitfield). Scudder, p. 28. 1886. Proscorpius osborni (Whitfield). Thorell, p. 269. 1899. Proscorpius osborni (Whitfield). Laurie, p. 577. 1901. Proscorpius osborni (Whitfield). Pocock, p. 309. 1904. Proscorpius osborni (Whitfield). Frit, p. 65, fig. 81. 1907. Proscorpius osborni (Whitfield). Frié, p. 6, pl. 3. 1912. Proscorpius osborni (Whitfield). Clarke and Ruedemann, pp. 387-400, figs. 81-83, pl. 88. 1913. Proscorpius osborni (Whitfield). Petrunkevitch, p. 32. 1944. Proscorpius osborni (Whitfield). Lehmann, p. 177. 1949. Palaeophonus osborni Whitfield. Petrunkevitch, pp. 129-131, fig. 189. 1953. Proscorpius osborni (Whitfield). Petrunkevitch, p. 12, fig. 118. 1954. Proscorpius osborni (Whitfield). Kjellesvig-Waering, p. 485. 1955. Proscorpius osborni (Whitfield). Petrunkevitch, p. P70, figs. 38(3), 39(A). 1962. Proscorpius osborni (Whitfield). Dubinin, p. 425, fig. 1220. 1966. Proscorpius osborni (Whitfield). Kjellesvig-Waering, pp. 361- 373, pl. 42, figs. 2-3; pl. 43, figs. 2-4; pl. 44; text-figs. 1-4, 11-18. Inasmuch as the earlier specimens were thoroughly covered in the 1966 paper, and all comparisons made therein, only the new Ciurca specimens® are covered below. Type information. —All specimens come from the Fiddlers Green Dolomite Member of the Bertie For- mation, commonly known as the “Bertie Waterlime”’, from the Upper Silurian, Ludlow age, of New York. The holotype, in the collections of the American Mu- seum of Natural History, New York City, comes from Waterville, Oneida County. All other specimens were found in Herkimer County: AMNH 28383, CUGM 41973’, and CIURCA 031966-1, 041771-1, 042570- 1A, 040564 and 040668-1,2 in the upper part of the Fiddlers Green Dolomite at Passage Gulf; CIURCA 072869-9B from near Litchfield. The specimens col- © Kjellesvig-Waering, in the first phase of this manuscript, set up anew genus and species for the specimens from the Ciurca collection, with CIURCA 031966-1 as the holotype, but later reconsidered and placed them under Proscorpius osborni (Whitfield). A.S.C. 7 Ed Landing (oral commun., May 7, 1985) reports that CUGM 41973 was transferred to the collections of the New York State Museum in Albany, NY in 1971, as NYSM 12948. P.R.H. 40 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 lected by Samuel J. Ciurca of Rochester, NY, are in his private collection, except for CIURCA 031966-1 which was deposited in the Buffalo Museum of Science as BMS E25162. Specimen I.—BMS E25162 [=CIURCA 031966-1 (Text-figs. 6, 12A)], collected by Samuel J. Ciurca, 1966, from the upper part of the Fiddlers Green Dolomite at Passage Gulf, NY. The specimen consists of a complete scorpion pre- served from the inside of the dorsal side, in light gray, typical waterlime. The specimen has been compressed, with little relief present, although considerable details of the morphology can be ascertained by use of very oblique lighting. Features such as the coxosternal re- gion can be made out faintly, particularly in the dry state. Otherwise, all details reported here have been determined by immersion in alcohol and use of dif- ferent intensities and direction of lights. The carapace is broad at the base, straight, with lat- eral margins that converge anteriorly. In life, the lateral margins were likely parallel. The anterior margin is produced into a wide snout that probably was origi- nally slightly curved downward as in certain eurypte- rids. Two prominent bulbous compound eyes are lo- cated at the anterolateral margins. These are typical compound eyes and the facets, or ommatidia, are pre- served in the right eye of the specimen. They are very small. The median ocellar node has two small median eyes on the lateral margins, and is located so that it lies on a plane slightly above or forward of the lateral compound eyes. The eye node is roughly elliptical and occupies the area of the snoutlike projection at the anterior of the carapace. The prosomal appendages are only partly preserved. The chelicerae are particularly well preserved and show that they were formidable organs approximately three- quarters the length of the carapace. They, therefore, jutted forward into two large grasping organs located above the coxae of the pedipalps, as in modern scor- pions. The chelicera is composed of four distinct joints; the last two are falcate and two large teeth were noted, at least on the free finger. It seems probable that more teeth occurred originally. The distinct pedipalps are noteworthy for the great length of the first segment, which represents the coxa. This is tubular and reaches almost to the base of the fixed finger of the chelicera. The trochanter is round along the anterior margins and otherwise triangular in shape. The femur is very long and contrasts with the very short tibia. The palpus has a long hand followed by a recurving fixed finger. Along the edge of this finger is a well-developed longitudinal ridge. The free finger is long and blunt at the end, slightly curved forward and ornamented with a longitudinal narrow ridge. This longitudinal narrow ridge may serve to separate this as a species from other scorpions in the Fiddlers Green horizon. The first walking leg consists of eight segments, all of which are very short and tubular, and terminates in two claws, the anterior claw being longer than the pos- terior. The other legs are also tubular, but are preserved only as fragments. On the first leg, and on what likely is the second leg, there are two spurs, which represent the tarsal spurs, on either side of the end of the pen- ultimate joint. The coxosternal region is particularly interesting. The sternum is elongate-scutelliform, pointed anteriorly and truncated at the base. The lateral margins converge anteriorly. The last three pairs of coxae abut against the sternum, whereas the first pair of coxae meets di- rectly in front of the sternum and abuts against the upper part of the sternum. The preabdomen consists of the usual seven tergites. Greatest width is reached at the fifth tergite. Each ter- gite is bounded by an ear at the anterior lateral margins and by a prominent transverse ridge along the anterior edge. Both structures occur in living scorpions. All tergites are rounded at the epimeral corners. Each ter- gite increases progressively in length posteriorly. The last tergite has two prominent carinae located on the middle of each half. The caudal segments are all par- allel, appear to be tubular, longer than wide, and each is surmounted on the dorsal side by two carinae. These carinae are in turn surmounted by granules. At least one ventral carina, composed of granules, occurs on the second to the fourth tergites and very likely on the first and fifth. The stinger is very well preserved. It reveals at least one dorsal carina on each side, although only one side has been noted. There is no sub-aculear node and the aculeus is highly curved and hooklike, which differs greatly from that present in the other scorpions of this horizon. The abdominal plates are only preserved at the edges, but, very interestingly, these edges show that there were five, rather than four, plates present. These are rounded and apparently broadly emarginate along the central posterior part. There is no sign of any teeth in the first abdominal plate, and therefore the first plate could not Text-figure 6.—Proscorpius osborni (Whitfield). From the Upper Silurian, Bertie Waterlime, Herkimer Co., NY. Specimen I, BMS E25162 (=CIURCA 031966-1). See also Text-figure 12A. See fold- out inside front cover for explanation of abbreviations. A. Anterior details of figure B, showing the carapace and first few appendages. Broken lines show underlying coxae of the pedipalps, which would be overlain, as in Recent scorpions, by parts of the chelicerae. B. Dorsal aspect of an unusually lovely specimen of fossil scorpionid. C. Mesosomal details. D. Coxosternal organization, tak- en from the counterpart. E. Left prosomal appendages. F. Distal postabdomen and telson. 42 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 be the anterior part of the combs. There is no sign of stigmata present on any abdominal plate. The species is definitely a holosternous scorpion. Measurements (in mm) of Specimen I, BMS E25162 (=CIURCA 031966-1).— Prosoma: Width at base: 4.8 Greatest width: 4.8 Length: 5.0 Location of median eyes: From anterior margin: 0.15 From posterior margin: 4.0 From lateral margin: 0.9 Diameter of median eyes: 0.1 Palpus: Chela length: 5.4 Hand width: 1.6 Hand length: 23 Free finger length: Femur length (dorsolateral): Tibia length (dorsolateral): 3.0 Mesosoma and metasoma: middle length width of tergite of tergite No. 1 0.7 4.2 No. 2 1.6 5.0 No. 3 2.0 6.2 No. 4 Qe 6.6 No. 5 Dal 6.9 No. 6 23 6.4 No. 7 4.0 anterior 5.4 posterior 3.2 No. 8 2.6 265) No. 9 3.0 2.2 No. 10 3.0 22 No. 11 3.2) 2a No. 12 3.1 21 Vesicle: Length: 3.2+ Width: 1.0 Spine: Not entire Remarks. —Occurs next to a fragment of the ventral shield of an eurypterid, probably Eurypterus remipes remipes De Kay. It was on a big slab on a bedding plane 115 mm from a horizon with large four-inch mud cracks; presumably below the mud cracks. The genital plates and combs were not preserved. Specimen IT, —CIURCA 041771-1, part and coun- terpart of a complete specimen (Text-fig. 7) from the Upper Silurian, upper part of Fiddlers Green dolomite at Passage Gulf, Herkimer County, NY. The specimen is preserved as a dorsal impression of the dorsal side, but with the edges of the abdominal plates, which are displaced laterally, seen from the dor- sal side. Remarks here are restricted to parts not pre- served in other specimens. The preservation of the chelicerae is noteworthy (see Text-fig. 7B); the chela of the right chelicera is open, revealing the entire structure. The free ramus (Ch4) is decidedly hooked, slender, with numerous teeth, of which a central tooth is most prominent. The fixed ramus (Ch3) is very narrow, also with numerous teeth. The manus is not as long as wide by about one third. The chelicerae are preserved in their entirety, showing all four joints. The first joint is seen whole for the first time. This basal joint is triangular, cup-shaped, and very short. The second joint is a narrow, cup-shaped band, showing denticles at the inner edge. A faint joint line was noted on the central part, which was inter- preted as the ventral side of the second joint (Ch2). This would give the joint vertical movement and con- siderable inside movement, as the joint line is on a diagonal, not as long as the dorsal side. The entire chelicera is covered with small punctae, which un- doubtedly are the bases for the setae. These setae are more densely packed toward the median line where they would rub against the other chelicera, and were probably used to convey food to the oral opening. The carapace is very poorly preserved; only the left posterolateral angle and the base are present. This part is typical of Proscorpius osborni. The legs are remarkable in their preservation. All are short, but not stout. The first two pairs are partic- ularly stunted. All legs are armed with very small dou- ble basitarsal and tarsal spurs (see Text-figs. 7D-G). The termination is in a double claw, the anterior being larger than the posterior, and in the posttarsus, which is very small. The legs are also noteworthy for the long, narrow tarsus. The tergites are typical for the species. The abdom- inal plates are rounded, and the third is of particular interest as it shows a wide slit at the junction of the doublure with the plate (see Text-fig. 7C). This dou- blure is like those described elsewhere (Wills, 1960, p. 299) and in this report where the gill-bearing holo- sternous forms are characterized. These tergites show the characteristic anterior prolongation at the antero- lateral angle. The cauda, preserved in its entirety, is rather stout, including a relatively large stinger, but the last tergite is very long, narrower than the preceding, and un- doubtedly denotes a male. The telson, or stinger, com- prises a large, bulbous, rounded vesicle followed by a curved narrow aculeus. Text-figure 7.—Proscorpius osborni (Whitfield). From the Up- per Silurian Bertie Waterlime, Herkimer Co., NY. Specimen II, CIURCA 041771-1. See foldout inside front cover for explanation of abbreviations. A. General aspect of the specimen. B. Detail of the chelicerae, showing denticles. C. Showing gill slit of third abdominal plate (AP3). D. Terminal details of first left walking leg (I). E. Terminal details of the third left leg (III). F. Terminal details of the fourth left leg (IV). G. Terminal details of the fourth right leg (IV). FossiL SCORPIONIDA: KJELLESVIG-WAERING 43 44 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 chs Text-figure 8.—Proscorpius osborni (Whitfield). From the Up- chs ff cha per Silurian Bertie Waterlime, Herkimer Co., NY. Specimen III, P2 CIURCA 040564. Overall length from chelicera to tail is 19 mm. i See foldout inside front cover for explanation of abbreviations. A. Whole specimen. B. Detail of telson. C. Postabdomen showing the elaborate carination of tergites T8-12. sup cr lat cr covered 1mm PiVeeses lat icr inf cr superimposed over inf cr latch —— covered FossiIL SCORPIONIDA: KJELLESVIG- WAERING 45 Remarks. —Sexual dimorphism in the form of the short (2) and long (¢) twelfth tergite is particularly well illustrated in this species. Specimen IIT.—CTURCA 040564 from the same Passage Gulf location and formation (Text-figs. 8, 9). The rather small scorpion, preserved only on the dorsal side, consists of a single specimen without a counterpart. It is possibly the best preserved of the Passage Gulf specimens, although, unfortunately, the central part of the carapace has been lost. Considerable information, however, has been obtained from the parts preserved. The prosomal legs are shown in great clarity (see Text-fig. 9) and it is clear that each leg contains two tibial and two basitarsal spurs. The tarsal spurs, of course, are the double ungues and are not greatly unlike those of living forms. The posttarsus is very short and also like that in living species. The short legs and the very long tarsus are clearly very different from those in living forms. The cauda (see Text-figs. 8A, C) is so well preserved that details of the crests can be determined. It appears 1mm that all segments retain two superior, two lateral (on each side) and two ventral crests. The ventral crests are very narrow, marked by elongated, setaceous pus- tules. The superior and lateral crests are wider knobby carinae, partly setaceous. The aculeus and vesicle are preserved in their en- tirety. The vesicle or stinger is slender, long, and curv- ing (see Text-fig. 8B). No other specimen shows the aculeus as well. Specimen IV.—CIURCA 072869-9B (Text-fig. 10) from the top of the Fiddlers Green Dolomite at con- fidential locality No. 1 of Mr. S. J. Ciurca. The specimen, poorly preserved, consists of only one side of the abdomen, but reveals certain anatomical features of considerable importance. The specimen shows only the last five tergites of the preabdomen along with the imprint of the inside or ventral side of three abdominal plates. Part of the ventral side of the seventh preabdominal tergite is also shown. Of greatest importance is the fact that a large comb is preserved. This is triangular, very long, without any segmentation or any rounded sclerites, and appears to have been an IN6S o> IV5S IV right Text-figure 9.—Proscorpius osborni (Whitfield). From the Upper Silurian Bertie Waterlime, Herkimer Co., NY. Specimen III, CIURCA 040564. See foldout inside front cover for explanation of abbreviations. A. Terminal details of the third right leg (III). B. Terminal details of the first left leg (I). C. Terminal details of the second left leg (II). D. Nearly complete fourth right walking leg (IV). 46 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 undivided rachis as in Branchioscorpio richardsoni of the Devonian. Although poorly preserved, several teeth can be faintly made out. These are relatively broad, but not of the broad type found in Branchioscorpio richardsoni. Large triangular fulcra separate the teeth. Although only a few of the teeth were preserved, it can be estimated that approximately 20 occupied the edge of each pectine. The discovery of the pectine in Pros- corpius has verified the interpretation given to the extra anterior “sternite” which appears on the holotype (Kjellesvig-Waering, 1966, p. 365). Itis thus confirmed that this extra “‘sternite”’ is the comb, the teeth of which had been covered by the first abdominal plate in the holotype. The pectines are very wide, and along with the narrowness of the abdominal plates, likely indicate that this specimen was a male. The two pectines would probably extend beyond the edges of the abdomen. The three well-preserved abdominal plates are very long with a narrow marginal rim, very likely raised in life, along the anterior edge, and all are rounded on the corners. Of particular interest is the fact that on the fifth abdominal plate can be seen what appears to be a large slit at the epimeral angles. This is much like the large slits present in the doublures of the Triassic scorpions (see Wills, 1947). The large slit also indicates a water-dweller, as it is much too large for anything that might have been useful for a pulmonate scorpion. Gills require a larger opening than lungs in order to receive enough oxygen and expel the volumes of water necessary. The last preabdominal tergite is preserved so that the dorsal surface could be stripped away revealing the underside. There is no ornamentation on the dorsal surface, nor were any dorsal carinae noted. The ventral side, however, has a strong diagonal double carina. There was no ornamentation on any of the tergites or abdominal plates. Another confirmatory feature presented by this scor- pion has been that all the abdominal plates are com- pletely holosternous. Measurements (in mm) of Specimen IV (CIURCA 072869-9B).— Pectines: Length of rachis (not including teeth) 2.0 Width of rachis (along line of fulcra) 331 Tooth length 25 Tooth width .09 Abdominal plate: Length (No. 3) 1.8 Width (No. 3) 3:2 Seventh tergite: Anterior width 4.6 Posterior width 21 Length 3.0 Specimen V.—CIURCA 042570-1A (Text-fig. 11A) from the Fiddlers Green Dolomite Member at Passage Gulf. Consists of a single, nearly complete, specimen pre- served as a dorsal impression. The entire specimen is estimated at an overall length of 17.5 mm from the anterior of the carapace to the estimated end of the cauda. It shows well-developed lateral compound eyes and the ocellar mound with well-preserved ocelli. The tergites are all much narrower than the underlying ab- dominal plates. The specimen merely verifies some of the details previously noted in other specimens. Specimen VI. —CIURCA 040668-1,2 (Text-figs. 11B-D, 12), from the Upper Silurian Bertie Waterlime, Herkimer Co., NY. The underside of this specimen is well preserved and, under alcohol, many details hitherto unknown or not fully understood can be discerned. The coxosternal region is preserved in its entirety and reveals various structures not previously noted (see Text-fig. 12B). The sternum is lacrimiform with a narrow notch at the Imm Text-figure 10.—Proscorpius osborni (Whitfield). From the Up- per Silurian Bertie Waterlime, Herkimer Co., NY. Specimen IV, CIURCA 072869-9B. Showing the abdominal plates and a gill slit on AP5; also the pectinal plate with a few teeth. See foldout inside front cover for explanation of abbreviations. FossiL SCORPIONIDA: KJELLESVIG- WAERING 47 Il right Text-figure 11.—Proscorpius osborni (Whitfield). From the Upper Silurian Bertie Waterlime, Herkimer Co., NY. Specimens V and VI, CIURCA 042570-1A and 040668-1. See also Text-figure 12. See foldout inside front cover for explanation of abbreviations. A. Whole of Specimen V, CIURCA 042570-1A. B. Second right leg (II) on Specimen VI, CIURCA 040668-1, a counterpart. C. Right pedipalp, dorsal side, Specimen VI, CIURCA 040668-1. D. Left pedipalp, dorsal side, Specimen VI, CIURCA 040668-1. 48 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 anterior where the small, triangular plate of the labrum adjoins it. This triangular plate seems to fit where the sternum would be continued if it were pointed. The two structures are separated, however. The mouth therefore lies between the two plates. The triangular plate is bowed anteriorly, and in life would have had a scooplike appearance. The last three pairs of coxae, all short and truncated, abut the lacrimiform sternum. The first pair is remarkable in having a distinct gnatho- base, a structure well known in the eurypterids, and, although expected in the early scorpions, that has not been found or noted until now. The posterior part of the first pair of coxae abuts the sternum and the la- brum, but the greater part meets at midsection in front of the pointed labrum. The gnathobase is composed of small triangular denticles that occur in a single row, larger anteriorly and decreasing in size posteriorly al- most exactly like the gnathobases adorning the coxae of the eurypterids. The mouth, therefore, was not at the end of a tube formed by the maxillary lobes, che- licerae and pedipalp bases, as in modern scorpions, although it occupied the same position relative to the carapace. Chewing occurred outside of the oral open- ing, as in the eurypterids, and very likely a pellet, made of the hard undigestible parts, was rejected. This seems to have been the method used by the eurypterids. The “coprolites” in the form of tubular bodies of apparently masticated chitinous skins associated with the euryp- terid Megalograptus ohioensis (Caster and Kjellesvig- Waering, 1964, p. 337) apparently are not fecal in or- igin, but are rejected masticatory pellets. The sternum is large, ovate, widest at the posterior end and, as mentioned before, has a deep triangular notch that accommodates the parrot-beaklike sclerite, which is the labrum. The chelicerae are composed of four joints, occur at a higher plane than the pedipalps, and the basal joints apparently have considerable lat- eral movement (see Text-fig. 7B of CIURCA 041771- 1). The basal joint or coxa of the pedipalp is very long and is in the same position vis-a-vis the mouth as in living forms. It occurs, therefore, at a higher plane than the legs, but on the same plane as the coxae of the legs but below the chelicerae. The legs are short, stout, with small triangular coxae that abut the sternum. The first pair abuts the upper part of the sternum and the labrum. Basitarsal and tibial spurs, apparently double, occur on each joint, but the first legs also show the presence of a femoral spur, possibly also doubled. The opercular plates are small and round and occur directly posterior to the sternum. There are five ab- dominal plates. Remarks. —The location of the mouth in Proscor- pius shows that it was well forward of where the mouth is in present-day scorpions. This still is not so far for- ward as in Stormer’s Waeringoscorpio, which is more primitive than Proscorpius although not so old in geo- logic age. Stormer’s excellent discussion of the probable evo- lution of the oral chamber (1970, pp. 345-348) can be largely substantiated and expanded. With the details furnished by the Ciurca specimens, in particular spec- imen VI (CIURCA 040668-1), and Waeringoscorpio hefteri Stormer, it is clear that the mouth migrated backward as in Limulus and possibly in the Eurypter- ida; the chelicerae occurred above the first or basal joint of the pedipalps, on the inner side of these joints. The base of this wide oral chamber, however, was open and here a labrum occurred, which functioned against the sternum. Furthermore, in such genera as Eoscor- pius, Kronoscorpio, etc., the maxillary lobes of the first pair of coxae were extended and later the maxillary lobes of the second pair of coxae squeezed the first pair of maxillary lobes away from the midline, so that the two pairs of maxillary lobes formed the base of the oral chamber, whereas the sides were composed of the first joint of the pedipalps and the chelicerae formed the roof of the chamber. Each of these steps can be demonstrated by the fossil record. Genus ARCHAEOPHONUS Kjellesvig-Waering, 1966 (emend.) Proscorpiidae (?) with large, subquadrate carapace; large oval compound eyes located at anterolateral mar- gins; central eye node anteriorly located with two very small ocelli; no raised cephalic area developed. Che- licerae of four joints, last leg with double trochanter. Preabdominal tergites with acute epimeral angles, not rounded. Type species. —Archaeophonus eurypteroides Kjel- lesvig-Waering, 1966. Remarks. —The discovery of another specimen of this genus has necessitated revision of the original ge- neric description. It should be noted that the differ- ences in comparison with Proscorpius are significant, but, because the underside of Archaeophonus is not well known, it is provisionally retained in the Pro- Text-figure 12.—Proscorpius osborni (Whitfield). From the Up- per Silurian Bertie Waterlime, Herkimer Co., NY. Specimen VI, CIURCA 040668. See also Text-figure 11. See foldout inside front cover for explanation of abbreviations. A. This reconstruction of the prosomal appendages was largely based on Specimen VI, CIURCA 040668, with the outline and size of the carapace in relation to the appendages from Specimen I, CIURCA 031966-1. B. Coxosternal area as shown by Specimen VI, CIURCA 040668-1. The left chelicera appears smaller because it is inclined. Claws on leg II taken from counterpart. C. Right fourth leg (IV), showing terminal detail, Specimen VI, CIURCA 040668-2. 50 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 scorpiidae, until more specimens are forthcoming from the Passage Gulf locality. Archaeophonus eurypteroides Kjellesvig-Waering, 1966 Text-figure 14 1966. Archaeophonus eurypteroides Kjelleswig-Waering, pp. 373- 375, pl. 42, fig. 1; pl. 43, figs. 1, 5; text-figs. 5-10. Type information.—Both the holotype, CUGM 411098, and CIURCA 062065-1 come from the Fid- dlers Green Dolomite Member of the Bertie Forma- tion, Upper Silurian, Ludlow age, upper part at Passage Gulf, Herkimer County, NY. Specimen. —CIURCA 062065-1. Collected on May 8 Ed Landing (oral commun., May 7, 1985) reports that the ho- lotype of Archaeophonus eurypteroides Kjellesvig-Waering, 1966 (CUGM 41109) was transferred to the collections of the New York State Museum in Albany, NY in 1971, as NYSM 12947. P.R.H. 20, 1965, by Samuel J. Ciurca. The specimen is well preserved, nearly whole, and consists of only the dorsal side, but reveals several morphological structures not preserved on the holo- type, which previously was the only known specimen. As in nearly all Bertie material, the specimen is pre- served as a carbonized imprint, showing little relief but no distortion. Also, as in other Bertie material, the specimen is best studied under alcohol or when mois- tened with a mixture of glycerine and water. In the dry state, and using light at an angle, certain joints and structures can be well delineated. The following de- scription is restricted to the parts not previously seen in the holotype or those which needed confirmation. Principal among these new morphological structures are the large, bulbous, compound lateral eyes. An un- usually large compound eye, the proportions of which suggest some of the eurypterids, is preserved on the (ed ee \\ VIS Text-figure 13.—Proscorpius osborni (Whitfield). From the Upper Silurian Bertie Waterlime (Fiddler’s Green Dolomite), Herkimer Co., NY. Reconstructions based on specimens in the collection of Samuel J. Ciurca, Rochester, NY. A. Dorsal side. B. Ventral side. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 51 anterolateral angles of the right side of the carapace. A median eye node is present midway on the anterior of the carapace, and the two small ocelli are barely perceptible. These ocelli are much like those present in eurypterids and not of the large size seen in Car- boniferous or later scorpions. The carapace is nearly square, very large, broadly rounded at the anteromarginal corners, and has a well- developed transverse ridge at the base. The chelicerae (see Text-fig. 14B) are well preserved, particularly the left one, which is seen from the dorsal Text-figure 14.—Archaeophonus eurypteroides Kjellesvig-Wae- ring. From the Upper Silurian, Bertie Waterlime, Herkimer Co., NY. CIURCA 062065-1. See foldout inside front cover for expla- nation of abbreviations. A. The whole specimen showing right lateral compound eye. B. Chelicera, showing ‘“‘over-bite” of the free finger. C. Distal part of the left fourth walking leg. side. This structure is composed of four segments, not three as in living scorpions. The first two are short and bandlike, but allow these structures considerable lat- eral movement, again unlike those of living scorpions, which are quite rigid, pointing outward and hardly capable of side movement. The immovable finger is elongated and armed with small denticles, but it was not possible to determine their exact shape or arrange- ment. The free finger is very long, hooklike, and also armed with several denticles. It is certain that the ac- tion between these two teeth is scissorlike and that no double tooth termination is present on the free finger such as is present in the Buthidae. The pedipalp is seen for the first time and this struc- ture is stout, short, and although likely displaced for- ward, the trochanter is apparently much longer than in other scorpions. The femur and tibia are short and stout. The hand is very stout, with the immovable finger curved inward against the free finger which is correspondingly curved backward. This is the opposite of what occurs in all other scorpions which the writer has seen and is unlike those living. The walking legs are preserved in excellent condi- tion. The last leg in particular confirms the presence of the unusually long tarsus, and the claws and post- tarsus are well preserved (see Text-fig. 14C). The post- tarsus forms a thick pointed heel, whereas the claws are curved and not greatly unlike those in living scor- pions. Two small basitarsal and tibial spurs, one on each side, occur on the last leg and possibly on others, although not seen in the specimen. If present on the other legs, the spurs are very small. The fourth walking legs, as all the others, are very short and do not extend past the posterior end of the fifth tergite. There are two dorsal median ridges on each tergite of the cauda. The telson, or stinger, is perfectly pre- served, slender, and with the aculeus bent in a gentle hook. No ornamentation was noted on the dorsal side of this scorpion. Measurements (in mm) of CIURCA 062065-1.— Carapace: Length 4.3 Width 4.3 Lateral eye: Length 135 Width 1.0 Pedipalp: Length of femur 2.0 Length of tibia 1.3 Width of hand 1.6 Length of palp 4.1 (incomplete) Tergite length: 1 0.8 2 183 a2. PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Measurements of CIURCA 062065-1 (cont’d).— Tergite length: 3 1.4 4 125 =) NEW) 6 telescoped “l 1.9 8 2.4 9 25 10 225, 11 2.7 12 2.8 (estimated) Vesicle width: 1.0 Length of stinger: 2:9 Complete length 30.1 (not including estimated: chelicerae) Preabdomen length: 10.4 Postabdomen length: 15.4 Remarks.—It seems that a correlation can now be made between the size of the median eyes and the size of the lateral compound eyes. In Silurian scorpions, such as Proscorpius and Archaeophonus, it appears that large lateral compound eyes are accompanied by small median eyes. In the Carboniferous and Triassic scor- pions the lateral compound eyes are greatly diminished in size with a corresponding increase in the size of the median eyes. Some of the scorpions without lateral compound eyes, such as Mazonia, have very large me- dian ocelli. Family WAERINGOSCORPIONIDAE Stormer, 1870 Proscorpioidea? with all legs abutting and radiating from the sternum. Labium and labrum present. Type genus.— Waeringoscorpio Stormer, 1970. Remarks. —The family Waeringoscorpionidae Stormer, 1970, is unlike any other in the arrangement of the coxae vis-a-vis the large pyriform sternum. The archaeoctonoids differ in that the first pair of coxae has moved anteriorly and the coxae abut one another and not the sternum. Although the waeringoscorpio- nids now appear more closely related to the archaeoc- tonoids than to any other group in their development, only more information can resolve the degree of their actual relationships. Genus WAERINGOSCORPIO Stormer, 1970 Waeringoscorpionidae with long slender postabdo- men; dorsal side little known; powerful chelicerae; pedipalps with slender hand and fingers, inner edges ° Although Kjellesvig-Waering had preliminarily proposed a new superfamily for Waeringoscorpio, new material of Proscorpius (see Text-figs. 6D, 12A, B) that came to light during the course of his study showed the assignment to the Proscorpioidea to be the better solution. A.S.C. cultrate; walking legs rather long; opercular plates ovoid; combs relatively small with 20-25 (?) teeth, median appendage present; ovate filamentous areas probably attached to ventral plates; postabdomen with long seg- ments, each with close-set longitudinal ridges; telson triangular in outline. Type species. —Waeringoscorpio hefteri Stormer, 1970. Geological range. —Lower Devonian. Remarks. —Stormer considered the abdominal plates to be moderately lobostern (?). Waeringoscorpio hefteri Stormer, 1970 Text-figure 111B 1960. Scorpionlike eurypterid ? gen. et spec. indet. Stormer, pp. 87- 91, text-fig. 1. 1970. Waeringoscorpio hefteri Stormer, pp. 336-343, pls. 1, 2, text- fig. 5. Type information. — Holotype and only known spec- imen, SMF VIII 31. Type horizon: Lower Devonian, late Lower Emsian, Nellenképfchen Schichten. Type locality: quarry in the Alkener Bach-Tal E. of Alken an der Mosel, MTB Miinstermaifeld r 03800 : h 68670, W. Germany. Family LABRISCORPIONIDAE, new family Proscorpioidea (?) with large pentagonal sternum!° and subcuneate coxae. Type genus. —Labriscorpio Leary, 1980. Remarks. —Regardless of whether the Labriscor- pionidae have a pentagonal sternum as suggested here, or a rhombic sternum with a triangular labrum, as in the original description (Leary, 1980, p. 1255), the genus Labriscorpio must be placed in a new family assignable to the Proscorpioidea. It is unique and com- parison with other families is superfluous. The abdominal plates have not been found, therefore doubt exists on the position of the family with regard to higher taxonomic categories. Labriscorpio is close to either the Holosternina Proscorpioidea or the Lo- bosternina Loboarchaeoctonoidea. It is more primi- tive than Loboarchaeoctonus and approximately at the same level of development as Proscorpius. For the time being, until further confirmation is forthcoming from other specimens, the family Labriscorpionidae is re- tained in the superfamily Proscorpioidea. Genus LABRISCORPIO Leary, 1980 Labriscorpionidae with first pair of coxae with an- '0 Leary (1980, p. 1255) says: ‘“‘Labrum triangular. Sternum trap- ezoidal with posterior edge slightly more than twice as wide as an- terior edge; sides slightly convex.”’ Kjellesvig-Waering never saw the actual specimen and his knowledge of the scorpion was based on manuscript copy. A.S.C. FossiL SCORPIONIDA: KJELLESVIG-WAERING teriorly-directed spurs, legs very short; free finger of pedipalp cultrate and straight. Type species. — Labriscorpio alliedensis Leary, 1980. Geological range. —Late Mississippian or early Pennsylvanian. Remarks. —The overall aspect of this scorpion is somewhat like Archaeoctonus glaber (Peach) and Lo- boarchaeoctonus squamosus from the Lower Carbon- iferous of Scotland. Labriscorpio alliedensis Leary, 1980 Text-figures 111C, D 1980. Labriscorpio alliedensis Leary, pp. 1255-1257, | pl., 2 text- figs. This unusual and taxonomically important scorpion has been adequately described, and my comments are restricted to another possible interpretation of the coxosternal area. The small piece in front of the ster- num is regarded by Leary as a labrum, and indeed, this may be correct. However, this would mean that the mouth would occur between this triangular piece and a sub-rhombic sternum, and nothing is mentioned of the presence of an opening, which should have been preserved. My interpretation is that the sternum is pentagonal (see Text-figs. 111C, D) with all subcuneate coxae adjoining this pentagonal sternum, and it seems to me to be worth considering as an alternative to Leary’s determination. Type information.—Early Pennsylvanian or Late Mississippian channel fill, associated with numerous plants and a fish scale in the quarry of the Allied Stone Company, between Rock Island and Milan, Illinois (SE% sec. 14. T. 17 N., R. 2 W., Milan Quadrangle). The plant fossils indicate an upland flora, but the scor- pion should not be considered as anything but a gill- bearing, water-dwelling scorpion, although it is prob- able that Labriscorpio alliedensis could take excursions out of the water, carrying water in its gill chambers. The holotype is deposited in the collections of the Illinois State Museum, Springfield, IL and is acces- sioned as ISM 416899. Superfamily STOERMEROSCORPIONOIDEA, new superfamily Holosternina with first pair of coxae abutting ante- rior to the sternum and with poor development of maxillary lobes; second and third pairs of coxae abut against the sternum, and the fourth against the genital opercular plates. Type family. —Stoermeroscorpionidae, new family. Remarks. —Among the Holosternina, it differs from the Archaeoctonoidea in having the fourth pair of cox- ae abutting the genital opercula, and from the Meso- phonoidea in having the second pair abutting against nn Ww the sternum, and in the primitive development of max- illary lobes. Family STOERMEROSCORPIONIDAE, new family Stoermeroscorpionoidea with very large, elongate- pentagonal sternum with deep, triangular basal notch and large, elongate elliptical opercular plates, and lat- eral compound eyes. Type genus. —Stoermeroscorpio, n. gen. Remarks. —There is no family that warrants com- parison as it is so different from all others. Genus STOERMEROSCORPIO, new genus Stoermeroscorpionidae with round ocellar node on an anteriorly-rounded, very elongate carapace, and lat- eral compound eyes. Derivatio nominis. —Named for Dr. Leif Stormer. Type species. —Stoermeroscorpio delicatus, n. sp. Geological range. —Silurian. Remarks. —This scorpion is so different from all oth- ers that there is no necessity for comparison. Having the characters mentioned above for the superfamily and family makes this a unique scorpion, likely rep- resenting a large group in the Silurian and older beds. It is a distinct pleasure to name this genus after my friend, Professor Leif Stormer of the Paleontological Institute of the University of Oslo. Stoermeroscorpio delicatus, new species Text-figures 15, 16 Type information. —The holotype consists of part and counterpart of a nearly complete scorpion, show- ing most of the dorsal and ventral surfaces. It is reg- istered as No. 041771-1,2, in the private collection of Samual J. Ciurca of Rochester, NY, and comes from the Silurian, upper Fiddlers Green Member of the Ber- tie Waterlime at Passage Gulf, Herkimer Co., NY. It occurs in the typical ‘‘waterlime’’, on the same layer as numerous eurypterids and several scorpions, as well as marine fossils. The ecological zone or biological community is what the writer refers to as the ‘““Euryp- teridae Zone’”’ (see Kjellesvig-Waering, 1961, p. 794). The holotype is a small scorpion measuring 12 mm in total length from the anterior of the carapace to the base of the stinger. Overall it is a delicately constructed individual. The carapace is rounded anteriorly and rather large, particularly in comparison with the common Proscor- pius osborni that occurs in the same horizon. The lat- eral compound eyes are round, marginally located on the anterior of the carapace, and have relatively coarse ommatidia. The ocellar node is well developed, an- 54 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 teriorly and marginally placed, round, elevated, and has two small, rounded ocelli on the area slightly an- terior to the middle of the ocellar node. The specimen is flat, but no ridges or ornamentation were discernible and it is assumed, probably correctly, that the surface of the carapace is smooth. There is no sign of a basal marginal rim, and the base is straight, with genal angles of 90°. The entire carapace is longer than wide by a ratio of 8 to 7. The mesosoma does not differ from that of most other scorpions. It is slender; each tergite has a poorly- developed anterior transverse ridge, and the tergites increase in length progressively backward. The last preabdominal tergite is large, triangular and apparently Text-figure 15.—Stoermeroscorpio delicatus, n. gen., n. sp. From the Upper Silurian Bertie Waterlime, Passage Gulf, Herkimer Co., NY. Holotype: CIURCA 041771-1,2. See foldout inside front cover for explanation of abbreviations. A. Dorsal side, CIURCA 041771-2 (counterpart), 12 mm long from part of carapace to the stinger end. B. Ventral side, CIURCA 041771- 1, showing coxosternal area and anterior preabdomen. FossIL SCORPIONIDA: KJELLESVIG- WAERING 55 smooth: crests, if present, are weak. The postabdomen continues as a slender organ with lateral, superior and inferior crests, but in their flat- tened state they could not be precisely determined. Each tergite is “normal” and only on one tergite were small pustules, probably setaceous, developed on one of the inferior crests. The telson, or stinger, is unusual in being very short and stout. The aculeus is particularly short. Three crests are developed on the vesicle and I interpret these as a superior, two laterals and an additional apparent line of scales which likely represents a single inferior “crest” that has become almost obsolete. The coxosternal region, of course, is the area which is very surprising and exciting because of its phylo- genetic implications. The sternum, in keeping with all Silurian and Devonian scorpions, is large —a large plate 1mm similar to that present (metastoma) in eurypterids. This large sternum is elongate-pentagonal with slightly ta- pering lateral margins, a rather obtuse anterior, and a deep, inverted triangular basal margin where the two, nearly elliptical, rather elongate, marginal plates are located. The first pair of coxae is of particular interest. These abut against each other, anteriorly to the triangular part of the pentagonal sternum. The inner anterior parts of these coxae are produced forward into a pair of short, wide, maxillary lobes. The second and third pairs of coxae are short, rounded at the inner part, and trun- cated in contact with the trochanters. Both pairs abut the lateral margins of the sternum. The fourth pair, similar in shape to the second and third pairs, abuts against the genital opercula. The chelicera is very slender, long, composed of four i Maat tadesiLUTLL DU Text-figure 16.—Stoermeroscorpio delicatus, n. gen., n. sp. Reconstruction based on the holotype, CIURCA 041771-1,2. From the Upper Silurian Bertie Waterlime (Fiddler’s Green Dolomite), Herkimer Co., NY. A. Dorsal side, based on CIURCA 041771-2. B. Ventral side, based on CIURCA 041771-1. 56 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 joints, with the chela small and curving. I was unable to note any teeth, which are undoubtedly present, but small. The hand is rounded on the margins but this may be due to flattening. The pedipalp is known in its entirety. The trochanter is roughly triangular, but long. The first joint is seen to lie under the first leg, and far from the area of preoral chamber formed by the first two coxae. It is therefore to be noted that the pedipalp coxae had not joined in the formation of the preoral chamber as in later scor- pions. The femur is very long and slender, devoid of any noticeable ridges, which should be seen even in the flattened state, since folds (the crests of the cauda) were preserved. The brachium is very short. The chela is large, having a well-developed, stout hand with two curved dactyls. The fixed dactyl is incurved, whereas the free one is curved forward and overlaps the other at the end. Both are cultrate along the edge. The four legs are present, but details of the ends and spurs will have to await better preserved specimens. It is certain, however, that the first pair is very short, that all the rest are slender, increasing in size backward, that the last pairs are longer than usual for the Bertie assemblage, and that the terminations are composed of a short posttarsus and two short ungues as in other Bertie scorpions. The legs are cylindrical. The combs are barely discernible and this only in highly favorable light and wet with alcohol. Fulcra are definitely present and are very small, but, surprisingly, as many as 20 elongated teeth occur. I suspect that the rachis is not divided into anterior or median lamellae, although this is more of an impression I received from not being able to see any joint lines. It is not a point to be stoutly maintained. The abdominal plates are the usual holosternous type. Derivatio nominis. —delicatus (L.) = delicate. Remarks. —The short maxillary lobes of the first pair of coxae are the oldest attempt at forming a preoral chamber and suggest that the labrum of forms such as Proscorpius and Waeringoscorpio is obsolete and has disappeared in this scorpion. Possibly this scorpion was becoming amphibious as early as the Upper Si- lurian, as the preoral chamber is considered an ad- aptation for land. The fact that Stoermeroscorpio was a Holosternina, having gills, meant that it could not live long out of water by carrying water in the pouches present in the inner part of the holostern plates. At any rate, this attempt at a preoral chamber was in its i1n- fancy, as the coxae or first joints of the pedipalps had not yet moved to the position that they hold today, namely on each side of the oral chamber, forming the lateral margins. For that matter, the chelicerae, al- though only one is present, show that they were far apart—as far apart as the pedipalp bases—and there- fore could not take part in forming the roof of the oral chamber as they do today. Another interesting feature of this scorpion is that it has the same type of short first legs found in Pro- scorpius and Archaeophonus, which lived in the same environment and walked over soft magnesium car- bonate muds (‘‘waterlimes”’ of eurypterid literature). There seems to be an adaptation for similar conditions. Of course, they could have lived on reefs or subaqueous rocks, but these are unknown in the Bertie mudflats. The type of rocks, as well as the nearly perfect complete preservation, precludes transportation of the fragile skins. Superfamily ALLOPALAEOPHONOIDEA, new superfamily Holosternina with first three pairs of coxae meeting in front of the sternum, without maxillary lobes; other pair of coxae abutting against the sternum; legs very thick, short, eurypteridlike, with posttarsus greatly de- veloped as a single spine. Type family. — Allopalaeophonidae, new family. Family ALLOPALAEOPHONIDAE, new family Allopalaeophonoidea with large pentagonal ster- num. Small carapace without lateral compound eyes. Type genus. —Allopalaeophonus, new genus. Genus ALLOPALAEOPHONUS, new genus Allopalaeophonidae with well developed, anteriorly located median eyes without apparent ocellar nodes. Anterior and lateral margins of carapace concave; walking legs without tibial spines. Derivatio nominis. —allo (Gr.) = another, different + Palaeophonus, a scorpion genus. Type species. —Palaeophonus caledonicus Hunter, 1886. Geological range. —Silurian of Scotland. Remarks. —Thorell and Lindstrém (1885, p. 24), in discussing Palaeophonus caledonicus, stated in a foot- note to their classification: “If this species is provided with eyes, and if P. nuncius should prove destitute of eyes, a new genus must of course be formed for the reception of the former species.” As it turned out, both species have median eyes, but P. nuncius has large lateral facetted eyes as well. The two species do have rather similar legs and share a general dorsal aspect of the body, but P. nuncius is a lobostern, and not closely related. FossiL SCORPIONIDA: KJELLESVIG- WAERING S7/ Allopalaeophonus caledonicus (Hunter) Plate 2, figures 1-4; Text-figures 17, 18, 1100 1885. Palaeophonus sp. Peach, p. 297, fig. 1. 1886. Palaeophonus caledonicus Hunter, pp. 169-170. 1888. Palaeophonus caledonicus Hunter. Hunter, pp. 185-191. 1901. Palaeophonus hunteri Pocock, pp. 291-311, fig. 3, pl. 9. 1904. Palaeophonus caledonicus Hunter. Fric, pp. 63-64, text-fig. 79. 1911. Palaeophonus caledonicus Hunter. Bather, p. 676. 1912. Palaeophonus hunteri Pocock. Clarke and Ruedemann, p. 395, fig. 85. 1913. Palaeophonus caledonicus Hunter. Petrunkevitch, p. 32. 1923. Palaeophonus hunteri Pocock. Versluys and DeMoll, p. 109, figs. 15, 38A. 1934. Palaeophonus caledonicus Hunter. King, p. 563. 1944. Palaeophonus hunteri Pocock. Lehmann, p. 177. 1949. Palaeophonus caledonicus Hunter. Petrunkevitch, pp. 128- 129, fig. 172. 1953. Palaeophonus caledonicus Hunter. Petrunkevitch, pp. 5, 8- 10, figs. 6, 116. 1955. Palaeophonus caledonicus Hunter. Petrunkevitch, p. 69, figs. 38(2), 39B. 1962. Palaeophonus caledonicus Hunter. Dubinin, p. 425, fig. 1222. 1963. Palaeophonus caledonicus Hunter. Stormer, pp. 92, 94. 1966. Palaeophonus caledonicus Hunter. Kjellesvig-Waering, p. 359. Specimen. —Holotype, KM unnumbered specimen. It is from the Silurian, Llandovery—Wenlock, at Dun- side, Logan Water, Lesmahagow, Scotland. The specimen was studied under alcohol, and as with the other chitinous forms preserved in the greenish black shales of Lesmahagow, details are beautifully revealed when immersed in a wet medium. Two sili- cone casts made by Dr. Ian Rolfe reveal much which can be discerned in much greater detail than on the original. The holotype specimen was kindly sent by the Kilmarnock Museum to Edinburgh, where I was able to study it. Considerable confusion exists as to whether the dor- sal or ventral side is preserved. Peach (1885) and Po- cock (1901) believed that the ventral side was pre- served with outlines of the eyes and carapace impressed through. Petrunkevitch (1953, p. 9) stated that the specimen was preserved as a dorsal impression with the ventral structures impressed through. There is no doubt that the specimen is preserved with the ventral side revealed and with the median eyes and outline of the carapace impressed through, as Peach and Pocock had stated. This can easily be proven because the eyes are preserved as molds of the internal part, imbedded in the shale as two small depressions. The carapace is deeply emarginate along the anterior margin, with the base also emarginate. The sides ap- parently are straight or slightly incurving or concave. The median eyes appear elliptical, but actually are round, small, and placed anteriorly on the carapace, very close to the margin. Undoubtedly they are located on a small ocellar mound of slight elevation, as two small wrinkles occur on each side of the median eyes, which represent the compressed ocellar node. The node is Clearly lacrimiform with the two round eyes placed on the anterior end. Small punctae are present on the anterior part of the node. Petrunkevitch (1953, p. 9) shows two orbital ridges on each side of the anterior part of the carapace. These were not noted by Hunter (1886, 1888), Peach (1885) or Pocock (1901), and I have been unable to find them. This scorpion is a holostern and, therefore, if lateral eyes were present, they would be compound eyes, and orbital ridges are never developed with such eyes. Lat- eral orbital ridges are developed only in modern, land- living, scorpions to protect the few (one to five) lateral ocelli. If orbital ridges had been developed, it would mean that the eyes had been small ocelli as in present- day scorpions, and had disappeared, leaving the orbital ridge as a vestige. This is hardly possible and I consider the structure to be nothing more than fortuitous wrin- kling of the carapace. The most remarkable part of this scorpion is the small size of the carapace in relation to the great size of the chelicerae, pedipalps, legs and sinuous body. The appendages are well preserved and much new infor- mation of taxonomic value has been obtained by studying this specimen under alcohol. The chelicerae are formidable structures, almost as long as the carapace, composed of four joints as in most other Paleozoic scorpions, with long fingers which are armed with teeth (see Text-fig. 17A). Inasmuch as the bases were covered, it would perhaps be safe to say that the chelicerae are actually as long as or longer than the carapace. The pedipalps are stoutly constructed, with a re- markable development of coarse pustules, each with a sensory setal opening. The pustulation is particularly well developed on the trochanters. Here the setaceous pustules are crowded so that the entire surface is dense- ly covered, anteriorly and posteriorly. Presumably the dorsal side would also be densely covered. The hand is short, mainly rounded and with curved fingers that are entirely cultrate on the inner edges. The walking legs are spectacular in being very eu- rypteridlike; they were probably mainly cylindrical in life, massive structures that are progressively longer from anterior to posterior. The last pair of legs (IV) is very massive—at least half as thick again as the pre- ceding. This estimate is made from the flattened legs, but in life would have been the same. Each leg had two spurs on the ends of the tibia and of the basitarsus, as the spurs were found on the first (I) (see Text-fig. 17B) and the third (IIT) (see Text-fig. 17C), as well as on the intervening second (II) leg. It does not seem speculative to conclude that all legs had 58 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 = FossIL SCORPIONIDA: KJELLESVIG- WAERING 59 double tibial and basitarsal spurs. The terminal part of the leg consists of a very long, curved spine that is considered to be the posttarsus (7). This structure is similar to the terminal joint in eurypterids, except that the posttarsus of this scorpion has a row of ventral denticles—a development that is very common with scorpions found in the Carboniferous. It is thought that these spines furnished traction for the claws against soft surfaces such as are found on water plants. On one side of the posttarsus (and presumably on both sides) is a spur, which would be a tarsal spine (6S) (see Text- figs. 17B, C), homologous to the claw of present-day scorpions. (See Remarks.) Text-figure 17.—Allopalaeophonus caledonicus (Hunter). Holo- type, KM unnumbered specimen. From the Silurian Llandovery— Wenlock, at Dunside, Logan Water, Lesmahagow, Lanarkshire, Scotland. See foldout inside front cover for explanation of abbre- viations. A. Entire holotype specimen, showing the ventral side revealed, but with the median eyes and outline of the carapace impressed through the flattened specimen, as Peach and Horne originally in- terpreted it. B. Terminus of the left first leg (I) showing the prominent spinelike terminal joint (17). C. Terminus of the left third leg (III) and the curious serration (enlarged) on the last joint (III7). The coxosternal arrangement can be accurately re- constructed with ease. The coxae of the walking legs are short and triangular. The sternum is very large, and pentagonal in shape, as is prevalent in living scorpions, such as the families Scorpionidae, Vaejovidae, etc., a shape developed as early as Middle Silurian. Petrun- kevitch (1953, p. 9) denies that the sternum is pre- served, although both Peach (1885, p. 297) and Pocock (1901, p. 301) described it correctly. The short coxa of the third left leg clearly abuts the opposing coxa in front of the sternum, which would mean that the other coxae of the first three pairs of legs abut one another in front of the sternum. The coxa of the fourth leg on the left side is preserved in place, along the lateral margin of the sternum, thus a coxosternal arrangement such as is shown on Text-figure 1100 seems indicated. This agrees with the interpretation given previously by Peach (1885) and Pocock (1901). The opercular plates are not preserved. Peach (1885, p. 297) misinterpreted the two “‘bifid’’ plates described below to be the opercular plates. The pectines are well preserved, showing an unjointed rachis, and have two elongate plates (Peach’s bifid plates) on their inner side. Text-figure 18.—Allopalaeophonus caledonicus (Hunter). Reconstructions based on the holotype, KM unnumbered specimen. From the Silurian (Llandovery—Wenlock) of Scotland. A. Dorsal surface. B. Ventral surface. The genital opercula are unknown and conjectural. 60 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 These are unknown in other scorpions. It is true, how- ever, that some living scorpions have a dilated plate on the inner side of the pectines, such as some species of Tityus, where the female has a greatly enlarged rounded plate, or in the North African buthid Leiurus quinquestriatus, which has rather narrow plates on the inner side of the pectines, or in the Madagascan Gros- phus, which is characterized by greatly elongated inner plates. The presence of these plates, therefore, could be merely a development of the inner plate of the pec- tine. There are no fulcra, although a linear series of small punctae was noted at the base of the rachis. The teeth are large, long, and about 15 to 20 would be present, if all had been preserved. The abdominal plates are holosternous and, of course, without the slightest trace of stigmata. Peach (1885) considered this scorpion to be an “Ancient Air Breath- er’, and indeed so named his paper, so that he en- countered no difficulty in finding stigmata. Pocock (1901) and Petrunkevitch (1955) did not find any stig- mata, and I thoroughly agree, although in Petrun- kevitch’s case, the stigmata would hardly be present as he thought that the dorsal side was present. Under alcohol, which would easily reveal stigmata, as well as on the silicone casts, nothing that remotely resembles a stigma is present. The holotype was viewed under excellent lighting at various angles. The specimen is clearly a holostern. Each abdominal plate is bounded on the anterior by the usual transverse ridge. The ventral part of the last preabdominal tergite (No. 7) does not show the abrupt contraction that many scorpions do, but grades into the cauda, giving this scorpion a sinuous appearance such as occurs in many living scorpions, like Broteo- chactas, and other Chactidae. Three longitudinal ridges are shown, although undoubtedly there were four, as one of them was not preserved. Two carinae are present on the median part, and another on each side. The cauda is preserved in its entirety, with superior, lateral and ventral carinae well developed. All carinae are surmounted with setaceous pustules. The superior carinae are roughly serrated. The telson is small, con- sisting of a small bulbous vesicle with a nearly straight aculeus. The vesicle is covered with setaceous pustules, particularly on the ventral part. Remarks. —The sinuous shape, very small eyes, and great development of setae, presumably sensory, on the anterior region, such as the trochanters of the pedi- palp, may indicate that this scorpion spent much of its time in burrows, a cryptozoic, but underwater, ex- istence. The presence of small spines on the underside of the posttarsus indicates that, at least part of the time, it needed traction against some object, perhaps un- derwater plants. Although Petrunkevitch believed that the dorsal, rather than the ventral side, was preserved, he stated (1953, p. 10), “I think it is safe to say that the coxae of the fourth pair of legs meet in the median line just on the edge of the carapace. The pentagonal area be- hind the fourth coxae may then be interpreted as the sternite of the first preabdominal segment later com- pletely lost and now wanting in recent scorpions.” It is difficult to understand this reasoning as a very tor- turous route has to be taken not to make the “‘pentag- onal area” the sternum and not to have all the coxae meet as shown below. The coxosternal region is diffi- cult to reconstruct. Although dislocated, nearly all parts are present. However, several factors are present that give a rather sound idea of the arrangement. The ster- num is pentagonal, very large, and in making the re- construction on Text-figure 1100 it was apparent that in order to fit all the massive legs under the unusually small carapace, the only logical accommodation was in having the fourth pair of legs abut the sides of the pentagonal sternum with the third pair meeting in front of the sternum. The first two pairs, of course, abut one another at midsection. This reconstruction is by no means without some solid foundation, as the left side of the specimen shows exactly this type of arrangement, namely, the fourth coxa in position against the sternum with the third lying also, in position, in front of the triangular anterior part of the sternum. This coxoster- nal arrangement is a very primitive one, and in recon- struction is highly possible and not without factual basis. There would have been no question about the pec- tinal teeth if the original specimen had been studied in alcohol. As was stated in the description, the terminal spine of the leg is considered to be the posttarsus, whereas the two small spurs at the base of this spine, or end of the tarsus, are considered to be the tarsal spurs which later, by moving forward, became the claws known in many Paleozoic and Recent scorpions. This trend can be seen, further developed, in legs of Palaeoscorpius devonicus Lehmann from the Lower Devonian of Ger- many. The legs of the latter are short and stout, very much like those that characterize the Palaeophonidae and the Allopaleophonidae, except that the termina- tion consists of three spines. The central one is con- sidered to be the posttarsus, whereas the two adjoining are the tarsal spurs or claws. The posttarsus moved to a rear position, giving a trifid arrangement where each spine is equidistant from the others. When preserved in a flattened condition, this trifid arrangement appears as if all three spines were pointing in one direction. In Kronoscorpio danielsi (Petrunkevitch) the trifid ar- rangement is greatly developed, probably into a bird- FossIL SCORPIONIDA: KJELLESVIG- WAERING 61 footlike impression, with the two long tarsal spines taking an anterior position and the posttarsus a pos- terior one. This would result in a cushioning effect, somewhat like snowshoes, which would enable the scorpion to walk on soft mud. Superfamily PALAEOSCORPIOIDEA Lehmann, 1944 (emend.) Holosternous scorpions with four pairs of coxae meeting in front of the sternum. Type family. —Palaeoscorpiidae Lehmann, 1944 (emend.). Remarks. — Petrunkevitch (1955, p. 70) defined this superfamily as follows: ““Preabdomen with median longitudinal fold continuous with tail and resembling it.” The X-ray photographs (see Pl. 3, fig. 1) show a distinct black part running through the central part of the preabdomen and continuing, not so clearly, through the caudal segments. In my opinion, this is merely the presence of the intestinal canal of this scorpion, and such interesting preservations are not unknown in the fossil record. For example, in the Silurian eurypterid Carcinosoma newlini (Claypole), Ruedemann (1919, fig. 33) showed a specimen in which the entire intes- tinal tract was preserved and this is very similar to that present in this Devonian scorpion. Heubusch (1962, p. 222) figures other eurypterids where the in- testinal canal is preserved. If the intestinal tract is full, chances of preservation are, of course, considerably enhanced. Nevertheless, ‘“‘a median longitudinal fold continuous with tail and resembling it” is hardly a character to be used for a genus, much less a super- family, particularly when one is ignorant of the struc- ture. Family PALAEOSCORPIIDAE Lehmann, 1944 (emend.) Sternum consisting of an equilateral pentagon. No maxillary lobes developed on the first two pairs of coxae. Legs with three short straight claws. Type genus. — Palaeoscorpius Lehmann, 1944. Genus PALAEOSCORPIUS Lehmann, 1944 (emend.) Legs very short, with very wide joints, and probably armed with very short double tibial and tarsal spurs. Pedipalps massive. Type species. —Palaeoscorpius devonicus Lehmann, 1944. Geological range. —Lower Devonian. Palaeoscorpius devonicus Lehmann, 1944 Plate 3, figure 1; Text-figures 19, 111E 1944. Palaeoscorpius devonicus Lehmann, pp. 177-185, figs. 1-4. 1953. Palaeoscorpius devonicus Lehmann. Petrunkevitch, pp. 5, 14. 1955. Palaeoscorpius devonicus Lehmann. Petrunkevitch, p. 73. 1962. Palaeoscorpius devonicus Lehmann. Dubinin, p. 428. Material. —Two excellent plaster casts of the holo- type, kindly made for the writer through the gracious- ness of Dr. H. K. Erben of the Institut fiir Palaontologie of the Rhine, Friedrich-Wilhelm Universitat, Bonn, West Germany, where the holotype, FWU Egr 263 of the collection of W. M. Lehmann, is deposited. The casts are now deposited in the Field Museum of Nat- ural History, Chicago, IL. Also used in this study is an X-ray photograph made by W. M. Lehmann, which Leif Stormer kindly sent to the writer. It must be stated that the casts of this fossil are exceptionally fine, showing many details that even re- veal certain minute, but important, structures. After the drawings were made, I was delighted to receive the X-ray photograph, which verified all the details that could be seen on the cast. Type information. —Lower Devonian, Hunsriick Shale, Pit I, Eschenbach, near Bundenbach (Hunsriick), West Germany. This is a marine shale, and the spec- imen was found in the same pit as specimens of Wein- bergina opitzi, a xiphosurid. Eurypterids also occur in the same strata. Inasmuch as the scorpion is a holo- stern, it is considered to be marine and to have been living in the habitat where found. It should be noted that the specimen is complete, suggesting little or no transportation, and that the type of shale hardly argues for the introduction of extraneous material by currents. My description is supplementary to that of Leh- mann, but in several important aspects differs from the original. This necessitates an emendation of the family and genus as defined by Lehmann. It also ne- cessitates complete emendation of the superfamilial, familial and generic definitions given by Petrunkevitch (1955) in his classification, as the characters used are erroneous and trivial with respect to the taxonomic category. These changes and reasons have been dis- cussed above. In my study of the casts, drawings were made from each half by use of the Wild Drawing Tube (a type of camera lucida), then each drawing was superimposed on the other with the result that many structures that had not been previously noted could be easily identi- fied. The drawing therefore is a composite, with the dorsal side eliminated. My main interest was in the determination of the important underside and the structure of the legs and abdominal plates. Many of the details are apparent when light is directed on the 62 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 ( int. oc. Abo TE le Zz B Imm Text-figure 19.—Palaeoscorpius devonicus Lehmann. Holotype, FWU Egr 263. From the Lower Devonian, Hunsriick Shale, Pit I, Eschenbach, near Bundenbach (Hunsriick), West Germany. See foldout inside front cover for explanation of abbreviations. A. Underside of the holotype, taken from a cast, in two parts. Note large pedipalps. Distal details of legs III and IV shown. X-ray photos suggest that some refinement of the configuration of the abdominal plates may be in order. B. Obscure cephalic indications. C. Possible restoration of the coxosternal region, based on the holotype. D. Alternative restoration of the coxosternal region, made on the assumption of lateral dissociation of the leg bases during fossilization. FossIL SCORPIONIDA: KJELLESVIG-WAERING 63 specimen at a very low angle. The degree of detail is remarkable in the casts, and even minute structures such as setal perforations are clearly apparent. The entire coxosternal region was completely revealed, as were all the joints of the legs. The left pedipalp was left out of the drawing as it caused confusion. The important arrangement of the coxosternal re- gion can now be described. The sternum consists of an equilateral pentagon. The coxae of all legs are short, mainly triangular, and appear most likely to lhe 1m- mediately in front of the sternum, in a condition some- what like that present in the Carboniferous Cyclo- phthalmidae. No maxillary lobes are present. The genital plates are nearly elliptical, narrow, and would lie directly posterior to the base of the sternum. Two reconstructions or interpretations of the coxosternal area are possible and are given in Text-figures 19C and 19D. I prefer the interpretation given in Text-figure 19D. Another point of disagreement with the original de- scription is the presence of three, distinct, straight claws on the legs. The middle claw would correspond, of course, to the posttarsus. All legs are stout and all comprise eight podomeres as in living scorpions (coxa to posttarsus). Spurs, also short and stout, were noted on the posterior of the left second leg, which would correspond to a basitarsal spur, and another on the left fourth leg, which represents a tibial spur. Very likely spurs occur on all legs on each side of the base of the tibia and the basitarsus. The combs are not well preserved, but a number of detached long and slender teeth were found. The well-preserved abdominal plates are holoster- nous and straight across the anterior, with well-round- ed posterolateral corners. No stigmata of any type are present. Since the entire plates are preserved, these structures would be easily preserved and noted if any were present. On the carapace, no eyes were noted in the position indicated by Lehmann (1944, p. 181), who stated that there were two very small eyes “the size of pin heads” in the posterior part of the carapace, much as occurred in Proscorpius osborni. We know now that there are no posterior eyes in P. osborni (Kjellesvig-Waering, 1966, and above). I was unable to see any eyes at the posterior part of the carapace in Palaeoscorpius de- vonicus as shown on the cast. The holotype of Palaeoscorpius devonicus Lehmann was studied at the Institut fiir Palaontologie of the Rhine, Friedrich-Wilhelm Universitat, at Bonn, West Germany, on November 16, 1970. Because it is pre- served in a thin plate of black slate, both the dorsal and ventral sides can be viewed. Study of the holotype shows that the carapace is not visible in its entirety, although the anterolateral angles are preserved, and these are broadly rounded. Two large median ocelli, which appear elliptical but probably are round in the inflated state, are visible on the anterior part of the carapace. These large eyes are heavily replaced by crys- tallized pyrite, but the complete outline of the right eye is clearly visible. A median ridge separates the two eyes. No other eyes are visible and there is no evidence whatsoever of eyes toward the back of the carapace as reported by Lehmann. The so-called eyes are rounded lumps of pyrite not connected with any scorpionid structure. The three-toed terminations reported above were checked on the actual specimen. These are distinct on each leg. Lehmann (p. 182, fig. 4) shows only the two claws and does not recognize the central claw, which actually is the posttarsus. The sternum, genital operculum, and coxae are of considerable interest. The equilateral, pentagonal ster- num has a short spur at the extremities of the base where, apparently, the elliptical genital plates fit. Family HYDROSCORPIIDAE, new family Palaeoscorpioidea (?) with median eye node located at the first quarter of the carapace; legs with segments as wide or even wider than long; tarsi terminating in two claws. No lateral compound eyes. Type genus. —Hydroscorpius, n. gen. Remarks. —Since the coxosternal area is not pre- served, the superfamily position is uncertain. Genus HYDROSCORPIUS, new genus Hydroscorpiidae with round, hemicircular carapace; very short, stout legs terminating in two long claws; chelicerae copiously supplied with hairs or setae on the inner side at the junction between the second and third joints. Derivatio nominis. —hydro (Gr.) = water + scor- pion. Type species. —Hydroscorpius denisoni, n. gen., n. sp. Geological range. —Lower Devonian of Wyoming. Hydroscorpius denisoni, new species Plate 3, figure 2; Text-figure 20 The species is based on a single nearly complete specimen, FMNH (PE) 6176, collected by Robert H. Denison and E. S. Richardson, Jr. in 1962 at the FMNH excavation site in Cottonwood Canyon, Wyoming. The specimen is broken, but excellent as to details and original cuticle, with the dorsal side exposed. It is preserved in light-gray, fine-grained, dololutite (dolo- mite) with the remains of plants, eurypterids, and three other scorpions. 64 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The carapace is incomplete, hemicircular in outline, measuring 4.9 mm in length and an estimated 5.5 mm at the base, which is rounded at the genal angles. The median eye node lies approximately 1.2 mm posterior to the anterior margin, but is too poorly preserved to show the median eyes. Median eyes must have existed because there is no known occurrence of eyeless scor- pions in the fossil record. No lateral eyes are present nor seem to have existed, but crucial parts of the car- apace are missing. The chelicerae, measuring 3.4 mm in length from the margin of the carapace, and 2.5 mm in greatest width at the second joint, are preserved nearly whole. The right one reveals the basal joint, and nearly all of the hand and free chela. The free finger is curved to a sharp point and edged with definite, but blunt, serra- tions (see Text-fig. 20B). The hand is rounded at the base where it articulates with the second joint. At the junction with the second joint, and placed inside the two chelicerae, is a mass of setae (see Text-fig. 20C). No ornamentation occurs on the chelicerae. The color changes from dark brown to black at the extremities. The second joint definitely shows a thin doublure at d,, suggesting that the base of that joint is there. The third joint, comprising the hand and immobile chela, is the area marked (3) on Text-figure 20C. The round area with the thin doublure marking its base is d3, which is covered over by the transparent cuticle of joint 2. The pedipalp measures 3.4 mm in greatest width at the palm, whereas the fixed finger has a width of 0.9 mm at the midsection (it is tapering) and the free finger is 1.1 mm in width at midsection and 4.3 mm in length. Estimated overall length of the pedipalp is 8.5 mm. Parts of all four walking legs are represented, and they are very short and stout (see Text-fig. 20A). They consist of the usual seven segments plus coxa. The third right walking leg, which is the only leg showing ter- minal details, ends in a short rounded posttarsus and two long, powerful claws. The thick legs of these scor- pions are considered eurypteroid in aspect, and suggest that they were adapted for digging under water. The habitat explains the long sinuous shape. The ventral side is not revealed, so there is no in- dication whatsoever of the coxosternal area nor of the pectines. Parts of two displaced abdominal plates show that this scorpion is a holostern. The preabdominal tergites show a narrow transverse ridge on the anterior margin. The stinger as preserved is visible from the side. In overall length it measures 3.2 mm on the dorsal side; the actual curved stinger is 1.5 mm long, whereas the bulbous part is 1.7 mm. The bulbous part measures 2.6 mm in length through the midsection, and 2.3 mm through the midsection from top to bottom. The cuticle does not reveal any sort of ornamenta- tion, except the two carinae on the dorsal side of the last tergite. The color pattern is clearly preserved. The carapace, as well as most of the abdomen, is shiny honey-colored; on the carapace, however, the eye node is darker brown, and the appendages show a gradation from the honey-colored cuticle to progressively darker brown, to end in brownish black to black at the ex- tremities. This is true of all appendages from the che- licerae to the walking legs. The tail also shows this gradation, the increase in brown is progressively darker to end at the stinger, which is dark brown to black. The latter has mainly flaked off, but enough of the original cuticle was present to reveal the color. Type information.—Lower Devonian, Beartooth Butte Formation, from the FMNH excavation site in Cottonwood Canyon, Big Horn Mountains, WY, U.S.A. The specimen is deposited in the Field Museum of Natural History, Chicago, registered as FMNH (PE) 6176. Derivatio nominis. —Named in honor of Robert H. Denison, former Curator of Fossil Fishes at the Field Museum of Natural History, Chicago, IL, and one of the collectors of the specimen. Measurements (in mm) of mesosoma and metasoma of FMNH (PE) 6176.— length width 1.2 1.6 incomplete 21 . 2) oh) 2.6 5:2 3.6 Tergite No. incomplete missing 3.9 2.8 (from side) 2.8 (from side) 2.3 (dorsal) 1 2 3 4 5 6 i 8 9 10 11 4.5 12 4.5 Remarks. —Hydroscorpius denisoni is easily differ- entiated from the other scorpions in the Cottonwood Canyon locale by its short, stout legs and round car- apace. It differs from Pa/aeoscorpius devonicus in hav- Text-figure 20.— Hydroscorpius denisoni, n. gen., n. sp. Holotype, FMNH (PE) 6176. From the Lower Devonian, Beartooth Butte For- mation, of Cottonwood Canyon, Big Horn Mts., WY, U.S.A. See foldout inside front cover for explanation of abbreviations. A. Entire specimen. B. Free chela of the left chelicera. C. Right chelicera showing the four joints. The curved line at the base rep- resents the edge of the cephalothorax. Note the mass of setae. D. Terminal end of the third right walking leg. E. Detail of left pedipalp and part of the second left walking leg. F. First left walking leg. G. Carapace with median eye node. H. Right pedipalp. chelicerae ( ) uv FossiL SCORPIONIDA: KJELLESVIG-WAERING —/ pedipalp ~ B 66 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 ing stubbier walking legs and not so bulbous a pedipalp hand. Superfamily ARCHAEOCTONOIDEA Petrunkevitch, 1949 (emend.) Holosternina with first pair of coxae meeting in front of the sternum, without maxillary lobes; other three pairs of coxae abutting against the sternum; legs very short. Type family. —Archaeoctonidae Petrunkevitch, 1949. Family ARCHAEOCTONIDAE Petrunkevitch, 1949 (emend.) Archaeoctonoidea with sternum large, campanulate, slightly tapering and pointed anteriorly; no lateral com- pound eyes. Type genus. —Archaeoctonus Pocock, 1911. Remarks. —There is no family that this very distinct scorpion group could be confused with except the Lo- boarchaeoctonidae, which is a lobostern, whereas 4r- chaeoctonus 1s a holostern and has many other obvious fundamental differences. Genus ARCHAEOCTONUS Pocock, 1911 (new definition) Legs very short, not reaching to the posterior of the preabdomen, and in general not very different in length; hand of pedipalp and tibia very short; fingers straight with contact along edge; free finger hooked at distal end. Carapace with anteriorly located median eyes and no lateral eyes. Type species. —Eoscorpius glaber Peach, 1883. Geological range. —Lower Carboniferous. Archaeoctonus glaber (Peach, 1883) Text-figures 21, 110K, 113B1 1883. Eoscorpius glaber Peach, pp. 400-402, pl. 22, figs. 2, 2a—21. 1885. Centromachus glaber (Peach). Thorell and Lindstrém, p. 25. 1911. Archaeoctonus glaber (Peach). Pocock, pp. 17-18, text-fig. 3. 1913. Archaeoctonus glaber (Peach). Petrunkevitch, p. 34. 1949. Archaeoctonus glaber (Peach). Petrunkevitch (partim), pp. 138- 139. 1953. Archaeoctonus glaber (Peach). Petrunkevitch, pp. 14-15, figs. 14-17, 126. 1955. Archaeoctonus glaber (Peach). Petrunkevitch, p. P73, figs. 40(2), 41(1). 1959. Archaeoctonus glaber (Peach). Wills, p. 266. 1960. Archaeoctonus glaber (Peach). Wills, p. 328. 1962. Archaeoctonus glaber (Peach). Dubinin, p. 428, fig. 1241. Not Archaeoctonus glaber (Peach). Petrunkevitch, 1949, pp. 138- 139, specimen BM(NH) In.988, figs. 138, 176 [=Loboar- chaeoctonus squamosus Nn. gen., n. sp.]. This unusual scorpion was first described by Peach, who did not fully appreciate that only the ventral side was showing. The description lacks the most important feature of this scorpion, namely the coxosternal area, which was preserved in considerable clarity. This fact was also not appreciated by Pocock (1911, pp. 17-18), who apparently worked only from Peach’s figures. Pe- trunkevitch (1953, pp. 14-15) gave a detailed descrip- tion with four figures. He recognized the coxosternal area and further described several parts, which are here redescribed. I disagree in several important respects with Petrunkevitch’s description and figures. The most serious point of disagreement is his statement that a “*kidney-shaped stigma (lung opening)”’ was present on the right side of the fourth “‘sternite”. Wills (1959, p. 266) subjected the holotype to careful review and con- cluded that no stigmata were present. In my study of the holotype at Edinburgh, I made a special effort to examine the holotype under different types of light and can state without the slightest hesitancy that there are no stigmata present and that the scorpion has holo- sternous abdominal plates such as are found in count- less other fossil forms. My interpretation of the coxosternal area differs widely from that given by Petrunkevitch (1953, p. 14), although he deserves credit for being the first to de- scribe this region. The sternum is very large, campan- ulate, straight at the base, with sides slightly tapering anteriorly. The anterior is produced into a point. The first pair of coxae abut each other above the sternum. There are no maxillary lobes present as was stated by Petrunkevitch (1953, p. 15). The coxae are all very short (see Text-fig. 21 A). All legs are very short, but they are not tubiform. They resemble, in a great sense, the legs of Palaeo- phonus, Allopalaeophonus, Palaeoscorpius and other older scorpions. The segments are not greatly differ- entiated in length (see Text-fig. 21 A). The right fourth leg is preserved so that the terminal parts are clearly seen. The ungues are short, falcate, without underlying spines, and the posttarsus (IV7) is rounded and very short. A short spine or spur occurs in the tissue between the two joints, or at the end of the basitarsus (IV5S) (see Text-fig. 21B), which of course is the basitarsal spur. Text-figure 21.—Archaeoctonus glaber (Peach). GSE 5858 and 5859 (part and counterpart) (holotype), and GSE 2039, a juvenile form. From the Lower Carboniferous (lowermost Bernician or S1 of the Viséan), Upper Bradon Group, Cementstone-sandstone Series, Glencartholm Volcanic Beds, River Esk, 5 miles south of Langholm, Dumfriesshire, Scotland. See foldout inside front cover for expla- nation of abbreviations. A. Coxosternal organization taken from GSE 5858 and GSE 5859. B. Terminal detail of leg IV, with two claws (from GSE 5859). C. Chela. Free finger and most of palm from GSE 5859, the rest from GSE 5858. D. A nearly complete pedipalp from GSE 5858. E. Ju- venile individual. Drawing made from the original specimen, GSE 2039, and a latex cast thereof. FossiL SCORPIONIDA: KJELLESVIG-WAERING 67 68 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The pedipalps (see Text-figs. 21A, D) have a short femur and tibia, but the chela is composed of a short hand with two very long, rather straight, fingers that are decidedly falcate at the ends. The inner edge is entirely cultrate without the slightest evidence of den- ticles. The trochanter, femur, and tibia are covered with squamous ornamentation, whereas the hand re- tains some scattered mucrones and several setal sites. The fingers also show setal sites, and some are in linear series, at least for a limited area (see Text-fig. 21D). The chelicerae are well preserved, and although only the last three joints were seen, there should be little doubt that four occur. The denticulation is not overly unusual (see Text-fig. 21C), however, the free finger (Ch4) is bifid and undoubtedly the teeth straddle the single termination of the fixed ramus. Another specimen, GSE 2039, in the collection of the Institute of Geological Sciences, Edinburgh, Scot- land (see Text-fig. 21E) can safely be included in 4r- chaeoctonus glaber (Peach) on the basis of the pedipalp, and I have no hesitance in referring it to 4. glaber. The specimen is a juvenile, preserved on dark gray to black shale in a flattened condition; nevertheless it is highly important as it reveals the dorsal side of this important species. The specimen consists of only one piece, with the carapace, most of the opisthosoma, the right pedipalp and fragments of the legs. The carapace is rather squarish, with an undivided raised cephalic shield, no glossate process, and with small, rounded, median eyes placed on a small, round- ed or elliptical, optical mound that is practically so far forward on the carapace as to be marginal. No lateral aggregate or compound eyes are present. The raised cephalic shield is covered with coarse pustules. The pedipalp reveals the short, stout femur and tibia with a large hand and fingers, which are opposable, incurving so that the inner edges do not meet. The ends undoubtedly must have been falcate. The prosomal walking legs are only preserved as small fragments of legs III and IV, but show that they are very short, which is a character of the holotype. The entire preabdomen is preserved, is widest at the sixth tergite, and except for scattered, fine pustules, devoid of ornamentation. No carinae are present. Only the anterior half of the cauda is preserved, and shows three unornamented tergites. Measurements (in mm) of GSE 2039.— Nn Carapace length: Carapace width: Length of preabdomen: Greatest width at sixth tergite: Overall estimated length: Neen OANWON Type information. —River Esk, 5 miles south of Langholm, Dumfriesshire, Scotland, in the Lower Car- boniferous, Cementstone-sandstone Series (lower Vi- séan) Upper Border Group, Glencartholm Volcanic Beds, on deposit in the Institute of Geological Sciences, Edinburgh. Specimen GSE 2039 is from the same ho- rizon and locality as the holotype (GSE 5858, 5859, part and counterpart). Remarks. —Peach (1883, p. 402) listed a specimen from the Lower Carboniferous (mid-Viséan), at Red- halls quarry, Water of Leith, near Slateford, Edinburgh, Scotland, which he determined as Eoscorpius glaber. I have not been able to locate this specimen and be- cause the ages involved are quite different, it is reason enough to question the determination until the original can be studied and verification made. Genus PSEUDOARCHAEOCTONUS, new genus Archaeoctonidae with trapezoidal carapace, ante- riorly-located median eye node, no lateral eyes and with a raised, partly divided cephalic shield. Pedipalps with very small teeth, apparently on a single row along the inner edge of the fingers. Coxae and legs, long, not tubiform. Pectine, small, long, composed apparently of an undivided rachis, no apparent fulcra and rela- tively few teeth—less than 15 on each side. Derivatio nominis. —pseudo (Gr.) = false + Ar- chaeoctonus, a scorpion genus. Type species. —Pseudoarchaeoctonus denticulatus, n. gen., n. sp. Geological range. —Lower Carboniferous. Remarks. —The coxosternal arrangement seems to be the same as in Archaeoctonus, but the coxae are much longer, as are the rest of the legs. The pedipalp of P. denticulatus, instead of having the fingers op- posable and falcate, is recurved on the fixed finger with the free finger adjoining the other along the entire cut- ting edge. It is also denticulate, as against the cultrate cutting edge of Archaeoctonus glaber. The genus Es- kiscorpio differs in being a Lobosternina and in having lateral compound eyes and cultrate and falcate fingers. Pseudoarchaeoctonus denticulatus, new species Text-figure 22 The holotype is a specimen preserved in dark gray shale, revealing mainly the dorsal side of the prosoma and parts of the appendages, all of the preabdomen, and one tergite of the cauda. The coxosternal region has been partly impressed through the carapace. The holotype is in the collections of the Institute of Geo- logical Sciences, Edinburgh, Scotland, where it regis- tered as GSE 2038. The carapace is trapezoidal, without an anterior glos- sate process, and with a raised cephalic shield that is partly divided at midsection. The lateral margins are convergent anteriorly. The eyes are round, small, lo- FossiIL SCORPIONIDA: KJELLESVIG-WAERING 69 cated on a small, rounded, anteriorly-located ocellar mound. The entire shield is covered by coarse scales. Scattered, but smaller, scales occur in the posterior area of the carapace. The base of the carapace is bounded by a thick marginal rim. The coxosternal arrangement is visible in part, since it is impressed through the carapace. The entire area is very probably like that in the Archaeoctonidae, al- though it must be admitted that the possibility exists that, instead, all four coxae abut the very large sternum. If this is so, it would require the establishment of a new family referable to the superfamily Proscorpioi- UU ae Text-figure 22.—Pseudoarchaeoctonus denticulatus, n. gen., n. sp. Holotype, GSE 2038. From the lower Carboniferous (lower Viséan), Glencartholm Volcanic Beds, River Esk, Langholm, Dumfriesshire, Scoland. See foldout inside front cover for explanation of abbreviations. 70 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 dea. The sternum is very large, the base is square and appears to have straight lateral sides that are abutted by the last three pairs of coxae and possibly also by the first. No maxillary lobes are developed. The an- terior of the sternum is unknown. The chelicerae are large, but are not sufficiently well- preserved to warrant detailed description. The pedi- palps are short but stoutly constructed. The femur is particularly thick and considerably longer than the tib- ia, which is very short. The hand and chela are large, with the fixed or immovable finger recurving, and the free finger incurving forward to fit both cutting edges together. The edges of the fingers are armed with a line of small denticles giving a serrated appearance (see Text-fig. 22). Coarse semilunar to submucronate scales cover the trochanter, femur and tibia. The rest of the legs are too poorly preserved for description, but are longer than those of Archaeoctonus glaber and are flattened as in modern scorpions, and not tubiform. The coxae are relatively long. The preabdomen is devoid of ornamentation of any kind, each tergite increases in length, is bounded an- teriorly by a transverse ridge, and the greatest width is reached at the fourth and fifth tergites. The seventh tergite seems to be smooth, without any carinae. Only the first tergite of the cauda is preserved. The pectine of the right side has been preserved and this is a small structure, as long as wide, without any divisions or areoles and without fulcra. The teeth are short, stout, and less than 15 occurred on each pectine. Nothing else is known of the underside. The specimen is completely flattened on the shale. Measurements (in mm) of the holotype, GSE 2038.— Pedipalp: Femur length (P3) 4.71 width (P3) DS], Tibia length (P4) 3.6 width (P4) 2.8 Hand length (P5) 3212 width (P5) 3.92 Free finger length 5.7 Type information. —Glencartholm Volcanic Beds, Upper Border Group, Lower Carboniferous (lower Vi- séan), River Esk, Glencartholm, Langholm, Dumfries- shire, Scotland. Derivatio nominis.— denticulatus (L. dim. of dens) = tooth: alluding to the small teeth on the pedipalp. Remarks. —The ornamentation of raised elongate scales is remarkably like that of mixopteroid euryp- terids. The genera Archaeoctonus and Loboarchaeoc- tonus have similar scales and occur in the same beds. Superfamily ACANTHOSCORPIONOIDEA, new superfamily Holosternina with the first pair of legs meeting at the midline, with well-developed maxillary lobes, last three pairs abutting the sternum. Type family. — Acanthoscorpionidae, new family. Remarks. —There is not much doubt in my mind that this holosternous scorpion represents a different superfamily than the Archaeoctonoidea, but at our present state of knowledge it seems best to include it in that superfamily. More specimens, perhaps better preserved, will resolve this matter.!! Family ACANTHOSCORPIONIDAE, new family Acanthoscorpionoidea with ovoid sternum and lat- eral compound eyes. Type genus. —Acanthoscorpio, n. gen. Genus ACANTHOSCORPIO, new genus Acanthoscorpionidae with ovoid sternum and square carapace with lateral compound eyes poorly devel- oped; epidermis covered with small triangular mu- crones. Derivatio nominis. —acantho (Gr.) = spiny + scor- pion. Type species. —Acanthoscorpio mucronatus, n. gen., Nn. sp. Geological range. —Devonian. Remarks. —There is no scorpion which resembles this one. Because of the robust cauda, as well as the square carapace with compound eyes, it superficially resembles the Scottish Dolichophonus loudonensis Laurie, but the latter is a bilobostern, whereas Acan- thoscorpio mucronatus is a definite holostern. Acanthoscorpio mucronatus, new species Text-figures 23, 24, 110N The species is based on a single, nearly complete specimen, which has been split laterally so that each half is composed mainly of dorsal or ventral tissues. The cuticle is unaltered, retaining the original color- ation, which is a light brown reminiscent of many of '! However, the Acanthoscorpionidae have maxillary lobes and the Archaeoctonoidea do not, and thus cannot be in the same su- perfamily. Moreover, the huge campanulate sternum of Archaeo- ctonus has no resemblance to the moderately-sized oval sternum of the Acanthoscorpionidae. A.S.C. Text-figure 23.—Acanthoscorpio mucronatus, n. gen., n. sp. Ho- lotype, FMNH (PE) 6177b. From the Lower Devonian, Beartooth Butte Formation, of Cottonwood Canyon, Big Horn Mts., WY, U.S.A. See foldout inside front cover for explanation of abbreviations. A. Dorsal aspect of the complete holotype. B. Anterior details. C. Terminal details of right legs III and IV. D. Terminal detail of first right leg. Either a single or double claw. FossIL SCORPIONIDA: KJELLESVIG- WAERING 71 the living Buthidae. The specimen unfortunately has been rolled over to the side, so that interpretation of important morphological details is extremely difficult. This is particularly true of the coxosternal region, and although an attempt at reconstruction has been made, 12 3 this has been accomplished with considerable misgiv- ings, and I believe firmly that confirmation from other specimens is needed before being fully accepted, as the structures are important phylogenetically, and mis- takes in interpretation are easy to make. right SIDE of ae WIS 12 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The carapace is nearly quadrate, measuring 2.6 mm in length and 2.65 mm at the base. The anterior is completely straight, the base parallel and rounded at the genal angles, which join the straight and parallel lateral margins. The small compound eyes occur at the anterolateral angles and contain facets of very small size. There is no sign of a median eye node nor of the median eyes, although it must be pointed out that this area is not well preserved. There is no doubt, however, that median eyes are present as no scorpion with lateral eyes is known without median eyes. I suspect that these are present and that they are placed well forward on the carapace. The pedipalps are only partly preserved, but enough is present to show that they were slender Text-figure 24.—Acanthoscorpio mucronatus, n. gen., n. sp. Holotype, FMNH (PE) 6177a. From the Lower Devonian, Beartooth Butte Formation, of Cottonwood Canyon, Big Horn Mts., WY, U.S.A. See foldout inside front cover for explanation of abbreviations. A. Ventral aspect of the holotype, counterpart of Text-figure 23A. Shows the laterally rolled nature of the specimen. B. Anterior details. FossiL SCORPIONIDA: KJELLESVIG- WAERING 73 and elongated. On the outer base of the brachium (P3) is a row of four large setal holes, with three other setal holes forming a triangle more toward the center. These are on the ventral side. The appendages are also poorly preserved, but enough is present for a fairly good description of these struc- tures. The chelicerae are very large, apparently four- jointed, but little can be made out of the denticles or the shape of the chelae. The four walking legs are rel- atively stout, all armed with at least double basitarsal and tibial spurs, all of which are well developed but not unusually long or robust. The ungues (I6S) are very long, slightly bent downward, slender and prob- ably covered with short setae. Setal holes are present. The posttarsus is short, robust and also, very likely, covered with setae (see Text-figs. 23C, D). Each of the coxae is armed at the end with a row of spines, and these short spines are probably found at the end of each joint of all legs. The coxosternal region, unfortunately, was broken apart and the interpretation given here needs verifi- cation. The sternum is ovoid, wider on the posterior half. Apparently the long triangular coxae of the sec- ond, third and fourth pairs of legs abut it. Anterior to this is the first pair which meets at the midline, and medium-sized maxillary lobes adjoin one another. The opercular plate is single, ankylosed, and with a median indentation at the basal midsection. It is truncato-ovoid (see Text-fig. 110N) in shape. The pectines are large, flipperlike, unjointed, but ap- parently with a poor development of small rounded areoles, although, for all practical purposes, the pectine may be considered as a single plate. Fulcra are pre- served and these are small and rounded. The teeth are very poorly preserved, but are the long type, not as in Branchioscorpio richardsoni, and there were probably about 25 on each comb. The abdominal tergites have been displaced later- ally, but show that they are not unusual except for the ornamentation. Each tergite is bordered by a narrow transverse ridge and each increases in size posteriorly. The important seventh tergite cannot be described as it was crushed. The postabdominal tergites, forming the cauda, are noteworthy for their great thickness, showing that the cauda was very robust. The abdominal plates are clearly holosternal and, of course, without the slightest indication of stigmata. Each overlaps the succeeding. A narrow, poorly de- veloped, anterior transverse ridge is present on each. The ornamentation is of extreme interest as it recalls that present in many of the eurypterids. The carapace has at least one row of sharp, pointed, triangular and elevated spines, which normally are referred to as mu- crones in eurypterid terminology. These single rows of large mucrones mark the base of each preabdominal tergite and, moreover, each tergite is covered with smaller mucrones. The abdominal plates likewise are covered with scattered mucrones. The overall length of this scorpion is estimated at 12.5 mm and contrasts with the nearby, enormous, free chela of the chelicera of the eurypterid Pterygotus mcegrewi Kjellesvig-Waering and Richardson, where a single denticle is considerably larger than the entire scorpion. Type information.—Lower Devonian, Beartooth Butte Formation, from the FMNH excavation site in Cottonwood Canyon, Big Horn Mountains of Wyo- ming, U.S.A. The specimen is deposited in the Field Museum of Natural History, Chicago, and is registered as FMNH (PE) 6177a and b. Derivatio nominis. — mucronatus (L.) = spiny. Family STENOSCORPIONIDAE, new family Acanthoscorpionoidea with last three pairs of coxae abutting the sternum, which is ovoid. First pair of coxae unknown. Carapace divided into halves, without lateral compound eyes. Abdominal plates with gill chamber opening plain or with sparse spines and setae, opening posterolaterally. Type genus. —Stenoscorpio, n. gen. Remarks. —It is with considerable confidence that the coxosternal area of Wills coll. specimen No. 164 of Stenoscorpio gracilis (Wills, 1947, text-fig. 40D) is reconstructed, revealing that the last three pairs of cox- ae abut the sternum, inasmuch as the bases of all three coxae are squeezed together naturally and not by pres- ervation. Surprisingly, this is somewhat similar to Acanthoscorpio of the Lower Devonian. In itself, this in not as strange as it seems, since the Triassic fauna of England includes such very “primitive” elements as the Spongiophonidae. Unfortunately, the first pair of coxae is unknown, as well as whether or not it and the second pair developed maxillary lobes. By early De- vonian, the Acanthoscorpionidae had developed both, so that it is not impossible that the reconstruction made in Text-figure 25B is correct. There is the possibility that the first pair of coxae developed only poorly formed maxillary lobes such as those found in the Spongiophonidae. At any rate, the Stenoscorpionidae differ from the Acanthoscorpioni- dae in lacking the compound eyes which were devel- oped in the latter. Genus STENOSCORPIO, new genus Carapace quadrate; no lateral eyes; anterior margin sharply glossate; round median eyes anteriorly located. Cephalic shield is divided longitudinally by median 74 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 groove; opercular plates elongated, subelliptical. Pec- tines with narrow, undivided rachis and without fulcra. Derivatio nominis.—steno (Gr.) = narrow, + scor- pion. Type species. —Mesophonus gracilis Wills, 1910. Geological range. — Triassic. Remarks. —The above generic diagnosis is based on the type species, except for the characteristics of the pectines (no fulcra) which are revealed by S. pseudo- gracilis. Stenoscorpio gracilis (Wills, 1910) Text-figures 25, 110M 1910. Mesophonus gracilis Wills, pp. 317-318, pl. xxv, figs. 2-4; not pl. xxv, fig. 10 (specimen 195) = Willsiscorpio bromsgro- viensis (Wills). 1947. Mesophonus gracilis Wills. Wills, pp. 116-120, pl. IU, figs. 2-9a; pl. V, fig. 2; pl. VI, fig. 8; pl. VIII, fig. 9; text-figs. 3B, 5G, 20, 23, 24, 40D, 45C, 48A-C; not pp. 121-122, pl. V, figs. 6, 9; pl. XI, fig. 4; text-fig. 48, J to M (specimen 0278) = Stenoscorpio pseudogracilis (Wills). 1953. Mesophonus gracilis Wills. Petrunkevitch, p. 36. The lectotype (Wills coll. No. 132) is a well-pre- served tergite, identified here as the fifth tergite, which has a well-developed anterior median keel (““notch’’). Determination of this tergite as the fifth indicates the probability that all the tergites are carinated, as even the seventh tergite has the same dorsal anterior median keel. Wills coll. specimen numbers included under this species are 132 (lectotype), 11, 025, 102, 137, 147, 164, 185, 0191, 203, 228, 0257, 283, and 327A. Type information. —Triassic Lower Keuper Sand- stone Series, Bromsgrove Quarry, near Birmingham, England. Remarks. —Specimen numbers refer to Wills’ orig- 1mm Text-figure 25.—Stenoscorpio gracilis (Wills). From the Triassic (Lower Keuper) sandstone, Bromsgrove Quarry, near Birmingham, England. Based on specimen SM 164 (of Wills, 1947). Two plausible restorations of the coxosternal area, although that of 25B is preferred in the light of present information. The coarsely stippled areas on the first pair of coxae, maxillary lobes of the second coxae, and parts of the sternum and operculum are restored. inal specimens, which are in the collections of the Sedg- wick Museum, Cambridge, England. Stenoscorpio pseudogracilis (Wills, 1947) 1910. Mesophonus sp. Wills, pl. xxiii, fig. 3; pl. xxvi, fig. 12. 1947. Mesophonus gracilis Wills. Wills, pp. 121-122, pl. V, figs. 6, 7; pl. XI, fig. 4; text-figs. 48, J-M. 1947. Mesophonus infans A (?) Mesophonus gracilis Wills, pp. 122- 123, pl. V, fig. 1; pl. XI, fig. 12; text-figs. 48, D-H, 49. 1947. Mesophonus pseudogracilis Wills, pp. 123-126, pl. IV, figs. 9, 10; pl. VI, fig. 3; pl. XII, fig. 15; text-figs. SH, 6, 9, 17, 19, 42C; not p. 33, pl. V, figs. 3, 4; text-fig. 13 (specimen 094) = Mesophonus perornatus Wills. 1947. Mesophonus infans D Wills, pp. 128-130, pl. XI, figs. 2, 3; text-figs. 45B, 50A, B. One of the pertinent specimens for taxonomy and morphology of this species is the specimen described as Mesophonus infans D, the so-called type specimen (Wills, 1947, pp. 128-130, text-fig. 50, A, B). This specimen reveals the important seventh tergite, dorsal and ventral sides, which determines the specimen (0232B) to be S. pseudogracilis (compare specimen 012, Wills, 1947, text-fig. 19D). The abdominal plates and tergites are also preserved; the former revealing the gill opening to be small, opening ventrally and placed between the doublure and the posterior edge of the abdominal plate (Wills’ Type A2). The fourth ter- gite (Wills’ tx) has a part missing along the anterior of the transverse ridge in the form of a notch which very likely represents the slight median carina (Wills’ “‘notch’’) that characterizes the species. The species M. pseudogracilis lacks fulcra in the pec- tines. Among the more than 100 genera of living scor- pions, this lack of fulcra is of generic importance and is uncommon. The absence of those fulcra, duplicated in Paleozoic scorpions, inclines me to refer this species to the genus Stenoscorpio. The seventh tergite, how- ever, retains four well-developed ventral carinae, sur- mounted with scales, and this is an important generic character, at least among living scorpions. The devel- opment of the four carinae 1s apparently greatly re- duced in S. gracilis, but nevertheless is present, as attested by the presence of four rows of scales. It seems rather logical at present to consider M. pseudogracilis another species of Stenoscorpio. Both species have a central carina (Wills’ “‘notch’’) on the transverse ridge of the tergites on the dorsal side. The dorsal carina is much more pronounced in S. gracilis where, even on the seventh tergite, it is extended in a ‘V-shaped, central prolongation of the anterior transverse ridge. In S. pseudogracilis the cen- tral carina is greatly reduced, being present only on the fourth tergite, possibly slightly on the third. Specimen 282, the only specimen of Mesophonus infans A, which Wills (1947, p. 122) suggested might questionably be a neonate of S. gracilis, instead is a FossiL SCORPIONIDA: juvenile of S. pseudogracilis. Wills, throughout his monograph, refers to ““neonates’’, but all these are much too large to be neonates. They are more properly ju- veniles, as a neonate would be composed of such fragile cuticle that preservation would be very unlikely. Past the first instar, the scorpion assumes the shape and major characters that are present in the adult. This is not the case in true neonates. The presence of a strong anterior median keel or carina (“‘notch”’ of Wills) on the fourth tergite of M. infans A clearly indicates that this juvenile is Stenoscorpio pseudogracilis (Wills). Specimen 0278, labelled Mesophonus gracilis (Wills, 1947, pl. XI, fig. 4; text-figs. 48J, K, L, M) is an im- portant specimen, revealing much of taxonomic value. There are a number of areas where my study of the original material is not in agreement with Wills (1947, p. 121, and above text-figures). The opercula are elon- gate-ellipsoidal, but not so long as the restoration on text-figure 48M. I was unable to see any bilobation of the first abdominal plate. The fulcra are not present. The notch, which is so important in the identification of this species, is a slight median keel or carina that occurs on the anterior of the tergite and, when flattened into a single plane, it appears as a “notch” on the posterior of the anterior transverse ridge of the tergite. The presence of the anterior median keel (“‘notch’’) on the fourth tergite and its absence on the other tergites is sufficient for the identification of this specimen as Stenoscorpio pseudogracilis (Wills). The specimens which are included under this species are: SM 231 (holotype), 012, 029, 052B, 065, 80, 184, 229, 231, 0232B, 257 (fide Wills), 278, 279, 282. Type information. —Triassic, Lower Keuper Sand- stone Series, Bromsgrove Quarries, near Birmingham, England. Remarks. —All specimen numbers refer to Wills’ original specimens, which are in the collections of the Sedgwick Museum, Cambridge, England. Superfamily GIGANTOSCORPIONOIDEA, new superfamily Holosternina with last two pairs of coxae abutting the sternum, and first two pairs meeting at the midline in front of the sternum, without the development of maxillary lobes. Type family. —Gigantoscorpionidae, n. fam. Remarks. —The lack of maxillary lobes in the su- perfamily separates it from all other superfamilies of the Holosternina with similar coxosternal arrange- ments. Text-figure 26.—Gigantoscorpio willsi Stormer. GSE 2137. From the Lower Carboniferous, Eskdale, Scotland. T. S. Stock collection (developed by B. N. Peach). See foldout inside front cover for ex- planation of abbreviations. KJELLESVIG- WAERING T9dor T10dor > 710 dor Tiidor T1i2dor 76 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Family GIGANTOSCORPIONIDAE, new family Gigantoscorpionoidea with large, subrounded pen- tagonal sternum; carapace elongate without lateral compound eyes, but with median eyes anteriorly but intramarginally located. Type genus. —Gigantoscorpio Stormer, 1963. Genus GIGANTOSCORPIO Stormer, 1963 Gigantoscorpionidae with a row of fine denticles on the edges of the pedipalps and with well-developed, long, serrated, flaring tibial and basitarsal spurs. Type species. —Gigantoscorpio willsi Stormer, 1963. Geological range. —Lower Carboniferous. Remarks. —The large tibial and basitarsal spurs are of particular importance and are responsible for the identification of this genus in other specimens. Gigantoscorpio willsi Stormer, 1963 Text-figures 26, 27, 11OH, 113B2 1963. Gigantoscorpio willsi Stormer, pp. 15-61, pls. 1-3, pl. 4, figs. 1-5, pls. 6-12, pl. 13, figs. 1-4, 7, 8, pl. 14, pl. 15, figs. 1-3, pl. 16; text-figs. 1-3, 5-23. 1972. Gigantoscorpio willsi Stormer. Kjellesvig-Waering, p. 39. Specimen. —GSE 2137 deposited in the Institute of Geological Sciences, Edinburgh, Scotland, consists of a nearly complete specimen showing mainly ventral features, but a few dorsal structures are preserved. In life the specimen must have measured about 75 to 80 mm in length, from the anterior of the carapace to the base of the telson. It is associated with several ostra- codes, and a smooth-shelled pelecypod lies on top of the second abdominal plate. The specimen had been “developed” by B. N. Peach, as stated (etched) on the shale slab along with “‘Eskdale, T. Stock collection”. The specimen is in the grayish black shale that char- acterizes the River Esk fauna. Unfortunately, the development performed by Peach obliterated many details, although I do not mean any criticism, as optical instruments and the coarse chisels used at that time did not permit delicate work. It is remarkable that development was as good as it was. The description that follows is also based on a latex cast that further brings out certain important details. The carapace is subquadrate, pointed along the an- terior margin, and with a large elliptical eye node sur- mounted on the sides with two rounded median eyes; the eye node is placed intramarginally but well forward of the center of the anterior half of the carapace. The anterolateral angles are rounded and without lateral eyes. The pedipalps are developed as stout, strong struc- tures, and, although the first three joints are present (P1, P2, P3), they show that the basal joint could not form part of the oral chamber as it does in present- day scorpions. The chelicerae are large, composed of the usual four joints, and occur above the basal joint of the pedipalps. The pedipalps, at least joints P2 and P3, the trochanter and brachium respectively, contain pustules, which on the ventral side of the brachium occur in a row along the back edge, as do the tricho- bothria in some living scorpions (for example, some of the Chactidae such as Broteochactas, Chactas, Teu- thrautes, Broteas, etc.). These pustules were likely se- taceous. The dorsal side of the brachium of the pedi- palp reveals two carinae surmounted by pustules. Of particular importance is the coxosternal arrange- ment. The sternum is very large, round, probably peaked at the anterior, giving it a rounded pentagonal shape. It has a deep median sulcus that seems to bi- furcate in the middle part of the sternum. The last two pairs of coxae abut against the sternum. The first and second coxae meet at the midline in front of the ster- num and neither pair has developed maxillary lobes, although both project slightly anteriorly to a blunt- pointed inner part that very likely represents a prim- itive development of the maxillary lobes. The abdominal plates are holosternous, with round- ed epimeral angles. The last postabdominal tergite, seen only in ventral view, is large with a very well-developed transverse ridge and with two short carinae surmounted with pus- tules. The cauda is nearly complete except for the telson. All tergites are long but stout, and in the case of the first two, both sides could be extracted and preserved. The dorsal side has three carinae, all surmounted with coarse granules, or pustules, whereas the venter seems to have only two carinae, also surmounted with coarse granules. There is at least one lateral carina, which also is surmounted by coarse pustules. The genital opercula, most of the legs and the pec- tines were not preserved. The latter undoubtedly were covered by the first and second abdominal plates. Specimen. —Holotype of Gigantoscorpio willsi Stormer. Part of counterpart, BM(NH) In. 42706a in the British Museum (Natural History), London, En- gland. Several small pieces of black calcareous shale, with parts of a large scorpion, were found while engaging in the study of the fossil scorpion types at the British Museum. I identified these pieces as parts of Gigan- toscorpio willsi Stormer, and later, Mr. Samuel Morris of the British Museum recognized them as part of the counterpart of the holotype of G. wi//si. Actually, the pieces represented the entire left half of the specimen as shown by Stormer in his figure (1963, fig. 1). Dr. Stormer and I had discussed several times the possi- bility of the large oval plate, which had been identified FossIL SCORPIONIDA: KJELLESVIG-WAERING a (Stormer, 1963, fig. 18) as the operculum, being the sternum instead, and the part with the small double plates, which had been identified (Stormer, 1963, fig. 10) as possibly the pregenital tergite, as being the oper- cular plates. These parts, particularly the so-called “sternum” as well as the terminal portion of the first walking leg are well preserved on the counterpart, and therefore it was anticipated that the question would be settled by the newly-found counterpart. Conse- quently, the counterpart was borrowed from the British Museum through the kindness of Dr. H. W. Ball, who sent the specimen to me, first in Norway and later in Chicago. In both places I have had long discussions with Dr. Stormer regarding the identification of these parts. Dr. Stormer and I agreed that the possible pregenital tergite (Stormer, 1963, p. 51, text-fig. 10) represents the two opercular plates, and they occur in their rightful or Original position and are seen from the dorsal sur- face. The part previously identified by Stormer as the operculum is preserved entirely from the ventral sur- face, and consists of two large semicircular plates, joined at midsection by a deep sulcus. These represent the large pectinal plates, or median anchor plates to which the large pectines were attached by a condyle, socket and intersegmental tissue. These plates are shown here on Text-figure 27A. Attachment is on the lower part of the anterior lamina of the pectines with the above- mentioned condyle and socket, as shown in the res- toration (Text-fig. 27B). Traces of the areoles of the middle lamina occur attached to the anterior lamina (see Text-fig. 27A), which prove that the structures in question are the pectines and that one is attached to the pectinal plate. Text-figure 27.—Gigantoscorpio willsi Stormer. From the Lower Carboniferous (lower Viséan), Glencartholm Volcanic Beds (the “‘River Esk Fauna’’), River Esk, Dumfriesshire, Scotland. A-C. Holotype, part and counterpart, BM(NH) In.42706a,b. See foldout inside front cover for explanation of abbreviations. A. Somewhat problematical impressions of the counterpart surface. B. Pectinal area partially revised and restored from Stormer (1963). C. Terminal details of leg II of the holotype. D. Carapace, GSE 2134. 78 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 This pectine, therefore, 1s the one of the right side, while the left pectine is folded over, and preserved with the dorsal side showing. This is the pectine figured by Stormer (1963, text-fig. 18) as the pectine of the mght side. In my interpretation, the very small narrow plate, with a median organ (Stormer, 1963, p. 55) is much too fragile to hold the enormous pectines, much less articulate against them, and this plate, therefore, is not the pectinal anchor plate, but the prepectinal plate, greatly reduced. The interrelationship of the plates can be seen in Stormer’s (1963, pp. 79-82, text-fig. 32) excellent figure of Centromachus euglyptus (Peach) (see Text-fig. 31B) and in Opsieobuthus pottsvillensis Moore (see Text-fig. 31A). In Stormer’s figure 32, the part labelled as the “‘basal plate of the combs” which carries the median organ, in my opinion, is the prepectinal plate (ppp) whereas the “‘basal plate of the combs” can be seen directly below, with the combs attached to it (see Text-fig. 31B here for changes). This implies that the opercula are prosomal organs and, because in many primitive scorpions the last pair of legs abuts against them, the possibility is worth considering. It is, how- ever, far from the purpose of this paper to dwell on comparative morphologies other than on the taxonom- ic levels. Based on this as well as the parts of the coxosternal area preserved in specimen GSE 2137 (see Text-fig. 26), the coxosternal region can be reconstructed as in Text-figure 110H. The coxosternal area, therefore, comprises a large pentagonal sternum with the third and fourth pairs of coxae abutting against it and with the first and second pairs meeting at midline, but with- out maxillary lobes, as shown on specimen GSE 2137 of the Institute of Geological Sciences, Edinburgh. The two small squarish opercular plates fit below the ster- num (Stermer, 1963, fig. 10). The terminal joints of the second leg were preserved in an excellent condition (Text-fig. 27C). Two claws are present, which are long and strongly unciform, without ventral denticles as commonly present in many Carboniferous scorpions. One of the claws is slightly larger and this is probably the anterior. Presumably, therefore, the leg is the second of the right side. The posttarsus (II7) is short, pointed and triangular. The tarsus (II6) is unusually short, with a terminal lobe, or emargination at the terminal part. The entire tarsus widens appreciably, flaring toward the terminal part. The basitarsus is partly preserved and this also is flar- ing toward the terminal part. It has two spurs at the usual part, and one of these seems to be rounded. The other was broken. A jagged piece of a tibial spur also was preserved, which greatly resembles that figured by Stormer (1963, fig. 16). The presence of this leg of the holotype confirms Stormer’s determination that GSE 2174 belongs to Gigantoscorpio willsi. Another specimen which is referred to Gigantoscor- pio willsi reveals the entire carapace in good condition, but flattened on the black shale. This specimen is GSE 2134 (see Text-fig. 27D) and is deposited in the col- lections of the Institute of Geological Sciences, Edin- burgh, Scotland. The rim is well developed in the posterior and lower half of the carapace, then narrows anteriorly so that it is not noticeable along the front. The anterior of the carapace is produced into a linguiform process. The eyes are elliptical with strong orbital ridges, but very likely originally round. The eyes occur on a lacrimi- form node. They measure about 1.6 mm in the con- dition they are in (flat). The anterior half is elevated into a cephalic shield, which was divided at midsec- tion. This outline of the cephalic area is accentuated by the coarse pustular ornamentation. Measurements (in mm) of carapace of GSE 2134.— Length: Si IRs) Width at base: 41.0 Width at midsection: 35.5 Width at anterior (at eyes): 27.0 The entire length of the scorpion from the anterior of the carapace to the end of the cauda is estimated at 240 mm, which is large for a scorpion as compared with modern ones, but is considerably less than half the size indicated by some other fossil specimens. The holotype of Gigantoscorpio willsi Stormer is broken, and lacks the raised cephalic shield. The type of median ocelli, on a mound at the anterior of the carapace, the ornamentation, the glossate process, the outline of the carapace and, significantly, the type of laterobasal rim of GSE 2134 are identical with those of the holotype, and I do not hesitate to identify this specimen with the latter. Type information. — All the specimens, the holotype and GSE 2134, GSE 2137, and GSE 2174, are from the Lower Carboniferous (lower Viséan) in the Glen- cartholm Volcanic Beds, Upper Border Group, at Riv- er Esk, Glencartholm, Langholm, Dumfriesshire, Scot- land. Remarks. —The large size of this scorpion is spec- tacular, estimated at 390 to 440 mm (Kjellesvig-Wae- ring, 1972, p. 39), but there is no doubt that these estimates can be doubled, based on the foot of GSE 2174. Nevertheless, other scorpions found in the Up- per Carboniferous, Eoscorpius carbonarius and Titan- oscorpio douglassi, as well as Brontoscorpio anglicus of the Upper Silurian, and Praearcturus gigas of the Devonian, are equally as large or larger. The most spectacular and taxonomically important aspect of G. willsi is the coxosternal arrangement. The FossiIL SCORPIONIDA: KJELLESVIG- WAERING 79 first two pairs of coxae are devoid of maxillary lobes but, surprisingly, this is combined with the condition of having the last two pairs of coxae abut the sternum as in modern scorpions. It was known, however, that the abutment of the last two pairs of coxae against the sternum was a very ancient development, as it was present in the Lower Devonian Branchioscorpio rich- ardsoni, which had greatly-developed maxillary lobes. It was a great surprise to have a supposedly advanced condition such as the last two pairs abutting the ster- num coupled with a very primitive condition of no development of maxillary lobes in the first two pairs of coxae. Superfamily MESOPHONOIDEA Wills, 1910 (emend.) Holosternina with first two pairs of coxae meeting at the midline in front of the sternum, and with max- illary lobes. Third pair abuts the sternum and fourth abuts the genital opercula. Type family. —Mesophonidae Wills, 1910 (emend.). Remarks. —The superfamily and family bear little or no resemblance to the original definition of Wills (1910, 1947) and Petrunkevitch (1953, 1955) but little can be gained by the introduction of new names when the older ones could be emended without causing any confusion. Mesophonoidea is the holostern “‘counter- part” of the lobosternous Isobuthoidea. Family MESOPHONIDAE Wills, 1910 (emend.) Mesophonoidea with narrow, pentagonal sternum: carapace subquadrate; median eyes placed well for- ward on the carapace, and with small lateral compound eyes. Type genus. —Mesophonus Wills, 1910. Remarks. —The family Mesophonidae Wills, 1910, is here emended (restricted) and seems to have clear priority over the Mazoniidae Petrunkevitch, 1913, and Centromachidae Petrunkevitch, 1953, both included under the Mesophonoidea. All three families, inciden- tally, are emended here and bear no resemblance to their original descriptions based on the type of orna- mentation. The description of the family Mesopho- nidae, as emended by Petrunkevitch (1955, p. 78) was meaningless, as most of it was erroneously based on the coxosternal arrangement in the genus Spongio- phonus, a scorpion that is completely different from Mesophonus. Apart from that, the family was based on characters only of species value, as for example, “Comb with 8 to 17 teeth’. As restricted here, the family Mesophonidae Wills, 1910, includes only the genus Mesophonus Wills, 1910. Genus MESOPHONUS Wills, 1910 (emend.) Mesophonidae with wide, subquadrate carapace; round median eyes located on prominent median mound that protrudes as a conspicuous, sharp lingui- form anterior process; dorsal shield undivided. Ab- dominal plates with gill openings truncated and located on inner posterior edge of plate. Type species. —Mesophonus perornatus Wills, 1910 (genolectotype by designation, Wills, 1947). Geological range. — Triassic. Remarks. —Apart from Spongiophonus pustulosus, Wills (1910, 1947) referred all British Triassic scor- pions to the genus Mesophonus.'* Petrunkevitch (1953) followed this treatment. However, the genus Meso- phonus included species with lateral facetted eyes and others without them. This condition was considered by Wills to be sexual dimorphism, although he cor- rectly pointed out that it was unusual for one sex to have lateral facetted eyes and the other to be without them. Although sexual dimorphism is present in scor- pions, the differences are ofa rather subtle nature; such as the differences in the punctation, color, particularly on the carapace and pedipalps, differences in the thick- ness of the cauda, relative length of the twelfth tergite, relative width, length and size of the pedipalp, etc. These sex differences seem to have been established early, as some of them have been noted in Silurian scorpions. However, there is never any difference due to sex in the number or type of eyes, even in genera in the Buthidae having the greatest number, or vari- ation, in size or number of lateral eyes. In the Car- boniferous, several genera of scorpions have lateral facetted eyes; other genera do not, but these are defi- nitely considered to be separate families and genera. Indeed, in many cases, those that have obvious dorsal similarity are found to belong to different families and superfamilies when the underside is known. The differences between the type species and Me- sophonus opisthophthalmus Wills, 1947, are due to sec- ondary sexual characters and therefore both species are identical with the type species, Z. perornatus hav- ing priority. Wills (1947, p. 107) mentioned that this might be the case, but elected to consider them as separate species. The chelicerae of Mesophonus have four joints as in all other Holosternina. '2 Dubinin (1962, fig. 1258) pictures a reconstruction of “*Meso- phonus sp.” (Werner, 1934), a scorpion about 42 mm long from the Tniassic, with a coxosternal arrangement like that of living scorpions, and holosternous abdominal plates. A.S.C. 80 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Mesophonus perornatus Wills, 1910 Text-figures 28, 110B, 113A3 1910. Mesophonus perornatus Wills, pp. 311-313, pl. xxii, figs. 1- 3, 5, 7-9; pl. xxvi, figs. 9, 13. 1947. Mesophonus pseudogracilis ? Wills, pp. 33, 123, pl. V, figs. 3, 4; text-fig. 13. 1947. Mesophonus perornatus Wills. Wills, pp. 102-106, pl. II, figs. 1, 2, 6-8; pl. IV, figs. 3, 4, 7, 8, 11, 12; pl. XII, figs. 1, 7-9, 16; text-figs. 3C, 5C-F, 11, 12, 21, 47D-H. 1947. Mesophonus opisthophthalmus Wills, pp. 106-108, pl. II, figs. 3-5, 9; text-figs. 3D, 5J. 1947. Mesophonus infans E Wills (partim), p. 130, text-figs. 51H—-J. 1947. Mesophonus sp. Wills, pl. VIII, figs. 1, 2; text-figs. 40A-C, 41. 1962. Mesophonus perornatus Wills. Dubinin, p. 432, fig. 1239. Mesophonus opisthophthalmus Wills, 1947, is a ju- nior synonym of M. perornatus Wills, 1910. On Wills’, 1947, plate II, figures 3 and 4 (SM spec. Nos. 0107 and 0243) are figured carapaces labelled as Mesophon- us opisthophthalmus, which show very sparsely scat- Ne apne “. We Text-figure 28.— Mesophonus perornatus Wills. From the Triassic (Lower Keuper) sandstone, Bromsgrove Quarry, near Birmingham, England. Based on Wills, 1947 (pl. VIII, figs. 1 (SM 040), 3 (SM 0110)). Showing the coxosternal area. Stippled areas are restored. tered scale-tubercles. These two carapaces are consid- ered here to be females and are identical with the female carapace of M. perornatus, the lectotype (Wills, 1947, pl. II, fig. 1, SM spec. No. 206). The male of M. per- ornatus, with dense granulation, is shown on Wills’ (1947) plate II, figure 2 (spec. No. 011). The type of granulation on the carapace and other areas is a well- known secondary sexual characteristic. Specimens referred to Mesophonus perornatus Wills are as follows: SM 206 (lectotype), 7, 008, 011, 017, 024, 034, 039A, 040, 85, 094, 095, 0110, 0112, 0122, 125, 0128, 156, 0176, 205, 210, 0212,'* 213 (fide Wills, 1947, p. 36), 0266, 277, 287. Also referred to this species by Wills, but not figured, are SM 050 (p. 42), SM 036 (p. 102), SM 0203 (p. 102) and SM 284 (p. 102). In addition, the specimens that formed the species Mesophonus opisthophthalmus Wills, a junior syn- onym, may now be added to this list: SM 0107, 0234, and 0161 (fide Wills, 1947, p. 107). Type information. —Triassic, Lower Keuper Sand- stone Series, from the quarries at Bromsgrove, near Birmingham, England. Remarks. — All specimens are in the Sedgwick Mu- seum, Cambridge, England. Mesophonus (?) pulcherrimus Wills, 1910 1910. Mesophonus pulcherrimus Wills, p. 318, pl. xxvi, figs. 2, 4, 6,7. 1947. Mesophonus pulcherrimus Wills. Wills, pp. 66-67, 72-74, 126, pl. VII, figs. 5, 9-14; pl. XII, figs. 19, 21, 22; text-figs. 32.39% It is not possible to correlate the lectotype (spec. No. 201) or other specimens (all caudal segments) referred to this species with any of the other known British Triassic scorpions and it is here referred questionably to Mesophonus. Type information. —Triassic, Lower Keuper Sand- stone Series, Bromsgrove Quarry, near Birmingham, England. Mesophonus (?) pulcherrimus var. immaculatus Wills, 1947 1947. Mesophonus pulcherrimus var. immaculatus Wills, p. 126, pl. III, fig. 12; pl. VII, figs. 8, 15, 16; pl. XI, fig. 7; text-fig. 36. The ‘‘variety” is questionably referred to Meso- phonus, inasmuch as the caudal segment comprising '3 Wills (1947) lists spec. SM 0212 twice. On page 102, 0212 (G227) represents M. perornatus, thirteenth segment of a young in- dividual, pl. IV, fig. 8, and text-fig. 47F. On page 108, 0212 (G22) represents M. bromsgroviensis, “other sternite”, pl. VI, figs. 2A and 2B, text-fig. 7. Likewise, 0203 (G230) is listed on page 102 as a thirteenth segment of M. perornatus, and on page 116, 0203 (G52) is listed as ‘“‘tergite with pecten” of M. gracilis. A.S.C. FosstL SCORPIONIDA: KJELLESVIG- WAERING $1 the lectotype cannot be correlated with any of the known Bromsgrove scorpions. Type information. —Triassic, Lower Keuper Sand- stone Series, Bromsgrove Quarry, near Birmingham, England. Mesophonus (?) maculatus (Brauer, Redtenbacher, and Ganglbauer, 1889) 1889. Periplaneta maculatus Brauer, Redtenbacher, and Gan- glbauer, p. 12, fig. 14. 1906. ? Ophismoblatta maculata (Brauer, Redtenbacher, and Gan- glbauer). Handlirsch, p. 527, pl. XLVI, fig. 2. 1928. Ophismoblatta maculata (Brauer, Redtenbacher, and Gan- glbauer). Grabau, p. 410, fig. 580 b. Brauer, Redtenbacher, and Ganglbauer (1889) de- scribed as a blattoid a single specimen consisting of most of the dorsal side of the prosoma and mesosoma of a scorpion from the Ust-Balei region of the Irkutsk Basin of Siberia. It occurred in Jurassic strata associ- ated with numerous insects and fishes. I have not been able to study the holotype, and presumably only spec- imen, as it was not available. From the original draw- ings, there does not appear to be much doubt that the specimen represents a scorpion. It seems from the fig- ure that most of the carapace is present with the suc- ceeding tergites of the preabdomen along with two frag- ments of the third and fourth legs. The mesosoma shows the increase in width toward the third and fourth tergites, which is common in many scorpions, along with a decrease in width of the tergites from that point, also a common feature in scorpions. The seventh ter- gite, although incomplete, is revealed as being much longer than the previous tergites. Carinae, a very com- mon feature of the seventh tergite in scorpions, are also present. The attachment of the prosoma to the mesosoma is characteristic of scorpions, not blattoids. All of these features, characteristic of scorpions, are not possible in a blattoid. Although little can be added to our knowledge from this specimen, while it remains lost, it is important because it is the second known Jurassic scorpion and the only one from the Far East. Family MAZONIIDAE Petrunkevitch, 1913 (emend.) Mesophonoidea with subequilateral pentagonal ster- num, second pair of maxillary lobes reaching halfway the length of the first pair. Median eyes close to anterior edge of carapace; no lateral compound eyes. Type genus. —Mazonia Meek and Worthen, 1868. Genus MAZONIA Meek and Worthen, 1868 Mazoniidae with carapace about as long as wide, with a pointed projection in front; proportionately large chelicera. Type species. —Mazonia woodiana Meek and Wor- then, 1868. Geological range. —Carboniferous. Mazonia woodiana Meek and Worthen, 1868 Text-figures 29, 110P, 113A2 1868b. Mazonia woodiana Meek and Worthen, pp. 563-565, figs. A, D. 1877. Mazonia woodiana Meek and Worthen. Miller, p. 224. 1883. Mazonia woodiana Meek and Worthen. Peach, p. 409, pl. 23, figs. 24, 25. 1889. Mazonia woodiana Meek and Worthen. Miller, p. 571. 1910. Eoscorpius (Mazonia) woodiana (Meek and Worthen). Gra- bau and Shimer, p. 416. 1911. Mazonia woodiana Meek and Worthen. Pocock, p. 11. 1913. Mazonia woodiana Meek and Worthen. Petrunkevitch, pp. 54-56, pl. 3, fig. 13. 1944. Mazonia woodiana Meek and Worthen. Lehmann, p. 178. 1944. Mazonia woodiana Meek and Worthen. Shimer and Shrock, p. 709, pl. 300, figs. 6-9. 1949. Mazonia woodiana Meek and Worthen. Petrunkevitch, p. 132: 1953. Mazonia woodiana Meek and Worthen. Petrunkevitch, pp. 12-13, figs. 11, 119. 1955. Mazonia woodiana Meek and Worthen. Petrunkevitch, p. 70, fig. 38(5). 1960. Mazonia woodiana Meek and Worthen. Wills, p. 321. 1962. Mazonia woodiana Meek and Worthen. Dubinin, p. 428, figs. 122591227. 1963. Mazonia woodiana Meek and Worthen. Stormer, p. 116, fig. 44. 1969. Mazonia woodiana Meek and Worthen. Kjellesvig-Waering, pp. 171-190, figs. 91-101. The holotype, in a typical Mazon Creek ironstone concretion (UI X-491), was thoroughly discussed in Kjellesvig-Waering (1969). That discussion will not be repeated here. Specimen. —Part and counterpart of a nearly com- plete scorpion except for part of the cauda and most of the walking legs, preserved as dorsal and ventral impressions in an ironstone concretion from the Mid- dle Pennsylvanian, Francis Creek Shale, collected in ths spoil heaps of Pit 11, on the Kankakee—Will County line, Illinois by Mrs. K. Taelle in 1970. The specimen is in the collection of the Field Museum of Natural History, Chicago, accessioned as FMNH (PE) 39253. This is an unusually good specimen, which reveals the important underside as well as further confirmation that the preabdomen comprised seven tergites, and not eight, as had been postulated by Meek and Worthen (1868b), and which later was confirmed erroneously by Petrunkevitch (1949, 1955), resulting in his estab- lishment of a suborder and family based on the ficti- tious presence of an eight-jointed preabdomen (see Kjellesvig-Waering, 1969, pp. 186-189). This speci- men is a very welcome addition to our knowledge be- cause of its great taxonomic significance. 82 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Little can be added to our knowledge of the dorsal side as that was fully described (Kjellesvig-Waering, 1969) in the holotype, and that discussion will not be repeated here. However, the cauda is partly preserved and is shown to be thick and surmounted with carinae that are not too clear. The underside also shows cari- nae. The pedipalp is unusually long, stout, with a very long manus and long curving fingers. The free fingers are bent forward and, although the fixed fingers are not preserved, these undoubtedly would be recurving backward. No denticles are developed and only small setal openings are present on the cultrate edge. The chelicerae are preserved and are composed of four joints, are hooklike and proportionately very large. Although teeth are developed, these are not preserved sufficiently for description. The first walking leg is preserved nearly in its en- tirety. Basitarsals (15) are well developed, and linear punctations occur on this joint. The other legs are large- ly unknown except for the tarsus of the fourth leg, which reveals a long (possibly double) tibial spur at the proximal end and two large basitarsal spurs at the distal end. Two rows of spines occur at the posterior. This fragment was dissolved from a piece of the con- cretion chipped off during cleaning of the specimen and was mounted on a slide (M.W. 1) (see Text-fig. 29F). A small seta is present at the end of the joint. The end of the fourth leg, consisting of the distal part of the tarsus, two claws and the posttarsus (IV7) also was dissolved from chips broken off during the cleaning process. This part was mounted on a slide (M.W. 2) and is shown on Text-figure 29E. The end of the tarsus shows two small setae. The tarsal spurs, which are the curved claws, are long, the anterior one being larger, and are bounded on the bottom edge by evenly spaced small spines. The transtarsus or post- tarsus (IV7) is small—a short pointed horn with a long- er area for the attachment of muscles. The important coxosternal area is revealed almost in its entirety. The arrangement is clearly that of Me- sophonoidea, a pentagonal sternum with the last pair of coxae abutting the genital plates, and the third pair the sternum; the first two pairs meet at midsection in front of the sternum, have well-developed maxillary lobes, but the second pair reaches only halfway the length of the first pair. The anterior part of the tubular area in front of the mouth therefore is made up of the bases of the chelicerae, the first joints of the pedipalps, and the maxillary lobes of the first pair of legs only. The sternum is pentagonal, with straight sides and a straight base. The opercular plates are elongate, ob- long, rounded at the corners, and fit directly below the sternum. The combs are covered, but enough is impressed through the second abdominal plate to note that they are narrow, small, with small fulcra and many, very small, narrow teeth, which are estimated at about 45 on each comb. The abdominal plates are distinctly holosternous and not lobosternous, as I reported in 1969 (p. 185), having been misled by the presence of two distinct halves of an abdominal plate. There is no doubt that this was a fortuitous crack, making it appear as if it were a lo- bostern (or meristostern) plate. A piece of the ends of a holosternous abdominal plate was extracted; the adhering rock was dissolved in acid with the result that the slitlike opening to the gills, such as is found in some species of Wills’ Triassic Mesophonus, was clearly discerned. These pieces were mounted on a slide (M.W. 4). The slit is bordered by a few very small spines, the lip being slightly thickened. The integument on the face and on the doublure is reticulated, in a cell-like netting such as is well-known in nearly all scorpions, including the living ones. On the doublure the definite reticulated pattern grades into what is generally known as intersegmental tissue, al- though this is actually very thin, flexible cuticle and reveals no structure, at least at 200 (see Text-fig. 29C). One of the pieces showing the slitlike opening showed structureless, very thin cuticle that likely rep- resents pieces of the gills (Text-fig. 29C). The tergites are covered with distinct ornamentation in the form of pustules along the axis of the mesosoma, with larger pustules on the seventh tergite. Each tergite is bounded by a row of spines at the base (Text-fig. 29D). The ornamentation was of particular diagnostic value in the identification of this species. Text-figure 29.—Mazonia woodiana Meek and Worthen. Speci- men collected by Mrs. K. Taelle (part and counterpart), FMNH (PE) 39253. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, at Pit 11, Peabody Coal Co., Will-Kankakee County Line, Illinois. See foldout inside front cover for explanation of abbreviations. A. Ventral aspect. The third holosternous abdominal plate is al- most completely telescoped between abdominal plates two and four. B. Dorsal aspect of the specimen, clearly showing seven tergites in the preabdomen, rather than eight as the original authors supposed. C. Enlargement of a corner of the fourth abdominal plate posterior margin, which preserves the gill slit, perhaps a fragment of gill tissue, and intersegmental tissue. The integument on the abdominal plate face and the doublure shows the cellular reticulation characteristic of nearly all scorpions. D. Tergite surface, showing pustulose surface and basal spines. Based on slide M.W. 3. E. Detail of the terminus of the fourth leg. Based on slide M.W. 2. F. End of the fourth leg, dissolved out of matrix by acid. Based on slide M.W. 1. CHE ——$—<—$ = Doublure ______~ Cc (grading into Intersegmental tissueh———> Dorsal f side of P tergite 4 Basal marginal spines__, Intersegmental tissue ee FossIL SCORPIONIDA: KJELLESVIG- WAERING /Setal site 1mm F 83 84 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Measurements (in mm) of the Taelle specimen (FMNH (PE) 39253).— Carapace: Length: 7.26 Width at base: 8.3 Pedipalp: Humerus (P3): UES Brachium (P4): 6.7 Hand length (P5): 4.0 Hand width (P5): sil Free finger (P6): 7.34 Tergite: length width No. 1 1.36 (incomplete) No» 2 1.66 (incomplete) No. 3 1.95 (incomplete) No. 4 2.41 (incomplete) No; 5 2a (incomplete) No. 6 27 (incomplete) No. 7 3.0 (?) (see underside) No. 8 3:3. No. 9 325 No. 10 Se Abdominal greatest plates: length width No. 2 Del. 7.4 No. 3 2.92 7.6 No. 4 2.92 7.6 Tergite No. 7 3.8 anterior: 6.9 (underside) posterior: 4.9 Mazonia wardingleyi (Woodward, 1907) Text-figure 30 1907. Eoscorpius (Mazonia) wardingleyi Woodward, p. 543, fig. 3. 1953. Alloscorpius wardingleyi (Woodward). Petrunkevitch, p. 29, figs. 35, 133. 1969. Mazonia wardingleyi (Woodward). Kjellesvig-Waering, p. 188. Type information. —Holotype of Eoscorpius (Ma- zonia) wardingleyi Woodward, Carboniferous, Middle Coal Measures, 135'+ above Arley Mine Coal Seam, Sparth Bottoms, Lancashire, England; registered as BM(NH) In.18561 in the collections of the British Mu- seum (Natural History). Presented to the Museum by W. H. Sutcliffe, 1906. Several important points escaped the descriptions of Woodward (1907, p. 543) and Petrunkevitch (1953, p. 29). The anterior of the carapace is produced into a pointed, triangular area, much as in Mazonia woodi- ana Meek and Worthen. The chelicerae are unusually large, armed with large teeth, and other than noting their presence, preservation does not warrant further description. The pedipalp preserved shows the usual trochanter, followed by an unusually thick femur and tibia. The termination of the first leg on the right side is described for the first time. This ends in two slightly curved claws with spines at the underside; the post- tarsus is also produced into a sharp spine. The basi- tarsal spurs are of two types: a flat, setaceous spur that is serrated at the end, and a flat, short, pointed spur. Petrunkevitch states that, ““There is nothing to recall Mazonia” (1953, p. 29), a statement with which this writer does not agree. Both Mazonia woodiana and M. wardingleyi were well known heretofore only from the dorsal side. Good casts of the holotype of M. woodiana in the British Museum collections were compared di- rectly with the holotype specimen of /. wardingleyi and nothing could be found on the basis of the dorsal side that would place this scorpion in any genus other than Mazonia. The writer has also studied the holotype specimen of M. woodiana. The holotype of M. war- dingleyi, with its inflated aspect, probably denotes a female—that is, if sex criteria are the same for fossils as for living scorpions, and there is no reason to think otherwise. The cephalic shield shows distinct coarse, short, transverse ridges that should serve as a characteristic of this species. These ridges develop into indistinct large granules toward the base of the cephalic shield. The intersegmental tissue between the tergites 1s par- ticularly well preserved. The anterior transverse ridges are also very well developed, recalling again the strong, massive construction of Mazonia woodiana (see Kjel- lesvig-Waering, 1969, p. 171). Family CENTROMACHIDAE Petrunkevitch, 1953 (emend.) Mesophonoidea with small, subtriangular or sub- pyriform sternum. Type genus. —Centromachus Thorell and Lind- strom, 1885. Remarks. — Petrunkevitch (1953, p. 17) restudied the holotype of Centromachus euglyptus (Peach) with the result that the coxosternal region was described as hav- ing a pentagonal sternum with only the first pair of coxae having maxillary lobes, whereas coxae II abutted against the sternum and III and IV against the opercula. This led Petrunkevitch to describe the family Centro- machidae based on the peculiar arrangement of the coxae, and in 1955 (p. 78) he used the same characters to establish the superfamily Centromachoidea. Storm- er (1963, pp. 79-83) made a detailed study of the same specimen and disagreed with the work done by Pe- trunkevitch. The writer also studied the specimen in 1971 and can verify Stormer’s conclusions, namely: The sternum was found to be small and subpyriform, with only the third pair of coxae abutting against it; these coxae had short maxillary lobes that met in front ofthe sternum. The fourth pair of coxae abutted against Text-figure 30.—Mazonia wardingleyi (Woodward). Holotype, BM(NH) In.18561. From the Carboniferous, Middle Coal Measures, 135'+ above Arley Mine Coal Seam, Sparth Bottoms, Lancashire, England. See foldout inside front cover for explanation of abbrevi- ations. A. Part. B. Counterpart. FossiL SCORPIONIDA: KJELLESVIG- WAERING ae ( counterpart B 16S 85 86 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the operculum. The second pair of coxae meet in front of the sternum and have well-developed maxillary lobes. The first pair of coxae, not sufficiently well-pre- served to show the maxillary lobes, was found to be in normal position. It seems safe to surmise that the first pair of coxae had developed maxillary lobes, since all scorpions with maxillary lobes on the second pair of coxae have the same developed on the first pair. Indeed, maxillary lobes were always first developed on the first pair in the evolution of the scorpions (see Eoscorpius carbonarius, Text-fig. 75). A restoration, based mainly on Stormer (1963, pp. 79-82) with slight modification, is given as Text-figure 31B. Genus CENTROMACHUS Thorell and Lindstrém, 1885 Centromachidae with elongate, subovoid opercular plates. Pectines large with many teeth (about 70 (Stormer, 1963, p. 82)) anchored to a narrow trape- zoidal pectinal plate with a lanceolate median organ. Type species. —Eoscorpius euglyptus Peach, 1883. Geological range. —Lower Carboniferous. Centromachus euglyptus (Peach, 1883) Text-figures 31B, 110J, 113A1 1883. Eoscorpius euglyptus Peach, pp. 402-424, pl. 22, figs. 3-3d. 1885. Centromachus euglyptus (Peach). Thorell and Lindstrém, pp. Dl. 1911. Centromachus euglyptus (Peach). Pocock, pp. 19-20, text-fig. 4. 1913. Centromachus euglyptus (Peach). Petrunkevitch, p. 33. 1949. Centromachus euglyptus (Peach). Petrunkevitch (partim), p. 141, fig. 173; not fig. 136. 1953. Centromachus euglyptus (Peach). Petrunkevitch, pp. 17-18. 1955. Centromachus euglyptus (Peach). Petrunkevitch, p. 78, fig. 43(3). 1962. Centromachus euglyptus (Peach). Dubinin, p. 428, fig. 1242. 1963. Centromachus euglyptus (Peach). Stormer, pp. 79-82, text- fig. 32. Reference is made to Stormer (1963, pp. 79-82) who gives a detailed and excellent description of the ho- lotype, GSE 2142, from the Lower Carboniferous of Langholm, Dumfriesshire, Scotland. I noted, however, the imprint of the middle of the anterior part of the carapace, and this was distinctly bluntly glossate, and also that the only abdominal plate present clearly is holosternous. My identification of part of the mor- phologies differs from that given, and inasmuch as these determinations have considerable bearing on tax- onomy, they are here listed. The area from the oper- cular plates to the first gill-bearing abdominal plate is preserved undisturbed. In text-figure 32, Stormer shows a wide, divided, trapezoidal plate with a median organ separating the two plates. This plate is identified as the “‘basal plate of the combs” or the pectinal plate. I be- lieve instead that this is the reduced first plate (of six abdominal plates). This plate occurs anterior to the pectines, and can be seen in other scorpions, for ex- ample, the well-preserved underside of Eochaerilus. Family HELOSCORPIONIDAE, new family Mesophonoidea with first and second maxillary lobes narrow, reaching to oral opening, with first pair squeezed away from the midline, leaving only the sec- ond pair meeting at the midline. Third pair abuts pen- tagonal sternum and fourth pair, the opercula. Cara- pace elongate, subquadrate and divided by a deep V-shaped sulcus. Large median eyes located on an an- teriorly-placed ocellar mound. No lateral compound eyes. Type genus. — Heloscorpio, n. gen. Remarks. —The only family in the Holosternina that bears any resemblance to the Heloscorpionidae is the Buthiscorpiidae, but the latter has median eyes located well within the anterior margin, lacks the deep V-shaped sulcus, and more important, has lateral ““compound” eyes, as against the Heloscorpionidae, which has an- teriorly-located median eyes, a deep and large V-shaped sulcus, and no ““compound” eyes. Genus HELOSCORPIO, new genus Heloscorpionidae with median eyes located forward on carapace, no lateral ““compound” eyes; prominent median carina throughout the preabdomen with a lin- ear depression on each side of the carina resulting in trilobation of the preabdomen. Derivatio nominis. —helo (Gr.) = marsh-dwelling. Type species. —Geralinura? sutcliffei Woodward, 1907. Geological range. —Carboniferous. Remarks. —The great central carina and rough tri- lobation are sufficient to distinguish this genus from any other scorpion, and were the specimen more fully preserved, it would almost surely have to be placed in anew superfamily. There is a possibility that this scor- pion is a Meristosternina as the abdominal plates do Text-figure 31.— Details of the organization of the coxosternal and pectinal regions. See foldout inside front cover for explanation of abbreviations. A. Opsieobuthus pottsvillensis (Moore). From the Upper Carbon- iferous (Pennsylvanian) of Indiana. Holotype, FMNH 37984. Show- ing the coxosternal area and right pectine. B. Centromachus eu- glyptus (Peach). Holotype, GSE 2142. From the Lower Carboniferous (lower Viséan), River Esk, Scotland. After Stormer, 1963, text-figure 32, with some changes in morphologic interpretation. C, D. Tityus trinitatus Pocock. From Mayaro, Trinidad. Showing sexual differ- ences in coxosternal organization. C. Male. FSU Vial 778, E. N. Kjellesvig-Waering collection. Note presence of prepectinal plate (ppp). D. Female. FSU Vial 779, E. N. Kjellesvig-Waering collection. FossIL SCORPIONIDA: KJELLESVIG-WAERING 87 Cll right Pc CIV removed 88 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 show a possible separation, which here has been in- terpreted as an impression of the dorsal central carina. Heloscorpio sutcliffei (Woodward, 1907) Text-figures 32, 1101, 113A4 1907. Geralinura? sutcliffei Woodward, p. 545, fig. 4. 1953. Trigonoscorpio? sutcliffei (Woodward). Petrunkevitch, pp. 31- 32, figs. 29, 121 (Trigonoscorpio sutcliffei in text, p. 31). The holotype was first considered a Thelyphonida by Woodward (1907, p. 545), but was correctly iden- tified by Petrunkevitch (1953, p. 31) as a scorpion and was referred to the genus 7rigonoscorpio Petrunke- vitch, 1913, an American genus of highly doubtful standing, as it was based on dubious characters (it is known here as a male specimen of Eoscorpius carbo- narius Meek and Worthen). The reference to 77rigono- scorpio was, however, considered with question by Pe- trunkevitch. The figures made by Petrunkevitch, as well as the main parts of his description, are highly erroneous and due entirely to misinterpretation of preservational factors, and failure to develop the spec- imen. The holotype is composed of two parts revealing only the dorsal side, with only a triangular piece of the carapace exposed. The coxosternal region 1s, however, impressed on the part. The coxae abutted this trian- gular piece, which Petrunkevitch mistakenly accepted for the entire carapace, resulting in his drawing of an improbable scorpion with a very elongated, triangular, narrow and small carapace (1953, fig. 29). Actually, this ‘‘carapace” was the triangular corner of the right side of the carapace, from the diagonal formed by the median sulcus, surrounding the median eyes, to the genal angles of the carapace —less than one-third of the actual carapace. Development revealed that this was the case, for it was obvious that such a small carapace simply could not support the great pair of pedipalps and coxae, not to mention the lack of room for the chelicerae. A complete pedipalp was also uncovered with the development of this side. A quite accurate description of this scorpion is now possible. The carapace is distorted, but shows a triangular depression or sulcus surrounding the median ocellar node. Only a small part of the latter is preserved, but the eyes are impressed through the carapace, revealing that they are round and large. These eyes can be per- fectly seen on a rubber cast. A reconstruction (see Text- fig. 32B) shows an elongate rectangular carapace with a deep triangular sulcus and median eyes located on a mound at the anterior. There are no traces of lateral eyes. The coxosternal region is present so that it is possible to reconstruct the entire region (see Text-fig. 32D). The sternum is clearly impressed through the carapace and can be seen under alcohol. It is pentag- onal, large, and has only the third pair of coxae abutting it. Petrunkevitch is incorrect in assuming that the fourth pair abuts it. The fourth pair is faintly visible below the third pair, and can be seen abutting the genital plate, which can be outlined faintly under light at a low angle. The second pair of coxae, however, can clearly be seen to curve upward from the anterior of the sternum. These are the maxillary lobes of the sec- ond pair of coxae, which means that the second pair met at midsection and that the first pair abutted the second pair. A reconstruction of this important area is therefore possible (see Text-fig. 32D) without danger of error. The pedipalp shows a short tibia, which is orna- mented with a spine anteriorly. The hand is large, with two narrow ridges on the face of the hand on the dorsal side. The tergites are noteworthy for the conspicuous cen- tral carina. No other ornamentation was noted, in fact the bald aspect of the dorsal side is a feature that in- stantly attracts attention, as few scorpions are like that. Small scales are, however, noted at the base of the carapace and along the posterior of the first tergite and it must be assumed that the other tergites also had this single row of scales along the basal edge. The cauda is represented by fragments of two tergites. The venter of the seventh tergite is ornamented with at least a central carina composed of coarse pustules or granules. The narrow linear depression running on each side of the central carina, midway between the carina and the lateral edges of the preabdomen, means that this scorpion had a rough trilobation of the preabdominal tergites as is present in some eurypterids such as the genera Mixopterus and Megalograptus (see Stormer, 1955, pp. 34-36). The abdominal plates, at least four of which are well- preserved, reveal that they are holosternous, with a well-developed anterior transverse ridge. It is possible that a central carina is present, although it is believed that the apparent carina is an impression formed by the dorsal carina. Type information. —Carboniferous, Middle Coal Measures, about 135 feet above the Royley coal, half a mile south of Rochdale Town Hall, Sparth Bottoms, Rochdale, Lancashire, England, and deposited in the Text-figure 32.—Heloscorpio sutcliffei (Woodward). Holotype, BM(NH) In.18564, part and counterpart. From the Carboniferous, Middle Coal Measures, Sparth Bottoms, Rochdale, Lancashire, En- gland. See foldout inside front cover for explanation of abbrevia- tions. A. Dorsal aspect of the holotype. B. Reconstruction of the pro- somal carapace. C. Ventral counterpart. Note that the abdominal plates might be considered meristosternous. D. Reconstruction of coxosternal area. 90 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 collections of the British Museum (Natural History) where it is registered as BM(NH) In.18564. Remarks. —There is no scorpion that may be con- fused with this species. The central carina and trilo- bation of the preabdomen are sufficient for identifi- cation and comparisons are thereby superfluous. Family LIASSOSCORPIONIDAE, new family Mesophonoidea with carapace having nearly cen- trally-located median eyes; lateral schizochroal eyes, bulging, elliptical and with narrow rim; gill openings without fringe of large spines, but covered with very small, blunt, spinelike pustules. Type genus. — Liassoscorpionides Bode, 1951. Remarks. —Details of the above morphologies are given under the new description of L. schmidti Bode, which follows. It differs from the rest of the Meso- phonoidea in the nearly centrally-located median eyes compared with the anteriorly-located eyes of all the Mesophonidae. Perhaps it is not too speculative to suggest that the Jurassic genus Liassoscorpionides is directly descended from forms like Mesophonus with its extremely forwardly-located eyes. The type of large glossate process of the carapace, and the large gill open- ings also point to that development. Genus LIASSOSCORPIONIDES, Bode, 1951 (emend.) Liassoscorpionidae, devoid of ornamentation and with very large chelicerae. Type species. —Liassoscorpionides schmidti Bode, 1951. Geological range. —Jurassic. Remarks. —The large chelicerae, composed of four distinct joints, have a development of the cheliceral flange suture or slot which is also present in the British Triassic Stenoscorpio gracilis (Wills) (fide Wills, 1947, pp. 76-77). I do not recall seeing the suture or slot, which always occurs on the dorsal side of the second joint of the chelicera, in any other scorpion. Perhaps the comparatively great size of the chelicerae of the Mesozoic scorpions required a suture or slot for greater flexibility, or muscle attachment. Liassoscorpionides schmidti Bode, 1951 Text-figure 33 1951. Liassoscorpionides schmidti Bode, pp. 58-65, fig. 1. 1953. Liassoscorpionides schmidti Bode. Petrunkevitch, pp. 38-39 (Scorpiones incertae sedis). 1955. Liassoscorpionides schmidti Bode. Petrunkevitch, p. 79 (Gen- era incertae sedis). 1962. Liassoscorpionides schmidti Bode. Dubinin, p. 433, figs. 1259A-D (Scorpionida incertae sedis). Bode’s (1951, pp. 58-65) original description is in- correct in many major points, and this was followed by Petrunkevitch’s (1953, pp. 38-39) analysis of Bode’s description and figure, which was equally erroneous in assuming a short tail, leading to his incorrect deter- mination that the scorpion was possibly “related to either Palaeopisthacanthus or Typhlopisthacanthus’’. No three scorpions could differ more, notwithstanding the fact that all three had fictitious short tails. Palae- opisthacanthus is a pulmonate orthostern (a Neoscor- pionina), whereas ‘‘7yphlopisthacanthus”’ is a junior synonym of the gilled lobostern Eoscorpius (Bran- chioscorpionina). Liassoscorpionides schmidti Bode 1s clearly a Holosternina, belonging to the superfamily Mesophonoidea, for which a new family, the Liasso- scorpionidae, has been erected. The specimen is not by any means so poorly preserved as stated (Bode, 1951, p. 58) and all parts, mainly dorsal, are clearly and incontrovertibly preserved. This becomes one of the most interesting scorpions known, as it is a further development of the “Meso- phonus fauna” of the Triassic and gives us an insight, although limited, into the Jurassic forms, at least the aquatic Branchioscorpionina. The specimen is asso- ciated with marine fossils on the same level. The lime- stone is slightly oolitic. Studies of this scorpion were made both in the dry state and under alcohol, revealing many morphological details. The original colors were preserved and are shiny dark brown throughout. Bode (1951, p. 68, fig. 1) shows morphological fea- tures that are not present, and were erroneously de- termined. These include: 1) The alimentary canal: The right edge of the tergites, in part, was mistaken for the canal, and I confess that the upper part shown is inexpl- icable. It is not present. 2) According to Bode, the pulmonate type of stigmata occur on the last three preabdominal tergites. This is completely erroneous as the scorpion retains abdominal plates and not sternites, Text-figure 33.—Liassoscorpionides schmidti Bode. Holotype, GUM 525-1, part and counterpart. From the Jurassic (Lias epsilon), marl pit at Hondelage b. Braunschweig, West Germany. See foldout inside front cover for explanation of abbreviations. A. One side of the two-piece holotype. The organism was squashed during fossilization so as to expose both dorsal and ventral features on the same surface of the specimen. In Bode’s description (1951), the right edge of the posterior tergites (T4-7) of this complex pres- ervation were mistaken for evidence of the gut, and the presence of both tergites and abdominal plates (AP) in the preservation was not realized. He extended the “gut” forward without any evidence for the extension being present on the specimen. He also erroneously identified stigmata, although gill openings are present at the base of each abdominal plate (AP3-5). B. A reconstruction of the holotype showing the slim mesosoma (rather than broad, as Bode thought), and the extraordinary prosomal carapace and gigantic chelicerae, which are faithfully drawn. C. Showing the facetted left lateral eye of the holotype, with some of the lenses preserved. D. Detail of the fourth abdominal plate (AP4) showing the gill opening on the right side of the specimen. FossiL SCORPIONIDA: KJELLESVIG-WAERING 92 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 and gill openings (not stigmata!) are perfectly devel- oped at the base of each abdominal plate. It is sufficient to note that stigmata were also placed on the seventh tergite (ventral), an impossibility in any scorpion, fossil or living. 3) Only two tergites of the cauda are pre- served. Obviously these two tergites have been crushed laterally, but are not narrow, as attested by the basal attachment to the preabdomen (No. 7 tergite), which is normal and not as previously thought (Bode, 1951, p. 65; Petrunkevitch, 1953, p. 38; 1955, p. 79). With these changes, and others noted, a new de- scription of the holotype is possible, which assigns the genus to the Mesophonoidea. The holotype is preserved so that most of the dorsal side is present with the interior of the venter of the right side exposed. The carapace is known for the first time in detail. It is subquadrate, wider than long, with a large anteriorly- pointed glossate process and with a raised basal mar- ginal rim that is wide at midsection and tapers to either side. Located on the central part of each half are raised, elliptical, cephalic cheeks. The median eyes are located on the prominent median eye node. These eyes appear elliptical and are only faintly preserved, but entirely visible under alcohol. In life, the eyes were probably round and became elliptical by compression. At the anterolateral corners are the many-facetted, bulbous, elliptical eyes, which are bordered by a slight ridge surrounding the eyes. This margin is seen on the left eye (see Text-figs. 33A, C) and the eye proper is com- posed of about 200 round, certainly biconvex, lenses. Unfortunately the coxosternal area is not known, occurring in an area that consists of a mish-mash of cuticle that is impossible, at least for me, to unravel with any degree of certainty. Only the second leg of the left side is preserved, and this consists of the tro- chanter, femur and patella (see Text-fig. 33A). The last five tergites of the preabdomen are pre- served and show a gradual increase in size posteriorly. The last (T7) is complete and reveals a large, wide, attachment area for the cauda. Two carinae are present. What remains of the cauda consists only of two seg- ments out of the five, but it should not be mistaken for a thin or short tail. It is merely distorted. The chelicerae are noteworthy both for their good preservation and for their enormity. They are, in fact, longer than the carapace, and clearly consist of four joints, as in all Holosternina. The first joint, or coxa, is rounded, distinctly calathiform, and truncated at the anterior. The second joint consists of a band-like seg- ment, sutured at the dorsal side, although this suture- like structure may represent the junction of a slot. In either case it is interesting to see that some of the British Triassic forms also have this development. The third and fourth joints are like other scorpions, with teeth developed on the chelae, but not good enough for description (see Text-figs. 33A, B). The venter reveals the third to fifth abdominal plates of a holosternous scorpion from the inside. The gill openings of each abdominal plate are perfectly pre- served. These consist of large slits at the junction be- tween the abdominal plate and the doublure; the edges of both, and the inside, are covered with very small, triangular mucrones. No spines were developed (see Text-fig. 33D). The ventral side of the seventh tergite is devoid of any carinae in the preserved part. The first two tergites of the cauda are preserved and, other than to note that two superior crests surmount each tergite, no further description is possible. The tergites have obviously been distorted laterally, giving a superficial and incorrect aspect of being thin rather than of normal thickness (see Text-figs. 33A, B). Measurements (in mm) of GUM 525-1.— Estimated total length restored (chelicerae to end of cauda): 18.0 Length of carapace at midsection: 3.1 Lateral eye length: 0.8 Lateral eye width: 0.3 Type information. —Jurassic: Upper Lias, Lias ep- silon, Mergelgrube, Hondelage b. Braunschweig, Ger- many. It is registered as No. 525-1 in the Geological- Paleontological Institute and Museum, Georg-August University, G6ttingen, Germany [abbreviated herein as GUM]. Remarks. —This species, being the last known record of the Branchioscorpionina, and in particular, the Ho- losternina, is important in showing the gradual back- ward displacement of the median eyes to a nearly cen- tral position such as is present in modern scorpions. The holosterns very likely never left the water for ex- tended periods as the gills had to be bathed with water. On the same surface with the holotype are a number of fragments of scorpion integument and joints. None is identifiable, but they indicate that if the exact ho- rizon could be located again, further specimens of scor- pions could be obtained. The importance of further finds is assuredly of great interest, and I urge any col- lector to try to determine the exact horizon. I know of few stratigraphic levels which could furnish anything as interesting as these Jurassic scorpions. The locality promises much and could surpass the British scorpion localities in interest, uniqueness, and the possibility of finding further important evolutionary data. These specimens will be found to be composed of original cuticle and therefore the Holm—Wills technique (see Wills, 1910, p. 303; Wills, 1947, pp. 136-137) is pos- sible, thus freeing the scorpionid parts from the matrix. FossiL SCORPIONIDA: Complete specimens, or those showing the important coxosternal areas, should not be subjected to dissolu- tion in acids, as they come apart, in many cases making reconstruction quite difficult. Nevertheless, slabs with obvious scorpion parts can be dissolved in the hope of freeing other and better specimens, as Wills did with specimens from the British Triassic (1947). Family PHOXISCORPIONIDAE, new family Mesophonoidea with elongate hexagonal sternum. Carapace without lateral compound eyes. Enormously developed pectinal plate without median organ. Type genus. — Phoxiscorpio, n. gen. Remarks. —The only family with which the Phoxi- scorpionidae may be confused is the Centromachidae. At one time (not published) I had included both in an emended superfamily Centromachoidea, but the pres- ence of maxillary lobes developed on the third pair of coxae not only abutting the sternum, but meeting in front of the sternum, is a major development and taxo- nomic consideration in the Centromachoidea, which the Phoxiscorpionidae does not show. Genus PHOXISCORPIO, new genus Phoxiscorpionidae with subquadrate carapace, glos- sate on mid-anterior margin, and large anteriorly-lo- cated median eyes. Pectines large with unusually nar- row and small teeth. Derivatio nominis. —phoxos (Gr.) = pointed, peaked. Type species. —Phoxiscorpio peachi, n. gen., n. sp. Geological range.—Lower Carboniferous (middle Viséan). Phoxiscorpio peachi, new species Text-figures 34, 110E A poorly-preserved specimen approximately 60 mm in length when alive. Most of the underside is revealed and, dorsally, patches of the carapace, some tergites, and all of the basal three joints of the right pedipalp. Although most of the dorsal side of the carapace was not preserved, much can be interpreted from the parts present (see Text-fig. 34A). The eye node occupies a frontal median position, is elliptical, with large ellip- tical eye sockets, bounded above by rather heavy or- bital ridges. The median eyes are elliptical, but prob- ably were round in life, as they are now in a compressed state. The large median eyes indicate that there should be no lateral compound eyes, nor does the right lateral angle of the carapace reveal any. It doesn’t seem spec- ulative to consider the carapace to be subquadrate, rounded at the anterolateral angles, produced to a me- dian point anteriorly, and truncated at the base, as in most scorpions. A restoration is given in Text-figure 34B. KJELLESVIG-WAERING 93 The pedipalp (Text-fig. 34A) is preserved in dorsal aspect and undoubtedly is a stout appendage. Only the first three joints are preserved. The trochanter (P2) is noteworthy for being very large, and seems to be divided into two joints. This is a common condition in many living scorpions, and I have always believed that the pedipalp trochanter was composed of two ankylosed joints. This may be the case here, but it is difficult to ascertain whether it is two separate joints or one. A ridge of granules occurs across the uppermost part of the “‘second”’ trochanter. The femur (P3), also seen in dorsal aspect, reveals a line of setaceous granules at the back and at least one row of setaceous granules along the central part. The coxosternal region shows an elongate hexagonal cil CIV Text-figure 34.—Phoxiscorpio peachi, n. gen., n. sp. Holotype, GSE 2140. From the Lower Caroniferous (middle Viséan), Dalmeny railway cutting, near Edinburgh, Scotland. See foldout inside front cover for explanation of abbreviations. A. Carapace, entire right chelicera and right pedipalp seen from the dorsal side; the rest seen from the inside of the ventral side. Note: the pectinal plate has not been described in the text. B. Re- stored prosomal carapace. 94 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 sternum and two long, rather elliptical, opercular plates that fit at the base of the sternum. The fourth pair of coxae abuts the opercular plate whereas the third pair abuts the sternum. The first two pairs of coxae meet at midsection and there are indications of a slight max- illary lobe on the second pair. This would indicate that both the first and second pairs of coxae have short maxillary lobes. Both almost certainly meet at the mid- line. Mainly the ventral side of the postabdomen is pre- served, and it is important to record that the abdominal plates are holosternous. All abdominal plates have a narrow transverse ridge along the anterior margin. The pectines were covered by the abdominal plates and are seen from the dorsal side of the plates, but the teeth of the pectines are impressed through and are very narrow and small, like those shown by Stormer (1963, p. 60), for the holotype of Centromachus eu- glyptus (Peach). The pectinal plate, devoid of a median organ, is enormously developed. Type information. —It is preserved in dark gray-black shale, along with ostracodes, in the Lower Carbonif- erous (middle Viséan) at Dalmeny railway cutting near Edinburgh, Scotland, and is registered as GSE 2140 in the collections of the Institute of Geological Sciences at Edinburgh, Scotland. Derivatio nominis.— Named in honor of B. N. Peach, whose contributions to Scottish paleontology, partic- ularly his early fossil scorpion work, need no further comment here. Remarks. —At first I had mistaken this specimen for Centromachus euglyptus (Peach), but the differences became apparent, particularly in the shape of the ster- num. The sternum in Phoxiscorpio peachi is hexagonal rather than subtriangular, and the coxae of the last three pairs of legs are distinctly longer. More important is the presence of maxillary lobes on the third pair of coxae and the presence of the trapezoidal pectinal plate with median organ in Centromachus euglyptus (Peach), in contrast to the lack of maxillary lobes on the third pair of coxae, and the extremely large pectinal plate without a trace of a median organ in the new species. The specimen had been badly developed by chisels and, although what is left is fragmentary, much can be revealed in the dry state, by the use of various types of light, by immersion in alcohol, and from a latex cast. Family WILLSISCORPIONIDAE, new family Holosternina, possibly referable to the superfamily Mesophonoidea, without lateral facetted eyes, and with anteriorly-located median eyes in line with or anterior to the anterior margin of the carapace. Gill chamber opening slitlike, located on the doublure and opening internally and dorsally to the abdominal plate. Gill lamella bounded by small spinelets. Type genus. — Willsiscorpio, n. gen. Genus WILLSISCORPIO, new genus Willsiscorpionidae with quadrate carapace with large median eyes placed on a large anterior, rounded glos- sate process. Elevated dorsal, shieldlike, undivided ce- phalic area is well-delineated by the ornamentation. Derivatio nominis. —Named in honor of Leonard J. Wills, who discovered and so ably described the British Triassic scorpions. Type species. —Mesophonus bromsgroviensis Wills, 1910. Geological range. — Triassic. Remarks. —Although included in the genus Meso- phonus by Wills (1910, 1947), the new genus differs greatly in the lack of lateral facetted eyes and in the undivided cephalic shield of the carapace, which in Mesophonus is divided into two elevated cheeks sep- arated by a median sulcus. The presence or absence of lateral facetted eyes, as stated above, is not a secondary sex character as proposed by Wills (1947, pp. 20-21), but a fundamental character of at least family level. Willsiscorpio bromsgroviensis (Wills, 1910) 1910. Mesophonus bromsgroviensis Wills, pp. 313-314, pl. xxiii, figs: 15 2245.9;5°7; ple xxvi, figs 1: 1910. Mesophonus gracilis Wills (partim), p. 318, pl. xxv, fig. 10. 1910. Mesophonus sp. Wills, pl. xxii, figs. 6, 10; pl. xxvi, figs. 3, 10. 1947. Mesophonus bromsgroviensis Wills. Wills (partim), pp. 108- 115, pl. I, figs. 1-6; pl. III, fig. 1; pl. IV, figs. 1, 2; pl. VI, figs. 1, 2, 4-7; pl. XI, figs. 9, 10; pl. XII, figs. 12-14; text-figs. 3A, 5A, B, 7, 10, 46A-C, E, F; not Mesophonus bromsgroviensis ? Wills, pp. 35, 39, 110; pl. VI, fig. 9; text-fig. 14 (specimen 0163). The Wills coll. specimens which are included under this species are: SM 158 (lectotype), 9, 13, 18, 23, 35, 47, 074, 080, 098, 101, 195, 208, 212, 213 (fide Wills, 1947, p. 108), 215, 220, 0235, 270, 273, 275, 0280, 295. Type information. —This fine species is known from the Triassic Lower Keuper Sandstone Series, Broms- grove, near Birmingham, and from a boring at Messrs. Southalls (B’ham) Ltd. works at Charford Mills, Bir- mingham, England. Superfamily EOCTONOIDEA, new superfamily Holosternina with the first two pairs of coxae in front of the sternum and with maxillary lobes developed; first pair extends further anteriorly than the second pair; third and fourth pairs abut the sternum. Type family. —Eoctonidae, new family. Remarks.—This superfamily is an intermediate group between the Lobosternina and the Holosternina. FossiL SCORPIONIDA: KJELLESVIG-WAERING 95 It is closer to the Holosternina, whereas the superfam- ily Pseudobuthiscorpioidea is closer to the Loboster- nina. In my opinion, these two superfamilies connected the two suborders, namely through the Eoctonoidea into the Pseudobuthiscorpioidea. The development of the coxosternal areas bears this out. These interme- diates are therefore among the most important and interesting of the fossil scorpions. Text-figure 35.—Eoctonus miniatus Petrunkevitch. Holotype, YPM 131. Upper Carboniferous (Pennsylvanian), presumably from the Carbondale Formation, Lower Francis Creek Shale, Mazon Creek area of Grundy Co., IL. This specimen is of particular interest since it proves that the gills were attached to the body wall, rather than to the abdominal plates, just as the lungs are in living scorpions. It is a male as indicated by the extra-long twelfth tergite (T12). See foldout inside front cover for explanation of abbreviations. A. Dorsal side. B. Counterpart, showing ventral features. Note the lobate abdominal plates which, although posteriorly lobate, appear at be at a grade intermediate between holosternous and lobosternous condition. 96 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Family EOCTONIDAE, new family Eoctonoidea with intramarginal, small, lateral schi- zochroal eyes and with median ocelli placed in the center of the anterior half of the carapace. Type genus. —Eoctonus Petrunkevitch, 1913. Remarks. —See remarks under description of Eoc- tonus miniatus Petrunkevitch (below). Genus EOCTONUS Petrunkevitch, 1913 Eoctonidae with elongate subrectangular lanceolate carapace with indistinct lateral, longitudinal ridges. Type species. —Eoctonus miniatus Petrunkevitch, 1913. Geological range. —Upper Pennsylvanian. Eoctonus miniatus Petrunkevitch, 1913 Plate 4; Text-figures 35, 37-39, 110G 1913. Eoctonus miniatus Petrunkevitch (partim), pp. 51-52, pl. 3, fig. 14, text-fig. 15; not p. 52, pl. 3, fig. 15 (paratype No. 132) (=Paraisobuthus virginiae, n. gen., n. sp.). 1949. Eoctonus miniatus Petrunkevitch. Petrunkevitch, p. 137. 1953. Eoctonus miniatus Petrunkevitch. Petrunkevitch, p. 16, figs. 40-43. 1955. Eoctonus miniatus Petrunkevitch. Petrunkevitch, p. 73, fig. 41(2). 1962. Eoctonus miniatus Petrunkevitch. Dubinin, p. 428, fig. 1229. The holotype, YPM 131, is preserved in a nodule consisting of two halves showing most of the dorsal and ventral sides. The part showing the coxosternal region is not preserved, nor are most of the appendages. The cephalothorax is very elongate. The genal corners form right angles and the anterolateral margins are rounded. The anterior margin is also rounded. In gen- eral aspect it is lanceolate. The median ocelli are prominent, round, and about one-third of the length from the anterior edge, or lo- cated midway in the anterior half of the carapace, and with strong interocular ridges. The lateral eyes are schizochroal, intramarginal and on the anterolateral margins. Although individual fac- ets cannot be described exactly as to shape (they are rounded in another specimen), approximately eight of these facets or individual eyes occur in a roughly re- niform cluster or aggregate. The long axis of the kidney- shaped lateral eye is obliquely located on the carapace. The individual eyes are much smaller than the simple lateral ocelli, such as occur in modern scorpions or in the terrestrial Carboniferous scorpion Palaeopisth- acanthus schucherti Petrunkevitch. The only ornamentation on the carapace consists of faint traces of two longitudinal linear ridges running from the lateral eyes to the base. A definite wide mar- ginal and basal rim occurs. The entire preabdomen is preserved, although the last segment has been partly turned over. Each tergite had been pulled apart at buri- al to reveal a considerable expanse of intersegmental skin, indicating great flexibility in life. All the tergites are progressively longer posteriorly, and each is bound- ed anteriorly by a well-developed marginal ridge. There is nO ornamentation on the tergites. The appendages of the prosoma are not well enough preserved, except at their bases, to merit description. The pedipalp, however, is largely preserved. The tibia of the pedipalp is relatively short and stout, and is covered with cuticle that might best be described as very finely rugose, although not punctate, on the dorsal side. The hand of the palp is relatively wide and round- ed. Enough of the free finger was preserved to show that it has a well-developed dorsal ridge. There are no denticles preserved on the inner edge of the fingers. The combs are almost obliterated and approxi- mately six teeth are noted on the left side of the nodule. These are of the long, slender type and probably rep- resent only a small part of what actually existed. Fulcra occur in this species as at least one of these plates was noted. By far the most interesting thing about this scorpion is the presence of large, paired, oval areas that are impressed as rounded ridges through the dorsal side of the tergites. These oval areas occur along the trunk of the scorpion in the area where the respiratory organs should be located. The abdominal plates have been displaced to the left side of the specimen and thus these areas would have to be structures that are not con- nected to the plates. It appears that these oval struc- tures are the remains of the gills and that nine of them were preserved in this particular specimen. There can be no doubt that there were five pairs originally. It therefore seems that the gills were, in part at least, rounded or oval structures, almost as large as an in- dividual half or lobe of the holosternous plate. The ventral part of the holotype, which is preserved as an impression of the inner side, clearly reveals these struc- tures as deep, rounded depressions on the abdominal plates, exactly as in the Eurypterida (see Text-fig. 36). In other scorpions, such as Eoscorpius, Palaeobuthus, etc., identical structures have been described where they occur as impressions on the inner side of the lobostern and meristostern abdominal plates. As shown by the holotype of E. miniatus, the structures were not attached to the abdominal plates, but to the body, and floated free from the abdominal plates. The covering abdominal plates have a round pouch or gill chamber to accommodate the large gills. The abdominal plates are holosternous, with a very slight emargination in the middle of the posterior margin, and with well- developed anterior marginal ridges. FossiIL SCORPIONIDA: KJELLESVIG-WAERING 97 The complete postabdomen is preserved and, at the upper part, the tergites have been turned on their sides; therefore it is not possible to determine the number of ridges. On the last two tergites, two well-formed ca- rinae can be noted on the dorsal surface. The vesicle was exposed for the first time (although Petrunkevitch shows a drawing of this, which he undoubtedly took from YPM 132, a small specimen which is described herein as Paraisobuthus virginiae, n. sp.). The vesicle is very long, with the aculeus slightly curved and with a well-formed carina on each side of the telson. Two other indefinite ridges were noted below this but were not of the prominent type found toward the dorsal part of the long vesicle. The only ornamentation occurs in Median macrofolds Macrofolds Hinge Mesoderm Anterior granular skin | | | Gill pouch Lateral Posterior granular skin the form of punctations, which are probably setaceous, along the carinae. The long twelfth tergite indicates, without a doubt, that the holotype is a male. Measurements of this specimen are given in Pe- trunkevitch (1913, p. 51). The vesicle, which was un- known at the time, is approximately 2 mm long. Type information. —Upper Pennsylvanian from the Francis Creek Shale, at Mazon Creek, Illinois, regis- tered as YPM 131 in the Yale Peabody Museum of Natural History. Another specimen (see Pl. 4; Text-fig. 37) consists of an ironstone concretion preserved in two parts, one of which contains the dorsal side and the other the ventral side. Both sides are very well preserved, show- Anterior granular skin | | Medial Gill tract Microfolds | | Squamae on ventral surface of Blattfuss | | | { Text-figure 36.—Tarsopterella scotica (Woodward). An eurypterid from the Devonian of Scotland. This reconstruction of the gill chamber by Charles E. Waterston (1975, fig. 3) is introduced for comparison. This type of gill apparatus is thought to be the same in primitive scorpions of the infraorders Lobosternina and Holosternina. Reproduced by kind permission of C. E. Waterston, and Stefan Bengtson, editor, Fossils and Strata. 98 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 ing excellent detail. The specimen has been studied mainly in a dry state, although some structures were visible under alcohol. It was collected by Ray Ban- dringa from Pit 11 (Essex Fauna), Upper Pennsylva- nian, Francis Creek Shale, on the Will Co.-Kankakee Co. line, Illinois, commonly considered as part of the Mazon Creek area. The specimen is now in the col- lections of the Field Museum of Natural History where it is registered as FMNH (PE) 16498 (part and coun- terpart). The carapace is broadly rounded on the anterior, truncated at the posterior and with the sides converging anteriorly. A prominent marginal rim occurs along the posterior and lateral sides. Two suturelike ridges sep- arate the carapace diagonally and a narrow sulcus is present directly behind the ocellar node. The ocellar W2dor = Imm fa0t }rn Text-figure 37.—Eoctonus miniatus Petrunkevitch. Ray Bandringa collection, FMNH (PE) 16498 (part and counterpart). Upper Carbon- iferous (Pennsylvanian), Carbondale Formation, Francis Creek Shale, Mazon Creek area, Pit 11 (Essex fauna), Will Co._Kankakee Co. line, IL. See foldout inside front cover for explanation of abbreviations. A. Dorsal view from the inside. Note the fine denticulation of the pedipalp fingers. B. Counterpart, showing ventral features seen from the inside and dorsal components from the outside. The abdominal plates are essentially holosternous. C. Restored right chelicera seen from the dorsal side. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 99 node is nearly round and is located in the center of the anterior half of the carapace, much as occurs in living scorpions. The median eyes are round and are separated from each other by prominent ocular ridges. Of particular interest is a verification of the lateral compound eyes. These are located well within the margin of the cara- pace and anterior to the median eyes. These lateral eyes are roughly reniform and approximately eight small rounded eyes occur within the confines of the reniform area. The greatest length is 0.11 mm and the width is about half that. The chelicerae are small, much as in living scorpions, but consist of four joints instead of three. The first two are very short, whereas the third comprises the hand with the immovable chela, which is armed with several large teeth. The free finger, armed with denticles, is curved much like in living scorpions. The preservation of the denticles did not permit a clear definition of their structure or arrangement other than to know that they were present. The pedipalp is a massive structure with a rather long and stout chela, and in all respects resembles that of many living scorpions. The trochanter is triangular in general outline, whereas the femur and tibia are stout with some ridges apparently present. These three joints are covered with numerous setal holes which may have been trichobothria. These trichobothrialike openings occur in distinct rows much as found in some of the living scorpions. The chela has a wide hand and again evidence of ridges is present. The two fingers are very long, the immovable finger being curved slightly backwards, whereas the free finger has a slight forward bend. Def- inite small denticles in the shape of linear structures occur along both edges. The hand also is covered with some setal openings that appear to be in series. The four walking legs are preserved only in part, and must have been highly hirsute. Only one spur, appar- ently the basitarsal spur, was noted, and that occurred on the third leg. However, little can be said about spurs on the legs as preservation did not permit description. The second leg, however, does have two tarsal spurs and one basitarsal spur present. It is quite possible that two basitarsal and two tarsal spurs were present in life on all four pairs of legs. The coxosternal region is of unusual interest. The sternum is very large, pentagonal in shape with an inverted rounded area at the posterior, and the first coxa extends forward with a rounded, elongate max- illary lobe. Maxillary lobes are also present on the sec- ond coxa, which extends almost to the end of the first pair of maxillary lobes. The third and fourth coxae abut against the large pentagonal sternum. The pectines are small, not very wide, but are quite long, making each pectine appear triangular with a rounded anterior edge. They were not very clearly pre- served, but enough was noted to ascertain that rounded sclerites cover the rachis and that the teeth were small and elongate, and about sixteen teeth probably oc- curred on each pectine. The genital operculum lies di- rectly posterior to the wide sternum and is composed of two teardrop plates that fit in the inverted circular part at the base of the sternum. The preabdomen consists of the usual seven tergites dorsally, with each tergite increasing in length poste- riorly to reach the greatest width at the fifth tergite. All tergites are smooth and without any ornamentation. Each of the tergites is bounded anteriorly by a trans- verse ridge. One of the peculiarities about this scor- pion, which was also noted on the holotype, is that a considerable area of intersegmental tissue occurs be- tween each of the tergites. It is thought that this is actual and not a condition of ecdysis, and that this scorpion was particularly elongated along the preab- domen, was quite sinuous in life, and had a highly flexible body. This characteristic is present in living Chactidae and Diplocentridae. The abdominal plates are entirely holosternous; each 1s quite long and bounded along the anterior by a trans- verse ridge. The abdominal plates also do not have any ornamentation whatsoever. The postabdomen was only partly preserved; only the anterior tergites were present. These show that all had numerous ridges, al- though it was not possible to ascertain how these ridges were in life. Nevertheless, their presence has been not- ed toward the lateral parts, and on the dorsal side, at least, there are a few setal openings. No other orna- mentation has been noted. The wide mesosoma indicates that this specimen is a female. Measurements (in mm) of FMNH (PE) 16498.— Prosoma: Width at base: 3.8 Greatest width: 322 Length: 3.4 Median eyes: Distance from anterior margin: 0.8 Distance from posterior margin: 2.0 Distance from lateral margin: 1.5 Diameter: 0.11 Chela length: 4.7 Hand width: 1.1 Hand length: 1.8 Femur length (dorsoventral): 2.8 Tibia length (dorsoventral): 2*5 Free finger length: 3:2 No. of rows of denticles: Present but not possible to determine if more than one row occurs 100 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Measurements of FMNH (PE) 16498 (cont’d.) Pectine: No. of teeth: 16 (estimated from the ones present) Mesosoma and metasoma: length of greatest width tergite of tergite No. 1 0.5 3.6 No. 2 0.6 (covered) No. 3 0.7 4.2 No. 4 1.0 4.6 No. 5 3 5.0 No. 6 1.6 4.4 No. 7 1.9 3.5 (anterior) 1.8 (posterior) No. 8 1.8 1.8 No. 9 1.8 INS) No. 10 2.0 183) Abdominal plate: length width No. 1 (covered) (covered) No. 2 1.4 Sha No. 3 17, 3:25 No. 4 17 383) Total body length (including chelicerae): 27.0 (estimated) Remarks. —My description of the holotype differs greatly from that of Petrunkevitch (1913) and nothing is to be gained by detailing the differences. Originally the species was based on a holotype and paratype, but later (1953, p. 15) Petrunkevitch noted that the para- type was not the same species and relegated it to A/- loscorpius granulosus (Petrunkevitch), which here is considered to be a junior synonym of Eoscorpius car- bonarius Meek and Worthen. Actually, the paratype is the holotype of Paraisobuthus virginiae, n. sp. (see be- low). These two specimens of Eoctonus miniatus Petrun- kevitch fortunately reveal both ventral and dorsal sides so that a reconstruction of the species was possible (Text-fig. 38). Again, fortunately, it appears that these two specimens represent both sexes. This is a very unusual scorpion. The median eyes are rather small and have receded back from the an- terior margin to a position not unlike that of some of the living scorpions (Chaerilus, Calchas). The small lateral compound eyes are noteworthy too, because weet hs, Text-figure 38.—Eoctonus miniatus Petrunkevitch. From the Pennsylvanian of Illinois (Mazon Creek). The last three joints of legs I, HI and IV are inferred; the rest is taken from actual specimens. A. Dorsal reconstruction. B. Ventral reconstruction of a male. FossiIL SCORPIONIDA: KJELLESVIG-WAERING 101 they occur a considerable distance from the anterolat- eral margin of the carapace rather than marginally as in other fossil scorpions. The most important fact revealed by the holotype specimen is that the large elliptical gills are attached to the underside of the body and are free of the ab- dominal plates. This is in contrast to the book-lungs, which in living scorpions develop in the embryo by invagination of a small section of the external face of the sternite. Therefore in living scorpions the lungs are attached to the sternite, and during ecdysis the entire chitinous part of the lung books is cast off, attached to the sternite. A third specimen, FMNH (PE) 32202 in the Field Museum of Natural History, Chicago, consists of one- half of an ironstone concretion collected at the Strip Mines (Braidwood fauna) in Will Co.—Grundy Co., IL, from the lower Francis Creek Shale. The specimen is poorly preserved, but merits illustration of this rare species. The eye node as well as undoubted holoster- nous abdominal plates are preserved. The typical facial grooves are also revealed, as well as a lateral eye which is composed of several rounded lenses, very likely schi- zochroal. The last tergite of the cauda is greatly elon- gated as in the holotype and indicates a male. Overall length is approximately 18 mm (see Text-fig. 39). Family BUTHISCORPIIDAE, new family Eoctonoidea with first and second pairs of coxae having well-developed, narrow, maxillary lobes; sec- ond pair meeting at midline, with first pair of maxillary lobes behind the second pair and extending anteriorly together. Last two pairs abutting against the sternum. Sternum pentagonal, large, longer than broad, with lat- eral margins and base emarginate, anterior obtuse. El- liptical intramarginal schizochroal eyes are present. Type genus. — Buthiscorpius Petrunkevitch, 1953. Remarks. — Buthiscorpius Petrunkevitch, 1953, with Anthracoscorpio buthiformis as type species, represents a genus of holosternous scorpions that Petrunkevitch had included with the Eoscorpiidae in the Treatise of Invertebrate Paleontology (1955). Petrunkevitch had incorrectly diagnosed the coxosternal arrangement be- cause he erroneously used the coxosternal arrangement of a protolobosternous scorpion (BM(NH) In.1555), which is described here as Pseudobuthiscorpius labio- sus, and Wills, in his elaborate description (1960, pp. 277-290) confirmed Petrunkevitch’s interpretation of this area. However, Wills (1960, p. 277) believed that Anthracoscorpio Kusta, Buthiscorpius Petrunkevitch (including all paratypes of B. buthiformis erroneously referred to it by Pocock, 1911) and Eoscorpius Meek and Worthen undoubtedly belonged to the Eoscor- piidae, but he recognized that until the diagnosis of the family Eoscorpiidae was properly defined, the problem could not be settled. Had either Petrunkevitch or Wills recognized that the holotype of Buthiscorpius buthi- formis had lateral schizochroal eyes, it is possible that neither would have included this scorpion with several others that were referred to it. The coxosternal region of Eoscorpius is now known and it differs from Buth- iscorpius, therefore the two cannot belong toa common family nor even in the same superfamily. Eoscorpius is a lobostern with large compound lateral eyes, and with the last pair of coxae abutting against the genital opercula. It is an isobuthoid of the suborder Lobos- ternina. The paratypes of Buthiscorpius buthiformis (Pocock) have been referred to several other genera and species. For example, the important specimen BM(NH) In.1555 is now the holotype of Pseudobuth- iscorpius labiosus, which is also the type of a new family and superfamily. Paratype specimen BM(NH) In.31262 is the holotype of Coseleyscorpio lanceolatus, n. gen., n. sp., and paratype specimen BM(NH) In.22832 is the holotype of Leioscorpio pseudobuthiformis, n. gen., n. sp. An almost identical species of Buthiscorpius was Text-figure 39.—Eoctonus miniatus Petrunkevitch. FMNH (PE) 32202. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, lower Francis Creek Shale, collected at the strip mines (Braidwood fauna) in Will Co.-Grundy Co., IL. This is a complex mold that preserves the inside of some dorsal features and the ex- terior of ventral ones. 102 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 recently found in the Mazon Creek beds of Illinois. This important specimen, which was named Buthi- scorpius lemayi, shows the ventral side and furnishes the solution to many of the problems concerning B. buthiformis and others that had been referred to it. Buthiscorpius buthiformis is holosternous, with lat- eral, intramarginal, schizochroal eyes. The first two pairs of coxae are quite similar to those of living scor- pions, inasmuch as both have well-developed maxil- lary lobes, the second pair of which extends anteriorly to crowd out the first pair, which therefore does not meet at the median line. The sternum is roughly pen- tagonal, large, with the lateral sides and base emargin- ate. The last two pairs of coxae abut the sternum. The family Eoctonidae differs from the family Bu- thiscorpiidae in that the second pair of coxae does not Ch1- Ch4a Text-figure 40.—Buthiscorpius buthiformis (Pocock). Holotype, BM(NH) In.18596. From the Carboniferous (Westphalian A), Mid- dle Coal Measures, Carbonicola communis beds, Sparth Bottoms, Rochdale, Lancashire, England. See foldout inside front cover for explanation of abbreviations. A. Complete specimen. B. Showing the terminus of the chelicera. reach to the anterior of the first pair, and in having the second pair of coxae developed as spatulate lobes. The sternum in the Eoctonidae is broader than long and has lateral margins that converge anteriorly, in contrast to the longer-than-broad sternum with emarginate sides in the Buthiscorpiidae. Genus BUTHISCORPIUS Petrunkevitch, 1953 (emend.) Buthiscorpiidae with lanceolate carapace, median eyes in center of anterior half of carapace, and lateral schizochroal eyes anterior to median eyes. Carapace divided by a median sulcus into two cephalic cheeks. Sternum longer than wide, pentagonal and with emar- ginate lateral margins. Type species. — Anthracoscorpio buthiformis Pocock, 1911. Geological range. —Carboniferous. Remarks.—The differences between the genera Buthiscorpius and Anthracoscorpio are numerous and of higher category than generic. For example, they be- long to different families as the maxillary lobes of the second pair of coxae in Anthracoscorpio have not reached full development—that is: they do not reach the same level as the first pair. On the generic level, the sternum is wider than long in Anthracoscorpio and the opposite is true in Buthiscorpius. Buthiscorpius buthiformis (Pocock, 1911) Text-figures 40, 110F 1911. Anthracoscorpio buthiformis Pocock (partim), pp. 24-28, pl. 2, fig. 1; text-fig. 6; not pp. 26-27, text-fig. 8 (paratype BM(NH) In.22832 = Leioscorpio pseudobuthiformis, n. gen., n. sp.); not p. 26, text-fig. 7 (paratype BM(NH) In.7883 = Comp- soscorpius elegans Petrunkevitch); not pp. 27-28, pl. 1, fig. 2 (paratype BM(NH) In.1555 = Pseudobuthiscorpius labiosus, n. gen., n. sp.). 1913. Eoscorpius buthiformis (Pocock). Petrunkevitch, p. 35. 1949. Eoscorpius buthiformis (Pocock). Petrunkevitch, p. 153. 1953. Buthiscorpius buthiformis (Pocock). Petrunkevitch (partim), p. 32, figs. 31-33; not p. 32, fig. 34 (paratype BM(NH) In.1555 = Pseudobuthiscorpius labiosus, n. gen., n. sp.). 1955. Buthiscorpius buthiformis (Pocock). Petrunkevitch, p. 74, fig. 43(3). 1960. Buthiscorpius buthiformis (Pocock). Wills (partim), pp. 277- 290; not pp. 278-284, pl. 46, 47; text-figs. 1-5 (BM(NH) In.31262 = Coseleyscorpio lanceolatus, n. gen., n. sp.); not pp. 284-290, pl. 48; text-figs. 6-9 (BU 720 = Allobuthus macrostethus, n. gen., n. sp.). 1962. Buthiscorpius buthiformis (Pocock). Dubinin, p. 431, figs. 1233, 1252: The holotype consists of two parts showing the dor- sal surface only, BM(NH) In.18596a and b. The original description of Pocock (1911, p. 24) is essentially correct, therefore my comments will be re- stricted to features either not described or overlooked, FossIL SCORPIONIDA: KJELLESVIG- WAERING 103 some of which are highly important and which neces- sitate a redefinition of the genus. 1. On the anterolateral angles, the presence of small, elliptical, schizochroal eyes were first noted on latex casts and verified on the actual specimen. The presence of lateral eyes was also verified in the American species B. lemayi. The individual eyes are rounded, and ap- proximately seven to eight eyes occur in an elliptical cluster at the same spot where the lateral eyes of pres- ent-day scorpions are found. The eyes are therefore intramarginal and schizochroal. 2. Pocock (1911, p. 24) mentions that the dorsal surface is finely- to closely-granulated. This does not seem to be the case, although it was on other specimens incorrectly referred to this species. In fact, the dorsal surface is quite smooth except for the anterior part of the carapace where it is granular. 3. The carapace is elliptical or oval in general outline, with a distinct but slight prolongation at the anterior midsection (see Text-fig. 40A). The marginal rim is very narrow. The basal rim is narrow but well marked, slightly protruding over the adjoining tergite. 4. The dorsal keels or carinae are surmounted by granules, each of which seems to be setaceous. 5. The pedipalps show numerous setaceous granules, not in any apparent order, on the femur and tibia as well as the anterior of the hand. The edge of the fixed finger of the pedipalp seems to be cultrate, without granules or denticles. 6. The twelfth tergite or last caudal segment was developed for the first time and revealed an unusually long tergite. The overall aspect of the tail is very thick, and does not decrease in size posteriorly. The last seg- ment measures 3.3 mm in length. The great length of this segment is indicative of a male; the coarse gran- ulation on the anterior of the carapace also verifies that the holotype is a male. This sex determination is fur- ther strengthened by the presence of a very short cor- responding tergite in the closely related American species B. /emayi, which would be indicative of the female. Type information. —From the Carboniferous (West- phalian A) Carbonicola communis beds, Middle Coal Measures, Sparth Bottoms, Rochdale, Lancashire, En- gland. Remarks. —As restricted here, the species Buthi- scorpius buthiformis (Pocock) is at present known only from one specimen (part and counterpart) exhibiting the dorsal side only. The other specimens that had been referred to this species have here been relegated to different families, genera and species (see generic de- scription above). Buthiscorpius lemayi, new species Text-figure 41 Morphologically and taxonomically this is one of the most important specimens known, as the underside of an undoubted Buthiscorpius is preserved. The type species, B. buthiformis (Pocock) from England, was known only from the dorsal side, albeit a very well- preserved form. It further verified the presence of lat- eral compound eyes, which previously had not been known in this genus despite the studies by Pocock, Petrunkevitch, and Wills. Preservation in both the British holotype and the new American species is so perfect that the setal openings, or sites, on the pedipalps can be used for the first time as specific differences, in much the same manner as the trichobothria of the pedipalps are used in descriptions of modern scor- pions. It may be that the setal openings are tricho- bothria, as they occur in precisely the same areas as the modern trichobothria, although, of course, the trichogen cells developed at the base of each seta have not been fossilized. The carapace is lanceolate, with well-developed round median eyes located on a prominent eye node slightly behind the center of the anterior half of the carapace. A deep median sulcus divides the carapace behind the eye node. Supraorbital ridges are well-de- veloped. The lateral compound eyes are reniform and probably enclose eight small, rounded lenses. A few punctae occur along the base of the carapace, which is bounded by a transverse basal ridge and a very thin marginal rim. The anterior is not as noticeably pro- truding as in B. buthiformis. The legs are not well preserved except for the im- portant pedipalps, which are stout, with long curved fingers. Setal openings are present in the humerus (P3), brachium (P4) and most likely the chela (P5, P6). In their distribution, these are in marked contrast to those present in B. buthiformis (Pocock) (see Text-fig. 41). The rest of the legs and the chelicerae are not sufh- ciently well preserved for a meaningful description. The important coxosternal region is preserved in its entirety and reveals an arrangement much as in living scorpions. The sternum is elongate-pentagonal, the base being incurved with the anterior triangular part hardly projecting forward. A deep sulcus, no doubt for the attachment of muscles, occurs in the posterior half of the plate, and two other indentations occur at the sides at midsection, also for the attachment of muscles. The third and fourth pairs of coxae abut against the ster- num, whereas the maxillary lobes of the second coxae meet at midsection and crowd out the maxillary lobes of the first coxae, as happens in living scorpions. The maxillary lobes of both the first and second coxae are narrow and reach to the anterior, thus forming the 104 P5S Ch3 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Inclined upward P4 ] f CS SL “59 ee AP2 aI Pa OVS ie =e 3 (/ fw ZA aps p AP4 gp ws AP5 Iv6 T7 ven — tmm T8 Text-figure 41.—Buthiscorpius lemayi, n. sp. Holotype, FMNH (PE) 21600 (part and counterpart), presented to the Field Museum of Natural History, Chicago, IL, by Stephen LeMay. From the Penn- sylvanian, Francis Creek Shale, collected at Pit 6 of the Peabody Coal Company near Braidwood, IL. See foldout inside front cover for explanation of abbreviations. A. Dorsal view. B. Ventral view of the counterpart. tubular approach or chamber to the oral opening. The genital opercula are laterally elongated, narrow, and transversely lacrimiform. The pectines are notable for their enormous size and the great size of the sickle-shaped teeth. Fulcra are developed but not very large. The middle lamella con- tains areoles of irregular, rounded, or subrounded shapes. The anterior lamella appears to consist of three FossIL SCORPIONIDA: KJELLESVIG- WAERING 105 well-defined joints. About 13 teeth occur on the great pectines, which had to spread beyond the margins of the opisthosoma in life. The opisthosoma is lanceolate without great con- traction at the eighth tergite. Measurements (given be- low) show a gradual increase in length in the meso- soma, and the greatest width is reached at the fifth tergite. Double keels are very prominent on the meta- soma. Apparently these are also developed on the ven- ter. The abdominal plates are holosternous and have well-developed gill chambers (see Text-fig. 41B). The telson is not preserved in its entirety but it is very long and not unduly curved. The holotype is no doubt a female, which is revealed by secondary sexual characters, such as the short twelfth tergite (in contrast to the long male segment in the holotype of B. buthiformis (Pocock)), the lack of coarse granulation (in contrast to the coarse granules in the male of B. buthiformis) and the wide mesosoma. Measurements (in mm) of the holotype, FMNH (PE) 21600.— Carapace: Length: Bis: Width at base: 3.4 Median eyes: Distance from anterior margin: 0.8 Distance from posterior margin: 2.26 greatest Tergite: length width No. 1 0.7 (incomplete) No. 2 0.9 (incomplete) No. 3 1.0 3.1 No. 4 1.24 4.0 No. 5 1.4 4.7 No. 6 1.5 4.0 No. 7 1.6 anterior 2.8 posterior 2.16 No. 8 1.6 posterior 2.0 No. 9 ea/ posterior 1.7 No. 10 1.8 posterior 1.6 No. 11 2.2 posterior 1.56 No. 12 DD posterior 1.56 Telson Dal (incomplete) 1.5 Pedipalp: Length of P3: 2.4 Length of P4: 2.3 Hand width (P5): 1.0 Hand length (P5): 2.0 Free finger length (P6): 3.0 Type information. — Holotype is an adult female pre- served in an ironstone concretion, comprising part and couterpart of a nearly complete specimen. It is from the Pennsylvanian, Francis Creek Shale, collected at Pit 6 of the Peabody Coal Company near Braidwood, Illinois. Derivatio nominis.—Named in honor of Stephen LeMay, who kindly alerted the author to the specimen and graciously presented the specimen to the Field Museum of Natural History, Chicago, where it is reg- istered as FMNH (PE) 21600. Remarks. —In contrast to Buthiscorpius buthiformis (Pocock) which is known only from the dorsal side, the differences, other than those considered due to sex- ual dimorphism (granulation, short twelfth tergite, and mesosomatic width), consist of the presence of a much more delicate and longer pedipalp, the different setal distribution on the humerus and brachium, the lack of the pronounced anterior prolongation of the cara- pace, and the double rows of dorsal carinae on the postabdominal tergites. These dorsal characters are sufficient to separate B. /emayi from the British B. buthiformis. It is an interesting addition to the fa- mous Mazon Creek fauna. Family ALLOBUTHISCORPIIDAE, new family Eoctonoidea (?) apparently without lateral com- pound eyes. Type genus. —Allobuthiscorpius, n. gen. Remarks. —Since the single specimen is known only from the dorsal side, its higher taxonomic position is in doubt. Genus ALLOBUTHISCORPIUS, new genus Allobuthiscorpiidae with lanceolate carapace, divid- ed by a median sulcus into two cephalic cheeks; median eyes in center of anterior half of carapace. Derivatio nominis. —allo (Gr.) = another, different + Buthiscorpius. Type species. — Buthiscorpius major Wills, 1960. Geological range. —Carboniferous. Remarks. — Buthiscorpius major Wills cannot re- main in Buthiscorpius as it seemingly lacks lateral eyes. Allobuthiscorpius major (Wills, 1960) 1960. Buthiscorpius major Wills, pp. 300-305, pl. 51, figs. 1-3; pl. 52; text-figs. 14-16. Holotype (GSM Za 2926) on deposit in the Institute of Geological Sciences, London, from the Coal Mea- sures (Ammanian), base of communis zone, Kilburn Coal, Trowell Colliery, Nottinghamshire, England. The specimen was in a half nodule revealing only the ven- tral surface of the dorsal side of a large scorpion (to quote Wills, p. 301): about twice the size of Buthiscorpius buthiformis Pocock; carapace ornamented with granules, some being large, mimicking lateral eyes which are absent; median eyes small and near to one another on an eye-tubercle without visible ridges between the eyes; eye tubercle about two-fifths of carapace length from the front; tergites with mu- cronate posterior margins; tail short and relatively shorter than in B. buthiformis; caudal ring X VIII not much longer than the previous one, and shorter than the flask-shaped sting. Pedipalp hand with fingers longer in proportion to the palm than in B. buthiformis. 106 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 For dimensions and detailed description that cannot be improved upon, see Wills (1960). Remarks. —Wills felt that the hairs actually still at- tached to the chelicera of B. major (pp. 287-288) sug- gested that some of the Carboniferous scorpions employed the same method of feeding as do their pres- ent-day descendants, and one that would appear more appropriate in terrestrial than in aquatic animals, but the undoubtedly aquatic Hydroscorpius from the De- vonian of Cottonwood Canyon was copiously supplied with hairs or setae on the chelicerae (see Text-figs. 20A, C) and couldn’t possibly have had maxillary lobes. Genus ASPISCORPIO, new genus Allobuthiscorpiidae with oval carapace, small round median eyes located in central part of the anterior half of the carapace; well-developed, raised cephalic shield, with raised median glabellalike ridge behind eyes, and raised posterior genal areas. No lateral compound eyes. Derivatio nominis. —aspis (Gr.) = shield. Type species. — Aspiscorpio eagari, n. gen., n. sp. Geological range. —Carboniferous. Remarks. —This scorpion has a superficial resem- blance to Buthiscorpius and is associated with it in the type locality. It can readily be distinguished from that genus by the lack of a frontal protuberance, the de- velopment of the strong cephalic shield, and no lateral schizochroal eyes. I had placed it in the Buthiscorpi- idae with considerable misgivings as the underside is not known, but the lack of lateral compound eyes makes a position within the Buthiscorpiidae untenable. Aspiscorpio eagari, new species Text-figure 42 1953. Eoscorpius sparthensis Baldwin and Sutcliffe. Petrunkevitch (partim), p. 28. Based on a single specimen showing the inside of the dorsal part of the carapace, mesosoma, and anterior two tergites of the postabdomen. The chelicerae are preserved from the dorsal part, and the right (in life) ventral side of the entire pedipalp. Inconclusive frag- ments of the walking legs are present. Carapace relatively long, oval in front, without traces ofan anterior protrusion. It is rimmed by a very narrow marginal rim, which is very prominent on the basal and basilateral margins. There are no lateral eyes and the median eyes are very small, round, located on a rather round ocellar mound, about in the center of the anterior half of the carapace. Orbital ridges are negli- gible. A raised, elongate cephalic area is very promi- nent, and this is divided by a medial raised ridge that extends to the median eyes, and has the form of the glabellar ridge in eurypterids. Two raised areas occur in the pleural part. The mesosoma is noteworthy for the unusually long transverse ridges on the tergites. Undoubtedly these transverse ridges include the intersegmental tissue be- tween tergites, but nevertheless the ridges are very well developed. Slight, but noticeable, crenulations orna- ment the basal margins of the tergites, and it is due to Text-figure 42.— Aspiscorpio eagari, n. gen., n. sp. Holotype, UMM L 8182. From the Carboniferous (Westphalian A), Carbonicola com- munis beds, Sparth Bottoms, Rochdale, Lancashire, England. Note that the rachis of the pectines is reflected as an imprint (dotted lines). See foldout inside front cover for explanation of abbreviations. FossiL SCORPIONIDA: KJELLESVIG- WAERING 107 the presence of these crenulations that the tergites can be easily separated. The last tergite has a raised low carina in the central part and two diagonal carinae marked by pustules on the sides. The central carina is marked by two de- pressed thin lines. The cauda, or postabdomen, is rep- resented by only two tergites, which reveal well-de- veloped double carinae. The pedipalp is preserved in its entirety, showing it to be very long, with a very short hand, and very long fingers. The fingers contain very small denticles, but it is not possible to determine the pattern. They may be in a straight line along the cultrate edge. Two very thin carinae, which continue into the hand, occur on the ventral side of the fixed finger. Except for the crenulations noted, and some sparsely distributed pustules in the front of the carapace, the rest is very smooth. Some setal openings were noted on the pedipalp tibia. Measurements (in mm) of the holotype, UMM L 8182.— Carapace: Length: 6.5 Width at base: 6.2 Eyes: Diameter: 0.25 Distance from anterior margin of carapace: 1235 Distance from posterior margin of carapace: 2.6 Distance from lateral margins of carapace: 1.65 Pedipalp: length width Femur: 6.5 123) Tibia: 5.0 17 Hand: 355) 2.0 Free finger: 6.4 Tergites: length width No. 1 12: 7.0 (estimated) No. 2 2.0 7.2 (estimated) No. 3 2:6 7.3 (estimated) No. 4 2.6 8.6 (estimated) No. 5 3.0 7.4 No. 6 3.8 7.0 No. 7 35) 5.4 (anterior) 2.7 (posterior) No. 8 3.3 2.65 No. 9 (incomplete) 2.50 Type information. — Preserved in an ironstone con- cretion from the Carboniferous (Westphalian A) Car- bonicola communis beds at Sparth Bottoms, Rochdale, Lancashire, England (pers. commun., M. Eagar). De- posited in the Manchester University Museum, where the holotype is registered as L 8182. Derivatio nominis. — Named in honor of Dr. Michael Eagar of Manchester University. Remarks. —As stated in the remarks regarding the genus, the only scorpion with which this scorpion might be confused is Buthiscorpius buthiformis (Pocock), which also occurs at Sparth Bottoms. The prominent raised cephalic shield of Aspiscorpio eagari is sufficient for identification, as well as the lack of lateral schi- zochroal eyes. Family ANTHRACOSCORPIONIDAE, new family Eoctonoidea with large pentagonal sternum, long maxillary lobes on the first and second pairs of coxae; the second pair meets at the midline from the sternum almost to the anterior of the first pair of maxillary lobes; no lateral compound eyes. Type genus. —Anthracoscorpio KuSta, 1885. Genus ANTHRACOSCORPIO Kusta, 1885 Carapace semicircular anteriorly with slightly con- cave posterior edge; median eyes sessile, sitting in ad- vance of middle. Type species. — Anthracoscorpio juvenis KuSta, 1885. Geological range. —Carboniferous. Anthracoscorpio juvenis KuSta, 1885 Text-figures 43A, B, 110D, 113B3 1884. Cyclophthalmus senior Corda. KuSta (partim), p. 48. 1885. Anthracoscorpio juvenis KuSta, p. 7. 1904. Anthracoscorpio juvenis KuSta. Fric, pp. 75-76, text-fig. 94. 1949. Anthracoscorpio juvenis KuSta. Petrunkevitch, p. 143. 1953. Anthracoscorpio juvenis KuSta. Petrunkevitch, p. 30, figs. 30, 130. 1955. Anthracoscorpio juvenis KuSta. Petrunkevitch, p. 74, fig. 43(4). 1962. Anthracoscorpio juvenis KuSsta. Dubinin, p. 431, fig. 1250. Type information. —Holotype, No. 836 (Fri¢’s Inv. P. 3520) in the National Museum at Prague, Czech- oslovakia, from the Upper Carboniferous (Westpha- lian B/C) Radnice Member of the Radnice Group, at Rakovnik (about 45 km west of Prague), Czechoslo- vakia. Holotype consists of one half only, which is flat- tened, complete, and preserved in light greenish-gray, slightly sandy shale. The cuticle has no doubt deteri- orated since Kusta (1884) and Fri¢ (1904) worked on this specimen, but enough is left to reveal some very interesting points of structure. KuSta did not figure this specimen, but Fri¢ (1904) and Petrunkevitch (1953) did, and both showed the dorsal side. However, this is not possible, since the scorpion is preserved on the ventral side. The anterior maxillary lobes of the first two pairs of coxae were mistaken by Fri¢ (1904, fig. 94) for the chelicerae. I am unable to determine what prompted Petrunkevitch (1953) to make the drawing that he did, showing a wide carapace with two eyes. This is not to be taken seriously as the outline of the carapace is not present and almost certainly would not 108 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 have been as shown by Petrunkevitch. In fact, the out- The discovery of parts of the coxosternal area is a line of the carapace drawn by Fri¢ is probably correct. welcome addition to our knowledge of this scorpion It is assumed, of course, that the outlines were present and, although fragmentary, enough is present to make and that today the specimen has deteriorated. a good restoration possible (see Text-fig. 110D). The Text-figure 43.—Species of Anthracoscorpio KuSta, 1885. See foldout inside front cover for explanation of abbreviations. A, B. Anthracoscorpio juvenis Kusta. Holotype, NMP Inv. 836. From the Upper Carboniferous (Westphalian B-C), Radnice Member of the Radnice Group, Rakovnik, Czechoslovakia. A. Holotype after Kusta, 1885, text-figure 94. B. Restoration of coxosternal region. C. Anthracoscorpio dunlopi Pocock. Holotype, RSM 1957.1.4995a and b. From the Carboniferous, Upper Coal Measures, shale above Drumgray Coal, Longrigend, Airdrie, Lanarkshire, Scotland. Coxosternal area. D. Anthracoscorpio dunlopi (?) Pocock. RSM 1957.1.5000 through 5002. From the Carboniferous of Airdrie, Scotland. Seventh tergite. Note that this is the venter, because the carinae are close together (see Paraisobuthus duobicarinatus and Eoscorpius carbonarius). FossiL SCORPIONIDA: sternum is large and pentagonal in shape. The first pair of coxae has well-developed, narrow but long, maxil- lary lobes that abut against the maxillary lobes of the second pair, as in modern scorpions. The second pair meets at midsection, anteriorly to the sternum, ex- tending almost to the anterior part of the first maxillary lobes. The last two pairs of coxae abut the sides of the sternum. This arrangement therefore is like that found in the Eoctonoidea, particularly in the Buthiscorpiidae and in modern scorpions. The genital opercula are very small, rather elliptical and adjoin the base of the large sternum. The scorpion is a holosternous form and the plainly visible abdominal plates confirm this. The pedipalp does not show any denticles on the edge of the fingers, which appear to be cultrate. The palm and fingers are narrow and long. The preabdomen narrows abruptly into the cauda, which shows well- defined double carinae that may be ventral and lateral carinae. The vesicle is large and bulblike, followed by an aculeus that likely was long. Anthracoscorpio dunlopi Pocock, 1911 Text-figures 43C, D 1911. Anthracoscorpio dunlopi Pocock, pp. 21-24, pl. 1, fig. 1; text- fig. 5. 1913. Eoscorpius dunlopi (Pocock). Petrunkevitch, p. 34. 1949. Eoscorpius dunlopi (Pocock). Petrunkevitch, p. 153. 1953. Eoscorpius dunlopi (Pocock). Petrunkevitch, p. 28, fig. 131. 1962. Eoscorpius dunlopi (Pocock). Dubinin, p. 429, fig. 1247. The two parts of the holotype, preserved in black shale, reveal only the dorsal side and counterpart of the same surface. The important coxosternal region can be faintly made out as it is impressed through the dorsal surface, and is therefore described for the first time. The first two pairs of coxae have well-developed maxillary lobes. The maxillary lobes of the second pair of coxae are very narrow and extend anteriorly to squeeze the first pair of maxillary lobes away from the midline, as in living scorpions. The sternum is large, pentagonal, and the last two pairs of coxae abut against it. Pocock’s (1911) description of the dorsal side is quite adequate, and little can be gained by repetition here. The development of the coxosternal region in the ge- notype Anthracoscorpio juvenis KuSta (Text-figs. 43A, B) reveals that Pocock was correct in referring the ho- lotype to that genus. The scorpion cannot be referred to Eoscorpius, as Petrunkevitch (1913, 1949, 1953) thought, since it is not a lobostern, but a holostern, a major difference among many others. Type information. — Holotype is from the Upper Coal Measures, shale above Drumgray Coal, Longrigend, KJELLESVIG-W AERING 109 Airdrie, Lanarkshire, Scotland. It consists of part and counterpart, showing only the dorsal side, and is reg- istered as No. 1957.1.4995a and b, in the collections of the Royal Scottish Museum, Edinburgh, Scotland. Remarks. — Royal Scottish Museum Nos. 1957.1.5000 through 5002, part and counterpart, here illustrated as Text-figure 43D, may be a specimen of A. dunlopi Pocock. It consists ofa single seventh tergite preserved in dark gray micaceous shale from Airdrie. The tergite (No. 7) of the preabdomen, reveals two closely-set median carinae, each of which is surmount- ed by rounded tubercles. Similar tubercles, but larger and longer, occur along the base. The anterior is bound- ed by the usual transverse ridge. Measurements (in mm) of RSM 1957.1.5000 through 5002.— Anterior width: 9.9 (estimated) Posterior width: 4.0 Length: 6.6 (estimated) Anthracoscorpio (?) species Text-figure 44 1913. Eoscorpius typicus Petrunkevitch (partim), pp. 39-43, pl. 1, fig. 1. Specimen No. YPM 126, in a typical Mazon Creek \ Ch <4 P2? P6 “- / ae a J ~~ 5% \ cl P5\| cr fen | ——— a Yi L a Cane : ee Str rs YH (C= cil i / / mn aes —z m3 y); {for ) / / : / aaa i aS civ| Weary, f AP3 \ \o™M1 +. —4 t— aan tom ) \ AP4 aE \ eee | AP5 | Text-figure 44.—Anthracoscorpio (?) sp. YPM 126, designated as a paratype of Eoscorpius typicus Petrunkevitch. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Mazon Creek area. See foldout inside front cover for explanation of abbreviations. 110 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 ironstone concretion, one of the paratypes of Eoscor- pius typicus Petrunkevitch, 1913, was assigned to Ma- zonia woodiana Meek and Worthern in a note Kjel- lesvig-Waering left, and illustrated as Text-figure 44. No description was found. The specimen is obviously not a Mazonia, which is a Mesophonoidea, in which, by definition, the fourth pair of coxae abuts against the operculum, whereas, in this illustration, the fourth pair abuts against the sternum, so that YPM 126 seems to fit in the Eoctonoidea. The specimen is preserved with only the ventral side showing on half the nodule. The sternum is pentagonal, and the genital opercula appear to be elongate and ovoid, thus sternum and opercula resemble the Anthracoscorpionidae. The abdominal plates are holosternous, without any sign of stigmata. The free finger of the pedipalp is fairly well preserved, shows a ridge for its entire length, and lacks denticles (A.S.C.). Genus LICHNOSCORPIUS Petrunkevitch, 1949 (emend.) Anthracoscorpionidae with parabolic carapace, two small median eyes anteriorly placed on a wide lacrim- iform eye node, and with the preabdomen tapering into the cauda without an abrupt attenuation, and with the cauda tapering posteriorly. Type species. —Lichnoscorpius minutus Petrunke- vitch, 1949. Geological range. —Upper Carboniferous. Remarks. —The genus Anthracoscorpio Kusta was established in 1885 for a small scorpion from the Mid- dle Bohemian Coal Basin. In 1949, Petrunkevitch de- scribed another small scorpion from the Upper Coal Measures at Coseley, England, as Lichnoscorpius mi- nutus. In 1953, Petrunkevitch, based on a wrong in- terpretation of both species, decided that his genus Lichnoscorpius was a junior synonym of Anthraco- scorpio. 1am at a loss to understand how Petrunkevitch (1953) could have made a drawing of the dorsal side of the carapace when only the ventral side is preserved. It is not possible to include two scorpions such as Anthracoscorpio juvenis Kusta and Lichnoscorpius mi- nutus Petrunkevitch in the same genus, when the preabdomen of one (A. juvenis) is abruptly constricted into a slender cauda with parallel sides, and the other (L. minutus) has a preabdomen that tapers into a ta- pering cauda. Lichnoscorpius has an opisthosoma much like that of the more sinuous genera of the Chactidae, such as Broteochactas, and likely was a burrower or a true cryptozoan in a water medium. Anthracoscorpio Juyenis, on the other hand, has the slender aspect of the buthids, such as Tityus and Centruroides, which led a more active or less cryptozoan life. Lichnoscorpius was proposed by Petrunkevitch (1949) because the genus Anthracoscorpio was not con- sidered to have median eyes, whereas the genotype of Lichnoscorpius had well-developed eyes. On studying the holotype of Anthracoscorpio juvenis KuSta, the ge- notype, Petrunkevitch decided that it had a pair of small “sessile” eyes, and therefore considered his genus Lichnoscorpius to be a junior synonym of Anthraco- scorpio Kusta, 1885 (Petrunkevitch, 1953, p. 30). The holotype of Lichnoscorpius minutus clearly shows large rounded eyes located on a definitely-out- lined, anteriorly-located eye node that is surrounded by a narrow sulcus, and where the eyes are also sep- arated by a sulcus. Petrunkevitch (1949, p. 148) stated, “The eyes are visible only on the piece representing the mold and the interocular ridges appear therefore as depressions, but with perfectly sharp outlines. There is no eye-tubercle, the level of the eye-circumference being the same as that of the carapace”’. A rubber cast of the mold would have given the correct relief, name- ly, a very well-developed eye node, quite normal on Paleozoic scorpions, with a narrow sulcus separating the eyes. There are strong supraorbital (interocular?) ridges, and it would be surprising, if not impossible, to find a scorpion with “the level of eye circumference being the same as that of the carapace’’. Compaction and consequent flattening of thin tissues is not a dif- ficult process to understand in fossils. The position of the eyes on the carapace, as shown here for Lichno- scorpius minutus, and the purported eyes shown by Petrunkevitch for Anthracoscorpio, do not mean that the two belong to the same genus. Other much more important differences occur, as given below. Inasmuch as the generic name Lichnoscorpius Pe- trunkevitch, 1949, is available and is in truth a valid name, the generic description has been emended. The holotype is known from both dorsal and ventral sur- faces and shows that the abdominal plates are holo- sternous. It is considered here that, although the cox- osternal arrangement is not entirely preserved, enough is present to show that it is typical of the Eoctonoidea. The same is true of Anthracoscorpio KusSta, 1885, and a detailed restudy of the two genotypes proves the point. Lichnoscorpius minutus Petrunkevitch, 1949 Text-figure 45 1949. Lichnoscorpius minutus Petrunkevitch, pp. 148-149, figs. 144, 145, 181. 1953. Anthracoscorpio minutus (Petrunkevitch). Petrunkevitch, pp. 30-31. 1962. Anthracoscorpio minutus (Petrunkevitch). Dubinin, p. 431, fig. 1249. Type information. —Holotype from the Upper Coal Measures in the Modiolaris similis and pulchra Zones, ten feet above the Thick Coal of Coseley, in the South FossIL SCORPIONIDA: KJELLESVIG- WAERING atl Staffordshire Coalfield, England, and deposited in the British Museum (Natural History) as BM(NH) In.31266. The original description and figures of the holotype are erroneous. Petrunkevitch (1949, p. 148) states that the holotype consists of two parts, both showing “‘the dorsal surface without any traces of the ventral sur- face’. The ventral surface is shown on one of the spec- imens, thus revealing several very important struc- tures. The restudy requires a full redescription, which is given here. The carapace is broadly semicircular with a narrow marginal rim at the base, which is straight, and a well- developed lacrimiform median eye node surmounted by two small round eyes with well-developed supraor- bital ridges. No visible rim is present. The tergites increase slightly in size posteriorly. Therefore the last three preabdominal tergites are approximately the same size. A strong transverse ridge occurs at the anterior of each preabdominal tergite. The cauda, the anterior four tergites of which are present, is stoutly construct- ed—two high carinae, diagonally placed, surmount each Text-figure 45.—Lichnoscorpius minutus Petrunkevitch. Holotype, BM(NH) In.31266. From the Upper Carboniferous, Upper Coal Measures (Modiolaris similis and pulchra zones), Coseley, Staffordshire, England. See foldout inside front cover for explanation of abbreviations. A. Dorsal side. B. Ventral side, counterpart. C. Outline of the carapace and ocular details. From a rubber cast. 112 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 tergite dorsally. The cauda tapers posteriorly and there is no great constriction at the eighth tergite. The underside shows the abdominal plates to be holosternous, rounded at the genal angles, and with a slight transverse ridge on the anterior. The coxosternal region has largely been broken away, but the genital opercula are present and are elongate, subtriangular, and rounded at the posterior part. More important is that the anterior of the operculum lies below the level of the last pair of coxae. Therefore this pair had to abut the sternum, and it seems probable that the coxo- sternal arrangement was typical of the family Anthra- coscorpionidae. The pectines are faintly preserved and are narrow, long, containing approximately 25 to 30 teeth, with fulcra and the rachis divided into irregular-sized rounded sclerites or areoles, which are coarser toward the center than at the edges. The anterior lamella also has rounded sclerites. Measurements and description of the opisthosoma were given in the original description (Petrunkevitch, 1949, pp. 148-149). Remarks. —\t can easily be separated from Pseu- dobuthiscorpius labiosus Kjellesvig-Waering by the dif- ferent opercular plates, as well as the holosternous ab- dominal plates as against the lobosternous plates in the latter. Both species probably had rather similar coxo- sternal arrangements. Genus ALLOBUTHUS, new genus Anthracoscorpionidae having a square carapace, slightly emarginate or straight anterior margin, and median eyes located one-third of the distance from the anterior, without elevated cephalic shield. No lateral compound eyes. Derivatio nominis. —allo (Gr.) = another + buthus (Gr.) = a scorpion genus. Type species. —Allobuthus macrostethus, n. gen., n. sp. Geological range. —Carboniferous. Remarks. —\t differs from Buthiscorpius in having a quadrate carapace as against a nearly semielliptical one, and in the lack of compound eyes. The semicircular carapace of Anthracoscorpio is sufficient to distinguish it from the quadrate form of the new genus. Allobuthus macrostethus, new species Text-figures 110C, 113B4 1960. Buthiscorpius buthiformis (Pocock). Wills (partim), pp. 284— 289, pl. 48, text-figs. 6-9. Type information. —Holotype, Birmingham Uni- versity BU 720, ‘talmost certainly obtained from the Coal Measures of South Staffordshire, probably from above the Thick Coal at Coseley” (Wills, 1960, p. 284), England. Included by Wills (1960, p. 284) under the descrip- tion of Buthiscorpius buthiformis (Pocock) 1s a scorpion from a different locality and very likely different age than the holotype of that species. The specimen BU 720 was successfully extracted from the matrix, but it cannot properly be assigned to Buthiscorpius buthifor- mis (Pocock) nor even to the same genus. The differ- ences are much too great to include BU 720 and the holotype of B. buthiformis (Pocock) in the same genus, and these differences cannot be ascribed to secondary sexual characters. Moreover, there seems to be no doubt that both specimens are males. This is based on the generally slender mesosoma of BU 720 and on the equally slender mesosoma and the great length of the fifth caudal segment of the holotype of B. buthiformis. Both of these characters are used for determining sex in living scorpions, and are quite reliable. Even if the sex determination were not correct, there are many other differences unrelated to possible sex differences. The differences between Allobuthus macrostethus and Coseleyscorpio lanceolatus, n. gen., n. sp., which Wills had successfully developed (1960, p. 277) as a paratype (BM(NH) In.31262) of Buthiscorpius buthiformis (Po- cock) are listed below: A, macrostethus (BU 720) C. lanceolatus (BM(NH) In.31262) Carapace: Sternum: Coxa) i: Coxa 2: Coxa 3: Trochanter of pedipalp: Seventh tergite: nearly square anterior margin slight- ly emarginate slightly wider than long eyes smaller eyes close-set without interocular ridge sides nearly straight, base straight proportionately wider much wider maxillary lobe not divided by “joint line” from the max- illary lobe less than half as long as coxa 4 flaring at distal end carinae not present, apparently smooth semielliptical anterior margin oval considerably longer than wide eyes larger eyes much farther apart with interocu- lar ridge sides converging for- ward, base deeply notched proportionately nar- rower, particularly anteriorly much narrower maxil- lary lobe well-developed “joint line” separating maxillary lobe from coxa more than half as long as coxa 4 rounded at distal end well-developed sharp, double carinae and strong median spine FossiL SCORPIONIDA: KJELLESVIG-WAERING 113 The sternum of the holotype of Ad/lobuthus macro- stethus is not complete along the base, but enough is present to show that this part had to be straight rather than deeply-notched as in Coseleyscorpio lanceolatus. Detailed descriptions and figures of the holotype of Allobuthus macrostethus have been given by Wills (see synonymy) and little can be gained by repetition here. Derivatio nominis. — macro (Gr.) = long, large + ste- thos (Gr.) = breast. Genus COSELEYSCORPIO, new genus Anthracoscorpionidae of small size; carapace lan- ceolate, without lateral eyes, small rounded eyes lo- cated at the base of the anterior quarter of the carapace, poor development of eye node and orbital ridges. Ster- num pentagonal, converging anteriorly, deeply-notched or flared at posterior and with genital opercula fitting into notch. First two coxae with well-developed, nar- row maxillary lobes. Pedipalp fingers straight. Derivatio nominis. —Named for Coseley, where the holotype was found. Type species. —Coseleyscorpio lanceolatus, n. gen., Nn. sp. Geological range. —Carboniferous. Remarks. —This is another genus developed from a paratype of Buthiscorpius buthiformis (Pocock). Coseleyscorpio lanceolatus, new species 1911. Anthracoscorpio buthiformis Pocock (partim), p. 27, pl. 1, fig. 2a. 1953. Buthiscorpius buthiformis (Pocock). Petrunkevitch, p. 32. 1960. Buthiscorpius buthiformis (Pocock). Wills (partim), pp. 278- 284, pls. 46, 47; text-figs. 1-5. The above synonymy refers to specimen BM(NH) In.31262 only. The description and illustrations given by Wills (1960) cannot be improved upon. Type information. —From concretionary shale about ten feet above the ten-foot-thick bed of coal in the Coal Measures, Coseley near Dudley, South Staffordshire, England. The specimen is deposited in the British Mu- seum (Natural History) as BM(NH) In.31262. Derivatio nominis. —Named for the lanceolate car- apace. Remarks. —This specimen differs from the holotype of Buthiscorpius buthiformis as follows: 1. The median eyes of BM(NH) In.31262 are further forward than in the holotype. 2. Specimen BM(NH) In.31262 lacks the ornamen- tation along the anterior of the carapace of B. buthi- formis, although both are males, and also lacks the anterior prolongation. 3. No lateral compound eyes are present in BM(NH) In.3 1262. The many points in which it differs from A//obuthus macrostethus have already been enumerated in the dis- cussion of that species (see above). Family GARNETTIIDAE Dubinin, 1962 (emend.) Eoctonoidea with compound lateral eyes; maxillary lobes unknown; last two pairs of legs abutting pentag- onal sternum; first two pairs meet in front of the ster- num; anterior pair of legs fossorial, flattened, and with large spurs and serrations; highly-serrated rachis of comb, unsegmented. Type genus.—Garnettius Petrunkevitch, 1953. Genus GARNETTIUS Petrunkevitch, 1953 Garnettiidae with small, anteriorly-located, median eyes and large, bulbous, lateral eyes on a nearly square carapace, which is much narrower on central part than anteriorly or posteriorly; chelicerae very large, and with short, stout pedipalps; opercular plates elongate, asym- metrical, lacrimiform. Type species. —Mazonia hungerfordi Elias, 1936. Geological range. —Missourian Series of the Penn- sylvanian. Distribution. — Kansas. Remarks. —There is little to be gained by comparing this peculiar and unique scorpion with any fossil or living genus. It is distinct on many counts, as a glance at the figures will show. The most surprising feature of this scorpion is the anterior pair of legs, which is adapted for digging. Inasmuch as it is a gill-bearing or holosternous type, it must be assumed that digging had to occur in water or at least in very moist areas. The last three pairs of legs are also adapted for a digging existence, as they are encircled at the basitarsus and tarsus with a ring of spines that opened when pressure was applied, and thereby gave more stability and trac- tion to any digging action by the front legs. This com- bination of fossorial legs and “‘stabilizers” is known in many other arthropods, such as the mole cricket, and countless beetles, particularly the coprophagous scar- abs (Phanaeus, Copris, Canthon), but is totally un- known in any of the scorpions, fossil or living. The living scorpions that burrow do not have legs devel- oped into fossorial types, but do have a development of setae which enables them to dig in soft or uncon- solidated sands and soft soils. 114 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Garnettius hungerfordi (Elias, 1936) Type information. —Holotype, USNM 125453a and Text-figures 46, 47B-K, 48 b, in the collections of the National Museum of Natural 1936. Mazonia hungerfordi Elias, p. 13, fig. 6; p. 15, fig. 8. History, Washington, DC, formerly KU 220.1 and 1937. Mazonia hungerfordi Elias. Elias, pp. 335-336. 220.2 in the University of Kansas Geological Museum. 1949. Mazonia hungerfordi Elias. Petrunkevitch, p. 133. Pennsylvanian, Rock Lake Shale Member of the Stan- 1953. SoU aes ae (Elias). Petrunkevitch, pp. 34-35, figs. ton Limestone Formation of the Lansing Group, 1955. Garnettius hungerfordi (Elias). Petrunkevitch, p. 75, fig. 44(8). Missourian Series, in Section 32, T. 19S., R. 19 E., six 1962. Garnettius hungerfordi (Elias). Dubinin, p. 432, figs. 1240, miles northwest of Garnett, Anderson County, Kansas. 1257a, b. Associated on the same slab with articulated brachio- |-right, '6 ae Cc jell is wy Oa ts I- Jett ii "14 ie We. HS AP4 ( \7 S ISO 1 _. broken away intt Text-figure 46.—Garnettius hungerfordi (Elias). Holotype, USNM 125453a,b; part and counterpart (formerly KU 220.1, 220.2). From the Upper Carboniferous (Pennsylvanian) Stanton Group, Missouri Series, 6 mi northwest of Garnett, Anderson Co., KS. See foldout inside front cover for explanation of abbreviations. A. Showing dorsal features only. B. Counterpart showing some suggestions of ventral features. FossIL SCORPIONIDA: KJELLESVIG- WAERING 115 pods, bryozoa, crinoid stems and plants. The specimen is preserved as a flattened skin that retains original coloration and numerous details such as setae and setal sites. A considerable part of the orig- inal cuticle has flaked off but much can be discerned from the imprint left on the fine-grained, buff-colored, highly calcareous shale. This paleobiotype is known as the Garnettius Assemblage (Moore, 1964, p. 297), which is considered to be a marine lagoon or bay with an admixture of marine and terrestrial animals and plants. It is important to understand how this scorpion is preserved. The rock has been split along the bedding plane, revealing numerous fossils on each surface. The scorpion 1s composed of cuticle, and both sides of the bedding plane retain parts of the cuticle or the imprints of the cuticle, which, on the edges, are of considerable value in revealing the structure. Thus each half of the fossil may retain the imprint of the ventral or the dorsal surfaces, along with parts of either the dorsal or ventral side of the original cuticle, or both, depending on how the cuticle broke when both sides of the bedding were exposed. For example, one half may contain the dorsal imprint of the anterior part of the carapace and the edges, as well as part of the ventral side, as shown by the cuticle. This is precisely the case with the anterior part of the holotype seen in Text-figure 46B. The cu- ticle of the legs represents the anterior ventral side, whereas the back legs are revealed mainly as imprints of the posterior side, except where the original cuticle is present on the end of the fourth leg, and there the anterior side is preserved. The imprint of this fossil reveals fine details in var- ious types of lighting, particularly a light source from a very low angle. The specimen has been studied dry, without the preservative covering necessary to prevent the remaining cuticle from flaking off. It has also been studied immersed in alcohol, a method which reveals much from the cuticle. By all these methods it was possible to discern many morphological structures not previously noted. The holotype, and still the only known specimen, is remarkable for the large size of the chelicerae, the lat- eral compound eyes, and the unusual, highly-serrated, anterior legs, resulting in the differentiation of the legs, a factor not encountered in other scorpions. The spec- imen was described by Elias in 1936 and later was studied by Petrunkevitch (1953, pp. 34-35, figs. 36— 39), who gave descriptions and figures that are not in agreement with the description given here. Major points of disagreement or of total omission will be pointed out in the description. The overall coloration of this scorpion 1s dark reddish-brown, shiny and very much like that of living scorpions such as Broteochactas, Euscorpius, Opisthacanthus and Hadogenes. The carapace is nearly square, rounded at the an- terolateral angles and with the anterior margin pro- duced into a pointed process. The lateral margins are incurving, so that the area at the base and at the an- terolateral angles is wider than along the narrowed middle. The base is straight and the genal angles form rounded right angles. A wide, raised marginal rim ex- tends along the base and halfway around the lateral margins. There is no anterior marginal rim such as Petrunkevitch (1953) shows in figures 36 and 37. The part which Petrunkevitch misinterpreted as an anterior rim is the first segment of the two chelicerae, which almost meet in front of the carapace. The most prominent features of the carapace are the small median eyes and the very large lateral compound eyes. The median ocelli are small, elliptical in shape, and are located laterally on a small, cordated, raised ocellar mound that is placed anteriorly on the carapace, close to the anterior projection of the anterior margin. The lateral eyes, previously undetected, are bulbous, very large, and located at the rounded anterolateral corners. No facets were noted as all of the original cuticle has been stripped away, but the outlines of the compound eyes are very clearly revealed. The small median eyes together with large compound eyes is a common combination in Paleozoic scorpions. The prosomal appendages are known in almost all detail, and they are among the most unusual in any scorpion. The enormous chelicerae (see Measure- ments) are features that are unknown in other scor- pions. The chelicerae are very well preserved and were noted in part by Elias (1936) in his figure, but were completely omitted by Petrunkevitch (1953). These structures are nearly as large as the chelae of the pedi- palps or almost half as large as the carapace. It is not known whether there are three or four joints as the carapace covers everything but the last three joints. Because other related Paleozoic scorpions have four- jointed chelicerae, it must be assumed that four joints were present. The very wide chelicerae have a very large free finger that is hooked and armed with several large teeth; the fixed finger is concave along the edge and also is armed with teeth. It was not possible to determine the actual arrangement of these teeth as the chelicerae were compressed into a single plane and little of the original cuticle was left. However, it is certain that the distal end of the free finger ends in a single hooklike tooth and not in a double tooth, an arrangement present in some scorpions such as the living Buthidae. To stress the great size of these che- licerae, they extend 9.8 mm beyond the frontal edge of the carapace, whose total length is only 16.7 mm. The chelicerae are 6.8 mm wide at midsection. The pedipalps are known in their entirety (see Mea- 116 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 surements) and are stout short structures characteristic of burrowing scorpions such as some of the living Chactidae and Scorpionidae. The femur and tibia are very short, stout, and slightly longer than wide. The chelae are short, stout, with a rather quadrate hand, and fingers that are quite unusual. The free finger is stout, but much longer than the immovable one, and is slightly bent inward with a very blunt, rounded end. The immovable finger is also incurving and very stout. Numerous setae occur on the ends of the fingers, in particular as a fringe along the dorsal side of the fingers, near the edge. No teeth are discernible along the edge, but this may be due to preservation and should not be discounted. This is the only known scorpion where differentia- tion of the legs occurs. Other scorpions have the four legs similarly developed except, of course, for relative length of the individual segments, or at the coxae. Some differentiation does occur in scorpions in the presence or absence of certain tarsal or basitarsal spurs, but this is minor in comparison to the differentiation of the first leg of Garnettius hungerfordi into a scooping, flat- tened, serrated organ admirably suited for digging, as against the other legs which, although varying in length, are similar to each other. In modern insects, the Gar- nettius type of front legs can be duplicated in many groups, as mentioned above. In these groups the front legs are curved inwardly and are not employed in walk- ing, but are specialized for burrowing. The first pair of legs is well preserved and almost entirely known. The right leg has been displaced and is present on the left anterior part alongside the left leg. This leg is preserved in its original cuticle with the dorsal side showing. The left leg is also preserved in its original cuticle and is shown ventrally. The exact podomere count is not known, but it is assumed that eight joints, coxa and posttarsus included, make up the entire leg. If this is correct, then in Text-figure 47J the long joint labelled I3 would be the femur, I4 tibia, I5 basitarsus, I6 tarsus, and I7 the posttarsus. All joints are naturally flattened, and only the tibia and basitarsus are serrated along the posterior edge, which would be carried ventrally in life, bowed inwardly, and with the dorsal side facing the anterior. It is interesting to note that in this leg, although far removed by specialization from the normal plan, the basitarsal and tarsal spurs are developed as in many other fossil scorpions. It is certain that at least two very large movable spines are present on the distal part of the tibia. These definitely appear to be in sockets of the tibia and not in the intersegmental skin between the tibia and basitarsus as in other scorpions. It may be possible that they are present in the intersegmental skin in a cove or depres- sion at the end, but this does not seem to be the case as definite sockets have been observed. The spines are flattened in conformity with the flattened and serrated edge of the rest of the tibia. A short spine, which def- initely occurs in the intersegmental skin between the tibia and tarsus, was noted, and this may be the actual basitarsal spur. The basitarsus, as stated above, is constructed along the same plan as the previous joint, being flat and with very large serrations along the posterior side. This edge would be ventral in life. The tarsus, surprisingly, is a joint of “normal” nature and not spinous as the previous one. Two short claws, slightly bent and hooklike, terminate the leg. The an- terior claw is slightly larger. Immediately below is the posttarsus, which is rounded, short, and unusually stout. The second walking leg is rather tubular and without the development of the spurs present in the first leg. This leg has eight short tubular joints. Basitarsal and tarsal spurs are developed, two in the basitarsus and an unknown number in the tarsus. Preservation was not very good in this part. The tarsus, with the terminal claws, and the posttarsus are very well preserved from the ventral side, in their original cuticle. The tarsus is very short, devoid of spines except at the interseg- mental anterior part. Two claws are present, of which only the shorter posterior one is fully preserved. This is tapering and not hooklike. The anterior claw is pre- sumably longer, but only the base is preserved, which shows that it is considerably larger at the base than the posterior one. The posttarsus is a short, padlike, round- ed structure that occurred in life under the claws. It is ornamented with a definite semicircle of large, rounded holes that very likely represented setal sites. The third leg of the right side is preserved in its entirety and in posterior view. Eight joints are present and appear not to be flattened, but mainly tubular. In Petrunkevitch’s drawings (1953, figs. 37 and 38) the Text-figure 47.—Species of Garnettius Petrunkevitch, 1953. See foldout inside front cover for explanation of abbreviations. A. Garnettius (?) sp. RSM 1957.1.5007 (Dunlop coll.). From the Lower Coal Measures (Westphalian A), Airdrie, Scotland. The basi- tarsus or tarsus of a large scorpion. The posterior striated spines recall the fossorial scorpion Garnettius hungerfordi (Elias). B-K. Gar- nettius hungerfordi (Elias). Holotype, USNM 125453a,b; part and counterpart. From the Upper Carboniferous (Pennsylvanian) Stan- ton Group, Missouri Series, 6 mi northwest of Garnett, Anderson Co., KS. B. Diagrammatic interpretation of the coxosternal-oper- cular pectinal area. Based on Text-figure 47D. Dotted areas not seen. C. Chelicerae and powerful chela of nght pedipalp. D. Enlargement of ventral features shown in Text-figure 47E. E. Showing ventral features only of coxal arrangement, and anterior dorsal aspect of the prosoma. F. Third right leg from dorsal side; had been turned over and shifted to the left side. G. Fourth leg showing basitarsus, tarsus and posttarsus. H. Third leg, showing the extraordinary distal spi- nosity. I. Joint 5 of the second leg, also unusually spinose. J. First right walking leg from ventral side. K. Distal part of second right leg, ventral view, preserved with original cuticle. FossIL SCORPIONIDA: KJELLESVIG-WAERING 117 118 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 first leg, with its characteristic serrations, was unfor- tunately placed in the position of the third leg. The same specimen is shown in Petrunkevitch’s figures 37 and 38 respectively, as in Text-figures 46A and 47J herein, but this leg does not occur where he shows it to be, but in the anterior part of the specimen. This dislocation must have occurred when Petrunkevitch assembled the various parts of the drawing, as he stat- ed, In studying the details for a drawing, I was forced to make camera lucida drawings at a magnification of ten diameters, piece them together on a large sheet of paper and then reduce the drawing to the present size photographically. The third leg, therefore, is relatively uncomplicated, although a fringe of numerous spines occurs on the distal part of the basitarsus and tarsus. At least two basitarsal spurs are present and the same number of tarsal spurs, but it must be admitted that this may only be a small part of the actual spurs (see Text-figs. 46A, 47¥,, H)- The tarsus 1s short, and the terminal claws are short, hooklike, and with a short, spinelike posttarsus. The entire end of the leg appears to be admirably suited for pushing, as the spines on the distal ends would open to give the leg added traction. This can be found in many of the burrowing insects (Phanaeus, Pinotus, etc): The complete fourth leg is known from the posterior side of the right leg. It has eight joints and 1s relatively uncomplicated. As in the previous leg, the joints ap- pear to have been mainly tubular, without noticeable compression, and it also had a very spinous distal end (see Text-figs. 46A and 47G). At least two basitarsal spurs are known and several serrations that might be spurs are present at the distal end. These spurs appear to fit into sockets of the tibia, and not in the interseg- mental skin below the tibia and above the basitarsus as in other scorpions.This fact was also noted in the first leg. The distal end of the basitarsus shows at least four spurs and these could well be in the intersegmental skin and thus would correspond to the tarsal spurs. All spurs seem to be the same size. The tarsus is very stout and rounded at the edges and has two small terminal hooklike claws, the anterior one being larger. The post- tarsus is short and is hidden from view by the presence of several short spines. A possible interpretation, taken from an imprint, is shown on Text-figure 47G. The coxosternal region, with operculum and pec- tines, is largely preserved in the original cuticle. This is shown in ventral view in Text-figures 47D, E. The region, however, is badly preserved, displaced to the right side, and the coxae are squeezed partly one above the other, whereas the opercular plates have been dis- located so that they rest on top of the sternum. The mass of cuticle is therefore very difficult to untangle, and without a doubt other interpretations are quite possible. In fact, the writer tried various interpreta- tions, but the most reasonable one is that given in Text- figures 47B, D. Because of the complexity of the pres- ervation, the coxosternal arrangement must be verified by other specimens as, admittedly, I have no over- whelming confidence in the correctness of the inter- pretation given here. It appears that the third and fourth coxae abut against a large pentagonal sternum that has a fringe of short spines along the posterior of the lateral margins. These spines also form a prominent fringe at the base of the sternum where they are stout and tri- angular. The other two coxae apparently lie anterior to the sternum and it is assumed that they meet at the median line and have maxillary lobes developed (not seen). The operculum is well preserved and consists of two elongate asymmetrical lacrimiform plates. Both plates have been pushed over the sternum and are slightly overlapping. This was clearly seen in the wet state under alcohol. The pectines are relatively small and, as seen under alcohol, are rather well preserved. The rachis is devoid of any segmentation or of the small rounded sclerites common to the Carboniferous scorpions. There are also no fulcra. The teeth, approximately 10 on each comb, are small, stout, and rather elongate-ellipsoidal. I have also noted this type of comb in the Lower De- vonian genus Branchioscorpio, n. gen. The preabdomen is composed of the usual seven tergites, and these increase in size progressively. Each tergite is bounded by the usual anterior transverse ridge, presumably raised. Two central carinae occur on the fourth to sixth tergites, whereas the seventh appears to have three carinae. The abdominal plates are preserved to the side and, although not completely exposed, there is no doubt that they are holosternous. Therefore, no stigmata are present; enough of the abdominal plates were exposed to determine this fact. The cauda is unusually thick. The first tergite has two dorsal carinae surmounted by elongate puncta- tions. The rest of the tergites, seen in lateral view, show two lateral carinae. The telson is very stout, short, and powerfully constructed. The vesicle is nearly round and devoid of a subaculear tubercle. The aculeus is only preserved as an impression, but appears to be of the usual shape, curved and tapering to a point. The integument of the holotype is mainly smooth and without noticeable granules or any other orna- mentation not already noted. It is, however, highly hirsute and numerous setae were found on many parts of the cuticle, although it was not possible to determine any pattern, as not enough of the original cuticle was preserved. The area around the fingers of the pedipalps was particularly setaceous. FossiL SCORPIONIDA: KJELLESVIG- WAERING 119 1cocm Text-figure 48.—Garnettius hungerfordi (Elias). Restoration based on the holotype, USNM 125453a,b. 120 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Measurements (in mm) of the holotype (USNM 125453a,b).— Prosoma: Length 16.7 Width at base 16.8 Width at lateral eyes: 16.8 Width at midsection: 14.2 Median eyes: Length: 1.0 Width: 0.6 Distance from anterior margin: 0.7 Distance from posterior margin: 14.6 Distance from lateral margin: 6.9 length width Compound eye: 4.2+ 2.0+ Pedipalp: Chela 17.0 Hand Ue) Sy) Free finger 5.0 Femur 6.0 Tibia 6.8 Tergite: No. 1 2.5 No. 2 3:7 No. 3 3.8 No. 4 5.0 No. 5 6.2 No. 6 7.4 No. 7 10.0 No. 8 10.2 No. 9 11.1 No. 10 12.6 No. 11 11.0 No. 12 12.0 Telson: Vesicle 9.7 6.7 Aculeus 4.0 (estimated) Greatest width of tergites at base of fourth tergite: 23.0 Total length of scorpion: about 125.0 Remarks. —This is one of the most peculiar scor- pions known. Obviously the first pair of legs is typical of a burrower as they are structurally similar to those found in many burrowing arthropods. The second to fourth legs, with their fringe of numerous side spines, also denote a burrower as these legs seem to be ad- mirably suited for pushing or giving the animal more traction along the burrow. It is the only scorpion in which such great differentiation occurs in the legs. Du- binin (1962) was correct in forming a family for this scorpion, although only a small amount of the infor- mation was then available in the literature and, part of this, erroneous. Certainly the many other unique characters of this scorpion, including the enormous chelicerae, could also have served to form a distinct family. The nature of the habitat of this scorpion can, of course, only be surmised. On the same slab, and same surface, are numerous articulate brachiopods, frag- ments of bryozoa, and pieces of crinoid stems. Ob- viously the specimen was buried in a marine environ- ment, possibly of lagoonal nature. As noted elsewhere, the pointed anterior margin is characteristic of water- dwelling chelicerates such as the eurypterids, and is considered to be used for the same function in scor- pions inhabiting a water environment. There is no question regarding the presence of stigmata—there are none, and the scorpion is clearly holosternous. There- fore it is considered to be a water-dwelling scorpion. Of course, it is not possible to determine whether or not the specimen was swept in from a fresh-water source. However, the specimen is complete and it is known that appreciable transportation is quite destruc- tive to these frail skins. The indications are that the scorpion lived in or near the burial site. I see no good reason why the scorpion could not have been a marine form, inhabiting a lagoonal environment, as suggested by the burial site. Garnettius (?) species Text-figure 47A Specimen data. —A single joint, probably the basi- tarsus or tarsus (No. 5 or 6) of a large scorpion, pre- served in dark gray micaceous shale, from the Lower Coal Measures (Westphalian A), possibly from the ‘Fern bed’’, 20’ above the Upper Drumgray Coal, Air- drie, Scotland, deposited in the Royal Scottish Mu- seum (Dunlop Collection), as RSM 1957.1.5007. The single joint is important because it reveals a row of movable, striated spines along the posterior part, and numerous movable spines at the end of the joint, such as are present in Garnettius hungerfordi (Elias), which was an obvious fossorial scorpion. If this rela- tionship is correct, the joint could be in the third or fourth leg and probably would correspond to the basi- tarsus, or less likely, the tarsus (5 or 6), and the movable spines would open to push the animal forward when burrowing, much as is found today in fossorial insects. Length of joint, incomplete, is 15 mm, and width is 4.3 mm. Superfamily SPONGIOPHONOIDEA, new superfamily Holosternina (?) with coxosternal area having last two pairs of coxae abutting the sternum, second pair partly abutting the sternum and each other and without maxillary lobes. First pair with short maxillary lobes and adjoining one another at the midline. Type family. —Spongiophonidae, n. fam. Remarks. —This is a highly primitive superfamily, which surprisingly occurs in the Triassic. The lack of development of maxillary lobes on the second pair of coxae is unique and would indicate a relict from the FossiIL SCORPIONIDA: KJELLESVIG- WAERING 121 Paleozoic. It is obvious that our knowledge of scor- pions is meager, otherwise a development such as in- dicated by Spongiophonus would not be such a sur- prise. No other superfamily known, regardless of the infraorder, has the characters listed for the coxosternal area. Family SPONGIOPHONIDAE, new family Spongiophonoidea having large, mainly triangular, sternum with anterior developed as a small subpen- tagonal area. The ornamentation is coarsely pustulose in the form of polygons. Type genus. —Spongiophonus Wills, 1947. Genus SPONGIOPHONUS Wills, 1947 1955. Spongiotarsus Wills. Petrunkevitch (in error—not Wills), p. 79: The generic description includes the characters listed above for the superfamily and family. Type species. —Spongiophonus pustulosus Wills, 1947. Geological range. — Triassic. Spongiophonus pustulosus Wills, 1947 Text-figure 110L 1947. Spongiophonus pustulosus Wills, pp. 133-135, pl. 3, figs. 13- 15; pl. 11, figs. 8, 13; text-fig. 53. 1953. Spongiophonus pustulosus Wills. Petrunkevitch, p. 37, fig. 47. 1955. Spongiotarsus pustulosus Wills. Petrunkevitch (in error—not Wills), p. 79. 1962. Spongiotarsus pustulosus Wills. Dubinin (in error), p. 432. Four specimens of this important and peculiar form are known. The holotype (SM 280), consisting of the superb specimen that forms the basis of the higher taxa described above, comes from a borehole at Messrs. Southalls (B’ham) Ltd. works at Charford, Birming- ham, England, whereas three tergites, SM 0286, 276, and 059, come from the Bromsgrove Quarry, near Bir- mingham, England. All are from the Triassic, Lower Keuper Sandstone Series, and are in the Sedgwick Mu- seum, Cambridge, England. Family PRAEARCTURIDAE, new family Spongiophonoidea of large size with very narrow, greatly-elongated, lanceolate sternum, with deep me- dian infolded sulcus; coarsely tuberculate dorsally; smooth ventrally. Type genus. — Praearcturus Woodward, 1871. Remarks. —Surprisingly the coxosternal arrange- ment is the same as the Triassic Spongiophonus Wills, 1947 (and so far unknown elsewhere), namely: only the first pair of coxae has maxillary lobes, which are poorly developed, and the second pair is without maxillary lobes, meeting at midsection and also abut- ting the anterior part of the sternum. The last two pairs of coxae abut the sternum. The completely different sterna relegate the genera to separate families. Genus PRAEARCTURUS Woodward, 1871 (new definition) Praearcturidae having very poor development in size of the first tergite (or considerable development of the pregenital tergite). Carapace anterior unknown, but with well-developed lateral cheeks and raised posterior area. Chela of pedipalp cultrate. Type species. — Praearcturus gigas Woodward, 1871. Geological range. —Lower Devonian. Distribution. —England,; Wyoming, U.S.A. Remarks. —The tuberculate ornamentation is very distinct and is responsible for the recognition of this genus in the Lower Devonian of Wyoming. Praearcturus gigas Woodward, 1871 (new description) Text-figures 49, 110A 1871. Praearcturus gigas Woodward, pp. 266-270, 5 text-figs. 1872. Praearcturus gigas Woodward. Woodward, p. 166. 1877. Praearcturus gigas Woodward. Woodward, p. 65. 1900. Praearcturus gigas Woodward. Kingsley in Zittel-Eastman, p. 668. 1922. Praearcturus gigas Woodward. Hennig, p. 144.'* 1969. Praearcturus gigas Woodward. Hessler, Treatise R(1), p. 393. 1969. Praearcturus gigas Woodward. Rolfe, Treatise R(2), p. 622, fig. 395. 1980. Praearcturus gigas Woodward. Rolfe in Panchen, p. 147. This interesting and spectacular scorpion is based on a number of large fragments from the Dr. D. M. McCullough Collection in the British Museum, and two other specimens in the collections of the Institute of Geological Sciences, Edinburgh, Scotland. All frag- ments represent different body parts, and although there is no documentation to settle the matter, I believe all may represent one individual. At any rate, all are sim- ilarly preserved in light-gray, calcareous micaceous sandstone, and are uncrushed. The dorsal side of the carapace is known only in part from the lectotype, BM(NH) I.534. Less than half of the carapace is preserved; only the posterior part. The parts that are preserved are in excellent condition. The lateral cheeks are present as raised structures sur- mounted with coarse tubercles. A median raised ridge, a structure not seen in other scorpions, separates the cheeks. At the base of the lateral cheeks (see Text-fig. 49A) are two shallow, but quite noticeable, large pits which are similar to but larger than those of the first '* Apparently, Kjellesvig-Waering got this reference from the Treatise on Invertebrate Paleontology, part R, Arthropoda 4(2) (1969, p. 618), but the entry does not appear in the bibliography of the Treatise and we have been unable to locate it. A.S.C. & K.E.C. 122 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 tergite. These shallow pits are interpreted as marking the site for muscles. The basal area of the carapace also is raised and surmounted with coarse tubercles, with remnants of the raised median ridge present. The posterior margin is emarginate with the genal angles rounded. The carapace is uncrushed and the lateral margins converge slightly anteriorly. A very good width mea- surement is possible by doubling the complete right side. Width at nearly midsection of the carapace is 100.0 mm; at the posterior margin 102.0 mm. The first tergite is known in its entirety from the lectotype. It is a very unusual first tergite, and the possibility of this being an internal tergite, namely, the pregenital tergite, is a possibility that must be consid- ered. The fact that the plate lies below the level of the carapace, is much narrower than the adjoining cara- pace, and lacks coarse tubercles (except for small ones at the lateral edges) greatly enhances the possibility. It cannot be a ventral structure, such as the prepectinal plate, which has been thrust upward during ecdysis, since the small tubercles on the margins of the plate show distinctly that the plate is a dorsal structure (these tubercles are dorsally located). Ifit is not the pregenital tergite, I am inclined to consider this to be the inside of the first tergite. The plate is straight anteriorly, rounded at the posterior angles, considerably narrower than the carapace, and rounded along the posterior edge. It is surmounted by a raised coarse transverse ridge at the usual position. The plate is entirely smooth except for small scattered tubercles at the lateral mar- gins. Intersegmental membrane is seen to join this ter- gite to the carapace. The anterior of the tergite, above the transverse ridge, grades gradually into the inter- segmental tissue. Length of tergite is 23.0 mm, the greatest width is 78.8 mm. Two well-defined large, but shallow, pits, closer together but nearly in line with the “facial pits”, are present in the anterior of the tergite, above the transverse ridge on either side of the mid- section. As stated above for the ‘facial pits’’, these pits are considered sites (dorsal) for muscle attachment (see Text-fig. 49A). The chelicerae are only partially preserved, namely, the first two joints (Ch1, Ch2 in Text-fig. 49A) and the dorsal side of most of the hand of the third joint. There are four joints in this scorpion, since it must be as- sumed that the chelicerae were normal and included the free ramus of the chela, which would be the fourth joint. The first joint is annular ventrally, and only the distal end is preserved, and was, no doubt, largely cov- ered by coxae of the first pair of legs. It is devoid of ornamentation. The second joint is also devoid of or- namentation, both ventrally and dorsally, and seems to be considerably longer on the ventral side than on the dorsal, where it seems to be a narrow band. The third joint is preserved, although known only from the dorsal side of the hand. The chelae are not preserved. All that can be said about this joint is that it is covered with very coarse raised tubercles, mainly round to el- liptical of variable sizes, some of which measure nearly 3 mm in greatest diameter. The pedipalp is preserved nearly complete in BM(NH) I.534, In.60443, and In.41782. In this pon- derous animal over 300 mm in total length, the basal or first joint (see Text-fig. 49A, P1) is of particular interest because it contains a large area on the dorsal side that is covered by fine grooves, resulting in ridges of cuticle, which undoubtedly is a stridulating organ. This is an unusually large organ, consisting of almost the entire dorsal surface of the first joint, or coxa, of the pedipalp. Stridulating organs are known in the Triassic Mesophonus perornatus Wills (1947, pp. 82- 84, pl. 8, figs. 1-5; text-figs. 4OA-—C, 41A-C) where they consist of spinules on the first three pairs of coxae. In Praearcturus gigas, the stridulating organs are grooves at the right angle from the inner edge. The grooves gradually become less pronounced in the pos- terior of the joint (P1). The location of the corresponding or opposite stri- dulation organ presents some problems as the known ‘“‘sounding boards” are dorsally located on the coxae. They could occur on the opposing pedipalp coxae or first two pairs of legs, which apparently can be brought to bear on the stridulating organs. There are no scleritic organs above the pedipalps. The grooves are very fine and consist of four grooves or three ridges to the mm. The entire first coxa measures 30 mm by 17 mm, but these are incomplete measurements. Regardless of where the opposing stridulating organ was located, it is interesting and perhaps significant, that adjacent to the coxa of the right side, described above, is a small piece of the coxa of the left side, which also shows the same stridulating organ. The close Text-figure 49.— Praearcturus gigas Woodward. Lectotypes in the collections of the British Musuem (Natural History). From the Lower Devonian (XI), Old Red Sandstone (Cornstones) of Rowlestone, Herefordshire, England. See foldout inside front cover for expla- nation of abbreviations. A. Dorsal side of posterior region of the carapace (Cp) of lectotype BM(NH) I.534, showing rugged pustulation, the first tergite (T1) and bases of prosomal appendages (as indicated) where anterior prosoma is lost. The very fine parallel grooves (much finer than drawn) on the first joint of the pedipalp apparently represent a stridulation organ. There are four grooves per mm. B. Pedipalp from BM(NH) In.60443, showing the dorsal side of the manus and P4 joint. The coarse nodes or granulation are apparently restricted to this side. At any rate, the ventral surface of the fingers (P5, P6) preserved as external ventral molds, is smooth with scattered elliptical pits. C. Counterpart of BM(NH) I.534, showing parts as labelled. D. Res- toration of the coxosternal area of Praearcturus gigas Woodward based on the lectotype, BM(NH) 1.534. ven FossIL SCORPIONIDA: KJELLESVIG- WAERING OS ostiG aS 124 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 proximity, in fact the two coxae are touching, means that the two organs could be brought together, thus effecting stridulation. If the stridulation organs are placed on the dorsal side of the coxa, it would mean that these coxae would have to be rotated so that the stridulating boards were opposite each other. This is possible even with the pedipalp coxae of present-day scorpions, where the side of the coxae are adjacent to one another. The type shown here is the oldest known, being Lower Devonian in age, but it is known in various living arachnids (see Lawrence, 1953, pp. 166-168, text-figs. 61, 62), in- cluding living scorpions. The trochanter of the pedipalp is only partly pre- served, showing that it is mainly smooth on the ventral side, with only a few coarse pustules anteriorly (see Text-fig. 49A, P2). The femur is known from specimen BM(NH) In.60445, from the same locality and horizon. It is a short stout structure, terete toward the junction with the tibia. The dorsal side is covered with the same round and elliptical coarse pustules. Some of the pus- tules, particularly the round ones, appear to be sur- mounted with a setal opening. The length through the midsection is 57.0 mm; the width at the base (junction with P2) is 27.7 mm; the width at the distal end (junc- tion with P4) is 31.0 mm; the greatest width is 41.5 mm. The rest of the pedipalp, tibia, hand and fingers, are known from BM(NH) In.60443 from the same locality and horizon. The pedipalp, except for the fingers, is preserved on the dorsal side; it is therefore the pedipalp of the right side. The tibia, preserved dorsally, is short and stout, cov- ered with large pustules (see Text-fig. 49B) and appar- ently with a deep sulcus in the central part. It is not complete, except lengthwise, but measurements are as follows: Length: 51.7 mm; greatest width preserved: 34.2 mm; greatest width (estimated): 45.0 mm. The hand is preserved only dorsally, but incomplete as to width. It is a very stout, massive structure, cov- ered dorsally with large pustules, either round or ellip- tical, but I have not been able to determine that any were piliferous. I do not rule out the possiblity, how- ever. Measurements are as follows: Length of the hand through midsection to the junction of the fingers: 81.0 mm; greatest width preserved: 53.0 mm; greatest width (estimated): 60.0 mm. The two fingers are preserved from the ventral side. The free finger (P6) curves inward, whereas the fixed finger is recurving. Both are smooth but sparsely pitted with shallow elliptical pits. The inner edges of the fin- gers are cultrate, in keeping with other Lower Paleozoic scorpions. The fingers are not complete, but the major part of each is present. Measurements are as follows: from the inner junction of the two fingers, the preserved length of PS is 65.4 mm; estimated length of P5 is 80 mm; preserved length of P6 is 76.5 mm; estimated length is 82.0 mm; width at approximate midsection: 12.3 mm. The coxosternal area is preserved (in the lectotype, BM(NH) I.534) so that all parts are seen mainly from the ventral side, but some, as for example the first pair of coxae, can also be seen from the dorsal side (see Text-figs. 49A, C). An accurate reconstruction, there- fore, is possible (see Text-fig. 49D). The sternum is unusual, being very elongate, lan- ceolate-triangular, with the third and fourth pair of coxae, as well as nearly half of the second pair, abutting against it. Although part of the upper midsection is not preserved, the forward edge is preserved as well as the entire lower half, which shows that the sternum expands slightly. Thus a rather accurate restoration is possible. The anterior half is very slender, shaped like a slender spear point or dagger, which expands pos- teriorly. The base is indicated by a short line on the right side of the sternum, which apparently marks the posterior limits (see Text-figs. 49C, D). The sternum is deeply infolded in midsection. The entire sternum measures 68.5 mm in length; the base is 14.4 mm in width and 11.8 mm in width at midsection. At the middle of the posterior half, the sternum is 9.9 mm in width. The midgroove of sulcus is 50.6 mm in length. The first pair of coxae is preserved on both ventral and dorsal sides. They are large thick structures that meet each other at the midline. Maxillary lobes are definitely developed, but are blunt and represent an early attempt at forming the oral tube. The coxa mea- sures 42.4 mm in greatest length, 50.7 mm in estimated width (47.0 mm are preserved), and 19.4 mm in width approximately in the center of the coxa. The maxillary lobe is 15.0 mm in length (greatest length) and 20.0 mm in width at the base of the lobe. The second pair of coxae is poorly preserved, but well enough to determine that there were no maxillary lobes developed and that the anterior half of each coxa meets at the midline whereas the posterior half abuts the narrow point of the sternum. The third pair of coxae is sufficiently well-preserved that the coxae can be reconstructed. The coxa, a flaring structure, abuts the sternum in the anterior half. Mea- surements are as follows: length: 50.0 mm along the anterior line; 60 mm and 31.0 mmat the juncture with the trochanter (see Text-fig. 49C, CHII and ITI1). The trochanter is ringlike, unusually short, and at midsection measures 11.2 mm in length and 30.6 mm in width. A rather close estimate can be made of the FossIL SCORPIONIDA: KJELLESVIG--WAERING 125 thickness at the junction with the femur (III2), which is 16.8 mm. The femur (III2) is preserved only at the basal part, but shows that the joint is rather elliptical in cross- section. It is 14.2 mm thick ventrally and 25.7 mm thick horizontally. The coxae of the fourth pair of legs are preserved almost in their entirety, both having the trochanters attached. The coxae are long, flaring, and abut the sternum from slightly anterior to the midsection to the base of the sternum. The coxae measure 57 mm along the anterior edge and 74 mm along the posterior mar- gin (both straight-line measurements, not following the curve). The anterior edges of the coxae are also seen from the dorsal side, which reveals that the anterior part tapers to a rather cultrate edge. The terminal end, at the junction with the trochanter, measures 36.3 mm. The trochanters (see Text-fig. 49C, IV1) are unusu- ally short; in fact, considerably shorter than those of the third pair of legs. The trochanter is ringlike and measures 6.7 mm in greatest length and 34.7 mm in greatest width. The femur of the fourth leg is only preserved as a fragment at its anterior end. The thickness is 26.0 mm in horizontal measurement and 19.6 mm in vertical measurement, thus elliptical in cross-section. The coxae, sternum, and ventral side of the tro- chanter (dorsal side not known) are completely smooth, contrasting with the extremely coarse pustules that characterize most of the dorsal side of this scorpion. The measurement of the entire coxosternal area from the tip of the maxillary lobes of the first pair of coxae to the end of the fourth pair of coxae is 155 mm, whereas the width at the anterior of the third coxae is 86 mm. The thickness of the scorpion, measured in the middle of the carapace to the sternum is 47.0 mm. Type information. —Lower Devonian, Old Red Sandstone (Cornstones) of Rowlestone, Herefordshire, England, collected by Dr. D. M. McCullough of Aber- gavenny. Remarks. —Praearcturus gigas Woodward was first considered to be a gigantic isopod. Rolfe (1969, p. R622) considered the fossils to be ““Doubtful taxa for- merly attributed to Arthropleurida’’.!° In Norway, in 1971, Dr. Leif Stormer asked me to study the specimen registered as BM(NH) I.534 for the possibility of its being a scorpion. For reasons (mainly the unusual ter- gite) not entirely clear to me now, I did not think the specimen represented a scorpion. It was due to Dr. Rolfe’s persistence, however, that we are finally able 'S Praearcturus gigas has long been a classic ‘“‘/ncertae Sedis”’. Hessler (1969), in the Treatise on Invertebrate Paleontology, consid- ered it to be a peracaridan malacostracan. K.E.C. to determine that this spectacular fossil is a gigantic scorpion. In 1979, Dr. Rolfe wrote to me and asked me to review the photographs (Rolfe, 1969, p. R622, fig. 395) in the Treatise on Invertebrate Paleontology. Reori- entation of the photographs clearly showed that the specimen BM(NH) I.534 represented a scorpion, a de- termination that was easily and conclusively proved by the photograph of the enormous pedipalp chela (specimen BM(NH) In.60443). Later Dr. Rolfe advised me that Dr. Stormer had identified these fossils as a scorpion as early as 1971. I gladly credit both my friends, Leif Stormer and Ian Rolfe for the final de- termination of this spectacular animal. In total size, Praearcturus gigas must have measured a full meter in length. It therefore rivals the Downton- ian Brontoscorpio anglicus Kjellesvig-Waering and the Lower Carboniferous Gigantoscorpio willsi Stormer. Comparison with these is superfluous as the cultrate pedipalp fingers of Praearcturus are sufficient to indi- cate an altogether different scorpion. A single tergite of a Praearcturus was found among the material from the Devonian of Cottonwood Can- yon, Wyoming. A name would have been given to this North American representative of the genus, for not only is the tergite perfectly preserved in its original flexible cuticle, but it is easily recognized by the dis- tinctive tuberculate type of ornamentation, had it been possible to locate the specimen again. It may still be found in the material at the Field Museum of Natural History in Chicago.'® Indeed, because of the unusual ornamentation, it is more easily recognized from small fragments than are the other scorpions present in the same horion: Branchioscorpio richardsoni, Hydroscor- pius denisoni and Acanthoscorpio mucronatus, which are preserved almost in their entirety. The specimen consists of a single tergite, very likely the second me- sosomatic, which was dissolved out of the limestone with the use of weak hydrochloric acid. There are further specimens of Praearcturus'’ avail- able. ‘© Dr. Derek E. G. Briggs, in November, 1983, located a slide of cuticle material from the Beartooth Butte Formation, Cottonwood Canyon, Wyoming, FMNH (PE) 26080 B, which might well be the specimen Kjellesvig-Waering referred to. The tergite is tuberculate, but only 8 mm across. A.S.C. '7 Dr. Ian Rolfe (written commun., April 13, 1984) quoted a letter from Kjellesvig-Waering, dated December 1, 1978. ‘*. . . with regard to Praearcturus, I forgot to mention Stormer’s Ringerike paper, 1934, pl. 12, fig. 2, “genus indet 1”’ is another Praearcturus and I remember it again in the paper I wrote on the eurypterids of the Downtonian at Stoke Edith, 1951, pl. 2, figs. 3 and 4. So we have two others— different species, but same genus. I will have to rework these. Nice! I'll start thinking about any specimens with smallpox [/.e. the tu- bercles of P.] in America.” A.S.C. 126 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Brontoscorpio anglicus '* Kjellesvig-Waering, 1972 1972. Brontoscorpio anglicus Kjellesvig-Waering, pp. 39-42, 2 text- figs. Specimens BM(NH) [n.31403, In.31404 and In.31405, the entire free finger and counterparts of an enormous scorpion from the Siluro-Devonian con- glomeratic sandstone referable to the Downtonian 1-9 bed, at the Birmingham Water Works, NNE of Eas- tham Bridge, Trimpley, Worcestershire, England. Infraorder MERISTOSTERNINA, new infraorder Branchioscorpionina with abdominal plates divided by a median suture, and retaining well-developed gill chambers and doublures, with the gill slit opening be- tween the doublure and the abdominal plate. Remarks. —The infraorder Meristosternina has been characterized throughout its known history by the small number of genera and species so far discovered. In contrast to the other infraorders of the Branchioscor- pionina, which are rich in genera and species, the Me- ristosternina are represented by only four genera: 77- phoscorpio from the Upper Devonian of New York State, Cyclophthalmus and Microlabis from the Car- boniferous of Czechoslovakia, and Palaeobuthus from the Carboniferous of Illinois, with only six species in all. These are too few forms to establish lines of evo- lution. Superfamily TTIPHOSCORPIONOIDEA, new superfamily Meristosternina with lateral schizochroal eyes. Type family. —Tiphoscorpionidae, new family. Remarks. —The trilobed ribbed gills probably should be considered a familial or even superfamilial char- acter, but not enough is known about these structures to merit their use as a major taxobase. The posteriorly- slit meristosternous plates are considered to be of in- fraorder importance. Family TIPHOSCORPIONIDAE, new family Tiphoscorpionoidea with pectines composed of a single undivided plate on each side. Type genus. — Tiphoscorpio, new genus. '8 Brontoscorpio anglicus is being listed after Praearcturus gigas merely because its age and unusual size suggest comparison. In 1972, Kjellesvig-Waering placed B. anglicus in the Eosocorpiidae, which seems meaningless, now that the ventral surface of Eoscorpius carbonarius is known. It probably belongs in Genera Incertae Sedis. A:S:G, Genus TIPHOSCORPIO, new genus Tiphoscorpionidae with smooth carapace, median eyes on rounded eye node placed anteriorly on cara- pace, and schizochroal eyes marginal, round and very large; no ornamentation on the carapace. No fulcra on the pectines. Tarsus with fringe of movable spines. Derivatio nominis.—tiphos (Gr.) = marsh or swamp + scorpio. Type species. — Tiphoscorpio hueberi, n. gen., n. sp. Geological range. —Lower Upper Devonian. Tiphoscorpio hueberi, new species Plates 5-8; Text-figure 50 This is one of the most interesting of all scorpions because of its unusual preservation. Although known only from a few fragments, these fragments fortunately represent important structures of considerable taxo- nomic value. The holotype consists of half a specimen, as only the carapace, ventral part of the abdomen, the seventh tergite, and a walking leg were recovered. The dorsal side of the abdomen and all of the cauda are missing, as well as the prosomal limbs, except for a tarsus. But of greatest importance is the remarkable preservation of a great many of the gills. Dr. Francis Maurice Hueber, in his study of the flora from the Onteora ‘“‘red beds” (lower Upper Devonian) of New York State, recognized the fragments as part of an arachnid (Hueber, 1960). Dr. John Wells of Cor- nell University advised me of these fragments, which were promptly and kindly lent to me by Dr. Hueber. Text-figure 50.—Tiphoscorpio hueberi, n. gen., n. sp. Holotype (USNM 252629). From the lower Upper Devonian (Lower Fras- nian), lowermost Onteora Formation (““Onteora red beds’’), Scho- harie Co., NY. Drawn from slides of macerations. See foldout inside front cover for explanation of abbreviations. A. Anterior of carapace (see PI. 5, figs. 1-3). The glossate process, which is turned under on the specimen, has here been restored, by dash-line, to its original position. The median eye sockets were drawn by reflected light, whereas the lateral eye details came from trans- mitted light. Postorbital margin inferential (dot-dash lines). B. The pectine of the right side showing an attached abdominal plate with doublure. One or two pectinal teeth (Pt) and a small portion of the pectinal plate (pp) can be seen. C. A left abdominal plate of a me- ristosternous pair seen from the dorsal side. D. Showing the ventral surface of the right plate of a paired meristosternous abdominal plate (AP) with gill slit showing at the base. Note the doublure. E. Showing the ventral surface of the left plate of a paired meristosternous ab- dominal plate (AP). This was very likely paired with the plate in Text-figure SOD to complete the “‘sternite’’. Note doublure and gill slit. See Text-figures SOF, G. F, G. Dorsal and ventral sides of part of the doublure shown in Text-figure 5OE, which bear small mu- crones. H. Ventral surface of a right meristosternous abdominal plate inclined downward. I. The dorsal (internal) side of one of a pair of meristosternous abdominal plates (AP), left side, showing a wide doublure and part ofa gill slit. J. Gill with attached abdominal plate (AP), which shows the doublure. K. Small outer anterior portion of an abdominal plate. FossIL SCORPIONIDA: KJELLESVIG- WAERING 127 Cc = fee dbl AP right AP right inclined downward ven ol 1 mm gs 128 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The specimens occurred in shale and there seem to be only two individuals, which are here designated holotype and paratype. The latter, although repre- sented by only one lamella of a gill, is approximately twice the size of the holotype. All specimens are mounted in Canada Balsam on glass slides. They were macerated out of the shale by Dr. Hueber using concentrated (48%) hydrofluoric acid and the residues were washed through ten changes of water (Hueber, pers. commun., Nov. 1, 1965). Each of the samples of shale macerated averaged 3 x 4 x 0.5 in. Only the anterior half of the carapace is preserved, but this is preserved in such detail that even the setae and eye facets of the lateral eyes are clearly revealed. The part preserved is sufficient to provide a good description (see Pl. 5, figs. 1-3; Text-fig. SOA). The anterior margin of the carapace is straight, with a well- developed median glossate process which, in the Can- ada Balsam mount, has been turned under the eye node. The median eyes are relatively small, located on a wide lacrimiform eye node that nearly touches the anterior margin. The lateral schizochroal eyes are com- posed of about 250 lenses, appearing as a pebbly sur- face on the lateral eyes. They are marginal, large and bulbous. The width of the carapace is probably 2.5 mm. The surface of the carapace is smooth, without any ornamentation except for small setae or open setal sites that seem to form a circular area around the eye node. These setae could therefore demarcate the sulcus, or the shoulders of the inflated cheeks of the cephalic shield that surround the eye node. The pectine is very poorly preserved, but enough is present to show that the pectines were composed of an unjointed plate, without rounded sclerites or fulcra, and with small, rather stout teeth (see Text-fig. SOB). Judging from the whole tooth preserved, it appears that 20 to 25 teeth were present on each comb. The pectine is preserved attached to an abdominal plate and its doublures (see Text-fig. 5OB). One small piece of a prosomal leg was preserved. This is the tarsus and shows a posterior fringe of strong movable spines. The abdominal plates are almost completely pre- served; unfortunately, the inner posterior parts on each are missing. However, the preserved parts reveal a meristosternous plate, bounded by a wide doublure. Each of these plates is split along the basal margin between the doublure and the abdominal plate face. The slit probably extends to the middle part. The inner parts are covered with small triangular scales pointing to the posterior (see Text-figs. 50F, G). There are eight parts (or halves) of the double abdominal plates pre- served (see Text-figs. SOB—E, H—K) out of the original ten halves. There is no ornamentation on the plates. The seventh tergite is preserved and this is consid- ered to be the ventral side. It is a triangular plate, with a posterior doublure and with four rather smooth ca- rinae. No ornamentation other than scattered setal hairs occurs. The remarkable gills are preserved in unusual con- dition. There are three types of gills, which are de- scribed as follows: 1. Rounded type (Pl. 6, figs. 1-4, 6-8; Pl. 8, figs. 1, 2, 4-7). These are oval bodies composed of white, spongy material in a single lamella with ribbing on the anterior parts. This ribbing consists of turned-in folds, some of which are anastomosing. The surface of the gills is covered by small subtriangular spines, rather indistinctly formed, but pointing toward the rounded area that does not contain the ribbing, thereby proving that the posterior of this type of gills is the rounded, non-ribbed part. There are a total of six of the rounded type. 2. Subrectangular type (Pl. 6, fig. 7; Pl. 7, figs. 1-7; Pl. 8, figs. 3, 8). These gills are long, straight along the inner edge, and bowed on the outer. Along the outer edge are at least three large, thick spines. These also show that the anterior part has the anastomosing ribs, whereas the subrounded posterior does not have the ribbing. The gill consists of a single lamella covered with small indistinct subtriangular spines that point to the rounded end, thus revealing the orientation of the gill. All are composed of white spongy material. There are five of these subrectangular gills preserved in the holotype, and one in the paratype (from the same ho- rizon and locality), which is twice as large as the five described above. The paratype (USNM 252630) mea- sures 3 mm in length (see Pl. 7, fig. 1). 3. Irregular type (Pl. 6, fig. 8; Pl. 7, figs. 6-7; Pl. 8, fig. 1). These gills are smaller and rounded with the same ribbing as described for the other gills, but either with large thick spines at the posterior end (PI. 7, figs. 6, 7) or a fringe of serrations as in Plate 8, figure 1. These gills are also composed of white spongy material, and are covered with short indistinct triangular spines which reveal that the coarse spines represent the pos- terior of the gill. The three types of gills are found associated or even attached together. The gills are also found associated with the abdominal plates. Although the preservation is perfect, it is not possible to develop any ideas on how all three types fitted together in the gill chamber. It seems clear from the ornamentation of small spines that the three types of gills occurred together, all with the rounded or spinous ends facing the gill slit, or posterior end of the abdominal plates. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 129 Type information. —Lower Upper Devonian (Lower Frasnian), lowermost Onteora Formation, in a dark, fine-grained shale lens at Quarry on northwest slope of South Mountain, Schoharie County, NY. 1.1 miles west of Schoharie Co.—Green Co. line (Livingston 7!’ quad.) at 74°16'30”E, 42°23'55’”N. Collected by Francis Maurice Hueber. The scorpion is associated with nu- merous plants and scales of a crossopterygian fish!? (see Hueber, 1960). The specimens are on deposit at the National Museum of Natural History, Washington, DC, as USNM 252629 for the holotype, and USNM 252630 for the paratype. Derivatio nominis. —Named in honor of Dr. Francis Maurice Hueber, who collected the specimens and suc- cessfully macerated them out of the shale. Superfamily CYCLOPHTHALMOIDEA Thorell and Lindstrém, 1885 (emend.) (nom. transl. Kjellesvig-Waering, herein) (ex Cyclophthalmidae Thorell and Lindstrom, 1885 not Petrunkevitch, 1913) Meristosternina with two pairs of coxae abutting one another in front of the sternum, the third pair abutting the sternum and the fourth pair abutting the opercular plates. Type family. —Cyclophthalmidae Thorell and Lind- stro6m, 1885. Remarks. —Petrunkevitch (1913) apparently was unaware that Thorell and Lindstr6m had established the family Cyclophthalmidae in 1885, and this error was carried throughout his monographs to the Treatise (Petrunkevitch, 1955, p. 77). The superfamily Cyclo- phthalmoidea, as envisioned by Petrunkevitch (1955, p. 77), was furthermore wrongly diagnosed, as he mis- took the basal segment of the chelicera (see Chl in Text-fig. 52) for the first coxa, thus his count of “‘three pairs of coxae in front of sternum’’. There are only two pairs of coxae anterior to the sternum. The main reason for retaining the superfamily Cy- clophthalmoidea within the Meristosternina is the presence of the detached meristosternous abdominal plate on the holotype. In this respect it is important to keep in mind the preservation of this scorpion. It is mainly uncrushed, but only the dorsal side of the abdomen is present on both halves of the specimen. ‘9 According to Dr. Harlan P. Banks of Cornell University (written commun., April, 1983) the Onteora red bed association in the “first greening” of the continents, also contains centipedes, a tarantula- like arachnid, and a mite that can be put in a Recent family (these are apparently Givetian in age). There is also an arthropod fragment that may be a silverfish (machilid), which would be the first-known record of the Insecta. This information is confirmed by Rolfe (1982). KeEC: In one half (Text-fig. 51), the carapace and the dorsal side of the abdomen are preserved as a counterpart, whereas on the other half, the ventral part of the pro- soma and the same dorsal side of the abdomen are preserved as the part with relief. The underside of the abdomen is not preserved. The determination of the meristosternous plate, however, needs further confirmation. My determina- tion was made because both halves of the plate are connected by a suture and are without carinae, both are rounded at the posterior corners, and both are of approximately the same width. Nevertheless, it could well be that this plate actually represents two tergites of the cauda. I do not consider this probable, but never- theless, it needs confirmation as this determination results in an important conclusion. Family CYCLOPHTHALMIDAE Thorell and Lindstrém, 1885 (emend.) Cyclophthalmoidea with no maxillary lobes on the first two pairs of coxae, or at most, only a slight lobate extension. Median eyes large, placed on a scutelliform eye node in the middle of the anterior half of the car- apace. No evidence of lateral eyes. Type genus. —Cyclophthalmus Corda, 1835. Remarks. —I can see no reason not to credit Thorell and Lindstrém with the family, albeit much emended, for their original description was far more correct than the later ones. Genus CYCLOPHTHALMUS Corda, 1835 (emend.) Cyclophthalmidae with carapace having a ring of large bosses surmounting the cephalic shield, and with pedipalp chela with round denticles, probably four deep along the edge, arranged in a row. Ventral side of pedi- palp smooth but very hirsute. Type species. —Cyclophthalmus senior Corda, 1835. Geological range. —Carboniferous. Cyclophthalmus senior Corda, 1835 Text-figures 51-52, 1111 1835. Cyclophthalmus senior Corda, p. 36, figs. 1-13. 1873. Cyclophthalmus senior Corda. Fri¢, pl. I, fig. 4. 1885. Cyclophthalmus senior Corda. Thorell and Lindstrém, p. 24. 1902. Cyclophthalmus senior Corda. Fri¢, p. 483. 1904. Cyclophthalmus senior Corda. Frié, pp. 66-68, pl. 7, figs. 1- 4; pl. 8, figs. 3-5; text-figs. 84-86. 1913. Cyclophthalmus senior Corda. Petrunkevitch, p. 33. 1949. Cyclophthalmus senior Corda. Petrunkevitch, p. 141. 1953. Cyclophthalmus senior Corda. Petrunkevitch, pp. 24-25, figs. 22-26, 128, 129. 1955. Cyclophthalmus senior Corda. Petrunkevitch, p. 77, text-fig. 44(4). 1962. Cyclophthalmus senior Corda. Dubinin, p. 429, figs. 1234, 1246. 130 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 It is important to determine how the holotype is preserved. It consists of two parts, one of which con- sists of the carapace, appendages and mesosoma in dorsal aspect. The counterpart shows the dorsal (in- side) side of the coxosternal region, the ventral side of the appendages, the dorsal side of the mesosoma and the first two tergites of the metasoma. The abdominal plate is preserved from the dorsal side. The holotype is a well-preserved scorpion of ap- proximately 100 mm in overall length. It is preserved so that all the structures are in an inflated state. The carapace is preserved on the counterpart, but P3 P4 AP2 ? Text-figure 51.—Cyclophthalmus senior Corda, 1835. Holotype, NMP Inv. 801. From the Upper Carboniferous (Westphalian B-C), Radnice Member of the “Lower Gray” Formation, Chomle, near Radnice, Czechoslovakia. The dorsal part of the two-part holotype slab. A dissociated meristosternous abdominal plate (AP2 ?) is exposed on this surface below. See foldout inside front cover for explanation of abbreviations. FossIL SCORPIONIDA: KJELLESVIG- WAERING HSH the entire outline cannot be seen. It is obvious, how- ever, that Petrunkevitch’s (1953, pl. 9, fig. 22) depic- tion of an unusually long, narrow carapace, is erro- neous. The most prominent feature is the well-defined lacrimiform, highly-elevated eye node, with a rounded anterior and emarginate posterior, and two round eyes with well-developed orbital ridges and a deep sulcus between them. It was not possible to see if this scorpion had lateral compound eyes, as the part where they would occur, if present, was covered on one side by the pedipalp base, and the carapace was incomplete on the other (see Text-fig. 51). The cephalic portion of the 1mm carapace is surrounded by a ring of prominent bosses. The anterior margin of the carapace protrudes, as is common in other Paleozoic scorpions. A description of the outline is not possible, but on the left side of the counterpart, an edge of the carapace is present. This would mean that the carapace is long, as in Mazonia, probably rounded at the anterolateral corner, and that the base is straight, as noted by the basal edge which is largely preserved (see Text-fig. 51). This would result in a carapace with the eye node well within the margin, about in the center of the anterior half, and the outlines would reveal a carapace slightly wider than long. Text-figure 52.— Cyclophthalmus senior Corda, 1835. Holotype, NMP Inv. 801. From the Upper Carboniferous (Westphalian B-C), Radnice Member of the ““Lower Gray” Formation, Chomle, near Radnice, Czechoslovakia. The ventral part of the holotype slab. See foldout inside front cover for explanation of abbreviations. 132 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The mesosomatic tergites show a well-preserved raised anterior transverse ridge, and increase in size to the posterior. The seventh tergite of the preabdomen does not show any carinae of prominence which would have been preserved if present. The imprint of the intersegmental tissues on the sides is particularly well preserved with all the longitudinal creases (see Text- figssoiliees2): The coxosternal region is seen from the inside and is well preserved. The sternum is small and subtrian- gular with the elongate elliptical opercular plates fitting at the base. The first two pairs of coxae meet at mid- section and each has very slight, short maxillary lobes developed. The third pair of coxae abuts the sternum and the fourth abuts the genital opercula. Petrunke- vitch (1953, pl. 9, fig. 24) correctly shows the sternum and genital opercula, but mistakes the third coxae for the fourth. He further mistakes the base of the left chelicera for the first coxa. Thus his restoration of the coxosternal area (1955, fig. 3) is wrong, and the diag- nosis of the superfamily as having three pairs of coxae in front of the sternum is also wrong. Also erroneous is Petrunkevitch’s determination of “lung slits’ (1953, p. 24, fig. 26) at the end of a sup- posed “sternite”. The latter is the fourth coxa, as can be proven by the arrangement of the legs. The so-called “lung slit” is nothing more than the open end of the coxa. The appendages are well preserved and those present on the counterpart reveal the ventral side, whereas the dorsal side is shown on the part. It is important to keep this relationship in mind. The chelicerae are also in an inflated state, so that it is possible to make a complete restoration of both dorsal and ventral edges of the fixed ramus (see Text-fig. 52). The lower row contains two large teeth that jut outward, nearly on a plane with the edge of the ramus. The tip clearly must end in a bifid point, although the upper point was broken off. The rest of the dorsal edge is armed with four small sharp teeth followed by a large central tooth and two smaller ones. Both sides of the chelicerae are hirsute, the ven- tral side more so. The pedipalps are preserved on both dorsal and ven- tral sides. These are massive structures having large, well-developed trochanter, femur and tibia. The tibia retains a row of granules, each of which is surmounted by a seta. The chela is long, with a bulbous hand and long fingers. Both fingers are covered with many large setaceous granules on the dorsal side. The ventral side is very smooth but extremely hirsute, as the setal bases are still present, embedded in the skin. The edges of the pedipalp are armed with round denticles, probably several thick, along the edge. Near the posterior part of the fixed finger, a row of crowded setaceous denticles or granules was noted. These occur close to the edge, but likely occurred at the edge, as they were on a part of the finger that formed a narrow platform close to, if not at the edge. The walking legs may show the usual eight joints (the posttarsus is covered). The first leg is complete and this, as well as the second, shows well-developed double tarsal and basitarsal spurs. The dorsal side re- veals rows of well-developed setal granules, but the ventral side is smooth and very hirsute. The entire ventral side of the chelicerae, pedipalps and walking legs is very hirsute, which is in marked contrast to the dorsal surface. It is therefore very likely that all of the ventral side was hirsute. The coxosternal area has been seen only from the dorsal side. The abdominal plates were not wholly preserved, except for a single detached one which is complete. It is meristosternous and probably represents the second abdominal plate. Only two segments of the cauda are preserved, the first two tergites in dorsal aspect. These reveal the pres- ence of three very strongly-developed superior carinae. Measurements are given by Fri¢ (1904) and Petrun- kevitch (1953, pp. 24-25). Type information. —The holotype, consisting of part and counterpart, is from the Upper Carboniferous (Westphalian B-C) Radnice Member of the Lower Gray Formation at Chomle, near Radnice, Czechoslovakia, preserved in white, hard, kaolinized, volcanic tuff with considerable quartz of magmatic origin, associated with many plants. NMP Inv. 801 in the National Museum, Prague, Czechoslovakia. Cyclophthalmus robustus, new species Text-figure 53 1949. Palaeomachus sp. Petrunkevitch, p. 138, pl. 48, fig. 156; pl. 55, fig. 180. 1953. Not Palaeomachus sp. Petrunkevitch. Petrunkevitch, p. 38 (incorrectly referred to as In.3976 instead of In.13976). 1962. Palaeomachus sp. Petrunkevitch. Dubinin, p. 433, fig. 1260. In 1949, Petrunkevitch referred the specimen, In.13976, consisting of a hand and fingers of a scor- pion, to Palaeomachus sp. Later in 1953 he stated that the specimen was not congeneric with Palaeomachus Woodward. In either case, the chela was incorrectly described, his having omitted diagnostic details that would have permitted good generic separation. The row of setal openings on the fixed finger shows that this is undoubtedly close to Cyclophthalmus, and it is here described as the only British representative of that well-known, but rarely encountered, genus. The holotype, BM(NH) In.13976, consists of an ironstone concretion, retaining dorsal and ventral sides of a well-preserved pedipalp chela of the right side. FossiL SCORPIONIDA: KJELLESVIG-WAERING 133 The main diagnostic feature of the chela is the thick row of setal openings on the inner half of the fixed finger, which was unnoticed prior to this study. This conspicuous row of setal openings is five to six open- ings wide at the base of the finger and gradually narrows to a double row at midsection. The fixed chela is broken here so that it is not possible to determine whether or not the double row continues or narrows further to a single row. The setal row probably also occurs on the free finger, but this is covered by the fixed finger be- cause of a scissorlike action. The hand is robust, about as long as the fingers, and no carinae are present. The fingers are relatively short, with the free finger falcate and incurving, and with the fixed finger recurved. The fingers are cultrate. Measurements in mm of the ho- lotype (BM(NH) In.13976) are as follows: hand length, 7.4; hand width, 5.3; free finger length, 8.2. Type information. —The holotype is from the Upper Carboniferous, Upper Coal Measures, at Coseley, Staf- fordshire, England, and is registered in the collections of the British Museum (Natural History) as BM(NH) In.13976. Derivatio nominis. —robustus (L.) = robust (referring to the overall robust aspect of the holotype chela). Remarks.—I agree with Petrunkevitch that this specimen is not congeneric with Palaeomachus, al- though it is difficult to understand why the two were ever considered to be the same genus in the first place. It differs greatly from the Czechoslovakian Cyclo- eye 1mm OO} a seta, bases Text-figure 53.—Cyclophthalmus robustus, n. sp. BM(NH) In.13976 (=Palaeomachus sp. Petrunkevitch, 1949, XXXVII, p. 138, pl. 55, fig. 180; 1953, p. 38). From the Upper Carboniferous, Upper Coal Measures, Coseley, South Staffordshire Coalfield, England. Showing two aspects of the pedipalp manus. The long concentration of setal bases along the fixed finger (P5) is very characteristic of the genus. phthalmus senior Corda on the species level in having short fingers with a long hand, just the opposite of the combination in the genotype. Cyclophthalmus senior shows some small denticles on the edge of the pedipalp, which are not present in the British form. This may require a new genus for the British form, but until more is known of the species, it is not considered judicious to separate them. It is, however, a distinct possibility that should be resolved in the future. Cyclophthalmus (?) sibiricus Novojilov and Stormer, 19637° 1963. Cyclophthalmus (?) sibiricus Novojilov and Stormer, pp. 84- 87, pls. Land II. Holotype No. 638/1 of the collection of Palaeon- tological Institute, Academy of Sciences, Leningrad, U.S.S.R. A single specimen lying on its back, with ventral sclerites to a large extent dislocated or removed exposing the impression of the dorsal surface; total length 32.5 mm. The holotype occurs in a dark argillite, probably from the lower part of the Stephanian, in a drill-core, borehole 3, depth 71 m, from the Tomju- sinsky District, Kemerov Region, Kuznetsky Basin, WESiSiRe Family MICROLABIIDAE, new family Cyclophthalmoidea with a pentagonal sternum and quadrate opercular plates. All five abdominal plates approximately of the same size, without reduction of the anterior plates. Type genus. — Microlabis Corda, 1839. Remarks.—The genus Microlabis had previously been included in the family Isobuthidae by Petrun- kevitch (1955, p. 78), but this is no longer possible as Microlabis is a Meristosternina, not a Lobosternina. Until something is learned about the maxillary lobes, the family must remain in the Cyclophthalmoidea. Genus MICROLABIS Corda, 1839 (emend.) Microlabiidae having an anteriorly-rounded cara- pace, with two large median eyes anteriorly located on a lacrimiform eye node, probably no facetted lateral eyes, very slender, short pedipalps with long, well-de- veloped chelicerae. All legs are unusually short and slender. 20 The only reference to this species found in Kjellesvig-Waering’s papers was this note to himself, ““For Scorpion opus; do not forget Novojilov’s and Stormer’s”. Although it was not mentioned in the description, the median line on the fifth abdominal plate, shown in plate I, fig. a, and also visible on photograph | in plate II, strongly suggests a meristostern. There is a slight indication of a similar median line on the second abdominal plate. A.S.C. 134 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Type species. —Microlabis sternbergii Corda, 1839. Geological range. —Carboniferous. Microlabis sternbergii Corda, 1839 Text-figure 54 1839. Microlabis sternbergii Corda, pp. 14-18, pl. 1. 1873. Cyclophthalmus sternbergii (Corda). Fric, p. 2, pl. I, fig. 3. 1885. Cyclophthalmus sternbergii (Corda). Thorell and Lindstrém, p. 24. 1904. Microlabis sternbergii Corda. Fric, pp. 69-70, pl. 9, figs. 1- 4; text-fig. 87. 1913. Microlabis sternbergii Corda. Petrunkevitch, p. 35. 1953. Microlabis sternbergii Corda. Petrunkevitch, pp. 23-24, figs. 215 27 1955. Microlabis sternbergii Corda. Petrunkevitch, p. 78, fig. 44(6). 1962. Microlabis sternbergii Corda. Dubinin, p. 428, fig. 1245. The holotype is preserved so that the entire meso- soma on the dorsal side has been removed except for a small amount on the right side. The ventral side is shown in its entirety from the inside. The carapace as well as the prosomal appendages are seen from the dorsal side. It is not preserved as Petrunkevitch (1953, p. 23) states, ““The specimen really lies on its back... .”’. It is a dorsal impression of both the dorsal and ventral sides. Thus the “ridges” described by Petrunkevitch (1953, p. 23) separating the “‘sternites” are in reality grooves (sutures). The carapace is very indistinct and only the central part is present. The outline as shown by Petrunkevitch (1953, fig. 21) does not exist. The median eyes, how- ever, are present, flattened and apparently elliptical and very large. Petrunkevitch states that it is not pos- sible to show what parts of the dorsal side were re- moved by Fri¢ (1904, p. 70), but it does not seem to have been much as the only fresh break on the spec- imen is in the region of the operculum. Nothing can be said about the coxosternal arrange- ment except that it is clear that the fourth pair of coxae abuts the genital opercula, the third pair abuts the ster- num and the first two pairs are above the sternum. The sternum is short but large and pentagonal in shape. The maxillary lobes are not known. The legs are relatively short and small and the pedi- palps are unusually slender and small. The chelicerae are large and long, hooklike and undoubtedly armed with teeth, although they were insufficiently well-pre- served for description. The chelicerae are composed of four joints, exactly as found in most other Paleozoic scorpions. The comb contains few teeth, probably about ten, and these are large and well inflated. The rachis is small, but the fulcra are comparatively very large. The venter is noteworthy for the meristosternous abdominal plates. There are five plates, very wide, and jointed by a suture at midsection. A transverse ridge occurs along the anterior edge and this ridge is also cut by the suture, which is groovelike. The last ventral preabdominal tergite is not cut by a suture and has no carinae. No ornamentation of any kind occurs on the parts of the tergite preserved to the right of the spec- imen nor on the abdominal plates of the venter. The dorsal side has been preserved on the right side of the specimen showing the right side of the abdomen. The first to sixth tergites are preserved and are pro- gressively longer toward the posterior; nothing more can be described. There is no ornamentation on the tergites. Type information. —The holotype is a single piece registered as No. 583 Sternberg collection, Inv. No. Text-figure 54.—Microlabis sternbergii Corda. Holotype, NMP Inv. 802 (Sternberg Coll. 583). From the Upper Carboniferous (Westphalian B-C). Lower Gray Formation, Chomle, near Radnice, Czechoslovakia. The ventral side seen from the inside, or dorsal side. Note the characteristic meristosternous abdominal plates (AP1- 5). See foldout inside front cover for explanation of abbreviations. FOssIL SCORPIONIDA: KJELLESVIG-WAERING 135 802 in the National Museum at Prague, Czechoslo- vakia, preserved in white, fine, volcanic tuff that has been kaolinized and “‘contains quartz of magmatic or- igin. The ash had been deposited in fresh water” (ac- cording to label). Small specks of carbonaceous ma- terial are on the same slab. Collected from the Radnice Member of the Lower Gray Formation of Upper Car- boniferous (Westphalian B-C) age at Chomle (a town situated 65 km southwest of Prague) in the neighbor- hood of Radnice, Czechoslovakia. Remarks. —Judging from the great size of the me- dian eyes, it might not be too great a speculation to suspect that this scorpion had not developed lateral eyes. The correlation of large median eyes and small lateral eyes, or vice versa, has been noted throughout this study. This is an important specimen that shows all of the abdominal plates in place, without having been jammed forward as occurs during ecdysis. Superfamily PALAEOBUTHOIDEA, new superfamily Meristosternous scorpions with the first and second pairs of coxae meeting at midsection anterior to the sternum, and with the third and fourth pairs of coxae abutting the sternum. Type family. —Palaeobuthidae, new family. Family PALAEOBUTHIDAE, new family Palaeobuthoidea with sternum pentagonal and with the first two pairs of coxae having well-developed, nar- row, maxillary lobes; second pair meeting at the me- dian line and with first pair on each side abutting the maxillary lobes of the second pair; last two pairs abut- ting the sternum. No lateral eyes. Type genus. —Palaeobuthus Petrunkevitch, 1913. Remarks. —Petrunkevitch (1913, p. 53, and con- tinuing through to the Treatise on Invertebrate Pa- leontology in 1955 [p. 78, fig. 44(7)]) erroneously di- agnosed the sternum as being triangular, and the coxae of the last pair of legs as abutting the opercula. This was taken from the holotype, which shows instead that the supposed triangular sternum was nothing more than the junction of the base of the second pair of coxae. The actual sternum, which was mistaken for the oper- cula, is clearly pentagonal with a deep cleft in the pos- terior midsection, and with both third and fourth pairs of coxae abutting the sides of the sternum. This is a rather advanced scorpion, considerably more so than the Cyclophthalmoidea (or the true Isobuthoidea). The determination of the sternum, and its correct relationship to the genital opercula and the coxae in Palaeobuthus distinctus Petrunkevitch, means that the genus can no longer properly be in the superfamily Isobuthoidea Petrunkevitch, 1913. Inasmuch as this is a meristosternous scorpion, with well-developed max- illary lobes of the first two pairs of coxae and having the last two pairs abutting against the large pentagonal sternum, it is essential that a new family and super- family be created to accommodate this genus under the Meristosternina. The genus Palaeobuthus Petrunkevitch, 1913 with P. distinctus Petrunkevitch, 1913, type species, reveals a coxosternal arrangement much like the living scor- pions, with a well-developed oral tube. Of course it differs greatly from living scorpions in being meris- tosternous rather than orthosternous. Genus PALAEOBUTHUS Petrunkevitch, 1913 (new definition) Palaeobuthidae with square carapace, small median eyes, anteriorly located; no lateral eyes; coarsely punc- tate; pedipalps finely denticulate on the edge; opercula subcuneate; telson very large, scimitarlike aculeus. Type species. —Palaeobuthus distinctus Petrunke- vitch, 1913. Geological range. — Pennsylvanian. Remarks.—The description of this genus is now based on three female specimens and one male and is, therefore, one of our better known fossil scorpions (see restorations, Text-fig. 58). Palaeobuthus distinctus Petrunkevitch, 1913 Text-figures 55-58, 111H 1913. Eoscorpius typicus Petrunkevitch (partim), pp. 39, 42. 1913. Eoscorpius granulosus Petrunkevitch (partim), p. 46. 1913. Palaeobuthus distinctus Petrunkevitch, p. 53, pl. 1, fig. 5; text- fig. 16. 1949. Palaeobuthus distinctus Petrunkevitch. Petrunkevitch, p. 136. 1953. Palaeobuthus distinctus Petrunkevitch. Petrunkevitch, p. 22. 1955. Palaeobuthus distinctus Petrunkevitch. Petrunkevitch, p. 78, fig. 44(7). 1960. Mazoniscorpio mazonensis Wills, pp. 290-300, pls. 49, 50, 51, figs. 4-6; text-figs. 10-13. 1962. Palaeobuthus distinctus Petrunkevitch. Dubinin, p. 428. Specimen I.—The holotype of Palaeobuthus dis- tinctus Petrunkevitch, 1913 (YPM 133), is preserved in a nodule consisting of two halves, neither of which had been properly cleaned. Most of the carapace and the coxosternal arrangements are revealed, as well as parts of the dorsal and ventral sides of the mesosoma and the dorsal side of the cauda. It is a poorly-pre- served specimen, consisting of an ironstone concretion showing most of the dorsal and ventral sides respec- tively on each part of the concretion. The carapace is subquadrate, forming a right angle at the genal angles, rounded at the anterolateral angles, and with the anterior nearly straight except at the mid- dle where a triangular glossate anterior process occurs. 136 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 16S 16 15 13 14 hKR—————H icm T12 ven 1mm A deep, narrow, median sulcus separates the carapace into two parts and continues anteriorly to form a Y-shaped sulcus on the carapace, with the anterior median ocellar node in the open part of the Y-shaped area. At least one median eye is barely visible, but it is so badly preserved that description is not warranted except to note that it is relatively small. Coarse granules occur on both sides of the narrow median sulcus, and T10dor | |] rete Text-figure 55.— Palaeobuthus distinctus Petrunkevitch. Specimen I. Holotype, YPM 133. Upper Carboniferous, Carbondale Forma- tion, lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. A, B. Part and counterpart. C. Ventral side of the first leg. D. Dorsal side of the left chelicera retaining the last two joints. also along the other parts of the carapace, although not so well nor so coarsely developed. In the pleural parts of the carapace the pustules are very small. Most of them seem to be setaceous as each is surmounted by a round hole indicating the site for a seta (see Text-fig. SSB). FossIL SCORPIONIDA: KJELLESVIG- WAERING 137 The well-preserved chelicerae (see Text-fig. 5SD) are very long and unusually large for a scorpion this size. Numerous denticles occur, but it has been impossible to determine whether or not they occur in two rows, as only an impression of the dorsal side is visible. The free ramus is particularly massive and very wide in contrast to the much narrower fixed ramus. The first walking leg (see Text-figs. 5SA—C), known almost in its entirety, has double tibial and basitarsal (pedal) spurs, whereas the claws are two short spines that prob- ably are slightly curved. The posterior spine appears longer than the anterior one (see Text-fig. 55C). I was unable to see a posttarsus, but this is probably present. The coxosternal region is of considerable interest (see Text-fig. 55A). The sternum is definitely pentag- onal, very wide, large, and with a deep median sulcus at the posterior half and continuing anteriorly as a suturelike division. The base is straight, the sides are parallel, the anterior forms an obtuse triangle and it is slightly wider than long. The genital opercula are com- posed of two long plates, but these have been com- pressed to an almost unrecognizable form. The first and second pair of coxae have long and well-developed narrow maxillary lobes. The second pair of coxae meets at the midline, in front of the sternum, whereas the first pair abuts against the maxillary lobes of the second pair of coxae, having been squeezed away from the midline. The third and fourth pair of coxae abut against the sternum. The “triangular sternum” that Petrunkevitch (1913) described is merely the filling in by sediments of the space between the bases of the second pair of coxae, and the long “anterior process” of the sternum is mere- ly the matrix or filling-in by sediments of the narrow space between the maxillary lobes of the second pair of coxae. The rounded genital plates, which Petrun- kevitch (1913) revealed, are merely the large sternum with the edges covered and therefore appearing as a rounded, sutured plate that was interpreted to be the genital plates. The genital opercula are so badly pre- served that description is not merited. Little can be said of the tergites except that all have well-developed anterior ridges. The ventral abdominal plates cannot be described as preservation is very poor, but they are definitely meristosternous, each half being rather square and joined by a median suture. The last preabdominal tergite shows two very prominently de- veloped ridges with knobby pustules arranged on the crest. Coarse pustules are present on the extremities of the lateral angles in no particular pattern, although there is a faint linear arrangement on part of the pus- tules. Some fairly coarse pustules occur along the center of the tergite. The cauda is known only from the dorsal surface, and only the first four tergites are preserved. The ter- gites show the cauda to be massive, with two crests developed on the dorsal surface. These two linear crests are widely separated and do not appear to be sur- mounted with knobby pustules, such as occur on the two crests of the last preabdominal tergite. Interseg- mental tissue from the lateral part of the mesosoma obscures part of the venter. The robust and wide mesosoma and metasoma sug- gest that this specimen is a female. Measurements (in mm) of Specimen I, holotype, YPM 133.— Total body length: 55.0 (estimated) Length of tergite No. 7: 4.0 Length of tergite No. 8: 3.9 Length of tergite No. 9: 4.0 Length of tergite No. 10: 4.1 Length of tergite No. 11: 4.5 Sternum length: 1.9 Sternum width: 2.0 Specimen II.—Paratype of Eoscorpius typicus Pe- trunkevitch, 1913 (YPM 127). The specimen is in two parts containing the dorsal and ventral sides of a well- preserved scorpion in a typical ironstone concretion. The carapace is badly crushed but reconstruction was possible. The anterior margin has a slight glossate process at the middle. On either side of the process, the anterior is slightly emarginate. The base was prob- ably straight in life with the sides slightly tapering an- teriorly. On the midsection is a narrow sulcus at the end of which occurs a small ocellar node, with the sulcus separating two small, round median eyes. The cephalic part of the carapace is covered with coarse pustules, some of which are rather tuberculate. The pleural areas of the carapace have a few scattered pus- tules of much lesser size (see Text-figs. 56A, B). The chelicera is known from the dorsal side (see Text-fig. 56C), revealing a double tooth arrangement at the apical end of the free ramus, much as in living Buthidae. It was impossible to determine whether the chelicera was made up of three or four joints. However, four joints appear to be the correct number as there is a small condyle present on the basal segment, which I interpret to be the second joint, for the articulation against another (basal) joint which presumably existed but is covered. The pedipalp is almost completely known, showing a stocky femur that may not be complete, followed by a tibia that is narrower and slightly longer (see Text- fig. 56A). Only the hand and the free finger were pre- served and reveal a crest or groove running along the length of both finger and hand. It is probable that this groove is only cuticular folding. No denticles were dis- covered on the edges of the fingers. The coxosternal arrangement is not entirely known. Only the base and one of the lateral margins of the 138 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 P6 Text-figure 56.—Palaeobuthus distinctus Petrunkevitch. Specimen II, YPM 127, part and counterpart. (A paratype of Eoscorpius typicus Petrunkevitch.) Female. Upper Carboniferous (Pennsylvanian), Car- bondale Formation, lower Francis Creek Shale, Mazon Creek, Grun- dy Co., IL. See foldout inside front cover for explanation of abbre- P5 siete viations. A. Dorsal aspect. B. Counterpart showing meristosternous ab- dominal plates. C. Last three joints of the right chelicera. P4 1cm P3 Imm FossiL SCORPIONIDA: KJELLESVIG- WAERING 139 sternum are known. This shows that it was probably large and hexagonal. The third and fourth pairs of coxae abut against the sternum. It is presumed, from the position of the parts of the first and second coxae, that the second coxae meet along the midline and should have long maxillary lobes developed, which extend to the anterior. From their position, the first coxae appear to abut against the maxillary lobes of the second, very much as in present-day scorpions. In fact, the entire arrangement would be very similar to the living scor- pions. The genital plates are long and subcuneate. The dorsal side is almost complete and shows that the tergites increase progressively in size toward the posterior. They are almost devoid of ornamentation, except on the seventh preabdominal tergite, which shows two obliquely-placed crests with very coarse knobby pustules surmounting each crest. The abdominal plates are well shown and are of the meristostern type. It is interesting to note that large round depressed areas occur in the middle of each of these plates. No structure is discernible within the round areas, but presumably they are the impressions of pouches or gill chambers for the large gills. Interseg- mental tissue from the lateral sides of the mesosoma has been preserved on the might side of the specimen. The first two tergites of the dorsal side of the cauda are present and show two crests surmounted by knobby pustules. The rest of the cauda is known from the ven- tral surface, which appears to have two crests without ornamentation. The last two tergites show what appear to be lateral crests, or very high carinae, increasing 1n size from the third to the fourth tergite, and presum- ably also increasing in the fifth caudal segment, which is not well developed, but shows a serrated edge. It is probable that these high crests are lateral crests on each side of the tergite, much as in some of the “‘fat-tailed”’ scorpions of North Africa and South America. The telson is almost complete and is unusually large, having a rather short vesicle and a very long, curved aculeus. I know of no scorpion, other than the British Eoscor- pius pulcher (Petrunkevitch), that has a stinger as large as this one. The anterior joints of the walking legs are almost unknown. However, parts of the terminalia of these legs were successfully exposed and show that the tarsus has two long claws. Double pedal (basitarsal) spurs also occur and, at least on the fourth leg, a single tibial spur occurs. However, preservation was not too clear at this point, and it is highly possible that double tibial spurs also occur. The walking legs themselves are not well preserved, except for the first, where measurements were possible. This leg retains two tibial and two pedal spurs. The terminations of the second and third walk- ing legs were successfully exposed and show that the scorpion had a tarsus that terminated in two long, slightly curved, claws. It also appears that the other legs, as shown by the fragment of what seemed to be the fourth walking leg, had two basitarsal and two tibial spurs, although in the fourth leg itself only one of the tibial spurs is preserved. From the wide and robust size of the body, this specimen appears to be a female. The posttarsus was not seen but undoubtedly should be present in better preserved specimens. Measurements (in mm) of Specimen II, YPM 127.— Total body length: 67.0 Total abdomen length: 20.8 Total caudal length: 35:2 Carapace length: es Median eyes: Distance from anterior margin: 0.6 Distance from posterior margin: S\s// Pedipalp: Length of chela: 11.6 Length of femur: 6.0 Length of tibia: 6.5 Length of hand: 3.8 No. 4 abdominal plate length: 5.8 No. 7 tergite length: 6.6 No. 8 tergite length: 4.8 No. 9 tergite length: Sbf/ No. 10 tergite length: St/ No. 11 tergite length: 5.7 Stinger length: 8.7 Aculeus length: 5:5 Specimen ITI.—Paratype of Eoscorpius granulosus Petrunkevitch, 1913 (YPM 130), shown in Text-figure Sh This specimen comprises only the mold of the dorsal side of a scorpion larger than the two previously de- scribed. It 1s preserved in an ironstone concretion, showing parts of the prosomal appendages, most of the prosoma and nearly all of the mesosoma. It adds but little to the description of the species, but the carapace and pedipalp are rather well preserved. The carapace is subquadrate, with base nearly straight and lateral edges tapering slightly anteriorly. The an- terior is mainly straight with a glossate process in front, which likely was bent downward in life. Two small, rounded, median eyes are present on a crushed mound directly behind the anterior process. A Y-shaped sulcus continues posteriorly from the ocellar mound. No lat- eral eyes were noted. A narrow doublure occurs at least on the base and lateral margins. The carapace is cov- ered with small tubercules. The appendages of the prosoma are preserved only in fragments, except for the right pedipalp, which is almost complete. The femur is about as long as the tibia, both of which are strongly developed. The hand and free finger are almost complete and reveal a strong chela with subquadrate hand and curved free finger. A 140 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 narrow crest or groove occurs along part of the length of the finger, which very likely represents an infolding of the chitin as this crest or groove does not extend to the hand as was noted in the previous specimen. No denticles were discerned on the edge of the finger. The almost complete mesosoma is preserved from the dorsal side. It shows that each tergite increases in size posteriorly, and that each is bounded anteriorly by a massive anterior ridge. The last (seventh) tergite of the preabdomen is preserved only as a fragment and, other than showing the typical anterior ridge, little else is preserved. The mesosoma is wide and suggests that this specimen was a female. Measurements (in mm) of Specimen IIT, YPM 130.— Overall length: 90.0 (estimated) Carapace length: 8.0 Pedipalp: Length of femur: 6.7 Length of tibia: 6.4 Length of chela: 13.2 Length of free finger: 8.5 No. | tergite length: eg No. 2 tergite length: 7191 No. 3 tergite length: 2.6 No. 4 tergite length: 3.0 No. 5 tergite length: 4.0 No. 6 tergite length: 4.4 \ \ Bes | \ 2 we 113 no 44 ap 4 LN) i Lesh Je WS : a v3 vi T4 fx ur S 7 Specimen IV. —The holotype of Mazoniscorpio ma- zonensis Wills, 1960 (BU 721A,B,C). The excellently preserved specimen which Wills de- veloped and minutely described (1960, pp. 290-300, pls. 49, 50 and 51, figs. 4-6; text-figs. 10-31) was de- scribed as a new genus and species, as at that time only highly erroneous figures and interpretations of the oth- er Mazon Creek scorpions were available in the exist- ing literature. A restudy of the holotype of Palaeobu- thus distinctus Petrunkevitch (see above), has revealed the carapace, true sternum, type of abdominal plates, and the details of the coxosternal region. The details of the morphology of this specimen, coupled with ad- ditional new information from the specimens YPM 127 and 130, unfortunately reveal that the superbly described Mazoniscorpio mazonensis Wills, 1960, is a junior synonym of Palaeobuthus distinctus Petrun- kevitch, 1913. My reasons for considering Mazoniscorpio mazo- nensis Wills as a junior synonym of Palaeobuthus dis- tinctus Petrunkevitch are as follows (it should be noted that they are from the same horizon and locality): 1. The two carapaces are identical. Both show the same surface contours and the same anterior, lateral and basal margins. The small, rounded median eyes, located on a small, anteriorly-placed mound, and the lack of lateral eyes, are identical in both. The peculiar coarse puStulation of the carapace also is similar. 2. Both lack ornamentation on the mesosomatic ter- gites. 3. The very coarse tubercles surmounting the two crests on the dorsal side of the last preabdominal tergite are identical. 4. The coxosternal arrangements of the two are the same. 5. The shape and relative size of the sternum are identical in both. 6. Both are meristosternous scorpions. 7. The appendages are the same in both. The only difference that I have been able to see is that the specimen described by Wills is more slender, having a mesosoma that is considerably narrower, at least on the abdominal plates (“‘sternites’’), than YPM 127, which shows these plates well. This is only a sex difference and would indicate that the holotype of M. mazonensis is a male, whereas YPM 127 is the female. I consider YPM 130 and the holotype also to be fe- males. Text-figure 57.— Palaeobuthus distinctus Petrunkevitch. Specimen III, YPM 130. (Paratype of Alloscorpius granulosus Petrunkevitch.) Female. Upper Carboniferous (Pennsylvanian), Carbondale For- mation, lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. FossIL SCORPIONIDA: KJELLESVIG- WAERING 141 The presence of denticles on the “‘cutting edges” of the fingers of the pedipalps was not noted in the YPM specimens, but was described by Wills in his specimen (which had been dissolved out of the ironstone matrix). The difference is easily explained because the two YPM specimens were preserved as internal molds of the in- side of the pedipalps, and, because the denticles are external structures, they would not be preserved in- ternally. In Wills’ preparation, only the original cuticle was used in the description of the minute denticles. Type information. —The holotype (YPM 133), YPM 127, YPM 130, and the holotype of Mazoniscorpio mazonensis Wills (BU 721A,B.C) are from the Penn- sylvanian Carbondale Formation, lower part of the Francis Creek Shale at Mazon Creek, Grundy County, Le Remarks. —It is possible to make a restoration of this species as all morphological parts are known (see Text-fig. 58). Infraorder LOBOSTERNINA Pocock, 1911 (nom. transl. Kjellesvig-Waering herein, ex Lobosterni Pocock, 1911) Branchioscorpionina with five gently to deeply bi- lobate abdominal plates with well-developed gill chambers and doublures; gill openings or slits between Text-figure 58.—Palaeobuthus distinctus Petrunkevitch. From the Pennsylvanian (Mazon Creek) of Illinois. Reconstruction of a female, based mainly on YPM 133 and YPM 127, shown in Text-figures 55 and 56, and BU 721, shown in Wills (1960). A. Dorsal side. B. Ventral side. 142 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the doublure and the abdominal plate. Remarks. —The Lobosternina, the second largest group in the fossil scorpions, was successful from the Silurian through the Triassic. The earliest form, Pa- laeophonus nuncius Thorell and Lindstrém, 1884, seems to have barely evolved from the Holosternina, with its gently-lobed protolobosternous abdominal plates. Before extinction, the infraorder expanded into at least 37 species, arranged in six superfamilies, 15 families, and 22 genera, and developed abdominal plates so deeply bilobed in Waterstonia airdriensis, n. gen., n. sp., from the Lower Coal Measures, as to al- most resemble the Bilobosternina. Superfamily PALAEOPHONOIDEA Thorell and Lindstrém, 1885 (new definition) Lobosternina with protolobosternous abdominal plates; legs very thick, short, eurypterid-like, with post- tarsus greatly developed as a single spine, and with very short tarsal spurs (ungues). Type family. —Palaeophonidae Thorell and Lind- strom, 1885. Remarks. —As explained above, this scorpion has been considered to be very primitive. This was due to its presence in Silurian beds (Wenlock) and, in partic- ular, to its eurypteroid legs. Whether or not this is a “‘primitive’’ scorpion or a highly specialized one is a moot question. The presence of protolobosternous ab- dominal plates does not support the “‘primitive”’ nature of this scorpion. For example, the presence of this scorpion in marine waters in an area with nearby com- mon reef development may indicate a specialization for a reefal existence. The holosternous Allopalaeo- phonus caledonicus (Hunter) of Scotland, however, has rather similar legs and leg terminations, but occurs in marine, dark-green to black shales. Both occur with common eurypterids. It is a difficult matter, if not im- possible, to determine scorpion habitats from the type of legs and leg termination. A glance at the leg ter- minations of some living crabs and their lack of cor- relation with their habitats, is sufficient to further il- lustrate the dilemma. Family PALAEOPHONIDAE Thorell and Lindstrém, 1885 (new definition) Palaeophonoidea with large facetted lateral eyes and short subtriangular cup-shaped coxae. Type genus. — Palaeophonus Thorell and Lindstrém, 1884. Genus PALAEOPHONUS Thorell and Lindstrém, 1884 (new definition) Palaeophonidae with quadrate carapace, median eye node anteriorly-located with sulcus dividing carapace into two large inflated lateral cheeks. Fingers of pedi- palps with cultrate inner edges and small denticles de- veloped only on extreme ends, both inner and distal. Type species. —Palaeophonus nuncius Thorell and Lindstrém, 1884. Geological range. —Silurian (Lower Wenlock). Distribution. —Gotland, Sweden. Remarks. —Allopalaeophonus caledonicus (Hunter) was previously included in this genus, but the two are not closely related. Al/opalaeophonus caledonicus is a holostern rather than a protolobostern and has no lat- eral facetted eyes. These are, of course, major differ- ences, although the two do have rather similar legs and general dorsal aspect of the body. Palaeophonus nuncius Thorell and Lindstrém, 1884 (new description) Plate 9; Text-figures 59-62 1884a. Palaeophoneus*' nuncius Thorell and Lindstrém, p. 984. 1884b. Palaeophonus nuncius Thorell and Lindstrém. Thorell and Lindstrém, in the Glasgow Herald, Dec. 19, 1884. 1885. Palaeophonus nuncius Thorell and Lindstrém. Thorell and Lindstrém, pp. 1-33, pl. 1. 1901. Palaeophonus nuncius Thorell and Lindstrém. Pocock, p. 296, fig. 1. 1904. Palaeophonus nuncius Thorell and Lindstrém. Fri, p. 63, text-fig. 78. 1913. Palaeophonus nuncius Thorell and Lindstrém. Petrunke- vitch, p. 13. 1949. Palaeophonus nuncius Thorell and Lindstrém. Petrunke- vitch, pp. 127-128, fig. 170. 1953. Palaeophonus nuncius Thorell and Lindstrém. Petrunke- vitch, pp. 5-8, figs. 1-5, 115. 1955. Palaeophonus nuncius Thorell and Lindstrém. Petrunke- vitch, p. 69, fig. 1, a. 1962. Palaeophonus nuncius Thorell and Lindstrém. Dubinin, p. 425, figs. 1221a, b, 1223. This famous specimen was described in great detail by Thorell and Lindstré6m in 1885, when optical in- struments were not well developed, particularly for incident light. As a consequence, many important mor- phological structures that are present on the specimen were not described. In 1953, Petrunkevitch restudied the holotype, but equally as many structures were again omitted. The holotype, because of its geological age and unique morphology, attracted world-wide atten- tion. This fame mainly depended on the fact that it was unquestionably accepted as the first known air- 21 Footnote in Thorell and Lindstrém, 1885, p. 9: ‘The name of the genus should be written, as it is here, Palaeophonus, not Pa- laeophoneus.” A.S.C. FossIL SCORPIONIDA: KJELLESVIG- WAERING 143 breathing, land-dwelling animal in the geological rec- ord. It is a rare Historical Geology text-book that does not still consider this scorpion in the same light and, because of this, it has possibly been figured in more text-books than almost any other fossil. This is curious and incongruous, as the premise that it was an air- breather was without basis, and that it was a land- dweller was equally without solid morphological foun- dation. The parts, or morphological organs, all of great value for diagnostic, taxonomic and morphological purpos- es, that have been totally unreported previously are as follows: 1. The presence of large facetted lateral eyes. 2. Protolobosternous rather than holosternous ab- dominal plates. 3. The four-jointed chelicerae. slight fold 3mm 4. The presence of two tarsal spurs—these are de- veloped in other scorpions into the ungues (claws). 5. The presence of two basitarsal spurs and very likely two tibial spurs on each leg. Other than the holotype (SRM Ar. 32235), which is the most important specimen, there are in the Riks- muséet in Stockholm, Sweden, several small fragments that add to our knowledge of this scorpion.These in- clude two fingers of the pedipalp, a complete proto- lobosternous abdominal plate, and three disjointed caudal segments. In view of the above major and hith- erto undescribed morphologic structures, as well as the undescribed specimens, it is necessary to redescribe the species completely, as well as to emend the genus and family. The taxonomic part has been done above and the new description of the species follows: Carapace subquadrate, rounded at the anterolateral Text-figure 59.— Palaeophonus nuncius Thorell and Lindstrom. Holotype (SRM Ar. 32235). From the Middle Silurian, lower Wenlock, Visby, Gotland, Sweden. One of the most famous fossil scorpions. See foldout inside front cover for explanation of abbreviations. 144 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 angles, gently but noticeably emarginate along the fron- tal margin, and straight at the base with a narrow pos- terior marginal rim. The cephalic area is divided by a narrow sulcus into two elevated or inflated pouchlike areas, and the basal part is again elevated in a wide, bowlike area that Petrunkevitch (1953, p. 7) mistak- enly assumed represented the covered or “hidden” first tergite, and that resulted in further “‘evidence”’ of his ill-fated theory concerning an eight-tergite preabdo- men (see Kjellesvig-Waering, 1969, pp. 171-176). At the rounded anterolateral angles are two large facetted lateral eyes surrounded by a narrow rim. These eyes are protuberant, and of the type that are present in the family Eurypteridae of the Eurypterida, namely, fac- etted with very small ommatidia. The small facets seem to be preserved on both eyes, but this is not entirely certain, as in Eurypterus and Baltoeurypterus, these facets are rarely large enough to be seen unless the skin is taken off and mounted on a slide and viewed with reflected light. However, the large lateral, bulbous, fac- etted eyes are definitely present and the very narrow rim enclosing them is clearly outlined. It follows that ommatidia have to occur. These eyes may be seen in the photograph (see Pl. 9) and in the drawings (see Text-figs. 59, 60A). The surface of the carapace is devoid of ornamen- tation, but most of the skin has been peeled away. It is known that the ornamentation could not have been composed of pustules of the type present on the che- licerae, or any kind of pustules, as these would have been preserved in the internal mold. The skin covering the carapace, however, could have had some fine setal openings around the ocellar node, which retains some skin, but no such setae or setal sites are present. Midway in the front of the carapace is a large ocellar node, oval in shape and well elevated. Eyes are not preserved sufficiently well to be certain of their pres- ence, but this is suggested by a rounded, small area on the left anterior part of the node. I do not doubt that this scorpion has well-developed median eyes. Previ- ously, it was thought to be blind, but now we know that large lateral, bulbous, facetted eyes do occur. It would be incongruous to have these large facetted lat- eral eyes and have a well-developed, elevated, ocellar node without median eyes. I consider it unreasonable to suspect such a condition, if not an impossibility. All scorpions with lateral eyes, simple or facetted, have median eyes. Scorpions without any type of lateral eyes always have well-developed median eyes. An excep- tion, however, is the modern troglobite scorpion, Ty- phlochactas of Mexico, which has neither lateral nor median eyes. But the morphology and habitat of Pa- laeophonus are quite different and not to be compared to the former, as it is anything but blind, having great facetted lateral eyes, and certainly is not a troglobite. The prosomal appendages are very well preserved. The chelicerae were described well by Thorell and Lindstr6ém, except that only three joints were noted. Actually there are four joints, as commonly found in Paleozoic scorpions. The first joint of the chelicera is short, probably cup- shaped as in other scorpions, although the base was covered in the holotype. The second joint 1s also short, with a well-developed longitudinal suture or flange on the dorsal side, as in some modern scorpions (see Het- erometrus). The hands of the chela also are short and this is followed by a nearly straight fixed dactyl. This dactyl has small teeth along the edge; in par- ticular, a larger, subterminal one that serves as a socket for the opposing free dactyl. The latter is very wide, curving and pointed at the end. When closed against the fixed dactyl, it overlaps at the end. Several large teeth are present on the edge, but their arrangement, except as shown in the drawing (see Text-fig. 60D), is not possible to know, since the entire chelicera occurs in a flattened state; thus it is not possible to discern the number of teeth on each side of the chela, ventral or dorsal. All parts of the chelicera are covered with small pustules, no doubt setaceous. The pedipalp is known in its entirety; it is a very stout, strong structure. The base of the pedipalp or coxa is completely covered, but the first joint, or trochanter, is beautifully preserved. This is largely triangular with a remarkable number of unusual, large, rounded pa- pillae (see Text-figs. 60A—C, G). These papillae are scattered about the dorsal surface (venter not seen) in no apparent arrangement. The papillae are also rounded in cross-section, but each has a stumplike protrusion and it is reasonable to assume that these are the bases of very coarse setae. These setae, in turn, occur in a rounded depression that is bordered by a very narrow rim, as in present- day trichobothria of scorpions. The trochanters, there- fore, must have been particularly hirsute (see Text-figs. 60B-C, G). Text-figure 60.—Palaeophonus nuncius Thorell and Lindstr6ém. Holotype (SRM Ar. 32235). From the Middle Silurian (lower Wen- lock), Visby, Gotland, Sweden. See foldout inside front cover for explanation of abbreviations. A. Right lateral compound eye. In stippled areas the cuticle is lost. B. Enlargement of a part of the pustulous area of the trochanter of the right pedipalp (P2). C. Cross-sections through pustules shown in figure B. No cuticle remains on the specimen. D. Details of the anterior carapace and chelicerae. E. Left eye. All the cuticle is peeled away, but evidence of some facets is left. F. Right pedipalp of ho- lotype showing the scissorlike overlapping of the free and fixed dac- tyls. G. Pustules on left brachium (P5) of holotype pedipalp. Cuticle intact. Pustules are slight holes with slight lip. H. A coxa from a topotype (SRM Ar. 32243). I. Protolobate abdominal plate, 8.4 mm wide, from topotype (SRM Ar. 32242). edge of edge of P1 0.5 mm area of overlap FossiL SCORPIONIDA: KJELLESVIG-WAERING 145 146 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The femur of the pedipalp is short and stout and covered with setal sites in the form of coarse rounded papillae as described above for the trochanter, except that they are far less common. The tibia also is short and stout and covered with sparse papillae. As expected from the heavily-constructed femur and tibia, the chela of the pedipalp is large and stout also. The hand is large, rounded anteriorly and rather gently rounded posteriorly, but devoid of the coarse papillae except for small setal openings (see Text-fig. 59). The fixed finger is curved backward, the edge is cultrate and the end round. Small setal openings in the form of small round holes occur on the middle of the dorsal side. The free finger is incurving, to fit the recurving fixed one. The two dactyls in the holotype, however, are seen to have a scissorlike action, and overlap along the edge for a considerable depth (see Text-fig. 60F). At least two short rows of denticles occur at the distal end of the free finger (see Text-fig. 60F). A free ramus (SRM Ar. 31818), or finger of the pedi- palp was found on a small slab next to the telson of Baltoeurypterus serratus (Jones and Woodward). This finger measures 7.5 mm in total length, thus about the same size as the holotype (see Text-fig. 61K). It is curved inwardly; the inner edge is cultrate except at the posterior end, near the junction with the articu- lation with the fixed finger where a small area of small denticles occurs. There are no other denticles present on the entire edge (except as noted for the holotype) and, inasmuch as preservation was perfect, there should be no question of further denticles. The midsection of the ramus, presumably the outer (anterior) part, has a row of small perforations, no doubt the sites for setae, and a very narrow ridgelike fold is present on the middle of the ramus, which likely was an inconspicuous carina. The free finger (SRM Ar. 31818) shows interesting ornamentation in the form of a wide row of setal sites composed of small rounded openings. These setal sites are in no particular arrangement, except that they are in a wide row, about four to five perforations across. The holotype has these also on the fixed finger. In the holotype, the fixed finger has covered the free dactyl, thus masking the row of setal sites on the anterior side of the free finger (see Text-fig. 59). The walking legs are spectacular in the sense that they were purported to represent very primitive legs, much like those found in some eurypterids. The com- parison to the eurypterids is, however, superficial and in reality these are specialized legs, very likely for movement or attachment over hard surfaces such as rocks or reefs. Previous to this study the legs had been described as being tubular, or cylindrical, and this is correct. They are terete, ending in a sharp long spine, which is the posttarsus. Thorell and Lindstrém (1885) recognized that the coxae of the legs were covered by the carapace, but Petrunkevitch (1953) stated that the coxae were visible under alcohol, and therefore the entire leg was present. This resulted, according to Pe- trunkevitch, in a leg with five joints and the coxa. This would indeed be a great departure from all scorpions. But furthermore, the so-called coxae were arranged in a linear series at the edge of the carapace and this would have resulted in a sternum of colossal proportions. I do not doubt that the sternum of Palaeophonus nuncius was large, perhaps as large as that of the Scottish A/- lopalaeophonus, or even the German Waeringoscorpio, but these enormous sterni would be mere trifles com- pared to what was indicated by Petrunkevitch for Pa- laeophonus. In reality, there are no coxae exposed, either seen under alcohol or in the dry state. The first leg of the left side reveals only seven joints—the coxa being covered. The ‘‘coxa’’ shown by Petrunkevitch (1953, fig. 1) is the third joint (13). In the other legs, the so- called coxae noted by Petrunkevitch are the second (112) joint in the second leg, the first (III1) joint in the third leg, and the first (IV1) joint of the fourth leg (see Text-figs. 61 A—D respectively). All legs therefore, have seven joints plus the coxa, and thus the implied gigantic sternum shrinks to a size comparable to those of early Paleozoic scorpions. The terete legs increase in thickness progressively posteriorly, so that the last or fourth leg is considerably thicker than the previous. An entire coxa is shown from specimen SRM Ar. 32243, and this shows it formed half an ellipse (see Text-fig. 60H). If all the coxae were like this one, then they would radiate outward from the sternum as in Proscorpius and Waeringoscorpio. All legs have short joints, but the curious factor in these legs is that the tarsus is relatively long, longer than that of Allopalaeophonus caledonicus and certainly much longer than that of Recent forms. Although both Thorell and Lindstré6m (1885) and Petrunkevitch (1953) showed one leg to have a single spur, all legs contain these movable spurs and are here described and figured in detail. The legs of Palaeo- Text-figure 61.—Palaeophonus nuncius Thorell and Lindstrém. Holotype (SRM Ar. 32235). From the Middle Silurian (lower Wen- lock), Visby, Gotland, Sweden. See foldout inside front cover for explanation of abbreviations. A. First left leg (I). (See Text-fig. 61E.) B. Second left leg (II). (See Text-fig. 61F.) C. Third left leg (III). (See Text-fig. 61H.) D. Fourth left leg (IV). E. Terminus of first left leg (I). F. Terminus of second left leg (II). G. Terminus of third right leg (III). H. Terminus of third left leg (III). I. Terminus of fourth right leg (IV). J. Fingers of the pedipalp; stippled area preserves original cuticle. From topotype (SRM Ar. 31591). K. Free finger of the pedipalp, with enlargements of indicated areas for pustular detail. From topotype (SRM Ar. 31818). FossIL SCORPIONIDA: KJELLESVIG- WAERING 147 covered covered IV IVS gS 148 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 phonus have the same basic plan as all scorpions have, namely, a long posttarsus with two well-developed tar- sal spurs underneath, two basitarsal spurs, and two tibial spurs. All spurs of the tibia have not been seen on all legs, but it seems safe to assume that each had double spurs. The tarsal spurs may have a short sup- plementary spur. These tarsal spurs are short, stout, covered with setal openings, and the anterior one is larger than the posterior, as in many other Paleozoic scorpions. The large posttarsal spine is straight and tapering, but no spines, such as occur in A/lopaleo- phonus on the ventral side, are present. On the third leg of the right side, Petrunkevitch (1953, fig. 4) shows a peculiar terete object that he considers to be a new organ. This is merely the dorsal projection of the distal end of the tarsus, directly above the ter- minal spine. There are no triple, rounded objects at the end of this projection, and they are here dismissed as fictitious. The four legs have very little ornamentation; most of it consists of very fine pustules or setal openings concentrated on certain local areas, which are shown on Text-figures 61 A-I. The sternum, opercular plates, and pectines were either covered or otherwise not preserved. Petrun- kevitch (1953, fig. 1) reveals a small, double, scutiform object at the posterior part of one of the middle body segments, which he states may be the operculum. This is a large ostracode, with both valves open, which lies under the body segment. This condition, of ostracodes being found under and inside cast-off arthropod skins, is a very common occurrence in eurypterid specimens. The first two tergites are mainly smooth, but with the usual narrow transverse ridge. The first is shorter than the second, indicating, as will be seen by the other tergites, that they progressively become longer poste- riorly. A very large bivalve occurs midway between the two parts of the scorpion, where it has broken apart prior to burial. I am unable to determine whether the bivalve is a pelecypod or one of the large ostracodes that are common at the Vattenfallet locality. This bi- valve has been indicated in the drawing (Text-fig. 59) and may be seen in the photograph (PI. 9). The fifth and sixth tergites are also mainly bald or without ornamentation, but with the transverse ante- rior ridge developed. They show the progressive lengthening mentioned above. The seventh tergite is triangular, with a large opening in the mid-area where the thick caudal segment fits. The anterior has a strong transverse ridge and is bordered by a narrow, flat area. There are definitely four carinae, each of which is sur- mounted by coarse, elongated pustules. Of greatest interest, however, is the underside of the abdomen, which has been mistaken as to type, form, and consequently, function. The second, third and fourth abdominal plates are preserved; this shows def- initely that the abdominal plates are protolobosternous (see Text-fig. 59). The plates are straight along the an- terior, without any transverse ridge, and without ap- preciable doublures. The posterior is deeply incurving, in an inverted subtriangular form. There are no stig- mata and of course none should have been expected. A single detached plate, perfectly preserved and previ- ously undescribed (SRM Ar. 32242) verifies the above protolobosternous nature of the abdominal plates (see Text-fig. 601). This is apparently the fifth abdominal plate, as it tapers posteriorly. Thorell and Lindstrém (1885, p. 14, fig. 13) de- scribed what they considered to be a slitlike stigma. Petrunkevitch agreed, in 1949 (p. 86) and 1953 (p. 5). Pocock (1911) on the other hand, stated that there were no stigmata present, but this was based on his study of ‘“Palaeophonus” (=Allopaleophonus) caledonicus (Hunter) from Scotland. There is a small, narrow in- dentation on one side of the third abdominal plate. However, this is only an indentation, possibly made Text-figure 62.—Palaeophonus nuncius Thorell and Lindstrém. Reconstructed dorsal aspect, based on the holotype (SRM Ar. 32235) and SRM Ar. 31818. From the Middle Silurian (lower Wenlock), Visby, Gotland, Sweden. FossiL SCORPIONIDA: KJELLESVIG-WAERING 149 by the edge of one side of a crinoid stem joint on the right side, and has no opening of any kind. As no protolobosternous or lobosternous plates have stig- mata, it is superfluous to discuss the matter further. The reason it is mentioned here is the insistence of many that the possibility of stigmata existed, thus con- venient evidence of the first land dweller. The cauda is preserved in its entirety, and the de- scription given by Thorell and Lindstré6m (1885) is correct and not to be improved upon. I merely add that it appears as if the crests present are double su- perior, double laterals, and double inferior crests. The lateral crests on the last tergite become fainter but not obsolete. The stinger or telson also has well-developed pustulose carinae. A fragment of the pedipalp of another specimen (SRM Ar. 31591), showing only the distal part, is re- corded here because it reveals that Palaeophonus nun- cius could be much larger than the holotype specimen indicated. The pedipalp (SRM Ar. 31591) is approx- imately twice as large as the holotype specimen, thus indicating that P. nuncius reached at least 125 mm in length (see Text-fig. 61J). Type information. —The holotype and other speci- mens reported here are from the famous Vattenfallet section: Middle Silurian, Lower Wenlock, top of Hég- klintslager beds, at the Vattenfallet, Visby, Gotland, Sweden, associated with many eurypterids, and a typ- ical marine fauna consisting of brachiopods, bryozoa, crinoids, ostracodes, etc. The eurypterids present be- long in the deepest water or most marine eurypterid assemblage (No. 1 of Kjellesvig-Waering, 1961, pp. 793-794), which is known as the Mixopteroidea-Pter- ygotina Assemblage. Remarks. —The holotype specimen has not suffered any appreciable deterioration, such as was reported by Petrunkevitch (1953, p. 5), at least not in areas where the internal molds cannot be used, as in each case, the imprint of the original cuticle is retained. As in the case with the famous Saaremaa eurypterids, the cuticle has a tendency to flake off, and all these specimens should be covered with a thin spray of a plastic fixative that will not obliterate any details, but will keep the original cuticle from being destroyed.*” There are three caudal segments in the collections of the Riksmuséet (SRM Ar. 32236, Ar. 32244, and 22 Like nearly all paleontologists, I have an aversion to covering fossils with shellac, etc., but in the case of specimens retaining orig- inal cuticle, it is absolutely necessary, and a single pass with an aerosol spray to cover the fossil fully, but with a thin layer of fixative, will preserve the fossil indefinitely. This is the method used with irreplaceable oil paintings, some as fragile as any chitinous fossil. The holotype has now been preserved, but should it be necessary to remove the fixative, this can easily be done by using the particular solvent recommended; in this case enamel thinner. E. N. K.-W. Ar. 32067), all of which are excellently preserved, but do not add to the description above. One other frag- ment of integument is present (SRM Ar. 32270), but this does not add anything to the above description. Palaeophonus (?) lightbodyi Kjellesvig-Waering, 1954 1859. Eurypterus ? possibly pygmaea Salter, p. 235, pl. 10, fig. 20. 1954. Paleophonus lightbodyi Kjellesvig-Waering, p. 485, fig. 1. A single pedipalp chela from the Downtonian beds at Ludford Lane, Ludlow, Shropshire, registered under No. 89424 in the Geological Survey Museum, London, England. Its precise taxonomic position is uncertain. Superfamily ANTHRACOCHAERILOIDEA, new superfamily Lobosternina, but with very primitive abdominal plates that have barely advanced to the bilobate stage (protolobosternous); first two pairs of coxae in front of sternum, but only the second pair meets at midline; first pair has been squeezed away from the midline; last two pairs of coxae abut the sternum. Type family. — Anthracochaerilidae, new family. Remarks. —This superfamily is remarkable for the presence of a coxosternal arrangement which is very like that in living scorpions, associated with a very primitive group of legs that are distinctly tubiform, much like the legs of the Silurian genera Proscorpius and Archaeophonus, and the Devonian Waeringoscor- pio, etc. The abdominal plates are also remarkable in being barely lobosternous, indicating the development of the lobosternous from the holosternous type (see Text-fig. 4F). The superfamily is therefore an interest- ing intermediate form, but still distinctly a Loboster- nina. The superfamily Pseudobuthiscorpioidea has the same basic coxosternal arrangement, but with fully- developed lobosternous abdominal plates. Family ANTHRACOCHAERILIDAE, new family Anthracochaeriloidea with first pair of coxae spat- ulate and projecting beyond the maxillary lobes of the second pair. Sternum pentagonal. Pectines without ful- cra, and with approximately 150 teeth on each comb. Type genus. —Anthracochaerilus, n. gen. Genus ANTHRACOCHAERILUS, new genus Anthracochaerilidae with long, cylindrical walking legs. Derivatio nominis. —anthraco (Gr.) = coal + Chae- rilus, a living scorpion genus with greatly-inflated first pair of maxillary lobes. Type species. —Anthracochaerilus palustris, n. gen., Nn. sp. Geological range. —Lower Carboniferous. 150 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Remarks. —As stated above for the superfamily, it is remarkable that so many advanced structures are found together with primitive tubiform legs reminis- cent of much older scorpions. Anthracochaerilus palustris, new species Text-figures 63, 112L 1949. Anthracoscorpio juvenis KuSta (partim). Petrunkevitch, p. 144, figs. 137, 175. 1953. Anthracoscorpio juvenis KuSta. Petrunkevitch (partim), p. 30. The holotype comprises a well-preserved scorpion of less than 18 mm in total length. A great part of the original cuticle is preserved, and where not present (indicated by stippled areas in Text-fig. 63A) the impressions of the edges of the structures are clearly discernible. Some original brown coloring has been preserved and in other parts the cuticle has been partly carbonized. Details of the structures are best seen un- der different angles and intensities of light, but, in par- ticular, either under alcohol, wet with glycerine, or dry. The holotype specimen is a ventral impression with only small patches of the dorsal side of the mesosoma showing through windows. It is preserved in grayish- black shale. Only the anterior half of the sternum has been pre- served, but there is an edge marked by cuticular ma- terial that indicates the base of the sternum, which is shown by a line on Text-figure 63A. A reconstruction of the sternum 1s therefore an easy matter. The sternum is long and pentagonal. The second pair of coxae meets at midsection, directly in front of the sternum, and has squeezed the first pair away from the midline. Max- illary lobes are well developed and extend to the an- terior. The first pair extends partly above the second pair anteriorly and is enlarged or spatulate, ina manner that characterizes the living Chaerilidae and many fos- sil forms such as Eoscorpius, Kronoscorpio, etc. The third and fourth pairs of coxae abut the sternum (see Text-fig. 63C). The legs are not entirely preserved, but at least the third and fourth walking legs are sufficiently well-pre- served to reveal several details of taxonomic impor- tance. The right fourth leg is preserved almost in its entirety, and is notable for being the type of cylindrical leg found in some of the Silurian scorpions such as Proscorpius osborni (Whitfield). The segments are therefore cylindrical and, for the most part, of the same length. Eight segments constitute the entire leg. The tarsus, unlike that in the Silurian scorpions, has been shortened and is less than half as long as the basitarsus. Well-developed tarsal and basitarsal spurs are present on each side of the corresponding segments. The claws are short, curved, and covered with small pits, very likely setaceous. The posttarsus is rounded and sub- triangular, and acts as a heel (see Text-figs. 63A and B). The opercular plates are relatively small and nearly round. The pectines are of unusual interest and have more teeth than hitherto known in any scorpion. They are elongate structures which, in life, would project well beyond the limits of the body. The basal lamella is probably composed of three elongate joints. The in- termediate lamella consists of small rounded sclerites, a surface that is described as perliform in scorpions such as Brachistosternus of the family Bothriuridae. Only a few of the rounded sclerites were seen, but it was noted that the first or basal segment of the basal lamella was scalloped on the inner edge in order to accommodate these sclerites. Fulcra are not present. The teeth are remarkable in being very regular in size, but extremely narrow and unusually numerous. Al- though the pectinal teeth were not complete, at least 135 were actually counted, and it is estimated that more than 150 were originally present. This is by far a greater count than present in any other scorpion. The teeth are so slender that they appear as hairlike stria- tions at the posterior edge of the pectine. Fulcra, of course, could not be present, as it is doubtful that the sclerites could be broken into such small segments as to fit in between the bases of the teeth. The abdominal plates are very well preserved, and Text-figure 63.—Anthracochaerilus palustris, n. gen., n. sp. Ho- lotype, BM(NH) In.39764. From the Lower Carboniferous (Lower Viséan), Calciferous Sandstone of the Glencartholm Volcanic Beds, River Esk, Glencartholm, Dumfriesshire, Scotland. See foldout in- side front cover for explanation of abbreviations. A. Ventral view. Cuticle missing in the stippled areas, but impres- sion present. B. Fourth left leg, dorsal view. C. Restoration of the coxosternal area, also showing the pectinal combs. The pectinal plate has not been seen. FossIL SCORPIONIDA: KJELLESVIG- WAERING 151 are of the protolobosternous type with a slight emar- gination at the center. Large rounded pouches, indicating large gill chambers, are noticeable on the abdominal plates, but these were insufficiently well- preserved for description. They are of the same type as noted in other Carboniferous scorpions. The basal segment of the preabdomen (No. 7) is preserved from the ventral surface and shows a well- developed anterior transverse ridge and two crests on the middle of the plate, both of which are surmounted by large tubercles. Large tubercles are also present on the edges and along the base (see Text-fig. 63A). The cauda, preserved in its entirety, is thick and has an unusually massive stinger. The first tergite is con- siderably shorter than the rest and all are marked with longitudinal ridges. It has not been possible to deter- mine the number of crests or ridges present on the cauda. The telson, or stinger, comprises a large vesicle with a strong stout aculeus. No ornamentation has been noted on the stinger. Measurements (in mm) of the holotype (BM(NH) In.39764).— length width Comb: 2.1 (maximum) 0.8 (maximum) Abdominal plate: 3 1.0+ 3.1 at middle 4 1.0 3:5 5) 1.0 3.1 Tergite: 7 1.4 2.3 at middle 8 0.8 15} 9 1e2, 1.4 10 12 Til 11 125, 1.1 12 ? ? Telson: 2.3 (incomplete) 1.0 (maximum) Type information. —Lower Carboniferous, Calcifer- ous Sandstone Series (lower Viséan), Upper Border Group, Glencartholm Volcanic Beds at River Esk, Glencartholm, Langholm, Dumfriesshire, Scotland. The holotype is registered as BM(NH) In.39764 in the British Museum (Natural History). Derivatio nominis. — palustris (L.) = marshy. Remarks. —Petrunkevitch (1949, p. 144) referred this specimen to the Bohemian Anthracoscorpio juvenis Kusta, an identification which is difficult to under- stand, inasmuch as the Bohemian specimen was thought to be preserved dorsally only, and the Scottish scorpion is a ventral impression. Equally difficult to understand is the description of the ventral side in the text when Petrunkevitch (fig. 137) illustrates only the dorsal sur- face and the pectines. It is not possible, as Petrun- kevitch later stated (1953, p. 30), to compare either the presence of eyes in Anthracoscorpio or the lack of eyes in the Scottish specimen, as the carapace is fully covered by the opaque cuticle of the underside. Noth- ing is known of the dorsal side of this scorpion. The same is true of Anthracoscorpio juvenis KuSta, an altogether different scorpion, belonging to a different infraorder, namely the Holosternina. Actually, only the ventral side of this species is known too (see Text-fig. 43A). Superfamily ISOBUTHOIDEA Petrunkevitch, 1913 (emend.) (nom. transl. Petrunkevitch, 1955, ex Isobuthidae Petrunkevitch, 1913) Lobosternina; coxae of first and second pairs of legs meet in front of the sternum; both meet at the midline; the third pair of coxae abuts the sternum and the fourth pair abuts the genital opercula. Type family. —Isobuthidae Petrunkevitch, 1913. Remarks. —Frié (1904) established the genus J/so- buthus, using Cyclophthalmus kralupenis Thorell and Lindstrém as type, and described the position of the coxae and sternum, although he did not use the coxo- sternal arrangement as a taxobasis in his classification. He also described the genus Eobuthus with E. rakov- nicensis Frié as type species, and the genus Feistman- telia with F. ornata Fri¢ as type species. These three genera were referred to Thorell and Lindstrém’s “‘Se- ries I’’—Anthracoscorpii, 1885, which was considered a family, and included the genera Cyclophthalmus Cor- da, 1835, Microlabis Corda, 1839, Eoscorpius Meek and Worthen, 1868, Mazonia Meek and Worthen, 1868, and Anthracoscorpio KuSta, 1885. Pocock (1911), using the “‘sternites” as a taxobasis, divided the known scorpions into two suborders, Lo- bosterni and Orthosterni. The genera Jsobuthus, Eo- buthus and Feistmantelia were referred to the Lobo- sterni, whereas Eoscorpius Meek and Worthen and Mazonia Meek and Worthen could not be classified as their ventral sides were then unknown. Petrunkevitch (1913) established the family Iso- buthidae on the basis of the fourth pair of coxae abut- ting the opercula, but ignored the structure of the “‘ster- nites’. Thus the family Isobuthidae consisted of Isobuthus (a lobostern), Palaeobuthus (a meristostern) and Eobuthus (a lobostern). Feistmantelia was assigned questionably to the family Eoscorpiidae Scud- der, 1884. The other genera (Cyclophthalmus, Anthra- coscorpio, Mazonia, and Microlabis) were referred to other families or were undesignated as to family. The 1913 treatment of the family Isobuthidae re- mained the same in Petrunkevitch’s monograph of 152 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 1949. In 1953, however, Petrunkevitch (pp. 18-19) decided that the genus Jsobuthus included the genera Eobuthus and Feistmantelia, having come to the con- clusion that the shape of the sternum was not a reliable taxobasis (p. 19) as “I prefer to disregard completely the shape of the sternum in view of its relative un- importance and unreliability in the case of fossils”. It is sufficient to dismiss that reasoning merely on the basis that preservation is not a taxobasis but a con- dition and has nothing to do with taxonomy. In the Treatise on Invertebrate Paleontology (1955), Petrun- kevitch essentially followed this reasoning, but estab- lished the superfamily Isobuthoidea to include the family Isobuthidae Petrunkevitch, 1913, which in- cluded the genera [sobuthus Frié, 1904, Microlabis Corda, 1839, and Palaeobuthus Petrunkevitch, 1913. The main contribution was the establishment of the coxosternal arrangement as a superfamily taxobasis; the importance of the coxosternal arrangement had been recognized by Fri¢ (1904) and Pocock (1911) in fossils. This taxonomic scheme suffered because the coxo- sternal relationships of many genera were unknown, but it also suffered from the abandonment of the use of other important taxonomic structures as taxobases, and, as shown above, by the misinterpretation of the morphologies preserved. The superfamily Isobuthoi- dea Petrunkevitch, 1913, however, was correctly de- scribed and established, namely: two pairs of coxae in front of the sternum, the third pair abutting against the sternum and the fourth pair against the genital oper- cula, but totally disregarded the structure of the “‘ster- nites’. At that time, the underside of Eoscorpius Meek and Worthen, 1868, was completely unknown. We now have firm knowledge of the coxosternal arrangement in Eoscorpius. The coxosternal arrangement of Eoscorpius is of the same type as that which was defined for the superfamily Isobuthoidea, namely that the two pairs of coxae meet above the sternum, with the third pair against the ster- num and the fourth pair abutting against the genital operculum. Actually, the coxosternal arrangement of Tsobuthus (with I. kralupensis as type species), the type genus of the family Isobuthoidea, was somewhat in- correctly understood. I have studied the holotype of Isobuthus in Prague, and I find that I am able to add to what Fri¢ had described in 1904. In that publication, Fri¢ described the sternum as rhombic. Petrunkevitch interpreted it as being truncato-oval, considering the holotype of Eobuthus rakovnicensis Fri¢ to be an [so- buthus, but the sternum has now been verified as being pentagonal. The first and second pairs of coxae occur anterior to the sternum; the third pair abuts the ster- num and the fourth pair abuts the genital operculum. The first pair has narrow maxillary lobes, whereas the second pair of maxillary lobes was developed to about half the length of the first pair of lobes (see Text-fig. 64A). This in itself would considerably change the meaning of the family, but not that of the superfamily Isobuthoidea. Because the coxosternal region of the Eoscorpiidae Scudder, 1884, is now known, the family Eoscorpiidae should be emended. This results in the Eoscorpiidae not belonging to the superfamily Scor- pionoidea Leach, 1815, but to the superfamily Iso- buthoidea Petrunkevitch, 1913. The families Isobu- thidae Petrunkevitch, 1913 and Eoscorpiidae Scudder, 1884, are therefore emended; the family Eobuthidae is described as new and the superfamily Isobuthoidea Petrunkevitch, 1913, remains practically the same, but it is restricted to lobosternous scorpions having a coxo- sternal arrangement as Petrunkevitch had defined it, but with the second maxillary lobes barely produced forward, or only to the middle of the first pair of coxae, which also meets at the midline. Family ISOBUTHIDAE Petrunkevitch, 1913 (emend.) Isobuthoidea with the first pair of maxillary lobes narrow, meeting anteriorly, and with the second pair of maxillary lobes extending only slightly beyond the middle of the first pair of maxillary lobes; sternum pentagonal, longer than wide. Type genus. —Isobuthus Fri¢, 1904. Remarks. —Petrunkevitch (1953, pp. 18, 19) con- sidered the genus Jsobuthus to be equivalent to Eo- buthus and Feistmantelia and accordingly considered the latter two genera as junior synonyms. I cannot agree that this is the case. It is obvious that the genus Eo- buthus has certain features that are present in /sobu- thus, but these are only on the superfamily level. The shape of the first pair of maxillary lobes, which are spatulate, and the lacrimiform sternum of Eobuthus would not only require a different genus but a different family from that of Jsobuthus. In this paper, therefore, the genus Eobuthus is considered to be a genus in good standing, and is designated the type genus of the family Eobuthidae, new family. Regarding Feistmantelia, little is preserved of di- agnostic value and, although the genus may be sepa- rated from Eobuthus and Isobuthus on the basis of the comb, it can only be referred to the Isobuthidae with considerable doubt. The great number of teeth on the very narrow type of pectines may be sufficient to con- sider this a good genus, although it might have been preferable to have referred the species questionably to one of the known genera in the first instance. Unfor- tunately it was not, but nothing is gained by further FosstL SCORPIONIDA: guesswork in assigning it to the synonymy of any par- ticular genus. The family Eoscorpiidae Scudder, 1884, differs from all other families of the superfamily Isobuthoidea in the combination of a very short pentagonal sternum with greatly expanded maxillary lobes of the first pair of coxae and with the developed maxillary lobes of the second coxae extending slightly less than half the length of the lobes of the first pair of coxae (see Text-figs. 75A, B). The coxosternal arrangement of the new fam- ily Eobuthidae is similar except that the maxillary lobes of the second coxae are a little longer, and the sternum is suboval in shape. The genus Palaeobuthus is placed in the new family Palaeobuthidae of the infraorder Meristosternina, not the Lobosternina, because the abdominal plates are meristosternous. In this new classification the genus Isobuthus may now be described as having a pentag- onal sternum and a coxosternal arrangement that is fully known and thus the family Isobuthidae Petrunk- evitch, 1913, is on a firm basis. As it is, because the maxillary lobes of the first and second pairs of coxae are known, it is possible to fit this well-known genus into a more properly described family. This ends the dilemma that was recognized by Pocock in 1911 when he used the genus Eobuthus extensively rather than Isobuthus. At the time, Pocock recognized that the coxosternal arrangement of Jsobuthus was not well known and that of Eobuthus had already been rather well described by Fri¢. Petrunkevitch followed his own erroneous definition and interpretation of Palaeobu- thus made in 1913, and continued to do so through his monographs of 1949, 1953 and 1955. I agree with Petrunkevitch that no stigmata such as Fri¢ shows for the Bohemian scorpions are present. As a matter of fact, important structures of the loboster- nous “‘sternites” (abdominal plates), which were not noticed before, are the well-developed doublures. These are shown very well in Eobuthus rakovnicensis and Isobuthus kralupensis. On the other hand, the pur- ported “lung slit” on Cyclophthalmus senior, which Petrunkevitch shows in text-figure 26 of his 1953 monograph, is the open end of the left coxa of the fourth leg. There are no stigmata on any of the Bo- hemian scorpions because all were lobosternous, mer- istosternous or holosternous, having abdominal plates, not sternites, and therefore did not have lungs. Genus ISOBUTHUS Fri¢, 1904 Isobuthidae with long pectines armed with 30 teeth on a side; coxae of pedipalps small, not forming oral tube walls or extending beyond coxal maxillary lobes. Type species. —Cyclophthalmus kralupensis Thorell and Lindstrém, 1885. KJELLESVIG-WAERING 153 Geological range. —Carboniferous. Remarks.*> Isobuthus kralupensis (Thorell and Lindstrém), 1885 Text-figures 64, 112D, 113C3 1873. Cyclophthalmus senior Corda. Frié (partim), p. 9, pl. 1, fig. ad ee 1885. Cyclophthalmus kralupensis Thorell and Lindstrém, p. 17. 1904. Isobuthus kralupensis (Thorell and Lindstrém). Frié (partim), pp. 70-72, text-fig. 88A. 23 The description of the genus was missing completely in Kjel- lesvig-Waering’s manuscript, and there were only a few fragmentary notes on J. kralupensis. The drawings were found. A.S.C. Text-figure 64.—Jsobuthus kralupensis (Thorell and Lindstrém). Holotype (NMP Inv. 837). From the Upper Carboniferous (West- phalian B-C), Radnice Member of the “Lower Gray” Formation, Cervena Hirka hill in Kralupy, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Ventral side, taken from a rubber cast. This is the equivalent of Frié, 1904, figure 88A. B. Obverse of holotype, showing terminus of pedipalp. 154 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 1911. Isobuthus kralupensis (Thorell and Lindstrém). Pocock, p. 16. 1913. Isobuthus kralupensis (Thorell and Lindstrém). Petrunke- vitch, p. 33. 1944. Isobuthus kralupensis (Thorell and Lindstrém). Lehmann, p. 178. 1949. Isobuthus kralupensis (Thorell and Lindstrém). Petrunke- vitch, pp. 136-137. 1953. Isobuthus kralupensis (Thorell and Lindstrém). Petrunke- vitch, pp. 19-20, fig. 124. 1955. Isobuthus kralupensis (Thorell and Lindstrém). Petrunke- vitch, p. 78, fig. 44(5). 1959. Isobuthus kralupensis (Thorell and Lindstrém). Wills, p. 266. 1962. Isobuthus kralupensis (Thorell and Lindstrém). Dubinin, p. 428, figs. 1228, 1243. Type information. — Holotype, Steinberg collection 575, NMP Inv. 837A and B in the National Museum, Prague. The specimen, in shaly sandstone with plants, is preserved from the ventral side only, and was col- lected by Jos. Stiaska in 1868 from the Upper Car- boniferous (Westphalian B-C) Radnice Member of the ‘*‘Lower Gray” Formation at Cervana Hirka hill in the city of Kralup, Czechoslovakia. Frié’s (1873, pl. 1, fig. F) illustration is quite correct, but his later line drawing (1904, text-fig. 88A), which was mainly taken from it, is not. Text-figure 64A is my interpretation of what is actually visible on the specimen. The lobosternous abdominal plates show a wide doublure on the margins but, of course, no sign of the stigmata which Fri¢ so prominently showed. The chela of the pedipalp is redrawn from the reverse of the holotype, as it is nearly perfect (see Text-fig. 64B). A reconstruction of the coxosternal region is shown on Text-figure 112D. Genus BOREOSCORPIO, new genus Isobuthidae having chelicerae developed with only small, even-sized denticles on the edge; very long basi- tarsal spines on fourth legs; posttarsus short, small and subconical. Carapace with median groove, Y-shaped sulcus with cordated median eye node anteriorly lo- cated. Lateral eyes unknown. Derivatio nominis. —boreas (L.) = northern + scor- pion. Type species. — Boreoscorpio copelandi, n. gen., n. sp. Geological range. —Upper Carboniferous. Remarks.—Differs from Eoscorpius Meek and Worthen in the lack of large teeth on the chelicerae, the different type of cutting edges on the chelicerae, consisting of the convex edge of the immovable finger and the concave movable finger, and in the presence of a very long basitarsal spine on the fourth leg. The carapace of Boreoscorpio has a median groove as in Palaeobuthus and not the elevated cephalic shield of Eoscorpius. The differences from Paleobuthus are readily apparent because the two genera belong to dif- ferent superfamilies. Boreoscorpio copelandi, new species Text-figure 65 1957. Eoscorpius sp. Copeland, p. 51, pl. 15, figs. 4-5. Holotype.—GSC 12778, part and counterpart, in the collection of the Geological Survey of Canada. Preserved in hard, greenish-grey shale in two parts, both of which retain parts of the original cuticle as well as imprints of the cuticular exoskeleton. The scorpion is nearly complete, but is preserved on its side. Pres- ervation of the cuticle that is left is good, but much has been lost and interpretation of all parts is not pos- sible. Nevertheless, a considerable part of this scorpion is known and description is possible for much of the anatomy. In fact, most of the diagnostic parts are pres- ent. The carapace is very poorly preserved as it has been squeezed from the lateral margins, but shows a lon- gitudinal median groove that separates anteriorly into a Y-shaped sulcus. In the open end of the Y is a cordate eye node with large elliptical median eyes that likely were round in an uncompressed state. A marginal rim surrounds the basal margin and the posterior lateral margins. No other eyes were noted but it is highly doubtful that preservation was sufficiently good for any to be present. Fragments of the appendages reveal several inter- esting details. The chelicerae (see Text-figs. 65C, D) were preserved entirely to reveal that, as in many other Paleozoic scorpions, they were composed of four joints. The first joint is bandlike, has a socket on the dorsal and ventral sides for the articulation of a condyle that occurs in the second joint, which also is bandlike. This gives considerable lateral movement to the chelicerae and is unlike that of present-day scorpions. The third joint of the chelicera comprises the hand and immovable finger. The hand is nearly square, wid- er than long. The immovable finger is straight on the outer edge and curved convex along the cutting edge. Small denticles, all of even size, occur along the edge. A socket occurs for the articulation of the free finger. The latter is stout, hooklike, larger than the immovable finger and armed with numerous, small, even-sized teeth along the concave cutting edge. There is a com- plete absence of large teeth. The concave cutting edge of the movable finger fits exactly into the convex edge of the immovable finger. The pedipalp of the right side is preserved nearly whole. The trochanter is poorly preserved, and only the edges on the anterior side are noted and these seem to be rounded. The femur is stout, about twice as long FossIL SCORPIONIDA: KJELLESVIG-WAERING 155 PS [Ss P5 Gh int i S€5, f Z VM ~ Tel Text-figure 65.—Boreoscorpio copelandi, n. gen., n. sp. Holotype (GSC 12778), part and counterpart. From the Upper Carboniferous, Pictou Group, presumably from the Stellarton Coalfield, Nova Scotia, Canada. See foldout inside front cover for explanation of abbreviations. A, B. Part and counterpart. C. Right chelicera from the ventral side. Stippled areas lack cuticle, but the impressions remain. D. Left chelicera from the dorsal side; very fragmentary, but shows fixed ramus. E. Fourth leg. 156 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 as wide, and approximately as long as the tibia, which is also stout.The hand is elongated, rather rectangular in shape, and the immovable finger seems to be straight, very long and with a convex inner edge. No teeth are present on the cutting edge. The entire pedipalp is covered with numerous setae. A fragment of what may be the first walking leg is preserved. This shows two basitarsal spurs and one tarsal spur, although very likely two tarsal spurs were originally present. A leg, which has been interpreted as the fourth (see Text-figs. 65B, E), reveals a very long basitarsal spur and a wide, slightly serrated, triangular tarsal spur. It is assumed that single spurs would occur on the other side, giving two basitarsal and two tarsal spurs. The claws and posttarsal joint are well preserved. These show that the posttarsus is very small, short and con- ical. The two claws are hooklike and of uneven size. The shorter, which is preserved entirely, has two large, wide spines or serrations on the inner part of the ven- tral arc of the claw. Little can be described of the coxosternal arrange- ment or the opercula. However, fragments present give some clues, even though the preservation is very poor. The sternum appears to be notched at the posterior edge and to have straight sides. This fragment would indicate that the sternum was pentagonal in shape, such as found in the Anthracochaerilidae. The opercula are oval, with the long axis aligned with the length of the scorpion. What appears to be the fourth coxa abuts against the opercula, thus restricting this scorpion to the Isobuthidae. The combs are also fragmentary, but enough are present to note that they are very long, have numerous small, rounded areoles or sclerites, and well-developed triangular fulcra. Larger, rounded sclerites occur to- ward the inner part of the median lamella, recalling the inflated basal segment of some of the living scor- pions (e.g., Tityus melanostictus Pocock), which de- note the female. It is likely that this specimen is a female as the overall aspect of the opisthosoma 1s rath- er wide and stout. At least 50 teeth occur on each comb. The individual teeth are long, slender and rather tu- bular. The tergites of the preabdomen are poorly preserved, but at least show the presence of a well-developed transverse anterior ridge on each one, and a progressive lengthening of each tergite from the anterior to the posterior. The seventh tergite apparently has two me- dian crests and is bordered laterally by coarse serra- tions. The abdominal plates are definitely of the loboster- nous type, but very poorly preserved. They show the usual large, rounded pouches. The cauda is very strongly constructed and appar- ently surmounted by two crests or carinae, at least on the first two tergites. The last three tergites seem to have two crests laterally, and the dorsal crests are high- ly arched. The telson or sting consists of a slightly curved aculeus with a rather elongate vesicle. No crests are present, but there are numerous setae on the ventral part. Setae are also present on the distal joints of the tergites. Apparently this scorpion was quite hirsute. Measurements (in mm) of the holotype (GSC 12778).— Estimated overall length: 30.0 Length of carapace: 333 Length of preabdomen: 12.8 (estimated) Length of cauda: 13.8 Length of telson: 2.8 Type information. —Upper Carboniferous, Pictou Group. No label with specimen but assumed to have been collected from the Stellarton Coalfield, Nova Sco- tia, Canada, by W. A. Bell in the interval between 1912 and 1914 (Copeland, pers. commun., 1966). Derivatio nominis. —The species has been named in honor of Dr. Murray Copeland. Genus FEISTMANTELIA Fric, 1904 (emend.) Isobuthidae (?) with long pectines, middle lamella with small, regular-sized areoles, well-developed fulcra and long teeth. Type species. — Feistmantelia ornata Fri¢é, 1904. Geological range. — Permian. Remarks. —See description of F. ornata Fric. Feistmantelia ornata Fric, 1904 Text-figure 66 1904. Feistmantelia ornata Fri¢, p. 75, pl. 11, figs. 1-5. 1913. (2?) Eoscorpius ornatus (Frié). Petrunkevitch, p. 35. 1953. Isobuthus ornatus (Frié). Petrunkevitch, pp. 21-22, fig. 13. The figures given by Fri¢ (1904, pl. 11, figs. 1-5) are essentially correct; those given by Petrunkevitch (1953, fig. 13) are incorrect as they show neither the correct shape nor the anterior transverse ridge. The tarsus and basitarsus are not divided as shown in Fri¢’s original figure. The only two abdominal plates, both loboster- nous, are shown in Text-figure 66. The smaller of the two is 14.8 mm wide, indicating a scorpion of about 90 mm in length. The genus probably should not have been described, but certainly there is little that could lead anyone to state that it is a junior synonym of Tsobuthus, as was done by Petrunkevitch (1953, p. 19). It could just as well be referred in present-day classi- fication to Eoscorpius, Paraisobuthus, Waterstonia, Eobuthus and others. All that can be stated is that it FossIL SCORPIONIDA: KJELLESVIG- WAERING LS7 definitely is a Lobosternina. Provisionally, it seems preferable to assign this species to the Isobuthidae. The Fri¢ drawing of the comb is correct, and this seems to be the only part that is diagnostic other than the two lobosternous abdominal plates. Type information. —Preserved in dark gray shale from the Permian Kounov Member (=Kounov Beds) of the Upper Gray Formation (Stephanian B in age) from Studnoves, in the neighborhood of Slany, about 30 km northwest of Prague, Czechoslovakia. Collected by Karl Feistmantel and deposited in the National Museum at Prague where it is registered as NMP Inv. 828 (CGH 1981, CGH 1984). Genus BROMSGROVISCORPIO, new genus Isobuthidae (?) with large gill-chamber opening lo- cated at the junction of the abdominal plate and the doublure, and bordered by thick triangular denticles, inwardly directed. Gill lamellae without spinelets. Derivatio nominis. —From the Bromsgrove Quarry near Birmingham, England. Type species. —Bromsgroviscorpio willsi, n. gen., n. sp. Geological range. — Triassic. Remarks. —It is not at present possible to determine the family of the Lobosternina to which this genus ee icm Text-figure 66.—Feistmantelia ornata Fri¢é. Holotype (NMP Inv. 828) (CGH 1981, CGH 1984). Two lobosternous abdominal plates. From the Permian (Stephanian B) Kounov Member of the Upper Gray Formation, from Studnoves, about 30 km northwest of Prague, Czechoslovakia. should be assigned, and it is referred questionably to the Isobuthidae until more material is known. No ab- dominal plate of known Paleozoic species of the Lo- bosternina has thick spines like those that occur on the gill-chamber opening of this Triassic genus. The spec- imen is unique among the other Triassic scorpions, both in the gross morphology and the ornamentation. This abdominal plate cannot be referred to any Triassic scorpion and, although it represents only the left half of the first abdominal plate, it cannot be compared to the so-called lobation of the first plate in Wills’ spec- imen 094 (1947, p. 33, text-fig. 13), which is much too poorly preserved. Furthermore, in my review I was unable to verify said lobation. There are no scorpions known in the fossil record in which a bilobate first plate is followed by holosternous abdominal plates. Bromsgroviscorpio willsi, new species 1947. Mesophonus bromsgroviensis ? Wills, pp. 35, 39, 110, pl. VI, fig. 9; text-fig. 14. The holotype (Wills, 1947, specimen No. 0163) comprises almost the entire left half of the first ab- dominal plate. I would not usually describe a new species on the basis of an abdominal plate, but in this case it is necessary because of the unique importance of the specimen. It is not only the only evidence of a lobosternous scorpion in the British Triassic, but at the same time it is the last known straggler of a major division of the Scorpionida, namely the infraorder Lo- bosternina, which was one of the dominant scorpionid groups in the Paleozoic. The specimen has been figured by Wills (1947, pl. VI, fig. 9, and text-fig. 14) and will not need to be figured again, as his text-figure 14 is completely ac- curate and the photograph adequate. The specimen reveals a wide gill slit at the central part of the lobe between the edge of the abdominal plate and the dou- blure. The gill slit occupies fully half of the posterior margin of the lobe and is bordered by very large tri- angular denticles. The gill appears to be composed of a single, very wide lamina. There are no spines de- veloped on the gill. For description of the ornamen- tation see Wills (1947, p. 39, text-fig. 14). Type information.—A single specimen from the Triassic Lower Keuper Rocks at the Quarry in Broms- grove, Birmingham area, England. Wills No. 0163 is designated as the holotype and is deposited in the Sedg- wick Museum, Cambridge University. Derivatio nominis. —This species is named because of the perfect preservation, making it easily recogniz- able again and therefore requiring a name, which has been given in honor of Dr. Wills’ great work. Remarks. — Although the holotype is compared to 158 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Willsiscorpio bromsgroviensis (Wills, 1947, p. 39), the differences are much too great, apart from the lobation. The coarse ornamentation along the anterior ridge is a feature that differs not only from W. bromsgroviensis but from all other known Triassic scorpions. Family EOBUTHIDAE, new family Isobuthoidea with pyriform sternum, and with great- ly anteriorly-expanded or spatulate maxillary lobes of the first pair of coxae. Opercular plates not separated, remain as a very small lobostern plate. Facetted lateral eyes present. Type genus. —Eobuthus Fric, 1904. Remarks. —As stated in the discussion of the su- perfamily Isobuthoidea, the genus Eobuthus was ex- tensively used by various authors until it was erro- neously suppressed by Petrunkevitch in 1953, who referred it to the synonymy of Jsobuthus. The shape of the sternum and the different maxillary lobes of the first pair of coxae are in each case sufficient to form not only a separate genus but also a family. It differs from the Eoscorpiidae in the shape of the sternum. The presence of a bilobed, but not separated, oper- cular plate (plates) is of particular interest as it is almost an exact copy of succeeding lobostern abdominal plates. The differences between the Isobuthidae, Eobuthi- dae and Eoscorpionidae are tabulated as follows: Isobuthidae Eobuthidae Eoscorpiidae sternum longer than sternum pyriform sternum wider than wide long second maxillary lobes extend ante- riorly to less than half the length of the first coxae second maxillary lobes extend ante- second maxillary lobes extend ante- riorly to more riorly to half the than half the length of the first length of the first coxae coxae first pair of maxillary lobes spatulate first pair of maxil- first pair of maxil- lary lobes long lary lobes spatu- and narrow late coxae of pedipalp large, form walls of oral tube and coxae of pedipalp large, hooklike, form walls of large chamber in front of the mouth and ex- tend past the coxal maxillary lobes coxae of pedipalp small, do not form oral tube walls and do not extend extend beyond beyond maxillary coxal maxillary lobes lobes Genus EOBUTHUS Fric, 1904 The characters of the genus have been discussed un- der the family Eobuthidae. Type species. —Eobuthus rakovnicensis Frié, 1904. Geological range. —Upper Carboniferous. Eobuthus rakovnicensis Fri¢, 1904 (new description) Text-figures 67, 112C 1904. Eobuthus rakovnicensis Fri¢ (partim), pp. 72-75, pl. 8, figs. 1-2; text-figs. 90, 92. 1911. Eobuthus rakovnicensis Frié. Pocock, p. 13. 1923. Eobuthus rakovnicensis Frié. Moore, p. 132, pl. 1, figs. 1, 2. 1949. Eobuthus rakovnicensis Fri¢. Petrunkevitch, p. 137. 1953. Isobuthus rakovnicensis (Frié). Petrunkevitch (partim), pp. 18, 20-21, figs. 18, 134. Not Eobuthus rakovnicensis Fri¢ (partim), 1904, p. 74, fig. 91; Petrunkevitch, 1953, p. 21, figs. 19, 123; Petrunkevitch, 1955, p. 78, fig. 46 (all refer to BM(NH) 1.2950 = Paraisobuthus prantli, n. gen., n. sp.). Specimen. —The holotype, Inv. 822 in the National Museum, Prague, consists of one specimen preserved from the ventral side in white, hard, kaolinitic volcanic tuff which contains quartz of magmatic origin, and associated with plants. The coxosternal region is perfectly revealed. The sternum is suboval, small, rounded anteriorly and rath- er truncated, but still rounded at the posterior where it meets two elongate, elliptical, genital opercular plates. The first pair of coxae is large, with well-developed, spatulate maxillary lobes that meet at midsection on the anterior half. Petrunkevitch (1953, pl. 7, fig. 18) apparently mistook these large spatulate lobes for the base of the pedipalps, but this is erroneous as can be shown by many other fossil scorpions (Eoscorpius, Waterstonia, Anthracochaerilus) and even in Recent scorpions (Chaerilus). He shows these coxae as narrow prolongations. There is a line, marking a fold, on the left side, which undoubtedly led to the wrong conclu- sion. It is not present on the right side. The second pair of coxae also has maxillary lobes and, although these meet at midsection, they only extend half the length of the first pair of maxillary lobes. The third pair of coxae abuts the ovoid sternum and the fourth abuts the opercular plates. The coxosternal arrange- ment is therefore typical of the superfamily Isobu- thoidea. Posterior to the opercular plates is a diamond-shaped organ that separates the two large combs. This is ap- parently the pectinal plate seen in Branchioscorpio, Gigantoscorpio and others, but preservation does not permit further description. The combs are large and have small, rounded areoles developed on the rachis, but I was not able to make out the exact divisions of the anterior lamella. Obviously the entire structure is elongated. Fulcra are developed and the teeth are large and elongate. At least 25 teeth occur and it is estimated that no more than 30 occur. The normal lobosternous abdominal plates are poor- ly preserved; possibly the third and fourth are present. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 159 A distinct doublure is noted, bordering the posterior part of each abdominal plate. The appendages are preserved in fragments and oth- er than to indicate the presence of the trochanters and stubs of the femurs, little may be said. The pedipalps are large, stout structures and, although poorly pre- served, it was noticed that a ridge occurs at least on the ventral side of the free finger, which has a double row of small granules, alternating, but not in a very uniform pattern, such as noted in Eoscorpius carbo- narius Meek and Worthen, where a somewhat similar row of setal openings occurs. No setal openings were noted in Eobuthus rakovnicensis, only granules. The edge of the free finger shows denticles at the base, but it is not known if these continue on the cultrate edge. The trochanters in particular are devoid ofany granules or setal openings and this seems to be the case with the rest of the pedipalps. Text-figure 67.— Eobuthus rakovnicensis Fri¢é. Holotype (NMP Inv. 822). From the Upper Carboniferous (Westphalian B-C), Radnice Mem- ber, Radnice Group, in a hard kaolinic volcanic tuff containing plants, Rakovnik, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Ventral aspect. B. Detail of the right pedipalp free finger. 160 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Type information. —Holotype is from the Upper Carboniferous (Westphalian B-C) Radnice Member, Radnice Group, in a hard kaolinitic volcanic tuff con- taining plants, Rakovnik, Czechoslovakia. Holotype, NMP Inv. 822. Eobuthus cordai, new species Text-figure 68 1873. Cyclophthalmus senior Corda. Frié (partim), pl. 1, fig. 1. 1904. Isobuthus kralupensis (Thorell and Lindstr6m). Frié (partim), pl. 10, figs. 1, 2, 7, 10, 11; text-fig. 88B. 1953. Isobuthus kralupensis (Thorell and Lindstrém). Petrunke- vitch (partim), p. 20, fig. 20. I have great difficulty in identifying some of the fig- ures in Fri¢’s 1904 paper with the specimens. However, this species is adequately founded on several speci- mens, which I studied in Prague. The holotype (Text- fig. 68A), registered as NMP Inv. 838, comprises a carapace, pectine, seventh tergite, abdominal plate, leg joint and cauda. Another specimen also referred to this species is NMP Inv. 834, considered a paratype, which comprises a well-preserved chelicera and two nearly complete legs (see Text-figs. 68 B—E). The longitudinal ribbing of the leg joint in NMP Inv. 838 fortunately furnished the identification for specimen NMP Inv. 834. This species is the only Czechoslovakian fossil scorpion so far known with longitudinal ribs on the legs. The holotype (NMP Inv. 838) reveals an almost complete carapace, which is squarish, slightly com- pressed at the midlateral margins, and bowed in front into a short, glossate, median process (see Text-fig. 68A). The facetted lateral eyes are well preserved; in- deed, these facets led Frié to assume that this specimen had three lateral eyes (1904, pl. 10, fig. 1 and text-fig. 88B), which error was perpetuated by Petrunkevitch (1953, fig. 20) whose drawing of the carapace was mis- takenly truncated at a point slightly behind midsection. The facetted lateral eye is preserved so as to show clearly the outline or marginal line of the anterior of the eye, enclosing several well-defined facets. Although less than half of the eye is preserved, an estimated 25 facets may not be far from the number present. There- fore, these are coarse facets, not greatly unlike those of Eoscorpius carbonarius Meek and Worthen. The median eyes are large, round, with well-devel- oped orbital ridges, located on a lacrimiform eye node that lies in the middle of the anterior half of the car- apace. A semicircular ridge of coarse tubercles is present in the middle of the carapace on each side of the posterior end of the eye node. The basal margin of the carapace is mainly straight. The pedipalp chela is preserved lying next to the carapace. The hand is narrow with two cultrate, falcate fingers, and with a ridge or carina running throughout the length of each finger. Both carinae are surmounted with coarse pustules. The bending inward or falcation of the fingers is of particular diagnostic value. Parts of the pectinal plate are preserved, enough to show that fulcra are present and the teeth are very numerous and of the long slender type. Thirty-six teeth are preserved, but this number is not complete as all were not preserved. Therefore, as in several other Czechoslovakian forms, such as Cyclophthalmus, each comb probably had about 50 teeth. A lobosternous plate, medially deeply-cleft, with very wide doublures is present, lying on top of the seventh tergite. The seventh tergite is important for diagnostic, generic and specific determination; it is tapering and probably preserved from the ventral side, where it re- veals four prominent rows of tubercles. The cauda is preserved on the lateral side, including the segments from the eighth tergite to the telson. The entire cauda is thick, has at least a lateral keel (pre- served on the eleventh tergite), and the telson has a very large vesicle, rounded, and the aculeus seems to be only slightly curved, if at all. This specimen is very likely a female. Specimen NMP Inv. 834, a paratype, adds to our knowledge as it shows the chelicera and parts of two legs, identified as the third and fourth (see Text-figs. 68B-E). The chelicera shows all four joints. The first joint is seen only as fragments with a dorsal flange. This che- licera, therefore, is from the left side. The bandlike second joint is nearly complete. The third joint has a relatively narrow hand with a fixed ramus showing several large denticles; the terminal is falcate and fits between the two terminal teeth of the free finger. The free finger is also armed with denticles, but the char- acteristics can be shown to better advantage by refer- ence to Text-figure 68B. The third and fourth legs are noteworthy for their good preservation. The third leg is preserved from the tibia to the distal end, showing that each joint was long and was reinforced by a narrow keel, at least the tibia was. The tibia has one spur, but very likely two were developed. The basitarsal joint has a flat, obtuse spine in the posterior, and one short spine in front. The wide flat spine was setaceous. The ungues or tarsal spurs are large, falcate, and with traction spines on the underside (see Text-figs. 68C, E). The posttarsus is very short, setaceous and triangular in shape, but inwardly curved. The fourth leg also has longitudinal keels on the femur, tibia, and basitarsus. The tibia does not reveal any spurs or spines, but this may be due to preservation FossIL SCORPIONIDA: KJELLESVIG- WAERING 161 Imm II16S leg joint Text-figure 68.—Eobuthus cordai, n. sp. Holotype (NMP Inv. 838); paratype (NMP Inv. 834). From the Upper Carboniferous (Westphalian B-C), Radnice Member of the Lower Gray Formation, Kralupy Hill, Cervena Huirka, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Holotype. The specimen is a slab with the dissociated fragments of one animal. Above, prosomal shield with enormous pedipalp claw. Middle, seventh tergite (+) and part of the anteriorly adjacent segment revealing the abdominal plate (AP) and part of a pectine. Bottom, distal postabdomen and dissociated prosomal leg joint. B. Paratype. Chelicera, showing the prominent serration of the dactyls. C. Paratype. Third and fourth legs. D. Paratype. Tarsus, posttarsus and unguis of the fourth leg. E. Paratype. Third leg tarsus, posttarsus and ungues. 162 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 (see Text-fig. 68C). The basitarsus has small spurs at the posterior distal end in the form of short, pointed spines. The tarsus is long with slight serrations on the posterior edge, and has large ungues with ventral trac- tion spines. The posttarsus is quite large, and undoubt- edly was heavily setaceous, as numerous setal openings occur (see Text-fig. 68D). Type information. —Both specimens preserved in shaly sandstone along with numerous plant fragments. Both were in a flattened condition and come from the Upper Carboniferous, Radnice Member of the so-called Lower Gray Formation, Westphalian B-C at Kralupy Hill, Cervena Hurka, Czechoslovakia (Ivo Chlupac, pers. commun., Dec. 3, 1970). The holotype (NMP Inv. 838) and paratype (NMP Inv. 834) are in the collections of the National Museum, Prague, Czech- oslovakia. Derivatio nominis.—Named in honor of A. J. C. Corda, the great Czechoslovakian paleontologist who described the first known fossil scorpions from the Chomle locality in 1839. Eobuthus holti Pocock, 1911 1911. Eobuthus holti Pocock, pp. 14-16, pl. II, fig. 2; text-fig. 1. 1913. Eobuthus holti Pocock. Petrunkevitch, p. 33. 1923. Eobuthus holti Pocock. Moore, p. 132, pl. 1, figs. 4, 5. 1949. Eobuthus holti Pocock. Petrunkevitch, p. 137. 1953. Isobuthus holti (Pocock). Petrunkevitch, p. 22. 1962. Isobuthus holti (Pocock). Dubinin, p. 428, fig. 1244. A single specimen from the Coal Measures of Sparth, near Rochdale, England, in the collection of F. Holt. Unfortunately, as of 1953, the specimen has been lost.*4 Eobuthus (?) species Wills, 1934 1934. Eobuthus (?) sp. Wills, pp. 101-103, 1 pl. 1949. Eobuthus rakovnicensis Frié. Petrunkevitch (partim), p. 137. 1953. Isobuthus rakovnicensis (Fri¢). Petrunkevitch (partim), p. 21. Specimen No. 1422 from 15 m above Seam B and 318 m below the surface, Oranje Nassau Mijn No. III, Heerlerheide, Netherlands. Type specimen lost as of 1979 inquiry.° Family EOSCORPIIDAE Scudder, 1884 (new definition) Isobuthoidea with pentagonal sternum wider than long, parallel sides, with greatly expanded or spatulate first pair of coxae; facetted lateral eyes present. *4 This is another species for which Kjellesvig-Waering left only a “do not forget” memo. From the shape of the sternum, it belongs to Eobuthus rather than [sobuthus. A.S.C. > Kjellesvig-Waering left no further notes. The only thing certain about the specimen, as Wills noted, is that it is a lobostern, from the Upper Carboniferous. A.S.C. Type genus. — Eoscorpius Meek and Worthen, 1868. Remarks. —The determination of the coxosternal ar- rangement, lobosternous abdominal plates and facet- ted lateral eyes in Eoscorpius carbonarius Meek and Worthen means that the family Eoscorpiidae Scudder, 1884, as hitherto described by Petrunkevitch, 1955, is erroneously diagnosed and can no longer be retained in the superfamily Scorpionoidea Leach. The descrip- tion of the family Eoscorpiidae Scudder, as emended by Petrunkevitch, is based on the coxosternal region of Eoscorpius typicus Petrunkevitch, which was, fur- thermore, incorrectly diagnosed by Petrunkevitch in 1913. From a restudy of the holotype of E. typicus, it was determined that the fourth pair of coxae definitely abuts the genital opercula and not the sternum. In the Treatise on Invertebrate Paleontology, Petrunkevitch (1955, p. 13, fig. 40(1)) used the coxosternal region of a British Pseudobuthiscorpiidae (BM(NH) I.1555) for that of Eoscorpius. Consequently it is necessary to emend the family Eoscorpiidae Scudder in view of the new morphological characters described here. The family Eoscorpiidae is therefore referable to the emended superfamily Isobuthoidea Petrunkevitch, 1913 (for comparison on the family level with Iso- buthidae and Eobuthidae, see ““Remarks” under family Eobuthidae). Genus EOSCORPIUS Meek and Worthen, 1868 (new definition) Eoscorpiudae with carapace subquadrate; median eyes placed on conspicuous elevated lacrimiform eye node near exterior margin. Two large raised cephalic ridges differentiate the elevated cephalic from the thoracic area of the carapace. Type species. —Eoscorpius carbonarius Meek and Worthen, 1868. Geological range. —Upper Carboniferous. Remarks. —In the Treatise on Invertebrate Paleon- tology (1955), Petrunkevitch referred the following genera to the family Eoscorpiidae Scudder, 1884: Eoscorpius Meek and Worthen, 1868 Alloscorpius Petrunkevitch, 1949 Trigonoscorpio Petrunkevitch, 1913 Buthiscorpius Petrunkevitch, 1953 Anthracoscorpio KuSta, 1885 Lichnophthalmus_ Petrunke- vitch, 1949 Typhlopisthacanthus Petrun- kevitch, 1949 Palaeopisthacanthus Petrun- kevitch, 1913 Compsoscorpius Petrunkevitch, 1949 Typhloscorpius Petrunkevitch, 1949 Europhthalmus_ Petrunkevitch, 1949 Garnettius Petrunkevitch, 1949 FOossIL SCORPIONIDA: KJELLESVIG- WAERING 163 Of these 12 genera, now that the lateral compound eyes and the true coxosternal arrangement of Eoscor- pius carbonarius Meek and Worthen are known, only the genus Eoscorpius Meek and Worthen, 1868, re- mains in the family Eoscorpiidae Scudder. A//oscorpius Petrunkevitch was incorrectly described on the basis of illusionary characters in the type species A. granu- /osus (Petrunkevitch) and it is merely a male of Eos- corpius carbonarius Meek and Worthen (see below). Trigonoscorpio Petrunkevitch, based on the species 7. americanus Petrunkevitch, 1913, is based on a single specimen which has been lost,*° but from a cast in the Peabody Museum at Yale it is possible to determine that the original was a lobosternous scorpion and, with- out a doubt, a male of Eoscorpius carbonarius Meek and Worthen. It would not have been possible for a scorpion to have such a narrow triangular carapace, as the carapace could not possibly uphold, much less fit, the massive appendages, chelicerae, pedipalps and four pairs of legs, under it. Trigonoscorpio Petrunkevitch is here referred to the synonymy of Eoscorpius Scudder. Buthiscorpius Petrunkevitch, 1953, is a Pseudobuth- iscorplidae, being a lobostern with a coxosternal ar- rangement completely different from Eoscorpiidae. Anthracoscorpio KuSta is a holostern and belongs to the Eoctonoidea. Lichnophthalmus Petrunkevitch is also congeneric with Eoscorpius. Typhlopisthacanthus Petrunkevitch is a junior synonym of Eoscorpius, a female, as proven by a study of the respective holotypes (see below). Palaeopisthacanthus Petrunkevitch has been shown by Vogel and Durden (1966) to possess stigmata and, therefore, is an Orthosternina and cannot remain in the Eoscorpiidae. Compsoscorpius Petrunk- evitch is also an Orthosternina. Typhloscorpius Pe- trunkevitch turns out to be a junior synonym of Eo- scorpius. Europhthalmus Petrunkevitch is known only from the dorsal side, was incorrectly described, and restudy of the holotype shows that this also is a junior synonym of Eoscorpius and identical to E. pulcher (Pe- trunkevitch). Garnettius Petrunkevitch has been re- ferred correctly to the family Garnettiidae by Dubinin (1962, p. 432) and a study of the holotype (see above) shows that this is a Holosternina, not a Lobosternina. Thus the family Eoscorpiidae Scudder, 1884, as emended here, retains only one genus, Eoscorpius Scudder, but this genus is very well known morpho- logically and its geographical range includes central North America and England. 26 Mr. B. S. Beall (written commun., 1984) reported that he had located the holotype of Trigonoscorpio americanus Petrunkevitch in the Museum of Paleontology at the University of Michigan, Ann Arbor, MI. It consists of part and counterpart of a siderite nodule listed as being from “Mazon Creek, Illinois’, and is catalogued as UMMP 7224. A\S.C. Eoscorpius carbonarius Meek and Worthen, 1868 (new definition) Plates 10-12; Plate 13, figures 5, 6; Text-figures 69-75, 112B, 113C2 1868a. Buthus? (Scorpio? Eoscorpius?) carbonarius Meek and Wor- then, pp. 22-24. 1868b. Eoscorpius carbonarius Meek and Worthen, pp. 560-562, with fig. [sic]. 1883. Eoscorpius carbonarius Meek and Worthen. Peach, p. 407, pl. 23, figs. 23, 23a, b. 1904. Eoscorpius carbonarius Meek and Worthen. Frié, p. 76, fig. 95. 1913. Eoscorpius carbonarius Meek and Worthen. Petrunkevitch, p. 37, pl. II, fig. 6. 1913. Eoscorpius typicus Petrunkevitch (partim), pp. 39-43, pl. I, figs. 2-4; text-figs. 5-7. 1913. Eoscorpius granulosus Petrunkevitch (partim), pp. 45-46, pl. II, fig. 10; text-fig. 9. 1913. Trigonoscorpio americanus Petrunkevitch, p. 47, pl. IV, figs. 17, 18; text-fig. 10. 1913. Palaeopisthacanthus mazonensis Petrunkevitch, pp. 49-50, pl. I, fig. 1; text-figs. 13-14. 1949. Typhlopisthacanthus mazonensis (Petrunkevitch). Petrunk- evitch, pp. 144-145. 1949. Eoscorpius carbonarius Meek and Worthen. Petrunkevitch, p: 152: 1949. Eoscorpius typicus Petrunkevitch. Petrunkevitch, p. 152. 1949. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 153. 1953. Eoscorpius carbonarius Meek and Worthen. Petrunkevitch, p. 27. 1953. Eoscorpius typicus Petrunkevitch. Petrunkevitch, p. 27. 1953. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 29. 1953. Trigonoscorpio americanus Petrunkevitch. Petrunkevitch, p. Silt 1953. Typhlopisthacanthus mazonensis (Petrunkevitch). Petrunk- evitch, p. 34. 1955. Eoscorpius carbonarius Meek and Worthen, Petrunkevitch, p. 73, fig. 42. 1955. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 73, fig. 43(1). 1955. Trigonoscorpio americanus Petrunkevitch. Petrunkevitch, p. 74, fig. 43(2). 1955. Typhlopisthacanthus mazonensis (Petrunkevitch). Petrunk- evitch, p. 74, fig. 43(5). 1962. Eoscorpius carbonarius Meek and Worthen. Dubinin, p. 429. 1962. Alloscorpius granulosus (Petrunkevitch). Dubinin, p. 429, fig. 1248. 1962. Typhlopisthacanthus mazonensis (Petrunkevitch). Dubinin, p. 431, fig. 1256A, B. 1962. Trigonoscorpio americanus Petrunkevitch. Dubinin, p. 431, figs. 1232, 1251. Specimen I.—The holotype of Eoscorpius carbo- narius Meek and Worthen, 1868, FMNH (UC) 8949; counterpart in the University of Illinois Geological Museum.?’ The holotype (Pl. 10; Text-figs. 69, 70) comprises part and counterpart of a well-preserved specimen in a typical ironstone concretion. FMNH (UC) 8949 re- tains more of the specimen than the obverse, which 2? Lost as of April, 1985. A.S.C. 164 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 unfortunately has been covered with a protective film, probably varnish, which with the years has hardened to a point that it is now almost impossible to remove. The description given here is based on both specimens and all figures are a composite of both halves of the nodule. The carapace is known in its entirety (see Text-fig. 69). It is subquadrate, rounded at the anterolateral and genal angles. The anterior is produced into a short triangular point, and directly posterior to this median process is a conspicuous, elevated, broadly lacrimi- form ocellar node. On the anterior of the eye node are two large, elliptical median eyes. These eyes are ca- pable of forward, lateral and vertical sight, but prob- ably not backward sight. The cuticle covering the eyes is transparent, but is light brown in color, clear and not covered by finely reticulated patterns such as are found in other parts of the cuticle. At the anterolateral corner of the carapace is a relatively small reniform facetted eye. The compound eye is present only on the left side, as the right eye has been buckled under the crushed right side. The ommatidia are relatively coarse, rounded and probably number less than 30. Behind the median ocellar node is a Y-shaped sulcus that extends posteriorly to bisect the two large cephalic ridges. The ridges are very prominent on the carapace and roughly divide it into an elevated cephalic area and a less-pronounced thoracic area. The rear of the carapace is bordered by a raised transverse ridge which articulates with the first tergite (see Text-fig. 69). The ornamentation of Eoscorpius carbonarius is dis- tinctive and consists of coarse pustules, probably se- taceous, concentrated along the anterior of the cephalic ridges in the area adjoining the median sulcus. The pustules are also more common in the posterior tho- racic area. A single prominent row of pustules, prob- ably setaceous, borders the posterior margin. Measurements (in mm) of the carapace of FMNH (UC) 8949, — Length at midsection: 10.2 Length (not including anterior process): 9.6 Width at base: 1222 Width at midsection: 11.8 Width at base of lateral eyes: 9.6 Length of median ocellar node: 4.1 Width of median ocellar node: Sy) Length of median ocellus: jes} Width of median ocellus: 1.0 Length of facetted lateral eye: 1.4 Width of facetted lateral eye: 0.6 Both chelicerae are well preserved and noteworthy for their robust size. The number of joints cannot be determined with certainty, but it appears that four comprise the complete chelicera—two short, ringlike joints, followed by the pincer of two joints. In the Text-figure 69.—Eoscorpius carbonarius Meek and Worthen. Specimen I, holotype, FMNH (UC) 8949. From the Upper Carbon- iferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. This is a composite drawing based on both sides of an ironstone nodule. The carapace is complete. Text-figure 70.—Eoscorpius carbonarius Meek and Worthen. (> Specimen I, holotype, FMNH (UC) 8949. From the Upper Carbon- iferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. A. Right chelicera from the ventral side, somewhat distorted, part- ly drawn from the left chelicera to make a composite. Small dots are setal bases concentrated on the inner sides of the chelicera, very much as in living scorpions. There are four joints in the chelicera, as shown; this is the same as in eurypterids. B. First right walking leg, dorsal view. There is no question of the presence of two terminal claws. Lines on pits on second and fifth joints are probably setaceous. C. First left walking leg from the ventral side. D. Second right walking leg from the dorsal side. E. Central part of fourth (last) right walking leg from the ventral side. F. Third right walking leg from the dorsal side. The dots represent pits where presumably tactile setae were attached. G. One of the pedipalps, probably the right one. The hand and free finger contain many small openings where presumed tactile hairs were attached; however, they do not appear to be in any par- ticular pattern that would resemble the trichobothna of modern scorpions. H. A pedipalp, presumed to be of the left side, which had been turned over to the right side of the holotype. The double row of round lobes is definitely a structure on the outer side of the free finger; if present on the fixed finger of the manus, they are not preserved. The holes were probably attachment sites for very coarse setae. The edges of the fingers are without denticles. I. The pentagonal sternum and parts of the pectinal comb, showing small rounded sclerites, fulcra and very long teeth. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 165 166 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 composite drawing of the two chelicerae (see Text-fig. 70A), the two ringlike joints represent the dorsal side. On the venter, the segment that corresponds to the second joint in the figure increases to double its length. This is in keeping with present-day scorpions, which have the ventral side considerably longer than the dor- sal. However, there are four joints in the chelicera of this scorpion, as has been determined here for nu- merous Silurian, Devonian and Carboniferous scor- pions, and which are now known to occur not only in eurypterids such as the Pterygotidae (Kjellesvig-Wae- ring, 1964, p. 334) but in the Eurypteridae (Kjellesvig- Waering, in preparation),*® therefore, likely in all Eu- rypterida. The hand of the chelicera is very wide and, although it is seen in a crushed state, is a massive structure. The immovable ramus is short and stout and is bordered with apparently two large, rounded teeth. On the inner and ventral side is an area with large setae that border the primitive oral tube on each side, as in living scor- pions. The free ramus is particularly massive, ends in a double tooth, as also occurs in modern scorpions such as the Buthidae, and is bordered on the masti- catory edge by at least three and possibly four rounded teeth. These teeth may have been more pointed in life as they are known from an internal cast of the inner part of the original cuticle. It is also quite possible that these teeth may occur as double rows, or that the mas- ticatory edge on the dorsal side may have other teeth. The chelicerae protrude beyond the margin of the carapace for at least the equivalent of two-fifths of the carapace length, and it appears likely that the entire last two joints are beyond the anterior edge of the carapace. Measurements (in mm) of the chelicera of FMNH (UC) 8949.— Length: First joint (incomplete) 0.65 Second joint at midsection (dorsal) 0.5 Second joint at midsection (ventral; incomplete) NEP) Width of hand 4.2 Length of hand and fixed ramus 4.6 Free finger width 22. Free finger length 1.8 In the reconstruction of the chelicera on Text-figure 70A, the hand (probably all that has been crushed) may be wider than it would have been in life. The fingers, however, would not be appreciably expanded by compaction. The entire chelicera would probably still be a very massive organ, not comparable, however, to Garnettius of the Kansas Pennsylvanian or Lias- soscorpionides from the German Jurassic. 28 Unfortunately, the Eurypterida paper was never completed. Ais:€: The walking legs have been interpreted as being com- posed of eight podomeres. The relationship to the ster- num is not known as it was displaced to the left of the specimen. The pedipalps are almost completely known and are described for the first time, as one had previously been mistaken for the fourth walking leg and the other pedi- palp had not been exposed on the specimen. The large first joint (trochanter), a curved and unusually-long trapezoidal structure, may be seen on the left side of the holotype. The second joint (femur) is not known in this specimen. The third joint (tibia) is very large, massive, and wider at the distal end (see Text-figs. 70G, H). The hand is very long, but the fixed finger is un- known. The free finger is nearly complete and is curv- ing, long, narrow and is distinctly devoid of any denticles or other serrations along the inner edge (see Text-fig. 70H). The pedipalp is very well marked by a double row of setal sites on both fingers, appearing as if there were a raised narrow ridge separating each row. This double row runs medially on the side throughout the length of the fingers. Several scattered single setal sites occur throughout the finger and have been noted on the free finger (see Text-fig. 70H). The sites for the rows of setae should not be mistaken for denticles which may have been displaced from the edge, as abundant pieces of the original cuticle were adequately preserved to show that the inner edge of the finger is devoid of any denticles or serrations, and therefore is cultrate. The inner row has coarser setal sites than the outer. The first and second coxae do not have maxillary lobes exposed, although the first pair seems to be slight- ly produced anteriorly. Only the edges of both coxae are preserved on the counterpart. These two coxae meet at the midline directly in front of the sternum. The outlines of the last two coxae are also present in this specimen, and the third pair appears larger than the sides of the sternum, therefore it has been interpreted that the third pair abuts against the sternum and the fourth against the genital operculum. This can be prov- en in other specimens. The first leg is known almost in its entirety, and is composed of seven joints and the coxa (see Text-figs. 70B, C). The first joint (trochanter) is not preserved. The second joint (femur) is flattened as in living scor- pions, wide, massive and with a row of pits along the anterior edge, presumably setaceous. The third joint (patella) is also flattened and wider at the distal end. The fourth joint (tibia) is long and has well-developed tibial spines at the end. The fifth joint (basitarsus), also long and narrow, retains two spurs at the distal end. The posterior part of the joint is lined with a row of small pits, probably setaceous. Pits, smaller in size, are present in a line along midsection. The large double FossIL SCORPIONIDA: KJELLESVIG- WAERING 167 spines at the end of the joint correspond to the pedal spines of present-day scorpions. The anterior of these spines is much longer than the posterior. This joint, as indicated by the fringe of setaceous pits, is also probably flattened in life. The sixth joint (tarsus) is short, and is distinctly divided by a narrow terminal depression into two tubular ridges that accommodate the double claw. The claws have been seen on the counterpart and only the outer edges are preserved sufficiently to show that they are claws—double, short and curved. However, preservation was not adequate to determine whether these claws may not have been more complicated, as Wills (1959) and Stormer (1963) showed in other Carboniferous scorpions. The post- tarsus or heel was not preserved. The second walking leg is preserved in its entirety (see Text-fig. 70D). The second pair of coxae meets at the midline in front of the sternum, and has short maxillary lobes (although not seen in this specimen). The first joint (trochanter) is short, whereas the second joint (femur) is long, and both are bordered at the anterior by a single row of pits, probably setaceous. Both joints are flattened in life. The third joint (patella) is long and flattened. A socket for the articulation of the condyle on the next joint occurs at the end of the joint. The fourth and fifth joints (tibia and basitarsus) are very long and narrower than the preceding. The ends are ornamented with two spurs, the anterior being larger than the posterior, corresponding to the tibial and basitarsal (pedal) spurs of present-day scorpions. Scattered setae occur on the fourth joint, and a con- centration of these setae occurs at the posterior tibial spur. The evidence for the setae is seen in the small round holes in the cuticle, and also in the presence of the setae that are embedded in the matrix, although in nearly all cases only the cross-section or embedded stump of the seta is seen. The end of the fifth joint has an extra small spine, which likely is all that remains of a fringe of similar spines. The posterior edge of this joint is bordered by a single row of pits, probably se- taceous. In the median part of the joint on the dorsal side is a row of numerous pits that represent the bases for single setae. The sixth joint (tarsus) is short, with a narrow median furrow at the end. It is long in com- parison to the corresponding joint in living scorpions. The two claws are only partially preserved, but one is complete, showing it to be short and curved (see Text- fig. 70D). Only the four central joints (femur, patella, tibia and basitarsus) of the third walking leg have been preserved and these show the leg to be constructed like the pre- vious ones, except that it is much larger (see Text-fig. 70F). The socket and condyle of the third and fourth joints are well preserved, and the tibial spurs are pres- ent at the end of the fourth joint. The second and third joints are flattened as in the previous leg. Scattered setae, indicated by the setal sites, are present on the third and fourth joints. The fifth joint, although known only from the base, is probably like the previous joint. Numerous setae, concentrated along the median part of the joint, are also present. The fourth walking leg is represented by a complete joint, which represents the third (patella) beyond the coxa (see Text-fig. 70E). The fourth walking leg that Meek and Worthen (1868, fig. 1) described and figured actually is the pedipalp. It had not previously been exposed or noted by those authors or by Petrunkevitch, who had accepted Meek and Worthen’s erroneous de- termination. Measurements (in mm) of the appendages of FMNH (UC) 8949. — Pedipalp: maximum midsection Joint No. length width 1 (trochanter) 5.9 2.9 2 (femur) (unknown) (unknown) 3 (tibia) 14.3 4.0 4 (hand) 9.6 3.6 5 (free finger) 10.0 (estimated) 1.1 Leg No. 1: Joint No. 1 (trochanter) (unknown) (unknown) 2 (femur) 6.8 (unknown) 3 (patella) 6.4 2.4 4 (tibia) 4.8 1 5 (basitarsus) 4.2 1.5 6 (tarsus) ies) 0.8 7 (claws) 1.0 = Leg No. 2: Joint No. 1 (trochanter) (unknown) (unknown) 2 (femur) 4.8 (covered) 3 (patella) 4.6 D3 4 (tibia) 8.0 2.8 5 (basitarsus) 5.6 1.3 6 (tarsus) 2:3, 0.8 7 (claws) igs) - Leg No. 3: Joint No. 2 (femur) (incomplete) (incomplete) 3 (patella) 8.4 3.1 4 (tibia) 10.0 NEW 5 (basitarsus) (incomplete) 1.3 Leg No. 4: Joint No. 3 (patella) 9.0 2.8 The tergites are all preserved and reveal that they are strongly constructed with a well-developed ante- rior, transverse, marginal rim. This marginal rim, a common feature in many living scorpions (7ityus, Bro- teochactas, Diplocentrus, Hadogenes, Urodacus, etc.) and among the eurypterids (Eurypterus, Buffalopterus, 168 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Erieopterus, etc.) has repeatedly gone unnoticed in the literature (see redescription of Mazonia woodiana Meek and Worthen in Kjellesvig-Waering (1969), and Kro- noscorpio danielsi (Petrunkevitch) here). Meek and Worthen (1868b) and Petrunkevitch (1913, 1949, 1953, 1955) have shown the tergites in these scorpions as long sclerites with rounded ends, and thought the prominent raised anterior ridge was intersegmental tis- sue. The sclerites are embedded in the skin. Instead, this prominent ridge serves to raise the posterior of the previous tergite in order that greater flexibility is at- tained. The anterior transverse ridge becomes pro- gressively thinner anteriorly to grade into the flexible intersegmental tissue that connects to the posterior doublure of the previous tergite. This type of articu- lation is known in countless other arthropods. Each successive posterior tergite increases in length and is rounded at the posterior genal angles. The an- terior marginal rim is one of the distinct and pro- nounced features of this scorpion, and is wider along the center and the ends than in the pleural areas. The anterior ridge is devoid of any ornamentation as it must slide underneath the previous tergite in any side movement of the opisthosoma. Toward the posterior and the sides of each mesosomatic tergite are concen- trated a number of pustules, all of which are probably setaceous, as each is surmounted by a small opening. Along the posterior margin, and at the genal angles and extremities, are single rows of similar setaceous pus- tules. This is a distinctive character, at least for pre- liminary determination of this scorpion. The anterior two-thirds of each tergite is mainly devoid of pustules. The seventh tergite is massively constructed with a well-developed anterior transverse ridge. The entire plate is trapezoidal, and the area that serves for the attachment of the cauda is very wide, in keeping with the wide and powerful tail. At midsection are two rows of granules, probably setaceous. At each side of the tergite is a raised, unornamented, straight crest. At the base and along the sides is the characteristic row of pustules. The extreme sides of the plate are flattened with a few scattered pustules occurring along the high- ly-impending sides. The underside of this scorpion is not well known as it is a dorsal impression. However, three structures of great taxonomic importance are known. These are the sternum, combs and parts of the abdominal plates. The sternum is displaced and lies to the left of the specimen. It is broadly pentagonal, broader than long, with the lateral sides parallel and short. The anterior is triangular and the base is straight. The posterior corners are right angles. The sternum measures 3.0 mm in length and 4.2 mm in width (see Text-fig. 701). The combs also have been displaced to the left of the specimen and are poorly preserved and incomplete. Enough is present to show that the teeth are long, with a longitudinal depression running throughout their length. Ten teeth are present, but considerably more must have been present in life. Fulcra are coarse and well developed. At least one row of rounded small sclerites or areoles is present above the fulcra. The rest of the comb is not preserved (see Text-fig. 701). The abdominal plates have been revealed on the left side of the specimen where they have been displaced, and can also be seen in place under the anterior part of the seventh tergite where a part is broken away to reveal these important structures. The plates are of rounded, bilobed or lobostern type. The plate that is in place shows a raised rim, much like that present in Isobuthus kralupensis Thorell and Lindstrém, and has a doublure (see Text-fig. 69). The cauda is known from three well-preserved seg- ments and the anterior part of the fourth. These show that the cauda is very strongly constructed, and each tergite is surmounted by two median carinae. The un- derside of the first caudal segment reveals that there are three median crests present on the venter; however, it has not been possible to determine whether these are present on the other segments, as all are covered by the dorsal side. Numerous scattered pustules, probably setaceous, are present along the sides of the cauda. Measurements (in mm) of the opisthosoma of FMNH (UC) 8949. — Tergite greatest number length width 1 1.8 12.6 2 2.8 1321 3 3.4 14.9 4 3.9 14.9 5 4.2 14.9 6 5:2 15.8 7 9.0 (anterior) 14.0 (posterior) 7.8 8 8.7 6.5 9 9.4 5.6 10 9.4 5.6 Specimen II.—Holotype of Alloscorpius granulosus (Petrunkevitch), 1913, pp. 45-46, fig. 9, pl. II, fig. 10; YPM 128. This specimen is an estimated 110 mm in overall length (see Text-fig. 71A) and is therefore larger than the holotype of Eoscorpius carbonarius Meek and Wor- then. Only one-half of the nodule is known, mainly consisting of the dorsal side, preserved from the inside, showing the carapace, mesosoma and parts of the first four tergites of the cauda. Most of the chelicerae, two pedipalps and parts of the legs are preserved and are seen in ventral aspect. The diagnostic ornamentation FossIL SCORPIONIDA: KJELLESVIG-WAERING 169 of Eoscorpius carbonarius, consisting of a single row tergites, as well as the double row of setal sites in the of pustules along the posterior fringe of the mesosomal form of round holes, biserially arranged on the dorsal Text-figure 71.— Eoscorpius carbonarius Meek and Worthen. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. A. Specimen II. Holotype of Alloscorpius granulosus (Petrunkevitch). YPM 128. B. Specimen III. Holotype of Typhlopisthacanthus ma- zonensis (Petrunkevitch). USNM 37977. This is a pedipalp of a female specimen. It shows the characteristic double row of setal sites on the free finger (see Text-fig. 70H). 170 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 side of the fingers of the pedipalp, leaves no doubt that Alloscorpius granulosus (Petrunkevitch, 1913) is a ju- nior synonym of Eoscorpius carbonarius Meek and Worthen, 1868. Little can be added to the known characteristics of Eoscorpius carbonarius Meek and Worthen from this specimen, as preservation is poor and known mainly from the dorsal surface. However, this specimen is larger and shows the median eyes to be proportionately smaller and placed slightly further back from the tr- angular, glossated anterior margin. The entire fixed finger of the pedipalp was exposed, revealing this struc- ture to be very long and curved. As in the holotype of E. carbonarius, the edge of the finger was found devoid of denticles. The cauda is preserved from the dorsal side, re- vealing two rows of crests, which are surmounted by coarse pustules; some of these appear elongated in a transverse manner across the crests. The third caudal segment was broken away to expose the ventral surface. A well-developed crest occurs on the side, suggesting, as evidenced in the holotype of E. carbonarius, that three crests occur on the ventral side. Measurements (in mm) of Specimen II, YPM 128.— Prosoma: Greatest width: 10.2+ Length: 10.2 Median eyes: Length: 1.4 Width: 0.9 Lateral separation: 0.4 Distance from anterior margin: 0.7 Distance from posterior margin: Te Distance from lateral margin: 3.0 Palpus (chela): Length: 25.44 Pedipalp hand: Width: 4.9 Length: 7.0 Free finger length: 18.4 Femur length: 13'S: Tibia length: 11.0+ Tergites (length): 1 1.8 2 2.6 3 352 4 5h5) 5 4.5 6 6.0 7 6.9 8 9.0 9 9.0 10 9.2 Width of seventh tergite: Anterior 13.3 Posterior Tet. Specimen III.—The holotype of Typhlopisthacan- thus mazonensis (Petrunkevitch), part and counter- part, USNM 37977 (see Petrunkevitch, 1913, pp. 49- 50, pl. II, fig. 1; text-figs. 13, 14. The holotype comprises two parts of an ironstone nodule, showing only the dorsal side. Petrunkevitch’s (1913) statement that the underside is preserved is not verified. Both parts of the nodules retain only the dor- sal surface. The details of the pedipalp, which is well preserved, show beyond question that it is conspecific with Eoscorpius carbonarius Meek and Worthen. This is confirmed by the free finger of the right pedipalp, which shows the characteristic setae in biserial ar- rangement (see drawing made of this pedipalp, Text- fig. 71B). The carapace clearly shows the elevated, rounded cephalic shield that characterizes Eoscorpius carbo- narius, and that Petrunkevitch completely overlooked. The anterior median eyes and node had been broken away and are not present. Details of the lateral eyes are not possible to see without considerable cleaning. Petrunkevitch, in his original figure (1913, p. SO, fig. 13), shows a very long carapace, but this interpretation was only possible because he had mistaken the large chelicerae for maxillary lobes and therefore the cara- pace was extended to cover the supposed maxillary lobes. The opisthosoma comprises the usual seven seg- ments, but the terminal segment has been partly bro- ken so that it appears to be rounded. This is, however, an illusion, a preservational artifact, and it is exactly the same as 1n any normal scorpion, with a large open- ing for a normal caudal segment. All of the segments of the mesosoma have been considerably telescoped forward, making it appear to be a much shorter scor- pion. It is, moreover, typical of the female of Eoscor- pius carbonarius. There are no coxae, sternum or gen- ital opercular plates preserved, as had been shown by Petrunkevitch, whether studied in a dry state or under glycerin, alcohol or water. The small cauda that Pe- trunkevitch (1913) shows in figure 13, is wholly illusory and represents only the left margin of the last preab- dominal segment. Two of the abdominal plates are present toward the posterior end, but these were not sufficiently well-preserved to warrant description. The entire specimen is so poorly preserved that it does not justify further measurements or description, nor further figures other than that given here, in order to prove that it is a junior synonym of Eoscorpius carbonarius. The holotype would have to have con- siderable cleaning to be of much use and even then it is doubtful that anything else could be added to our present knowledge of the species E. carbonarius from this specimen. It should be noted that there is no pos- sibility of this scorpion being a gigantic Pseudobu- thiscorpiidae, an opinion mentioned to me by some FossiL SCORPIONIDA: KJELLESVIG- WAERING 171 who have accepted Petrunkevitch’s erroneous inter- pretation of the specimen, but who have not studied the holotype. The presence of the typical double row of setae on the pedipalp precludes any interpretation other than that the holotype of Typhlopisthacanthus mazonensis (Petrunkevitch) is a poorly-preserved specimen, probably a female, of Eoscorpius carbon- arius Meek and Worthen. Specimen IV.—USNM 37986, holotype of Eoscor- pius typicus Petrunkevitch, 1913, p. 39, pl. I, figs. 2, 3; text-figs. 5, 6. The specimen (Text-figs. 72, 73 A—D) comprises half Ml T10 1 cm 11 SS T12 Text-figure 72.—Eoscorpius carbonarius Meek and Worthen. Specimen IV. Holotype of Eoscorpius typicus Petrunkevitch, USNM 37986. Male specimen. From the Upper Carboniferous (Pennsyl- vanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. Figure taken from a rubber mold, therefore reversed. Outline of the impression of the ocellar node indicated by dotted line. See foldout inside front cover for explanation of abbre- viations. of an ironstone concretion, retaining most of the ven- tral structures. The obverse, retaining the dorsal side, which was available to Petrunkevitch in 1913, has not been found at the U.S. National Museum where Pe- trunkevitch (1913, p. 39) states that the two halves were deposited. Fortunately, the more important half, having the impression of the ventral anatomy, was available for this study. Redescription and reinterpre- tation of the structures present in the holotype reveal that the species Eoscorpius typicus is a junior synonym of Eoscorpius carbonarius Meek and Worthen and it is probably a male of that species (see Text-fig. 72). This conclusion is based on several distinct mor- phological structures, for example: the similar eye node; the fine line of pits along the femur and patella of the Ae} \ \ i \ \ | Hh V4? Text-figure 73.—Eoscorpius carbonarius Meek and Worthen. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See fold- out inside front cover for explanation of abbreviations. A-D. Specimen IV. Holotype of E. typicus Petrunkevitch, USNM 37986. Male individual. A. Femur and patella of first right leg from the ventral side. B. Trochanter, femur and patella of the second left leg from the ventral side. C. Part of the femur and patella of the third left leg, and femur or patella of the fourth leg, from the ventral side. D. Coxosternal area. E, F. Specimen V. A paratype of E. typicus Petrunkevitch. USNM 37987. E. The slightly restored carapace. Only the mound of the median ocellar node was preserved. F. A pectine, drawn from the original and rubber casts. 12 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 legs (see Text-figs. 73 A-C); the fine, close-set carinae with granules of the seventh tergite (see Text-fig. 72); the presence of three carinae on the ventral caudal segments (see Text-fig. 72); also features such as the lobosternous abdominal plates, similar sterna, oper- cula and coxosternal arrangements (see Text-fig. 73D). The only differences between the holotypes of Eoscor- pius carbonarius Meek and Worthen and E. typicus Petrunkevitch are that the latter is more slender in mesosomatic girth, and has a longer cauda, and that its ornamentation, although of the same type as E. carbonarius, is coarser. All of these characteristics are well-known secondary sexual characters in living scor- pions and there seems to be no good reason to consider them to be otherwise in fossil forms. It should also be noted that the paratype, USNM 37987 (see below), shows the presence of the pectines, which are identical to those of Eoscorpius carbonarius Meek and Worthen, except that the teeth are finer. The specimen USNM 37987 is also a male, and the presence of finer or more numerous and slenderer denticles in the pectines is a well-known secondary sexual character denoting the male. Of the four specimens designated by Petrunkevitch as the holotype and three paratypes of Eoscorpius typ- icus, USNM 37986 (the holotype) and USNM 37987 are males of Eoscorpius carbonarius Meek and Wor- then. YPM 126 is a poorly-preserved specimen of An- thracoscorpio (?) sp., whereas YPM 127 has been re- described as a female of Palaeobuthus distinctus Petrunkevitch (see above). It should be noted that the original descriptions by Petrunkevitch of both E. typicus and Alloscorpius granulosus are mainly incorrect, and that the original description of the holotype of Eoscorpius carbonarius Meek and Worthen is incorrect as well. The carapaces of all three species are identical in length-width rela- tionship, except for normal secondary sexual charac- teristics. Petrunkevitch’s claim (1949, p.153) that the eyes of A/loscorpius are much nearer the anterior than in Eoscorpius, 1s based on a comparison that was not possible at that time, because the eyes of Eoscorpius carbonarius Meek and Worthen were still covered and unknown. They have not been exposed until this study. The eyes are identical in A//oscorpius and Eoscorpius. Apart from this, the type species of A/loscorpius (A. granulosus) has no validity, as it is a large female of Eoscorpius carbonarius Meek and Worthen. The half of the nodule of the holotype of Eoscorpius typicus that was available for this study reveals several important morphological structures, which were either overlooked or erroneously described by Petrunkevitch (1913). The carapace is not preserved on the surviving specimen except for an imprint of the outlines of the cephalic shield and the median sulcus, showing the base of the eyes and the eye node. This is shown in dotted outline on Text-figure 72 and in reconstructed outline on Text-figure 73E. This reveals that the car- apace comprised an elevated horseshoe-shaped ce- phalic shield, divided by a deep Y-shaped sulcus, and that the eyes were well forward on the carapace as in Eoscorpius carbonarius. Petrunkevitch (1913, fig. 6) shows an altogether different carapace and position of the eyes. The coxosternal arrangement is clearly preserved. The sternum is pentagonal, broader than wide, but small. The first two pairs of coxae are incompletely preserved and meet in front of the sternum, but it cannot be seen from this specimen whether or not maxillary lobes were present (in the Herdina specimen (FMNH (PE) 32084), it can clearly be seen that max- illary lobes are present). The third pair abuts the entire lateral side of the sternum, a condition that was also apparent in the holotype, since, as stated in the rede- scription given above, the sternum was too small to have the last two pairs of coxae abutting it. Petrunke- vitch shows both pairs abutting the sternum (1913, fig. 6), an impossibility considering the short sides of the sternum. The fourth pair clearly abuts the large pyri- form opercula, and this is easily verified in the speci- men (see Text-fig. 73D). The basal segments of the comb are preserved, but these reveal only that both segments are robust. The abdominal plates are very long and definitely of the lobostern type. This can be proven by the last plate, which reveals the characteristic rounded lobes. The seventh tergite shows the characteristic close- set carinae surmounted by a line of fine granules. The sides are covered with some scattered granules and the base has a characteristic fringe of granules. The dorsal side is incomplete, but does show the characteristic line of granules on one crest and also a fringe along the base of the tergite. Five caudal segments are preserved ventrally and each reveals three carinae. An anterior row of granules occurs on at least the first tergite. Some granules are present in the depression between the carinae. How- ever, preservation is very poor in the entire cauda. The appendages of the prosoma show fragments of the legs, which mainly represent the femur and patella (see Text-figs. 73A—C). These are important and show the single row of pits, very likely sites for setae, which occurs on the venter (dorsum also?) of each. The pedi- palp is poorly preserved but the outlines reveal a long and strongly-constructed organ with the femur longer and presumably slenderer than the tibia (see Text-fig. 72). Proportionately, and as expected in a secondary sexual character, it is longer than in the female. FossIL SCORPIONIDA: KJELLESVIG- WAERING 173 Measurements have been given by Petrunkevitch (1913, p. 41). Specimen V.—Paratype of Eoscorpius typicus Pe- trunkevitch, 1913, p. 40, pl. I, fig. 4; text-fig. 7 (USNM 37987). The specimen (Text-figs. 73E, F) is preserved in a typical Mazon Creek concretion, but is in three pieces (two large and one small). It comprises a large, slender scorpion that is considered to be the male. Only the rear part of the carapace (Text-fig. 73E), the mesosoma (turned sideward), the anterior of the mesosoma in ventral impression, part of the last leg and nearly all of a pectine are preserved. Determination is estab- lished mainly from the characteristic close-set double row of tubercles on the central part of the venter of the seventh preabdominal tergite, the single row of punctations on the posterior of each preabdominal ter- gite and the three carinae of the caudal segments. Noth- ing is added to our knowledge of the species except for details concerning the morphology of the pectine, which was previously known as a fragment in the holotype. Petrunkevitch (1913, fig. 7) had figured the pectine of this specimen but the interpretation given here is con- siderably different (see Text-fig. 73F). The organ con- sists of approximately three segments on the anterior lamella, although cracks in the integument make it appear as if five were present. The distal segment can- not be clearly discerned, but is considered likely to be as in the figure. The middle lamella is highly interesting in showing numerous short lacrimiform sclerites (ar- eoles) that become greatly elongated toward the inner part of the pectine. A lower row of the lacrimiform areoles occurs just above the fulcra, which are roughly triangular, well-developed and large. The teeth of the pectines are very large, long, and apparently not more than 25 occur on the male. Petrunkevitch (1913, p. 42) gives 18 teeth, but this is based on the incomplete fragment that formed the basis of his figure 7. A small fragment of the end of a tooth reveals coarse perfo- rations that likely represent sites for the peg organs. A small diamond-shaped sclerite represents the pectinal plate, but it is not known whether this is complete — it may possibly represent the median organ of the pec- tinal plate. Specimen VI.—The cast of the holotype of Trigo- noscorpio americanus Petrunkevitch, 1913, YPM 139. Holotype specimen in the Daniels collection has been lost (see footnote p. 163). The review given here is based upon plaster casts of the original specimens. These casts were made by Pe- trunkevitch and are in the Peabody Museum. Photo- graphs of the original were used in part. The plasto- types (see Text-fig. 74) reveal important data, which leave no doubt that the holotype is a juvenile male of Eoscorpius carbonarius Meek and Worthen. My inter- pretation of this scorpion differs greatly from that of Petrunkevitch (1913, p. 47, pl. IV, figs. 17, 18; text- fig. 10). The scorpion was purported to have a triangular carapace, although how it was possible to accommo- date the chelicerae and large pedipalps under this nar- row triangle was not easy to understand. It is important to determine how this scorpion is preserved. The car- apace obviously is broken and in the obverse, it is preserved as a triangular fragment — leading to the mis- interpretation of a triangular carapace. The carapace reveals anteriorly-located median eyes, two inflated cheeks, separated by a sulcus and a lacrimiform eye node—all typical of E. carbonarius. The carapace and first tergite are preserved in dorsal aspect. The rest of the preabdomen is preserved showing the inside, or dorsal part, of the ventral side. The only dorsal parts of the preabdomen preserved are the first and fifth tergites. The legs, pedipalp and cauda are too poorly pre- served in the plastotype for detailed description, but lack of knowledge of these structures would not mean- ingfully change the identification. The two pedipalps are present in part, however; the trochanter (P2) and the femur of both pedipalps are preserved as dorsal P6 tom Ye Be Text-figure 74.—Eoscorpius carbonarius Meek and Worthen. Specimen VI. Holotype of Trigonoscorpius americanus Petrunke- vitch. Daniels collection, YPM 139, plaster casts of a (lost) original part and counterpart of the ventral surface. From the Upper Car- boniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. The drawings were made from the plaster casts, supplemented by photographs of the (lost) originals. This is a juvenile male of EF. carbonarius Meek and Worthen. Note the broad carapace, rather than the narrow, triangular one postulated by Petrunkevitch. There are five well-developed lobosternous abdominal plates. See foldout inside front cover for explanation of abbreviations. 174 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 and ventral impressions, thus appearing on the same side of the nodule. The crucial determination of the abdominal plates as being lobosternous is based on the presence of dis- tinct bilobation and the good development of a wide doublure rimming each lobe. There are five well-de- veloped abdominal plates. The chelicerae are well preserved and, even in the plastotype, showed that they were composed of four distinct joints. The overall narrow aspect of the preabdomen in- dicates that this specimen is a male. I suspect that the overall size of this specimen measured from the an- terior of the carapace is about 30 mm, thus a juvenile of the species. Specimen VII.—Part and counterpart of a nearly complete scorpion measuring about 80 mm long, from the anterior of the carapace to the end of the telson. The specimen (FMNH (PE) 32084) came from the private collection of Jerry Herdina, where it was reg- istered as H-35. The scorpion (Pl. 11; Text-fig. 75) has been parted so that the coxosternal region and the base of the pedi- palps are seen from the dorsal side. It is therefore of great interest to see the attachment of the basal segment of the pedipalps and its relation to the other coxae. This part is rarely seen in fossil scorpions and is of taxonomic and phylogenetic importance. The coxa of the pedipalp is a very wide, rounded segment, with a long anterior process that occurs in front of the max- illary lobes of the first pair of legs. In life these maxillary lobes of the pedipalp coxae would form the sides of the chamber leading to the mouth, whereas the spat- ulate maxillary lobes of the first pair of walking leg coxae would form the floor or base. Unlike living scor- pions, however, the coxae of the pedipalps did not form a narrow tubelike entrance to the mouth, which in Carboniferous time, as well as today, had a central position on the carapace. The anterior entrance was therefore composed of the maxillary lobes of the pedi- dorsal clV ventral Text-figure 75.—Eoscorpius carbonarius Meek and Worthen. Specimen VII. Herdina collection, FMNH (PE) 32084. From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale (Mazon Creek area), Pit 1, Peabody Coal Company, Grundy Co.-Will Co. line, IL. See foldout inside front cover for explanation of abbreviations. A. Coxosternal area, pedipalp base and opercular plates seen from the dorsal inner side. B. Restoration of the ventral side of the coxosternal area, with slight restoration. The anterior of the second maxillary lobe overrides the first. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 7/5) palps, which formed the sides, followed by a wide chamber with the rest of the coxae of the pedipalps forming the sides, the base (floor) being formed mainly by the maxillary lobes of the first pair of walking leg coxae and the short maxillary lobes of the second pair of walking leg coxae. It has been a matter of conjecture as to the function of the very wide maxillary lobes of the first coxae, but now it is easy to see that this was present because the coxae of the pedipalps formed a very wide chamber, posterior to the anterior opening (see Text-fig. 75). The second pair of maxillary lobes, also not clearly seen in other specimens, is very well shown in this specimen. They extend only approximately halfway anteriorly to the first pair of maxillary lobes. Both pairs of lobes meet at center. However, of considerable in- terest is the fact that on the dorsal side, the second maxillary lobes extended further anteriorly, whereas they are partly covered by the spatulate first maxillary lobes on the ventral side, which is nearly always pre- served (see Text-fig. 75). Therefore, the ends of the second maxillary lobes (also dorsal) override the first pair of maxillary lobes. The third pair of coxae, as noted elsewhere, abuts the small pentagonal sternum. The fourth pair abuts the genital opercula, which are subtriangular and elongated. The overall aspect of this scorpion is robust, wide and stout, therefore probably a female. The determi- nation of the species is based on the setal track, com- posed ofa double row of alternating setal openings that parallel the cultrate edge of the fingers of the pedipalp, a character that, among the Mazon Creek scorpions, is only present in Eoscorpius carbonarius Meek and Worthen. This specimen had been sent to me in Norway in February, 1971, by Mr. Jerry Herdina. The drawing of the coxosternal region, described above, was made at that time. Dr. Leif Stormer was shown the specimen and my drawing at the time, because of the exceptional preservation of a part that was not well known. In August, 1977, I borrowed the specimen again, to see if the pregenital plate was preserved. Unfortunately, in the interim, some unauthorized person, obviously lacking knowledge of scorpion morphology, had irre- sponsibly removed the entire coxosternal region, the base of the pedipalps, the opercula, and other parts of the venter, in order to expose the carapace and the anterior of the dorsal tergites, morphological parts which are well known in this species. The main knowl- edge that we have of the entire coxosternal area and opercula, as well as the peculiar coxae of the pedipalps is the drawing in Text-figure 75A, which was made with the Wild Drawing Tube, and is therefore correct. Unfortunately, due to the above incident, I find myself in the unenviable position of having to hope that another specimen will be found that will verify my description and drawing. Specimen VIII.—A very large chela of a pedipalp, FMNH (PE) 32083 (formerly No. H-19 in the Herdina collection). The specimen consists of the usual ironstone nodule showing part and counterpart of the chela of a very large individual. This chela reveals the great size that this species may attain. The overall aspect of the chela is slender, with the characteristic double row of setal sites along the face of the fingers, which accounts for the identification of Eoscorpius carbonarius. The hand is 28.5 mm long at midsection and 18 mm wide at its greatest width. The fixed finger is 6 mm wide at the midsection. The free finger is 5.5 mm wide at the mid- section and 34 mm long (see PI. 12, figs. 1, 2). A specimen of Eoscorpius carbonarius with a pedi- palp chela of this size would measure slightly more than 30 cm in length from the anterior of the carapace to the end of the cauda. Type information. —The holotype of Eoscorpius car- bonarius Meek and Worthen 1s from the Pennsylvanian Carbondale Formation, lower Francis Creek Shale at Mazon Creek, Grundy County, Illinois. The holotypes and paratypes (described here) of Trigonoscorpio americanus Petrunkevitch, Eoscorpius typicus Petrun- kevitch, Alloscorpius granulosus Petrunkevitch and Typhlopisthacanthus mazonensis (Petrunkevitch) are from the same horizon and locality as the holotype of E. carbonarius. The Herdina specimen (FMNH (PE) 32084) from Pit 1 of the Peabody Coal Company, on the Will Co.—Grundy Co. line, Illinois, is also from the lower Francis Creek Shale. The Herdina pedipalp spec- imen (FMNH (PE) 32083) is from approximately the same area (Pits 1-6 undifferentiated) and same hori- zon. Remarks. —In 1868, Meek and Worthen described the first American fossil scorpion, a well-preserved specimen from the ironstone concretions of the Francis Creek Shale at Mazon Creek, Grundy County, Illinois. This specimen was described as Buthus? (Scorpio? Eoscorpius?) carbonarius in a preliminary note (Meek and Worthen, 1868a) and later changed to Eoscorpius carbonarius (Meek and Worthen, 1868b). The descrip- tion of this fossil was made against a meager back- ground, as the only known fossil scorpions were Cy- clophthalmus senior Corda and Microlabis sternbergii Corda from the Carboniferous rocks of Czechoslova- kia. The holotype of Eoscorpius carbonarius Meek and Worthen consisted of two halves of a concretion re- vealing the obverse and reverse of a scorpion preserved mainly asa dorsal impression. Meek and Worthen gave 176 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 an elaborate and careful description which at the time was exceptionally good. Only the dorsal and more ob- vious parts were described, and no attempt at devel- oping or cleaning the fossil was made. As a conse- quence, many organs were misidentified, not reported, or not exposed. In 1913, Petrunkevitch borrowed both halves of the concretion and surprisingly reported (1913, p. 37) that Meek and Worthen’s inadequate and erroneous description was “‘quite correct and that nothing of importance could be added to it”’! He there- fore accepted the description without question, except for changing the measurements from inches to mm. In Petrunkevitch’s monographs of 1949, 1953, and finally in the Treatise on Invertebrate Paleontology in 1955, no further attempt was made to correct the original error, namely that Meek and Worthen’s description was incorrect in many important details. Consequent- ly, scorpion systematics has been seriously handi- capped because for a hundred years we have followed an erroneous interpretation of one of the most impor- tant fossil scorpions. The importance to scorpion tax- onomy of Eoscorpius carbonarius is obvious—if for no reasons other than those of priority, and also because Scudder formed the family Eoscorpiidae as early as 1884. The family Eoscorpiidae, having little meaning, quickly became a convenient place to refer any genera whose underside was unknown. For example, because nothing was known of some of the details reported here concerning the holotype, fully 12 genera were placed in the incorrectly-redefined (by Petrunkevitch, not Scudder) family Eoscorpiidae in the Treatise on Invertebrate Paleontology (1955), and of these 12 gen- era it was seen that only one genus, namely Eoscorpius, survived within the correct definitions of the family. It has also been possible to determine some of Wills’ genera which were excellently-described morphologi- cally, but, because in a great sense Eoscorpius (and Mazonia) had not been correctly defined, were not referred to correct genera and families (see Wills, 1959, p. 267). The review and development of the holotype has resulted in the determination or exposure, for the first time, of: A —facetted lateral eyes; B—the large median ocelli; C —the pedipalps; D—presence of lobosternous ventral plates; E—the sternum; F —presence of ante- rior transverse ridge of the tergites; G—a fragment of the fourth walking leg; H —double tibial and basitarsal spurs; I—the terminal claws; J —the arrangement of the coxae. It is obvious that the lack of knowledge of many of these basic morphological structures has been a decided drawback in the establishment of a workable taxonomy. Petrunkevitch (1949, p. 153) erected the genus A/- loscorpius with type species Eoscorpius granulosus Pe- trunkevitch, 1913, for the following reasons: For whereas in the species to which the genus Foscorpius is restricted here, the carapace is either as long as wide or longer than wide, in both species of Alloscorpius it is distinctly wider than long. The position of the eyes is also different. In Alloscorpius they are much nearer the anterior edge of the carapace than in Eoscorpius. In the first place, Petrunkevitch could not have re- stricted Eoscorpius to the species Eoscorpius carbo- narius, as that was done in 1868 by Meek and Worthen. In the second place, Petrunkevitch could not compare the position of the eyes in the holotype, as at that time the holotype, and only specimen known, did not reveal any eyes, as they had not yet been exposed. As a matter of record, the eyes (now exposed) are identical. This leaves as the only difference between 4//oscorpius and Eoscorpius the alleged dissimilarity in the width of the carapaces, a very weak and highly doubtful character for the separation of genera. Nevertheless, Petrunkevitch is incorrect regarding the measurements of the carapace of Eoscorpius car- bonarius which, including the anterior process, mea- sures 10.2 mm in length or 9.6 mm in length without the process, and 12.2 mm in width at the base. There- fore, the alleged differences in length and width of the carapaces in the genera are erroneous. The genus A/- loscorpius Petrunkevitch, 1949, is a junior synonym of Eoscorpius Meek and Worthen, 1868. Furthermore, the type species 4//oscorpius granulosus 1s identical to Eoscorpius carbonarius Meek and Worthen and should be relegated to the synonymy of the latter. All known morphological parts are identical. The ornamentation also is identical, including the characteristic row of pustules along the base of each mesosomatic tergite, as well as the pustules bordering the sides of the last preabdominal tergite. Of greater importance is the characteristic double row of setal openings along the fingers of the pedipalps. This can only indicate an Eo- SCOrplUus. The holotype and paratype of Eoscorpius typicus Petrunkevitch, the holotype of Typhlopisthacanthus mazonensis (Petrunkevitch) and the holotype of 77ri- gonoscorpio americanus Petrunkevitch are identical to Eoscorpius carbonarius. The other paratypes, YPM 126 and 127 (Petrunkevitch, 1913, p. 39) are referred to Anthracoscorpio (?) sp. and Palaeobuthus distinctus Pe- trunkevitch. Alloscorpius granulosus and Typhlopis- thacanthus mazonensis are considered to be females. The holotype and paratype of Eoscorpius typicus and the holotype of 7Trigonoscorpio americanus Petrun- kevitch are considered to be males. This determination is in keeping with knowledge of secondary sex char- acters in living scorpions. Alloscorpius wardingleyi (Woodward) has been re- ferred to Mazonia by Kjellesvig-Waering (1969). Al- FossIL SCORPIONIDA: KJELLESVIG- WAERING 177 B edge of Cp me, 7 wee ge eta” TES, eg ee a ’ Text-figure 76.—Eoscorpius pulcher (Petrunkevitch). Holotype, BM(NH) In.39772. From the Upper Carboniferous, Crow Coal, Phoenix Brickworks, Crawcrook, near Ryton-on-Tyne, Durham, En- gland. See foldout inside front cover for explanation of abbrevia- tions. A. The entire holotype, a nearly complete prosomal carapace in original relief. B. Detail of the left compound eye. C. Reconstruction of the pectinal area. Based on text-figure 4 of Wills (1959) with some revision from the holotype. The structure which Wills called opercula is really the prepectinal plate (ppp). loscorpius danielsi (Petrunkevitch) has been referred to Kronoscorpio here, and Alloscorpius tuberculatus has been referred to Benniescorpio by Wills (1960, p. 322). This completes the list of species that Petrunkevitch had, at one time or another, referred to the illusory genus A//oscorpius. Eoscorpius pulcher (Petrunkevitch, 1949) Text-figures 76, 77 1949. Lichnophthalmus pulcher Petrunkevitch, pp. 147-148, figs. el Re Wie 1949. Europhthalmus longimanus Petrunkevitch, pp. 154-155, figs. 140, 141, 174. 1953. Lichnophthalmus pulcher Petrunkevitch. Petrunkevitch, p. 33. 1953. Europhthalmus longimanus Petrunkevitch. Petrunkevitch, p. 33. 1955. Lichnophthalmus pulcher Petrunkevitch. Petrunkevitch, p. 75, fig. 43(7). 1955. Europhthalmus longimanus Petrunkevitch. Petrunkevitch, p. 75, fig. 45(1). 1959. Lichnophthalmus pulcher Petrunkevitch. Wills, pp. 269-282, pl. 49, figs. 1-9; pl. 50, figs. 1-15; text-figs. 2-8. 1960. Lichnophthalmus pulcher Petrunkevitch. Wills, Addenda p. 331. 1962. Europhthalmus longimanus Petrunkevitch. Dubinin, p. 431, figs. 1236, 1254A, B. 1962. Lichnophthalmus pulcher Petrunkevitch. Dubinin, p. 431, fig. 1238. Wills (1959, text-fig. 4) was able to free what he determined to be the opercula (see Addenda, 1960, p. 331). This plate he stated “‘can be matched closely with the genital opercula of Mazoniscorpio”’ (the latter ge- nus has been shown (above), to be a junior synonym of Palaeobuthus). The plate in ““Mazoniscorpio”’ is not a single plate, but two normal opercular plates (“el”’ on text-fig. 13 of Wills, 1959) with the prepectinal plate jammed forward and over (under in the text-fig.), an action that may have occurred during ecdysis. This also is intimated by the close proximity of the pectines to the opercular plates (see discussion of the prepectinal plate, above). The part labelled “il’’, which Wills (1959, text-fig. 13) considers the internal lobes of the opercula, is also part of the prepectinal lobe, as the operculum does not have internal lobes, either in fossil or living scorpions. The median organ also is part of the pre- pectinal plate. The opercular plates of Palaeobuthus distinctus can be seen in another specimen (Text-fig. 56B) and will bear out the above. Therefore, the large, undivided, elongated plate of Lichnophthalmus 1s also the prepectinal plate (p! and p of Wills, 1959, text-fig. 4). The opercular plates of Lichnophthalmus pulcher should be long on a perpendicular axis as in other Eoscorptus. The conclusion above (see Text-fig. 76C) means that in Lichnophthalmus, instead of having the fourth pair of coxae forward of the opercular plates as in Wills (1959, text-fig. 4), the coxae must abut the opercular 178 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 plates, which remain unknown in Lichnophthalmus. If Wills’ interpretation is correct, that the fourth pair of coxae is forward of the opercula, it would necessitate a new family under the Pseudobuthiscorpioidea. How- ever, with the new determinations, it is obvious that the genus Lichnophthalmus Petrunkevitch, 1949, hav- ing a carapace with well-developed cephalic cheeks, lateral schizochroal eyes, anterior glossate process, and anteriorly-located median eyes on a lacrimiform eye node, is a junior synonym of Eoscorpius Meek and Worthen, 1868. The differences are on the species level, such as differences in outline of the carapace, cephalic cheeks, pattern of ornamentation and numerous other details. EKoscorpius pulcher (Petrunkevitch) is a well- established species differing from all others in the ge- nus. The holotype (BM(NH) In.39772), consisting of a single piece showing the carapace in its entirety in orig- inal relief, is exceptionally well preserved. Petrun- kevitch (1949, p. 147) gives a reconstructed dimension of 11 mm long by 13.7 mm. This does not include the anterior, pointed, glossate process, which would have resulted in a measurement of 12.5 mm as the length. My conclusions from study of the holotype differ in sO many important ways from that of Petrunkevitch (1949, p. 147) that it is superfluous to point out the differences. It is better to redescribe the specimen. The carapace is quadrate, with a distinct anterior, pointed, glossate prolongation on the midanterior mar- gin, which is bent downward. The median eyes are large and round, located on a prominent elliptical eye node, well forward on the carapace. Petrunkevitch (1949, p. 147) states that the eyes are ellipsoidal, but this is due to compaction, as was shown by Wills (1959) concerning a specimen developed by acid. It should be noted that both eyes are not as shown by Petrun- kevitch (1949, fig. 139), where the eyes are placed so that their axes form an angle of 90°. The two eyes are slightly crushed so that neither is as shown in that figure. A small interorbital ridge separates the two eyes. Of greatest importance is the presence of distinct large, slightly intramarginal, lateral schizochroal eyes that had been completely ignored in the original de- scription, although they are sufficiently well-preserved to be noted in Petrunkevitch’s photograph (1949, pl. 55, fig. 177). The excellent photograph even shows the presence of small individual schizochroal eyes. In my drawing (Text-fig. 76B), the eye has been greatly en- larged to show the detail. The ocelli are all rounded and about 30 occur in the elliptical (crushed) area en- closing them. A median sulcus separates the median eye node from the cephalic area, which is unusually elevated and split at the center line. The thoracic part of the carapace also is inflated, but not to the extent present in the cephalic area. A rounded narrow marginal rim is pres- ent on the base and posterior half of the carapace. The cephalic area of the carapace is covered with scattered, sparse tubercles, but not in a linear row as was shown by Petrunkevitch (1949, fig. 139). It is interesting to note the reasoning used by Pe- trunkevitch (1949, p. 148) in denying the presence of lateral eyes as: distinct ridge visible even in the photograph . . .. However, careful examination failed to show imprints of eyes, and the ridge itself is wanting on the right side. I am not certain if the absence mentioned by Petrun- kevitch meant that preservation was incomplete or that the ridge was not present as should be expected in a bilaterally symmetrical animal. His drawing (139) shows a carapace without the “ridge” (=schizochroal eyes) on either side, and reveals that he thought most, if not all, of the anterior of the carapace was preserved. In the specimen the right side is not preserved. Wills (1959, pp. 269-282, pls. 49-50) successfully extracted most of a topotype specimen, which greatly enlarges our knowledge of this genus and species. I have studied the specimen (BM(NH) In.39770) and my remarks will be restricted to information where differences may occur with Wills’ interpretation, as his description is exceptional and can hardly be improved. However, there are some points which I think worth recording. Wills showed that the median eyes are round, and I agree. The lateral schizochroal eyes are too poorly preserved for a good description, but they are present, although not previously recorded. These lateral eyes are seen in Wills’ prepared specimen and in a latex cast of the same specimen which I made in 1950. The anterior transverse ridges, which are promi- nently developed, were mistaken for intersegmental skin (Wills, text-fig. 2). The basal and postlateral mar- ginal rims of the carapace were not discerned, although they may be seen in his excellent photograph (pl. 49, fig. 2). In his text-figure 6, the basitarsal spurs are mis- taken for tarsal spurs. Europhthalmus longimanus Petrunkevitch was de- scribed in 1949 (pp. 154-155, figs. 140, 141, 174), as a new genus and species, which was considered to be the type of the genus Europhthalmus Petrunkevitch, 1949. Petrunkevitch had not developed the holotype of E. longimanus, as will be shown. Furthermore, he mistakenly thought that two specimens occurred on the same concretion. Scorpions are so rare in the geo- logic column that it would have been such an improb- able condition as to make one question the observa- tion. There is only one specimen, showing mainly the dorsal side. FossiL SCORPIONIDA: KJELLESVIG- WAERING 179 The holotype of Europhthalmus longimanus Pe- trunkevitch is poorly preserved. The carapace is con- siderably distorted, leading to the interpretation that Petrunkevitch gave, namely, that it was round on the anterior margin. Obviously this is not the case with this scorpion. Furthermore, it has lateral schizochroal eyes, 1s a distinct lobostern and the telson is enor- mously developed. All of these important structures, as well as the third and fourth legs, were entirely missed or otherwise unmentioned, and in the case of the sting- er or telson, erroneously reported. Europhthalmus longimanus Petrunkevitch is a ju- nior synonym of Eoscorpius pulcher (Petrunkevitch) alt \ because the carapace has identical ornamentation, the median eye node and median sulcus are the same, both are lobosternous, both have lateral schizochroal eyes, and the claws (ungues) of both have spines on the underside. Both occur at the same locality and horizon. The daggerlike posttarsus and long basitarsal spurs found on the first and second legs (Wills, 1959, pp. 277-279) have not been seen on the holotype of E. longimanus. However, the legs preserved are not the same, those of the latter being the third and fourth. The part preserved, the ungues of the fourth leg, is identical. I do not think there is much doubt that they are Synonymous and, although both were described in \V7 Text-figure 77.—Eoscorpius pulcher (Petrunkevitch). BM(NH) In.39766, part and counterpart (the holotype of Europhthalmus longimanus Petrunkevitch, 1949). From the Upper Carboniferous, Crow Coal, Phoenix Brickworks, Crawcrook, near Ryton-on-Tyne, Durham, England. See foldout inside front cover for explanation of abbreviations. A. Dorsal aspect. Note the presence of lateral compound eyes and other details not shown in Petrunkevitch (1949; 1955, fig. 55). B. Ventral aspect, showing the long pedipalp; below, legs III and IV preserved on the same slab. C. Termination of leg IV. The spur (IV6s) is very interesting. 180 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the same publication, Lichnophthalmus pulcher Pe- trunkevitch, 1949 (p. 147), clearly has “page priority” over Europhthalmus longimanus Petrunkevitch, 1949 (p. 154), and therefore the latter species name must be suppressed. This has the added advantage that speci- mens of E. pulcher Petrunkevitch have been extracted from the matrix and are therefore far better known than the single specimen of Europhthalmus longima- nus. At any rate, as first reviewer, I hereby select Lich- nophthalmus pulcher Petrunkevitch, 1949, to be the correct name of the species. As noted above, the holotype of Europhthalmus lon- gimanus Petrunkevitch (see Text-figs. 77 A—C) has been inadequately and erroneously described and much in- formation of basic taxonomic value is still to be re- vealed. Furthermore, the specimen had not been pre- pared properly, so that the walking legs were unknown (one had been previously revealed, the other exposed by me). The cauda was barely exposed and further- more, misconstrued in the original interpretation and figure (Petrunkevitch, 1949, fig. 140). The stinger shown on the original figure was merely the fourth caudal segment; the fifth segment as well as the enormous and unusual stinger (telson) were all covered (see latex casts of the original at the British Museum made before exposure). These have now been exposed as well as most of the fourth leg. It is therefore necessary to redescribe the specimen in order to correct these misconceptions. The carapace, distorted, was originally quadrate, not rounded anteriorly, as shown by Petrunkevitch (1949, fig. 140). Two large round median eyes occur on a lacrimiform eye node. The surface retains large gran- ules on the sides of the median line. A large compound lateral eye is present at the anterolateral angle. This contains small eyes that are rounded-subhexagonal and each is approximately 0.08 mm in diameter. The entire eye is about 0.8 mm wide, measuring from the margin. The facets, therefore, are relatively large and each lat- eral eye would contain approximately 30 ocelli—this being a very rough estimate. The mesosoma is telescoped and highly distorted. There is no such regularity of tergites as shown in the original figure (Petrunkevitch, 1949, text-fig. 140), but of great importance is the presence of deeply-bilobed lobostern abdominal plates. These seem to be the fourth and fifth abdominal plates, and reveal the character- istic deep notch at the posterior middle part. The cauda was covered in part so that Petrunke- vitch’s original figure (1949, fig. 174) shows only a part of the base of the cauda up to the fourth caudal seg- ment. Careful exposure revealed all the segments, and this showed that instead of being a scorpion with a very narrow tail, it had a very robust cauda. The cauda is turned sideward and shows that the superior carinae were developed into high, prominent, coarsely-serrat- ed ridges (see Text-figs. 77A, B). The stinger or telson is of particular interest as this is unusually large and narrow (see Text-fig. 77A). The vesicle is relatively small in size, but this is followed by an enormous scimitarlike aculeus that curves in almost a circular manner. The scimitariform aculeus is extremely long and narrow, gradually tapering in width to end in a point. No tubercle occurs below the aculeus, but midway on the inside of the curve of the stinger is a small protrusion, actually a widening, which likely lent support to the great scimitarlike stinger. No scorpion, to my knowledge, has a stinger that compares with this formidable weapon, although Palaeobuthus distinctus Petrunkevitch approaches this one in length. Measurements (in mm) of the cauda of BM(NH) In.39766.— width from side Tergite: length (height in life) 8 9.0 7.0 9 7.5 7.0 10 9.0 6.3 11 9.6 5.9 12 8.3 6.0 (estimated) Telson: 13.0 5.4 (at base) The prosomal appendages are not preserved except for fragments of two legs and the pedipalp. The latter has been described. I have not been able to see well- preserved granules on the edge of the finger, but they are present, although insufficiently well-preserved for detailed description. The entire pedipalp is in keeping with the rest of the scorpion—it is very stout, although the hand of the chela is long. The tibia and finger are thick and are surmounted with carinae that have coarse pustules in rectilinear series. A fragment of one of the legs is present, but it is not possible to state definitely what leg this was (possibly the third). This is noteworthy for the very long spine present at the end of the tibia. This is an unusually long tibial spine. The fourth leg (see Text-fig. 77B) is present from the femur to the posttarsus. The segments are all long and narrow, in particular the tarsus and basitarsus. The posttarsus is a small (see Text-fig. 77C) short triangular stub, whereas the only claw that was preserved of the two original ones is long, curved at the tip and armed with small denticles on the underside. The overall length of this scorpion from the anterior of the carapace to the end of the cauda is estimated at 100 mm. Type information. —All specimens of Eoscorpius pulcher come from the Upper Carboniferous, Crow FossiL SCORPIONIDA: KJELLESVIG-WAERING 181 Coal, at the Phoenix Brickworks at Crawcrook, near Ryton-on-Tyne, Durham, England, including the ho- lotype (BM(NH) In.39772), the holotype of Euro- phthalmus longimanus (BM(NH) In.39766), and the specimen used by Wills (1959, pp. 269-282) (BM(NH) In.39770). Eoscorpius mucronatus, new species Text-figure 78 1959. Lichnophthalmus pulcher Petrunkevitch. Wills, not Petrun- kevitch (partim), pp. 269, 272, pl. 49, fig. 10. The holotype is represented by a carapace, nearly whole, and in good condition. It is preserved in green- ish-gray, soft shale with numerous plants. Previously, Petrunkevitch had considered the specimen as Lich- nophthalmus ? (unpublished), and it was identified as L. pulcher Petrunkevitch by Wills (1959, p. 269). The differences are much too great to warrant placing both in the same species and it is herein described as new. It comes from the same zone (which has considerable thickness), although from a different horizon and lo- cality. The carapace is preserved in two parts: the specimen proper, and a Marco-cast (see Wills, 1959, pp. 263- 265; Wills, 1960, pp. 331-333; Wills, 1965, p. 97) made of the same. Thus both specimens should be studied together, and in fact, the figure (Text-fig. 78) 1mm Text-figure 78.—Eoscorpius mucronatus, n. sp. Holotype, GSM PF 1392-1393 (formerly Wills’ No. Za.2842). A nearly complete carapace. From the Upper Carboniferous, Coal Measures (Anthra- conaia modiolaris Zone), Parkgate Coal, Dodworth, near Barnsley, Yorkshire, England. is a composite of the two. Actually, most of the skin adhered to the Marco-cast when prepared by Wills, and hardly improved the specimen. Non-calcareous specimens should not be subjected to the “‘Wills tech- nique’. Nevertheless, both specimens reveal sufficient information to add to our knowledge of the British scorpions. Carapace typically squarish, rounded at the anterior angles, with the middle of the anterior consisting of a prominent, triangular linguiform process. Median eyes anteriorly, but intramarginally, located. A median eye node, lacrimiform and with well-developed orbital ridges, contains the median eyes, which are round, although slightly distorted. Lateral schizochroal eyes, relatively small, but definitely present in the Marco- cast. These eyes are slightly reniform and are present at the anterior angles of the carapace, but not touching the margin. The carapace 1s divided by a median sulcus that separates two large, inflated, cephalic cheeks. The base is not present, but the smaller basal cheeks are prominent and inflated. Both the cephalic and the basal cheeks are covered by numerous prominent pustules. Estimated width at anterior of carapace: 9 mm. Es- timated overall length of scorpion about 80 mm. Type information. —Carboniferous, Coal Measures, Anthraconaia modiolaris Zone, Parkgate Coal, Dod- worth, near Barnsley, Yorkshire, England; L. Moysey, collector, ca. 1910; GSM PF 1392-1393 (formerly Wills’ No. Za.2842). Derivatio nominis. — mucronatus (L.) = pointed. Remarks. —The differences between this species and Eoscorpius pulcher (Petrunkevitch) are readily appar- ent from the presence of very dense pustulation on the carapace of the new species. However, this difference could well be sexual, since it is a well-known secondary sexual character in the male to have considerably more granulation on the carapace and other areas than the female. There are more important differences: the lat- eral eyes are placed further back on the carapace in EF. mucronatus, the anterior linguiform process is longer on E. mucronatus; the eye node is lacrimiform, rather than subrounded as in EF. pulcher, and the lateral ce- phalic cheeks are larger and more pronounced ante- riorly in E. mucronatus. Eoscorpius distinctus (Petrunkevitch, 1949) (new description) Plate 13, figures 1-4; Text-figures 79A—B, 80 1949. Typhloscorpius distinctus Petrunkevitch, pp. 146-147, figs. 146, 178. 1953. Typhloscorpius distinctus Petrunkevitch. Petrunkevitch, p. 33. 1955. Typhloscorpius distinctus Petrunkevitch. Petrunkevitch, p. 75, fig. 44(2). 1962. Typhloscorpius distinctus Petrunkevitch. Dubinin, p. 431, fig. 1237 182 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Specimen I.—The holotype, BM(NH) In.13990. Typhloscorpius distinctus Petrunkevitch, from the Upper Coal Measures of Coseley, Staffordshire, En- gland, should be referred to the genus Eoscorpius, as will be shown by new evidence. The original descrip- tion (Petrunkevitch, 1949, p. 146) is erroneous, and it is necessary to redescribe the holotype. A rubber cast of the holotype, obverse and reverse, sent to me by Dr. H. W. Ball of the British Museum (Natural His- tory), reveals a scorpion with features of Eoscorpius. Petrunkevitch (1949, p. 146) stated that one of the features that made the scorpion distinct was that “it has no eyes’’. However, no scorpion known has more eyes than 7. distinctus! Distinct, large, facetted lateral eyes occur at the anterolateral angles. These facetted eyes are protuberant and were part/y embedded in the matrix, but fully half of each eye was visible before mechanical development. Later the entire eyes were exposed (see Text-figs. 79A, B). A narrow, raised pal- pebral rim surrounds the eyes, which have retained fine ommatidia. The ommatidia are much finer in size than those present in the eyes of Eoscorpius and Kro- noscorpio. I estimate that about 75 ommatidia com- pose the eye, approximately twice as many as in the latter genera. I do not consider the number of om- matidia as of more than trivial importance. The fac- etted lateral eyes measure 1.2 mm in length. The me- dian eyes, which Petrunkevitch thought were also absent, but are now exposed (see Pl. 13, fig. 4, and Text-figs. 79A, B), are as large as those in Eoscorpius and Kronoscorpio, and are placed anteriorly on a raised, lacrimiform eye node. The anterior margin is straight, with a triangular protuberance at the center. The eye is elliptical and measures 1.3 mm in length and 0.9 mm in width. A Y-shaped sulcus, with the median eye node strad- dling the open end, divides the carapace. Two large, wide cephalic ridges continue on both sides of the sul- cus almost to the base of the carapace. Scattered large pustules surmount the ridges and follow the contours of the Y-shaped sulcus. The thoracic part of the car- apace rises diagonally from the cephalic ridges to end in a raised, round, basal marginal rim. It should be noted that the central protuberance at Text-figure 79.—Eoscorpius distinctus (Petrunkevitch). Specimen I, holotype, BM(NH) In.13909. From the Upper Carboniferous, Middle Coal Measures, Coseley, South Staffordshire Coalfield, England. See foldout inside front cover for explanation of abbreviations. A. Reconstruction of the holotype. All parts shown here have been seen on the obverse or the reverse of the concretion. B. Cross-section through the anterior portion of the carapace to show the position of the eyes and the elevation of the eye node. C. Hughmilleria banksi Salter, introduced for comparison in text. FOssIL SCORPIONIDA: the anterior margin 1s curved downward and resembles the glossated front of some of the eurypterids, such as Eocarcinosoma, Mixopterus, Rhinocarcinosoma, cer- tain hughmilleriids such as Hughmilleria banksii (Salt- er) and others. This feature is well known in many other scorpions, such as Proscorpius, Kronoscorpio, etc. Eoscorpius probably was not a scorpion that lived un- derground, whether in water or not, as the pedipalps are much too long for a burrowing scorpion. However, the bent “‘plow-nose” or glossate anterior is, in a water medium, an admirable tool for pushing soft sediments over the carapace for concealment. This would cor- relate with the anteriorly-located and highly-elevated median eyes, as this part would remain exposed, in wait for unwary prey. The median eye node is bounded on the posterior end by a flattened area. The two large ocelli are sep- arated by a wide sulcus. The median eye node and ocelli were uncovered by Dr. H. W. Ball and Mr. R. Rixon at my request, as it was obvious from the pho- tograph of Petrunkevitch (1949, fig. 178) that the me- dian eyes were present in the counterpart, but had not been exposed by Petrunkevitch in his original study. He had not noted the presence of either the median eyes or the prominent facetted lateral eyes and con- sequently, believing that the scorpion was blind, erect- ed the genus 7yphloscorpius. I consider the genus 7yphloscorpius Petrunkevitch to be a junior synonym of Eoscorpius, on the basis that the holotype specimen does not show any characters that distinguish it from the latter genus. The pedipalp described below further strengthens that determina- tion, although my study of the cast was made in 1965. Specimen II.—I refer to this species a single pedipalp (BM(NH) I.3172), on the basis that both the holotype and the pedipalp are clearly parts of an Eoscorpius. The fact that both are from the same horizon and general locality strengthens that determination. The double row of setal sites on the pedipalp fingers is indicative of the genus Eoscorpius. The pedipalp is from the left side, has most of the trochanter to the free finger preserved, all in excellent condition. The trochanter, however, is too poorly pre- served for description. The femur is very long, narrow and wider at the end than elsewhere. Setal sites are concentrated on the an- terior, dorsal side (see Text-figs. 80A, B). The tibia is preserved so that the posterior half rep- resents the dorsal side, whereas the anterior represents the ventral. The tibia is long, subquadrangular in shape and retains a concentration of setal openings on the ventral side. The hand is narrow, rounded on the inner side and nearly straight on the outer edge. The hand seems to KJELLESVIG-WAERING 183 be devoid of setal openings, a condition that may be due to lack of preservation (see Text-figs. 80A, B). The fixed finger is preserved nearly whole. It is very slender, cultrate, slightly recurving and with a distinct double row of setal sites on the outer (dorsal) face of the finger. The inner row is composed of coarse setal sites which, in many parts of both rows, are filled with a dark material that seems to represent the broken stubs of the setae. The anterior or outer row contains much thicker setal openings than the inner row. Measurements (in mm) of pedipalp (BM(NH) 1.3172).— length width Femur: 18.9 5.0 Tibia: NES 6.0 Hand: 11.5 7.8 Type information. —The holotype is from the Upper Carboniferous, Middle Coal Measures, from Coseley, South Staffordshire Coalfield, England. The pedipalp (BM(NH) I.3172) described above is also from the Upper Carboniferous, Middle Coal Mea- . \ : | es =a Text-figure 80.—Eoscorpius distinctus (Petrunkevitch). Specimen II. BM(NH) 1.3172. From the Upper Carboniferous, Middle Coal Measures, South Staffordshire Coalfield, Dudley, England. See fold- out inside front cover for explanation of abbreviations. A. Enlargement of setal rows on the pedipalp finger. B. The same pedipalp from the counterpart of the nodule. 184 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 sures, South Staffordshire Coalfield, but from Dudley, England. Both the holotype (BM(NH) In.13909) and the pedi- palp (BM(NH) I.3172) are in the collections of the British Museum (Natural History). Eoscorpius sparthensis Baldwin and Sutcliffe, 1904 (new description) Text-figure 81 1904. Eoscorpius sparthensis Baldwin and Sutcliffe, p. 395, figs. 2- 3: 1911. Anthracoscorpio sparthensis (Baldwin and Sutcliffe). Pocock, pp. 20-21. 1913. Eoscorpius sparthensis Baldwin and Sutcliffe. Petrunkevitch, p. 34. 1923. Anthracoscorpio sparthensis (Baldwin and Sutcliffe). Moore, p2132; pl; fig? 6: 1949. Eoscorpius sparthensis Baldwin and Sutcliffe. Petrunkevitch, pp. 152-153. 1953. Eoscorpius sparthensis Baldwin and Sutcliffe. Petrunkevitch (partim), pp. 27-28, fig. 120; not spec. L 8182 in the Man- chester Museum. Specimen. — The holotype of Eoscorpius sparthensis, UMM L 6271A, B. The holotype consists of two parts of a large nodule composed of hard, shaly ironstone. Most of the cara- pace, opisthosoma and the left pedipalp are preserved. Preservation is good in some respects but requires the use of oblique lighting and investigation under alcohol and in a dry state. Each of these methods will reveal some Structures that other methods do not show. The specimen is fully preserved; however, along the cara- pace, which has been crushed and compressed laterally, the outlines and structures are by no means clear. Nevertheless, on the counterpart, by use of rubber casts as well as the methods described above, considerable detail has been discerned. The carapace is elongate with a glossate process along the central part. Behind this is a large eye node, teardrop-shaped, with two large median ocelli. At the anterolateral corners, seen par- ticularly on the counterpart, is a small, reniform lateral eye that shows small, rounded facets, which probably represent only the impression of the lenses (see Text- fig. 81C). A large, deep sulcus separates the eye node from the elevated area of the carapace, which occurs as inflated cheeks toward the genal angles. Apparently an elevated ridge occurs along the base of the carapace. Large pustules are present along the elevated areas and also on the teardrop-shaped ocellar node, all of which are probably setaceous. What appears to be an ocular ridge, in the form of an inverted ‘*V’’, occurs at the base and between both median eyes (see Text-figs. 81A, B). The seven tergites of the preabdomen, each of which increases in length posteriorly, are revealed in dorsal aspect. It appears that their greatest width occurs around the fifth or sixth tergite. Two strong ridges are devel- oped on the seventh tergite. Each tergite is bounded anteriorly by a strong transverse marginal ridge and remnants of small punctations are present on all of the tergites. The venter is not preserved, although on the sixth tergite, part of the tergite has been removed and reveals a round area that undoubtedly is a reflection of a lobostern plate. Also impressed through the third tergite are two rounded areas that represent part of a lobostern abdominal plate. The cauda is preserved on its side and shows that it has highly elevated, rounded dorsal ridges. On the lat- eral part of each tergite are two prominent ridges. Each tergite seems to be approximately the same size, al- though only the eighth to the eleventh are preserved. Only a fragment of the twelfth tergite 1s present and the telson is unknown. The entire pedipalp is preserved, and is a slender structure overall. The trochanter is unusually long, with a femur and tibia of approximately the same size. Punctations in linear series are found in two lines across the femur. These undoubtedly occurred on ridges. On the tibia are at least three ridges. The hand is narrow, elongated, and probably retained three ridges in life. Only the free finger is preserved and this is long and curving, but shows a structure that is of considerable importance in the identification of this scorpion as Eoscorpius. There is an impression running along the inner part, but not at the edge, of the free finger, which has small setal perforations arranged on either side of the line in which a double row occurs on the inner side, whereas only a single row of these punctations is pres- ent on the outer side. This is also found in Eoscorpius carbonarius, but there only a single row of setal open- ings occurs on the inner side of the narrow linear groove (see Text-fig. 81D). Measurements (in mm) of holotype (UMM L 6271A,B).— Length of carapace through middle: 11.4 Greatest length of pedipalp trochanter: 2. Length of pedipalp femur: 8.4 Length of pedipalp tibia: 9.2 Length of hand (midsection): UP Width of hand (greatest): 4.2 Length of free finger (inner edge): 9.6 Text-figure 81.—Eoscorpius sparthensis Baldwin and Sutcliffe. Holotype (UMM L 6271, A,B), part and counterpart. From the Upper Carboniferous, Middle Coal Measures, Sparth Bottoms, Rochdale, Lancashire, England. See foldout inside front cover for explanation of abbreviations. A. Counterpart. B. Part. C. Details of the right lateral facetted eye. The marginal rim (mr) is indicated. D. The right pedipalp and en- largement of the basal part of the free finger to show details of setal arrangement. 185 186 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Measurements of UMM L62714A, B (cont'd). Length of tergites: l 23, 351 3.8 4.2 4.2 4.8 7.0 6.5 6.5 6.1 529, KK OOWAAIAHAUN HW NW — Type information. —Middle Coal Measures, Sparth Bottoms, Rochdale, Lancashire, England. Remarks. —This specimen was first described as Eoscorpius by Baldwin and Sutcliffe. Later Pocock (1911) referred the species to Anthracoscorpio. Pe- trunkevitch (1953) states, ‘“‘but the shape of the por- tions of the carapace and the general appearance speak strongly in favor of its being an Eoscorpius”. Eoscor- pius carbonarius was incorrectly known at that time as such important structures as the facetted lateral eyes, the median eyes, the anterior protuberance of the car- apace, the type of the abdominal plates, termination of the legs, and the presence of the spurs on the legs were totally unknown. It became, therefore, rather common to refer any specimen that looked like a scor- pion to the genus Eoscorpius. No one could with cer- tainty refer a living scorpion to any particular genus on the basis of ‘shape and proportions of the carapace and the general appearance’’. And, much less, should this be a criterion with fossil forms. I agree with Pe- trunkevitch that the species sparthensis should be as- signed to Eoscorpius, but for altogether different rea- sons. For example, I place this specimen in the genus Eoscorpius because it agrees in the presence of the anterior linguiform process, it has median eyes, placed on an eye node exactly as in Eoscorpius, and facetted lateral eyes that are also the same as in that genus. The median sulcus of the carapace is identical. It is also important that both are lobosternous scorpions. Again, another factor that shows the species to belong to Eo- scorpius is the presence of the peculiar and distinct ornamentation, consisting of rows of setae on the free finger of the pedipalp. This is almost exactly as found only in the genus EFoscorpius. Baldwin and Sutcliffe (1904, p. 396) mention that on one side of the specimen is a discoloration of the rock that probably represents the intersegmental tissue between the abdominal plates and the tergites. This discoloration is also present on the left side and I agree with Baldwin and Sutcliffe that this likely represents the intersegmental tissue. Pocock (1911) referred Eo- buthus rakovnicensis to the synonymy of Eoscorpius sparthensis, which he considered to belong to Anthra- coscorpio. This is incorrect, as Eobuthus rakovnicensis is the type species of the genus Eobuthus. The figure which was used in the original description by Baldwin and Sutcliffe (1904, fig. 2) was a rough sketch, which Pocock made for the authors (1911, p. 21). This sketch showed only the side containing the relief. In the figures shown here both part and coun- terpart are reproduced for the first time, as previously only photographs had been available of this interesting specimen. It is not possible to see many clear details from photographs of scorpions. In this particular case, the counterpart is the specimen which shows by far the greater detail. Eoscorpius species 1963. Lichnophthalmus sp. Laurentiaux-Vieira and Laurentiaux, pp. 23-25, pl. 3, fig. 1; text-fig. 1. Under the name Lichnophthalmus sp., Laurentiaux- Vieira and Laurentiaux (1963, p. 23) describe and figure the anterior part of a cephalothorax of a scorpion from the Upper Carboniferous (Lower Autunian) of the Aumance Basin in France. This specimen exhibits two large areas in the shape of half-moons in the antero- lateral angles. The writers refer to these areas as “‘al- veolate areas’’, as Petrunkevitch (1913, p. 43) had done in reference to Kronoscorpio danielsi. These lunate areas are undoubtedly the lateral compound eyes, and the ‘“‘alveolate areas” are the imprints of the ommatidia. The Aumance specimen is properly referred to Eo- scorpius, and likely is a form having affinities with, but distinct from Eoscorpius distinctus (Petrunkevitch). The repository of the specimen is not indicated. It may be either in the Ecole Nationale Supérieure des Mines de Paris or in the Geology Department of the Faculté des Sciences de Reims. Eoscorpius casei, new species Text-figure 82 1972. Eoscorpius sp. Carroll et al., p. 64. Ventral surface of a scorpion (PU 87266) comprising the coxosternal region, the mesosoma through the sev- enth tergite, one side of the pectines, and portions of the walking legs.° 29 Kjellesvig-Waering named the specimen and made a detailed drawing of it, thus indicating his interpretation of what is present, but never got around to writing up the description. A.S.C. Text-figure 82.—Eoscorpius casei, n. sp. Holotype, PU 87266. From the Carboniferous, Lower Pennsylvanian (Westphalian A), from a cliff exposure at West Bay, southwest of Parrsboro, Cum- berland County, Nova Scotia, Canada. See foldout inside front cover for explanation of abbreviations. FOssIL SCORPIONIDA: KJELLESVIG-WAERING 187 188 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The dimensions are taken from the drawing, by the compilers, and are approximate at best. The length of the specimen from the anterior end of the maxillary lobe of the first coxa to the posterior end of the seventh tergite is 48 mm. The seventh tergite itself is 10.5 mm long, but is too incomplete in width to measure. The incomplete width of the mesosoma is 19+ mm. The coxa of the fourth leg is complete and is 15.5 mm in length and averages 4.2 mm in width. One of the pec- tines is well preserved and measures 4 mm in length and 10.9 mm in width. About 30 teeth were preserved, but more were present in life. Type information. *° Derivatio nominis. —The specimen is named after its discoverer, Gerard R. Case. Genus TRACHYSCORPIO, new genus Eoscorpiidae of large size, carapace with deep, nar- row median sulcus, dividing the cephalic area into two narrow, inflated parts; facetted lateral eyes well devel- oped, surrounded by a narrow rim. The pleural area of carapace is very wide; ornamentation comprises large nodular scales. Pectines very large, no areoles, serrated edge; apparently a fossorial adaptation. Derivatio nominis. —trachy (Gr.) = rough (pertain- ing to the skin). Type species. — Trachyscorpio squarrosus, n. gen., n. sp. Geological range. —Lower Carboniferous. Remarks. —The narrow cephalic area and the wide pleural areas of the carapace will distinguish this genus from all others. The lack of fulcra on the large, finlike pectines, devoid of areoles, is an important generic distinction. The fossorial adaptation of the pedipalp is also very important. The maxillary lobes of the second pair of coxae are very short. 30 The following information was kindly sent me by Dr. Donald Baird of Princeton University. “‘The holotype of Eoscorpius casei (PU 87266) was discovered by my field assistant, Gerard R. Case, on 21 August 1967. It is preserved on a split section of fine-grained, reddish-brown sandstone cobble containing comminuted plant ma- terial. The cobble lay below high-tide level near the western end of the Carboniferous cliff exposure that forms West Bay, southwest of Parrsboro, Cumberland County, Nova Scotia. We were unable to determine the exact source bed, but the cliff between Union Valley and Watering Brook is composed of vertically-standing reddish- brown sandstones and shales belonging to the lower or “red” facies of the Parrsboro Formation of the Mabou Group. This formation has been dated on megafloral and microfloral evidence as Westpha- lian A, Lower Pennsylvanian. The sedimentology, stratigraphy and paleontology of the locality is discussed by Carroll et al. (1972, pp. 58-64), where the Eoscorpius specimen is cited on p. 64.”’ A.S.C. Trachyscorpio squarrosus, new species Text-figures 83A—D Specimen I. —The holotype is shown on Text-figures 83A, B (BM(NH) In.25985). This specimen reveals about half of the carapace, but includes important structures such as the facetted lateral eye, the cephalic shield, the pleural area of the carapace, rims, both lateral and basal, and the nodular ornamentation. The parts preserved permit a reasonable reconstruction (see Text-fig. 83A). The facetted lateral eye is reniform, and clearly re- veals subrounded facets, is surrounded by a raised rim, and possibly retains as many as 150 ommatidia (see Text-fig. 83B). The carapace is bounded at the base by a raised rim that is thickest at the center and narrows toward the genal angles. The lateral margins of the carapace are bounded by a narrow marginal rim that tapers anteriorly and presumably disappears in the area anterior to the facetted lateral eyes. The pleural area of the carapace is quite strongly elevated, and orna- mented with small granules. The cephalic area also is highly raised, rather narrow in comparison with other Lobosternina, and split in the center by a narrow, smooth sulcus. Both lobes of the raised cephalic area are covered with conspicuous large nodules that vary in shape from rounded to irregular-elongated (see Text- fig. 83A). The rest of the pleural area of the carapace is covered with nodular ornamentation. The median eyes are unknown, as they were not preserved. How- ever, they undoubtedly were well developed as all Lo- bosternina with facetted lateral eyes always have the accompanying median eyes, which are placed on a tear- shaped mound anterior to the median sulcus. The sul- cus was preserved. The facetted lateral eye is 26.5 mm from the base of the carapace, measures 2.4 mm in width and pos- sibly 4.0 mm in reconstructed length. The entire car- apace probably measured 44 mm in width at center, and slightly less than that in length. This would indicate Text-figure 83.—Trachyscorpio squarrosus, n. gen., n. sp. From the Lower Carboniferous (Tuedian), Cementstone Group, Calcifer- ous Sandstone Member, near Newton Farm, Foulden, Berwickshire, Scotland. Collected by T. M. Owens, 1924. See foldout inside front cover for explanation of abbreviations. A. Holotype, Specimen I (BM(NH) In.25985). Note the posterior median sulcus of the carapace. B. Detail of the partially preserved right lateral compound eye of the holotype (above). C. Part of the left coxosternal area and left pectine, nearly complete. Paratype, Specimen III (BM(NH) In.25986). D. Distal end of tibia of pedipalp. Paratype, Specimen II (BM(NH) In. 25984). E, F. Trachyscorpio (?) sp. (BM(NH) In.39765). From the Lower Carboniferous (lowermost Bernician or Viséan S-1), Glencartholm Volcanic Group, Glencar- tholm, River Esk, Eskdale, Langholm, Dumfriesshire, Scotland. E. A lobosternous abdominal plate, presumably a 7rachyscorpio. F. Showing part of the doublure. FossiL SCORPIONIDA: KJELLESVIG-WAERING 189 a scorpion of approximately 250 mm in length, a for- midable scorpion, but, nevertheless, not comparable to some of the giants among scorpions such as Gigan- toscorpio willsi Stermer or Brontoscorpio anglicus Kjellesvig-Waering. Specimen II.—A fragment (BM(NH) 1In.25984), tom Cc inner art dor art lat cr measuring about 18 mm x 20 mm, dark-brown in col- or, may represent the third joint past the coxa, or tibia, of the pedipalp, and is noteworthy for revealing the large nodular ornamentation as well as the serrated carina or edge, as in the anterior legs of Garnettius, showing that this scorpion was an agile digger. Two 190 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 rounded areas, marked by large, elongate nodules, re- veal the articulating end that adjoins the chela. The rounded area that occurs above the other represents the inner edge, which has been flattened during fos- silization. The dorsal and lateral carinae, or crests, are well developed and surmounted with large elongate nodules. These crests indicate that this fragment is part of a pedipalp, and is here considered a paratype. Specimen ITT. —BM(NH) In.25986, here designated as a paratype, retains the right pectine along with the right side of the first three coxae. Unfortunately, the opercular plates, sternum and last pair of coxae are not preserved. The second pair of coxae reveals a very short maxillary lobe, indicating that the first pair of coxae would also have very short maxillary lobes, which were not preserved. The third pair of coxae is elongate and cup-shaped. The fourth pair is unknown. The en- tire coxosternal arrangement, therefore, would be like that in the Eoscorpiidae, except more primitive, as the maxillary lobes of the second pair of coxae are only slightly developed. The second coxa retains orna- mentation of small pustules, which probably was pres- ent on the other coxae. The pectine is very large, with elongated, large, rounded teeth. There is no development of areoles in the median lamella, although small semilunar scales do occur and the anterior lamella is divided into ap- proximately three parts. Important on the generic level is the absence of fulcra. The pectine measures 26.0 mm in width, approximately the same in length; the teeth are 4.6 mm in length and approximately 18 were pres- ent on each comb. Type information. —Lower Carboniferous, Tuedian; Calciferous Sandstones, Cementstone Group, near Newton Farm, Foulden, Berwickshire, Scotland. All three specimens were preserved in greenish-gray shale along with fragments of plants on the same bedding plane. Derivatio nominis.—squarrosus (L.) = rough with stiff scales or processes. Remarks. —The very wide pleural area of the cara- pace is a feature that is unknown in other scorpions of this family. The distinctive coarse ornamentation like- wise is a unique characteristic, as well as the thickened edge of the tibia of the pedipalp, an adaptation for digging, such as in many other arthropods. The three specimens, all from the same locality, are considered to represent one individual because: |. The size of the three is compatible with a single individual; 2. The rough, nodular ornamentation is the same on all three specimens; 3. The preservation and charac- teristics of the bedding plane are the same on all. As stated above, the three pieces indicate a scorpion of approximately 250 mm in length (carapace to telson inclusive). This scorpion must provisionally be considered an Eoscorpiidae because the coxosternal area is not com- pletely known. Moreover, from what is known of the coxosternal area, it is evident that this scorpion has a more primitive development of this area than Eo- Scorpius, as the maxillary lobes of the second pair of coxae are barely developed. Trachyscorpio (?) species Text-figures 83E-F A single lobostern abdominal plate, unusually well- preserved, gives some light on the breathing apparatus of these early scorpions. The abdominal plate is pre- served with the doublure displaced away from the edge. This is not a torn edge, but is a straight edge that was free and not attached to the underlying abdominal plate. Thus the lobostern abdominal plate had a long slit between the edge of the plate and the doublure, as Wills (1947) determined for some of the Triassic scorpions, in particular, a lobostern which he figures on plate VI, figure 9 (see Bromsgroviscorpio willsi above). The an- terior transverse ridges are well developed. There are no spines on the lip of the doublure or abdominal plate, but there is a single row of small rounded scales. The abdominal plate measures 18.5 mm wide and 5.7 mm long at midsection. Ornamentation consists of very small, sparse, semilunar scales. Type information. —Glencartholm on the River Esk, Eskdale, Langholm, Dumfriesshire, Scotland. From the Lower Carboniferous, Glencartholm Volcanic Group (=lowermost Bernician or S1 of the Viséan); BM(NH) In.39765. Genus ESKISCORPIO, new genus Eoscorpiidae (?) with elliptical carapace, large bul- bous, anteriorly-located, facetted marginal eyes, main- ly in front of the median eyes, which are small and located on a small ocellar mound. Carapace smooth, without development of an elevated cephalic shield. Derivatio nominis.—The name is taken from the River Esk, Scotland. Type species. —Eskiscorpio parvus, n. gen., n. sp. Geological range. —Lower Carboniferous. Remarks. —This genus, although unknown from the ventral side, retains certain characteristics that allow it to be classified with a great degree of certainty as to the generic level. For higher category, it should be ac- knowledged that the opposite is the case. The shape of the carapace indicates a forerunner of the Buthiscor- plidae. It seems preferable, however, tentatively to re- fer this genus to the Eoscorpiidae, although this is high- ly speculative. On the other hand, its being included in the superfamily Isobuthoidea is not so speculative. The species, although firmly established on the generic level, will have to await knowledge of the underside FossIL SCORPIONIDA: KJELLESVIG- WAERING 191 before its true taxonomic position on higher levels is established. Eskiscorpio parvus, new species Text-figure 84 The holotype specimen (GSE 2139) is preserved as a dorsal impression of a nearly complete scorpion lack- ing only the legs and most of the chelicerae. The overall aspect is that ofa stout, but small, scorpion, indicating a female. The carapace is rounded along the anterior margin, and long and truncated at the base by a narrow basal rim. No marginal rim, nor ornamentation of any kind is present on the carapace. It is also devoid of a raised cephalic shield. Large, bulbous, marginal com- pound lateral eyes occur at the anterolateral areas, and facets can be determined on the left eye, which is the better preserved of the two. A round ocellar mound occurs between the lateral eyes and is located very close P6 Ch Ch f aes fo Noe cm P3 ae ad P4 eee \] Tl2 T/ ee Wa) = Till Toes —= TE Be T4 \ TIO T5 eres | T6 T9 WAS T8 ——) 1mm Text-figure 84.— Eskiscorpio parvus, n. gen., n. sp. Holotype, GSE 2139. From the Lower Carboniferous (lowermost Bernician or S1 of the Viséan), Glencartholm Volcanic Group of Calciferous Sand- stone Measures, at Glencartholm, 31 miles SSW of Langholm, Dum- friesshire, Scotland. See foldout inside front cover for explanation of abbreviations. to the midanterior margin. The median eyes are very small, in keeping with the correlation noted elsewhere in this paper, that large compound eyes are associated with small median eyes, and vice versa. The chelicerae are too incomplete for description, but they appear to have been rather large. The pedi- palps are perfectly preserved and are rather stout with the fingers incurved, that is, the fixed finger is bent backward to accommodate the forwardly-bent free fin- ger. The edges of the fingers are cultrate. The preabdomen is peculiar in having all of the ter- gites of approximately the same length. Each is bor- dered by the usual anterior transverse ridge. Again, like the carapace, none of the seven tergites of the preabdomen shows any ornamentation, including the important (for species determination) seventh tergite. The cauda reveals at least two superior carinae and two more lateral ones that are well developed even on the last tergite. The telson is short, with a rounded short vesicle and a very short, not curved, aculeus. The latter is not a feature commonly encountered and is in great contrast to some of the great scimitar-shaped stingers of other Paleozoic forms. Measurements (in mm) of the holotype (GSE 2139).— length width Carapace: 2.36 2.81 Pedipalp: Humerus: 1.26 Brachium: Chela: Hand 1.03 1.0 Movable finger 1.09 Tergites: 1 0.63 2 0.63 3 0.8 4 0.8 5) 0.8 6 0.82 7 0.91 8 1S) 9 Iles) 10 15) 11 1S) 12 15) Telson: Vesicle: 0.61 0.7 Aculeus: AST Greatest width of preabdomen at fifth tergite: 3.82 Derivatio nominis. —>! Type information. —The holotype (GSE 2139) is a 31 Kjellesvig-Waering did not provide us with a reference to the trivial name of Eskiscorpio parvus, n. sp., although it clearly is de- rived from (L.) parvus = little. I suspect Kj.-W. may have been think- ing small in terms of the anomalously short telson and aculeus of this species. P.R.H. 192 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 single specimen preserved in dark-gray finely-mica- ceous shale from the Lower Carboniferous Glencart- holm Volcanic Group of Calciferous Sandstone Mea- sures (=lowermost Bernician or S1 of the Viséan), at Glencartholm, 313 miles SSW of Langholm, Dumfries- shire, Scotland (Waterston, pers. commun., March 14, 1967). This locality is on the River Esk and is also known as Eskdale. Remarks.—Gigantoscorpio willsi Stermer, Ar- chaeoctonus glaber (Peach), Pseudoarchaeoctonus den- ticulatus Kjellesivg-Waering, Centromachus euglyptus (Peach), and Anthracochaerilus palustris Kjellesvig- Waering occur at the Esk River locality. The first three are fully known from the dorsal side, as a glance at the respective carapaces proves (Stormer, 1963, fig. 23 for G. willsi, Text-fig. 21E here for A. glaber, and Text-fig. 22 here for P. denticulatus). However, Anthracochaer- ilus palustris and Centromachus euglyptus (Peach) are known only from the ventral sides, whereas Eskiscor- pio parvus is known only from the dorsal, thus criticism might be made that E. parvus may represent one of those two forms. However, this is not possible. The telson of A. palustris is very large with a great vesicle and curved aculeus as against the short bulbous vesicle and short straight aculeus of E. parvus. The caudae are different; that of A. palustris is very stout, whereas the cauda of E. parvus is very slender. Nevertheless, dif- ferences might conceivably, but not logically, be con- sidered sex differences. In this respect the sex differ- ences would be of the opposite of what is commonly considered, as the wide preabdomen of E. parvus clear- ly denotes a female, whereas that of the more slender A. palustris is almost certainly a male. However, there are greater and more obvious dif- ferences. Although the seventh tergite of EF. parvus is known only from the dorsal side, it is devoid of carinae and is very short, whereas that of A. palustris has well- developed carinae and is much longer. If a scorpion has carinae developed on the venter of the seventh tergite, it follows that these carinae are developed much better on the dorsum. This is a feature that 1s present in all fossil and even living scorpions. Furthermore, no scorpion would have prominent development of scales on the underside of the preabdomen as in 4. palustris and be smooth on the dorsal as in E. parvus. It would be a complete departure, if not impossibility, for a primitive scorpion like E. parvus with marginal, compound lateral eyes to have an advanced coxo- sternal arrangement such as is present in A. palustris. With regard to Centromachus euglyptus (Peach), it is sufficient to state that Stormer’s restudy of the ho- lotype showed that it is a holosternous scorpion (Stormer, 1963, p. 80, text-fig. 32), whereas Eskiscorpio parvus with its lateral, marginal compound eyes surely is a lobostern. Family PAREOBUTHIDAE, new family Isobuthoidea with elongate hexagonal sternum; pec- tine with unjointed or undivided rachis and basal la- mella. Type genus. —Pareobuthus Wills, 1959. Remarks. —Clearly this scorpion, although known only from fragments, is a lobosternous form, with coxal arrangement typical of the superfamily Isobuthoidea, namely, having the third pair of coxae abutting the sternum, whereas the last pair abuts against the genital operculum. Wills (1959, p. 267) proposed the genus Pareobuthus for the specimen, which he had previously (1925) described as Eobuthus sp. and which he rightly wanted to separate from the genus /sobuthus. He stated that the coxosternal arrangement and shape of the ab- dominal plates closely resembled Jsobuthus. I agree that among numerous important differences, the above two are significant, but they are of greater than generic value. The coxosternal arrangement of Pareobuthus agrees with that of Isobuthidae, but differs in the shape ofthe sternum, which in the latter family is pentagonal, whereas in Pareobuthus it is elongate-hexagonal. Re- gardless of whether maxillary lobes may or may not be present in the first and second pairs of coxae, at present unknown, the genus Pareobuthus necessitates the erection of a family separate from the Isobuthidae. Perhaps as great a difference is found in the type of combs of Pareobuthus and Isobuthus. In Pareobuthus, they comprise an unjointed base or rachis, well-de- veloped fulcra and numerous teeth, the exact number of which is only important on the generic or, better, species level. In [sobuthus, as shown by J. kralupensis, the rachis is composed of a basal, jointed lamella fol- lowed by an area containing many sclerites, well-de- veloped fulcra, and teeth that are very elongated and numerous. The peculiar unjointed character of the pectines is known only in some of the Lower Devonian scorpions (see Branchioscorpio, Text-fig. 101A). Genus PAREOBUTHUS Wills, 1959 Characters as described for the family and, further- more, lobosternous abdominal plates that are deeply bilobed and without trace of doublures. Type species. — Pareobuthus salopiensis Wills, 1959. Geological range. —Carboniferous. Remarks. — Although the first two pairs of coxae are unknown, the genus can easily be distinguished from other genera by the hexagonal sternum and peculiar unjointed pectines. FossIL SCORPIONIDA: KJELLESVIG- WAERING 193 Pareobuthus salopiensis Wills, 1959 Text-figure 112K 1925. Eobuthus sp. Wills, pp. 87-97, pl. 3; text-fig. 3. 1949. Eobuthus holti Pocock. Petrunkevitch (partim), p. 137. 1959. Pareobuthus salopiensis Wills, pp. 267-269, text-fig. 1. The species has been fully described and figured by Wills (1925, 1959) and there is no necessity for du- plication here. Type information. —Coal Measures, Upper Anthra- conaia modiolaris Zone, probably near the Maid Coal of North Wales Coalfield, Preesgweene Colliery, Wes- ton Rhyn, Shropshire, England; GSM 87231. Remarks. —This is a highly important specimen for phylogenetic purposes. The lobosternous plates are deeply cleft in the middle and completely devoid of doublures. This genus and species therefore seems to be intermediate between the Lobosternina and the Bi- lobosternina, although definitely of the Lobosternina stock. Family KRONOSCORPIONIDAE, new family Isobuthoidea with first pair of maxillary lobes spat- ulate and greatly elongated; lacrimiform sternum, the anterior of which is very narrow and functions as a maxillary lobe, forming part of the base of the pre-oral chamber; tarsal spurs (ungues) and posttarsus greatly elongated; carapace with facetted lateral eyes. Type genus. — Kronoscorpio n. gen. Remarks. —The strange elongation of the sternum into what may be considered a “maxillary lobe”’, ac- tually forming part of the ventral part of the pre-oral chamber, is sufficient for considering this family unique and comparison is therefore superfluous. It is clearly an Isobuthoidea having the fourth pair of coxae abut- ting the genital opercula. Genus KRONOSCORPIO, new genus Kronoscorpionidae with subquadrate cephalotho- rax, anteriorly-located median eye node, relatively small, forwardly-directed, rounded median eyes and large, compound lateral eyes at the anterolateral angles of the carapace. Derivatio nominis. —kronos (Gr.) = sea god. Type species. —Eoscorpius danielsi Petrunkevitch, 1913. Geological range. —Pennsylvanian of Illinois. Kronoscorpio danielsi (Petrunkevitch, 1913) Plates 14-15; Text-figures 85, 112H 1913. Eoscorpius danielsi Petrunkevitch, p. 43, pl. 4, fig. 16; text- fig. 8. 1949. Alloscorpius danielsi (Petrunkevitch). Petrunkevitch, p. 153. 1953. Alloscorpius danielsi (Petrunkevitch). Petrunkevitch, p. 29. Specimen I.—The holotype (UMMP 7216) is well preserved, retaining the carapace, parts of the prosomal appendages and the entire preabdomen, in a typical Mazon Creek ironstone concretion, and is mainly in an undistorted and uncrushed condition. The carapace is elongate, wider than long, but with a decidedly square aspect. The genal angles are round- ed; the lateral margins converge slightly anteriorly and the anterolateral angles are rounded. The anterior mar- gin is nearly straight or slightly emarginate. The base is recurved. The most prominent feature on the carapace is the ocellar mound, which is centrally located at the ante- riormost part. This contains two relatively small, round median eyes, which are anteriorly-located on the ocel- lar mound and directed forward. The median eyes are located 0.7 mm from the anterior margin, 9.5 mm from the posterior margin, 3.7 mm from the lateral margins, and are 0.8 mm in diameter. A small depression sep- arates the median eyes. Embracing the scutelliform eye node is a deep Y-shaped sulcus whose stem extends posteriorly almost to the base of the carapace. Ridges adjacent to the sulcus are prominently raised. From these ridges, the carapace slopes away to the margins, except at the basal margin, which is bordered by a pair of narrow curved cephalic ridges that raise this basal part of the carapace considerably. The carapace mea- sures 14.3 mm in width along the base, 11.9 mm in width at midsection, 10.8 mm in width at the center of lateral eyes, and 11.0 mm in length. Two bulbous, elongate, reniform facetted eyes, with distinct om- matidia, occur at the anterolateral angles of the cara- pace (see Pl. 14). Each facetted eye is composed of about 25 ommatidia. The compound lateral eyes are relatively large, and bulge over the edges of the cara- pace, both vertically and laterally, at a level much low- er than the wide cephalic ridges of the carapace. A narrow palpebral lobe may exist, and is faintly shown in the enlargement of the photograph on Plate 14. This would result in an eye much like that present in many eurypterids. The facetted eyes measure 1.6 mm in length and 0.75 mm in width and protrude above the carapace about 0.2 mm (probably more in the original). These eyes are located 7.5 mm from the base of the carapace, and slightly behind the anterolateral margins. The me- dian eyes are in line with the anterior end of the bul- bous lateral eyes. Ornamentation of the carapace consists of scattered granules in no discernable pattern. There is no orna- mentation such as was noted on the inner part of the cephalic ridges by Petrunkevitch (see Pl. 15). The prosomal appendages are well preserved, al- though not all are present. A part of a stout chelicera is present at the anterior of the carapace, but because of its poor preservation does not merit further descrip- tion. The left pedipalp is present, although parts of the 194 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 hand and fingers are not. The pedipalp is a very elon- gated organ, and the part of the hand preserved indi- cates a narrow chela. The width of the hand is 2.7 mm. The tibia is strongly constructed and considerably longer than the femur. The femur of the pedipalp is concave along the an- terior margin and convex on the posterior margin, re- sulting in a slightly curved joint. The trochanter is unusually long, in keeping with the obviously long fe- mur and tibia. Only one of the walking legs is preserved on the left side, and parts of two on the right. Probably the leg that is almost completely preserved is one of the an- terior ones, possibly a first leg, with fragments of a second leg beside it. The leg on the left side is possibly a fourth leg. This opinion is based on the fact that the leg on the left side has much longer joints than those of the most nearly complete leg on the right side, and also on the short coxa, which does not retain a max- illary lobe. Of considerable morphological importance is the presence of some flattened joints, and paired basitarsal and tibial spurs or spines, and the termi- nation of the legs in three spines, the mesial one being the largest, and considered to be the transtarsus. The nearly-complete walking leg reveals seven joints past the coxa. The coxa is definitely present and is short, nearly oblong, and without maxillary lobes. The first joint past the coxa of the first (?) walking leg is short; the second joint is long and wide at the center, and was apparently flattened in life. The third joint is also flattened and not so long as the first, fol- lowed by two joints nearly equal in length. The fourth joint shows a constriction at the distal end, where two sockets occur, each of which carried a spur. Petrun- kevitch (1913, fig. 8) shows one of these spurs, but no sign appears in the cast. These are the tibial spurs. This joint is also shown with the spurs in place on the cor- responding joint on the leg of the left side, which is followed by the fifth joint with two spurs (basitarsal), one on each side. The sixth joint is very short and terminates in three very long spines; the mesial is con- sidered to be the posttarsus. The spines are very nar- row, long and straight. This tridactyl termination is quite unlike anything known in other scorpions, either fossil or Recent. The leg preserved on the left side, which has been interpreted as a fourth walking leg, is similarly con- structed except that the joints are longer and much slenderer. The spurs are more prominently developed, but the terminal spines seem to be shorter. This may be due to their not being entirely exposed and therefore appearing to be shorter, in the plaster cast. The preabdominal tergites are all present. The first four are preserved in one part and the remaining three, as well as parts of two abdominal plates, in another segment. The tergites increase in length posteriorly. The anterior tergites are rounded at the corners. The last three are too poorly-preserved and incomplete to merit description. Some of these tergites are bordered anteriorly by a transverse ridge. No other ornamen- tation is apparent on the dorsal side of the tergites. A prominent transverse ridge, however, is lacking on the seventh tergite, indicating that the seventh tergite is preserved as a ventral impression of the dorsal tergite. This tergite is pentagonal in outline, comparatively short, and without crests. The base indicates that the cauda is wide and strongly-constructed. To the right of and anterior to the seventh tergite are remnants of the last two abdominal plates, which are lobosternous, and rounded on each half. A prom- inent raised marginal rim, denoting a doublure, bor- ders the plates. No openings are revealed, although these would not be expected, and possibly not pre- served in the plaster cast. Specimen IT. —FMNH (PE) 32085, collected from the Francis Creek Shale in Pit 11, on the Kankakee Co.—Will Co. line, Illinois (formerly in the private col- lection of Jerry Herdina, No. H563a,b). Part and counterpart of a scorpion, including dorsal and ventral sides, preserved from the inside and con- sisting of the prosoma and basal parts of the append- ages as well as the entire preabdomen. The tail and distal parts of the appendages are missing, although most of the basal joints of the legs are preserved, but poorly. The dorsal side is not well preserved, but the ventral side reveals the entire coxosternal area, pec- tines, genital opercula and all abdominal plates, in- cluding the seventh tergite of the preabdomen. Judging from the wide girth of the abdominal plates, this spec- imen is apparently a female. The dorsal side is sufficiently well-preserved to as- sure identification with Kronoscorpio danielsi (Pe- trunkevitch), as the carapace shows the characteristic raised cephalic shield and, in particular, the large fac- etted lateral eyes and relatively small median eyes, placed very far anteriorly and pointing forward, which are typical of the species (see Text-fig. 85B). The seven tergites show nothing that has not been described for the holotype, and the small section of the appendages remaining do not show anything of interest, as pres- ervation is much too poor for description. However, the great doublures of the abdominal plates are pre- served and, coupled with the total lack of ornamen- tation on the tergites, assure determination with Kro- noscorpio danielsi (Petrunkevitch). The ventral side is well preserved, however, and is of great interest, since this rare and unusual scorpion was previously known only from the dorsal side. The FossIL SCORPIONIDA: KJELLESVIG- WAERING 195 important coxosternal area is well preserved, showing a sternum that is uniquely different from any other scorpion in being greatly elongated anteriorly so that the sternum may be said to have developed an anterior “maxillary lobe’. This anterior prolongation actually occupies a position similar to the maxillary lobes, in this case, separating the two maxillary lobes of the second pair of coxae; moreover, it forms part of the base of the “tube” leading to the mouth. Otherwise, the sternum is lacrimiform, pointed at the posterior where it fits against the two opercular plates. The first pair of coxae meets at midsection; the maxillary lobes are greatly expanded into a spatulate area. The second pair of coxae abuts against the aforementioned ‘“‘max- illary lobe” of the sternum, and further forward meets at midsection. The two lobes extend to less than one- 17 / / half the length of the maxillary lobes of the first pair. However, a great part of the first pair of coxae has already been squeezed away from the midline. The third pair of coxae abuts the main body of the sternum as in many scorpions, whereas the fourth pair abuts the genital opercular plates, as is characteristic of the superfamily Isobuthoidea. The genital opercular plates are large, elongate, lacrimiform, with the anterior tri- angular, where the sternum fits between them, and the posterior rounded. The enormous maxillary lobes of the first pair of coxae are armed with a single large spine or tubercle on the anterolateral angle of each lobe. This is seen, from the inside or dorsal side of the maxillary lobe, as a deep hole that presumably would be the inside of a spine or very large tubercle. This has not been seen on Text-figure 85.— Kronoscorpio danielsi (Petrunkevitch). Specimen II, FMNH (PE) 32085 (formerly Jerry Herdina Coll. H563a,b). From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Pit 11, Peabody Coal Company, Will Co.-Kankakee Co. line, IL. See foldout inside front cover for explanation of abbreviations. A. Showing superbly the organization of the ventral surface of prosoma and mesosoma. The organization is unique among scorpions. The anterior stippled areas are mud-filled depressions in the surface of the anterior lobes of the first coxae. Each lobe bore a single spine, the base of which is a circle. B. Dorsal surface. 196 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 other scorpions either fossil or living, but this scorpion has other more important unique features, such as the termination of the legs and the type of sternum. The pectines are poorly preserved, but enough is present for a good reconstruction —at least all essential details are known. The pectines are broad and very long, flipperlike, with a broad anterior lamella that is jointed, presumably as in other Isobuthoidea; the mid- dle lamella is broad and long with small, even-sized, rounded areoles. The fulcra are large and well devel- oped. The denticles are thin, very long and numerous— an estimate is possible since the width of the pectines is known—approximately 48 teeth on each pectine. The abdominal plates are lobosternous, deeply bi- lobed with a very large doublure developed. They re- semble the abdominal plates of the Czechoslovakian forms, where the doublure is also apparently massively developed. Measurements (in mm) of FMNH (PE) 32085.— length width Prosoma: 9.2 10.6 Preabdomen: 23.4 (estimated) 15.0 (estimated) Overall: 80.0 (estimated) This important specimen furnishes and completes much of the knowledge of this peculiar scorpion which was needed, for when this study was begun in 1965, the underside was completely unknown. Therefore it was a pleasure to see the underside, which, as predicted from the unusual dorsal side, was unique and up to the author’s expectations. Type information. —The holotype was in the L. E. Daniels collection and came from the lower Francis Creek Shale of the Middle Pennsylvanian, Carbondale Formation (Westphalian C) overlying Coal 2, at Mazon Creek, Grundy County, Illinois. The plaster cast used here is in the U.S. National Museum. The holotype, thanks to Mr. Bret S. Beall, is now known to be in the University of Michigan Museum of Paleontology, cat- alogued as UMMP 7216. It is a single nodule half from **Mazon Creek, near Morris, III.” The other specimen described is from the lower Francis Creek Shale in Pit 11 of the Peabody Coal Company on the Kankakee Co.—Will Co. line, Illinois, collected by Jerry Herdina, and filed in his collection as No. H563 (now FMNH (PE) 32085). Remarks. —This scorpion is one of the most unusual and significant fossil scorpions known, although Pe- trunkevitch failed to recognize and describe certain very important morphological structures. Facetted eyes at the anterolateral angles of the carapace, tridactyl termination of the legs, and rounded lobosternous ab- dominal plates in this species have important bearing on scorpion taxonomy and phylogeny. Originally de- scribed by Petrunkevitch as Eoscorpius danielsiin 1913, it was referred in 1949, on the basis of morphological characters of far less importance, to the newly-named genus Alloscorpius, with A. granulosus (Petrunkevitch) as type species. In his papers of 1953 (p. 29) and 1955, nothing is said of the above-mentioned important morphological characters. As a matter of record, he had illustrated these facetted eyes very well in his text- figure and photograph of 1913, which is reproduced here as Plate 14, and Plate 15, figure 1, with enlarge- ment of the compound eyes to show the distinct om- matidia. The presence of compound lateral eyes in Chelicer- ata is well known in the Merostomata. In the Scor- pionida, as early as 1885, Whitfield showed the pres- ence of these structures in Proscorpius osborni (Whitfield), but for various reasons, unfortunately all fallacious, this discovery was not taken seriously. A review of the holotype of Proscorpius osborni by Kjel- lesvig-Waering (1966) has verified Whitfield’s original determination. Wills (1947) also reported well-devel- oped facetted eyes in the Triassic scorpions of England. However, in the Treatise on Invertebrate Paleontology, Petrunkevitch (1955, p. 54) stated: Compound eyes such as are found in Xiphosura, Eurypterida, Crus- tacea and Hexapoda are never present in any Arachnida. The organs of scorpions belonging to the Triassic family Mesophonidae and those of the Palaeocharinidae (Devonian members of the order Tri- gonotarbida) claimed to be compound eyes, are unlike true com- pound eyes and are probably sense organs of some unknown func- tion. This statement is clearly erroneous as pertaining to both the scorpions and the Palaeocharinidae. Concerning the holotype of K. danielsi, the facetted lateral eyes are bulbous, protrude considerably above the surrounding surface, and are distinctly composed of about 25 roughly-hexagonal ommatidia. In modern scorpions, the group of lateral eyes, which number from one to five on each side, occur in the same position as the compound eyes of K. danie/si. Petrunkevitch had figured these eyes as facets in his 1913 paper (fig. 8), but unfortunately would not accept that the lateral eyes were compound eyes and referred to them as “alveolate areas’. Facetted lateral eyes are not as rare in fossil scorpions as has been thought. In addition to the instances cited above, I have recently found exceptionally well-pre- served facetted eyes, which previously had not been reported, in the holotype of Eoscorpius carbonarius (Meek and Worthen), Archaeophonus eurypteroides Kjellesvig-Waering, Garnettius hungerfordi (Elias), Eoscorpius distinctus (Petrunkevitch), etc., all of which are redescribed here. The facetted eyes are much like those found in the FossiL SCORPIONIDA: KJELLESVIG-WAERING 197 eurypterids. It is probable that the facetted eyes evolved into the present-day series of lateral eyes, inasmuch as they occur approximately at the same spot in living scorpions. The compound eye of the Carboniferous scorpions is reniform or bulbous in shape, and the margins are rounded and clearly delineated, sometimes by a distinct rim (see photographs of Eoscorpius dis- tinctus). The individual facets are sharply defined and hexagonal in shape. It also appears (see Wills, 1947, p. 20) that by Triassic time, the individual facets were becoming less sharply defined, and the entire facetted eye lost its reniform or bulbous symmetry. Possibly later some ommatidia became detached from one another, whereas others became obscured or obliter- ated, and the separate, small, round lateral eyes of living scorpions eventually developed. This could ex- plain why there is variation in some species of living scorpions in the number of individual lateral eyes. There is no question that many fossil scorpions did not have facetted lateral eyes.*? Petrunkevitch failed to note in his description of Kronoscorpio danielsi that the two posterior ventral abdominal plates are preserved and that these are of the bilobed type, which designates the scorpion as one of the Lobosternina. Equally important to the understanding of this fossil is the presence of three long terminal spines in the legs, along with paired spurs on the two penultimate joints. The tridactyl termination has previously gone unrec- ognized in the scorpions and, although Petrunkevitch showed these structures (1913, fig. 8), he failed to grasp their significance. These trifid claws alone would have been sufficient for the separation of this scorpion from any other known in Petrunkevitch’s classification as given in the Treatise on Invertebrate Paleontology (1955). The trifid termination (see Stormer, 1963, p. 86) is probably a very primitive one, and like the presence of facetted lateral eyes, is a character in common with the eurypterids. This scorpion appears to be an anach- ronism in the Pennsylvanian and in many respects is perhaps more primitive than some of the Silurian forms such as Proscorpius osborni (Whitfield), which had de- veloped double claws not greatly unlike most Penn- sylvanian and Recent scorpions. The presence of a tridactyl claw is rare in living Arthropoda. R. F. Lawrence (pers. commun., Oct. 26, 1965) states that these are known in the family Triae- 32 Whenever acids are used to free scorpions from the matrix, it is quite possible that the individual facets may be destroyed from the cuticle that is left as a residue. In this connection, experiments with acids on the eyes of various arthropods show that the ommatidia, occurring below the exocuticle, are mainly or altogether destroyed, leaving little or no trace on the exocuticle. E. N. K.-W. nonychidae, the dominant family of Opiliones in South Africa, and in the larvae of some cantharid beetles. It also commonly occurs in living and fossil Acari, in- cluding the Devonian Protocarus crani Hirst. Among the primitive trilobitomorphs, the tripartite termination is common in Middle Cambrian genera such as Marrella of the marrellomorphs, Burgessia of the order Burgessiida and Naraoia of the order Nec- taspida. In the trilobites the tridactyl claw is common and may be found in such divergent genera as Ole- noides, Triarthrus, Cryptolithus, Asteropyge, Phacops, Ceraurus, etc. Although unknown in the living scorpions, the tri- dactyl claw, in a considerably modified form, is known in the Carboniferous Eoscorpius pulcher (Petrunke- vitch). Wills (1959, pp. 279-282, pl. 50, figs. 9-10; text-figs. 5-8) shows where the mesial spine or dactyl is turned over to serve as a heel for two curved, rather “normal” scorpionid claws. In this case, the mesial joint, as well as the two claws, is greatly reduced in comparison with the three long, straight spines of the scorpion described here. A further reduction of the posttarsus or heel would result in a termination not greatly unlike that of living scorpions. I agree with Wills (1959, p. 281) and Stormer (1963, p. 87) that the modified tridactyl claw of Eoscorpius pulcher would be useful to an aquatic animal, but not to a land-dweller. The tridactyl claw of K. danielsi is even more suited for an aquatic existence. The posttarsus of modern scorpions therefore rep- resents the mesial spine of the tridactyl claw of scor- pions like K. danielsi, except that it has been reduced to a small heel or tubercle that rests on the ground. The posttarsus has been turned over so that the dorsal surface now rests on the ground, ventral to the claws. This development is suggested by the morphology of Eoscorpius pulcher. It is therefore highly probable that the posttarsus of K. danielsi was turned under and posteriorly, so that two long claws occurred in front and one large “‘claw” (posttarsus) occurred in back. This would give a ““snow- shoe” effect and indeed would be excellent for walking over soft bottom muds such as surely occurred in the Mazon Creek environment. The presence of forwardly-directed median ocelli may be a good indication of the predaceous habits of this scorpion. This characteristic is not known in other scorpions that normally have the median eyes pointing upward and laterally from the carapace.** 33 Waering’s search for the holotype of this unusual scorpion (now known to be in the Univ. of Michigan Museum of Paleontology) proved fruitless, and all that was available on which to base a de- scription was the original photograph (see Pl. 4, fig. 16, taken by Petrunkevitch in 1913) and a plaster cast of the original concretion 198 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Superfamily PARAISOBUTHOIDEA, new superfamily Lobosternina with the maxillary lobes of first two pairs of coxae reaching the anterior margin, first pair crowded out of the midline; second pair meets at mid- line; third pair abuts sternum and fourth pair abuts the genital opercula. Type family. —Paraisobuthidae, n. fam. Remarks.—The differences in the coxosternal ar- rangement between this superfamily and the Isobu- thoidea rests in the great elongation of the maxillary lobes of the second pair of coxae; in the latter, the lobes reach only midway the length of the first pair of coxae, as against reaching to the anterior margin. In the Para- isobuthoidea they have squeezed the first pair of coxae away from the midline. This is a fundamental change. Thus, the arrangement of the first three pairs of coxae is much as in modern scorpions. Of course, the simi- larity to modern terrestrial scorpions rests there, as the fourth pair of coxae abuts the opercula, and the preab- domen retains lobosternous abdominal plates rather than true sternites. This is a well-defined, easily rec- ognizable group, which undoubtedly, from all evi- dence, developed from the Isobuthoidea. Family PARAISOBUTHIDAE, new family Paraisobuthoidea with well-developed schizochroal lateral eyes and small pentagonal sternum. Type genus. — Paraisobuthus, n. gen. Remarks. —This family differs from the Isobuthidae in the development of long maxillary lobes of the first two pairs of coxae, and the squeezing away of the first pair from the midline with the second pair abutting one another. The difference is of superfamily impor- tance. Genus PARAISOBUTHUS, new genus Paraisobuthidae with well-developed paired ce- phalic cheeks, pectines with apparent double row of fulcra, and greatly-dilated inner areoles. Pectines very hirsute along the edge. Free finger curving over the recurved immovable finger with both edges meeting throughout the entire cultrate edge. Derivatio nominis. —para (Gr.) = near + Isobu- thus = a scorpion genus. Type species. — Paraisobuthus prantli, n. gen., n. sp. Geological range. —Upper Carboniferous. Remarks. —This represents an unrecognized genus made by L. E. Daniels in 1898, now in the U.S. National Museum (USNM 27921). The description given here, in the absence of the holotype specimen, is based on this plaster cast and several enlarge- ments of Petrunkevitch’s glass negative, now at Yale University. A.S.C. that had previously been included in the Isobuthidae because the fourth pair of coxae abutted the opercu- lum. It is an important genus, which now is accorded its proper taxonomic position. Paraisobuthus prantli, new species Text-figures 86-88, 112A, 113C4 1904. Eobuthus rakovnicensis Frié (partim), p. 74, fig. 91. 1911. Anthracoscorpio sparthensis (Baldwin and Sutcliffe). Pocock (partim), pp. 13, 20. 1923. Anthracoscorpio sparthensis (Baldwin and Sutcliffe). Moore, p: 132; pl. 1, fig. 3: 1953. Isobuthus rakovnicensis (Frié) (partim). Petrunkevitch, p. 21, figs. 19, 123. 1955. Isobuthus rakovnicensis (Frié). Petrunkevitch, p. 78, fig. 46. The holotype (BM(NH) I.2950), comprises a well- preserved, rather large scorpion preserved so that both dorsal and ventral sides are revealed (see Text-figs. 86A, B). It is preserved in light-greenish-gray shale, appearing to be an underclay. The part comprises the specimen on the dorsal side with almost the entire surface preserved, with some details of the underside. The counterpart reveals al- most the entire underside with the left side of the dorsal surface showing. This has been partly overturned dur- ing deposition so that this part of the dorsal area and its relationship to the upper part can be easily noted (see Text-fig. 86A). Coloration. —A considerable amount of the original cuticle is preserved, which shows that this scorpion was mainly light-brown in color with the tips of the fingers black as in many living scorpions. The carapace (see Text-figs. 86B, 88E, F) is subquad- rate, with two well-developed (elevated in life) cephalic cheeks separated by a deep sulcus. The base is bounded by a wide rim that continues into the lower half of the carapace as a lateral rim. The median eyes appear as elliptical, but undoubtedly were round in an uncrushed state, and they surmount a small ocellar node, which in turn is divided by a narrow sulcus. The ocellar node or mound, is sublacrimiform and located intramargin- ally on the anterior of the carapace. Two schizochroal lateral eyes are located at the anterior part of the ce- phalic cheeks; they are relatively small, reniform and consist of about 200 lenses. These are schizochroal in that each is small, rounded, and separated by sclerotic walls (see Text-figs. 88D-—F), all of which are contained in a bulbous, reniform structure. The anterior of the carapace is rounded at the anterolateral margin and enough is preserved of the middle to show that a glos- sate process was present. The coxosternal arrangement is clearly shown, re- vealing a small pentagonal sternum that is slightly long- er than wide (see Text-figs. 86A, 87G). The first pair FossIL SCORPIONIDA: KJELLESVIG-WAERING 199 of coxae does not meet at the midsection, but meets against the maxillary lobes of the second pair of coxae. The coxae of the first pair have massively-developed maxillary lobes. The second pair of coxae has well- developed narrow maxillary lobes that reach to the anteriormost part of the carapace. The oral chamber of this scorpion, at least on the ventral surface, would be much as in Recent scorpions where the maxillary lobes of the second pair of coxae have crowded out the first pair of maxillary lobes. The third pair of coxae abuts against the sternum, whereas the fourth pair abuts against the genital operculum. The genital operculum appears to be composed of two round parts, although it was not possible to determine whether or not these were two round sclerites that have been coalesced into one. The pectines (see Text-figs. 86A, 87C) are fairly well preserved and show that they are very large with a thickened basal lamella composed of at least five dis- tinct segments followed by the median lamella, which is a very wide area in which many irregular-sized, rounded setaceous sclerites occur. Fulcra are very well developed, with another row of sclerites occurring im- mediately above the fulcra, which appear as a double row of fulcra. The sclerites of the median lamella be- come smaller toward the tips of the pectines. A large, rounded, dilated lobe is present on this scorpion, which shows that it is very likely a female. This is common in some living scorpions (for example, Tityus mela- nostictus Pocock), and is known as the basal pectinal lobe. The teeth are very well preserved. They are long and rounded at the ends, and at least 35 are estimated to be present. The entire base of the pectines (see Text-fig. 88A) is preserved on the larger specimen. This shows that the pectines are attached to a small trapezoidal sclerite, separated by a small, median, lanceolate, two-jointed organ. The small median organ can be seen best after the removal of the coating of shellac and in a dry state. It is very similar to that found in Branchioscorpio rich- ardsoni and Gigantoscorpio willsi. Text-figure 86.—Paraisobuthus prantli, n. gen., n. sp. Holotype, BM(NH) 1.2950, part and counterpart. From the Upper Carboniferous (Westphalian B-C), Radnice Group, Rakovnik, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Ventral view of holotype. B. Dorsal view of the holotype with some abdominal plate impressions and other ventral features. 200 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Text-figure 87.—Paraisobuthus prantli, n. gen., n. sp. Holotype (BM(NH) 1.2950), part and counterpart. From the Upper Carbon- iferous (Westphalian B-C), Radnice Member of the Radnice Group, Rakovnik, Czechoslovakia. See foldout inside front cover for ex- planation of abbreviations. A. The preabdomen; partly restored, but with tergites in place as preserved. B. Basal part of fixed finger of nght pedipalp. The thin edge, with most of the setal openings, represents the cuticle that is preserved in the impression. C. Pectine detail. D. Ventral part of the fixed finger of the left pedipalp. E. Ventral part of the fixed finger of the right pedipalp. F. Telson, drawn from the counterpart. G. Coxosternal organization. Text-figure 88.—Paraisobuthus prantli, n. gen., n. sp. Holotype, -=> BM(NH) I.2950. From the Upper Carboniferous (Westphalian B- C). Radnice Group, Rakovnik, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Ventral details, preserved through the dorsal side of the ho- lotype. B, C. Tergites 10 and 11, seen on the ventral holotype slab. D. Showing the schizochroal facets of the left lateral eye. E. Resto- ration of the prosomal carapace. F. Somewhat distorted prosomal carapace of the holotype. FossiIL SCORPIONIDA: KJELLESVIG-WAERING “ts. sup er vs a “ ? Ne . ( Ge : 201 202 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The tergites of the preabdomen (see Text-figs. 86A, B; 87A) increase in size gradually posteriorly and each is bounded on the anterior by a strong transverse ridge. The seventh tergite has two faintly-developed carinae dorsally and apparently none ventrally. The pectines undoubtedly cover the first and second abdominal plates, as shown by the counterpart. These are followed by three others in excellent preservation, each bounded on the anterior margin by a transverse ridge. All are distinctly deeply-bilobed and rounded at the posterior angles. A total of five gill-bearing lobo- stern abdominal plates are present. The well-preserved pedipalps are of unusual interest (see Text-figs. 86A, B; 87B, D, E) with the trochanter, femur, tibia and all of the chela present. The tibia and the femur are approximately the same size. The tro- chanter shows many small setal openings ventrally, and some scattered setal sites are present on the femur and tibia. The fingers of the chela are very long. The movable finger is bent backward and shows a single series of openings along the edge that undoubtedly were occupied by the setae, as some of them are still pre- served. The tips of the fixed fingers possibly termi- nate in a double or bifurcating tip, between which the falcate end of the movable finger would fit. This is the first time that this structure has been noted in the pedipalp of any scorpion. The bifurcating end is known in the chelicerae of scorpions, but only on the movable or free finger. Small pointed denticles very definitely occur on the edge of one of the bifurcations at the tip. No denticles have been noted on any of the rest of the fingers, although preservation is very good and reveals small bristles along the cultrate edge. The free finger curves inward, is pointed or falcate and very long. This appears to be a very smooth scorpion as neither the ventral nor the dorsal side shows any granules ex- cept very sparse and minute ones on the tergites. The walking legs do not merit description since the ends were not preserved. They had been obliterated by care- less mechanical development. This seems to have been done long ago by a chisel that broke away the parts where the distal joints of the legs were preserved. The cauda reveals four well-preserved tergites, but only fragments of the twelfth along with the well-pre- served telson or stinger. The cauda is stout with the last three tergites greatly inflated and arched (see Text- fig. 86B). Superior lateral and ventral crests seem to be developed on all the tergites. These are slightly ser- rate but none noticeably so. The telson (see Text-figs. 86B, 87F) is rather long, with a bulbous vesicle and a curved aculeus, similar to many living scorpions. The vesicle is 4.7 mm wide (thick) and 5.7 mm in length. The aculeus is 3.0 mm (estimated) in length. Measurements (in mm) of BM(NH) I.2950.— length width Chela: 23.8 Hand 13 4.9 Free finger 16.5 Femur (dorsolateral): 11.0 Tibia (dorsolateral): 11252 Abdominal plate: No. 1 (covered) (covered) No. 2 6.9 (incomplete) No. 3 6.9 13.0 No. 4 6.9 1222 Tergite:3* No. 1 2.3 No. 2 D5 No. 3 Past) No. 4 4.0 No. 5 4.2 No. 6 522 No. 7 deal No. 8 6.0 Derivatio nominis. —*° Type information. —Holotype (BM(NH) I.2950) consists of two parts, ventral and dorsal of a complete scorpion, preserved in light-green shale from the Upper Carboniferous Radnice Member of the Radnice Group (Westphalian B/C), Rakovnik, Czechoslovakia. Remarks. —It seems incongruous that after the ho- lotype had been figured repeatedly, first by Fric (1904, 34 There is considerable intersegmental tissue between the tergites. E. N. K.-W. 35 Dr. H. W. Ball kindly sent the holotype to me in Trinidad in July, 1968. At the time, I did not question the generic and specific determination previously made (Eobuthus rakovnicensis Fri¢) by Frié and (Isobuthus rakovnicensis (Fri¢)) by Petrunkevitch, both of whom had studied the material. In November, 1970, I visited the National Museum in Prague where all of the Bohemian fossil scorpions were deposited. By study- ing the holotype of Eobuthus rakovnicensis, | recognized that another species was present in the material from Rakovnik. It was later determined, on my return to Oslo, Norway (Dec. 1970) that the beautiful specimen that had been presented to the British Museum (Natural History) by Fri¢ was different, and it is now named Para- isobuthus prantli in honor of the well-known Czech paleontologist, Ferdinand Prantl. It was a pleasant surprise in Mozambique, in January, 1972, when I was restudying the entire ““Jsobuthus prob- lem’’, to learn that Pocock (1911, p. 13) had noted that the British Museum syntype of Eobuthus rakovnicenis represented a different genus and species from the Prague Museum specimen, which Pocock considered the holotype (=lectotype). Unfortunately on page 20 of the same publication, he considered it to be conspecific with Eo- scorpius sparthensis, although on p. 13 he did consider it as provi- sional (“‘for the time being’’). It would have saved me considerable trouble had I not waited so long to determine what Pocock previously had thought, but in moving around so much, I have had to work from microfilm, and one knows that this method, although having its advantages, is highly inconvenient for ready reference. Actually, it was not until August, 1977, when Dr. Ball again sent me the original, that most of the figures were made and the “/sobuthus problem” was resolved. E. N. K.-W. FossIL SCORPIONIDA: KJELLESVIG- WAERING 203 fig. 91), and later by Petrunkevitch (1955, fig. 46), that it should emerge as not only an important new genus and species, but also the type genus of a new family and type family of a new superfamily.*° Paraisobuthus frici, new species Text-figure 89 1873. Cyclophthalmus senior Corda (partim). Fric, pl. 1, fig. 2. 1904. Isobuthus kralupensis Thorell and Lindstrém (partim). Fri¢, pl. 10, fig. 11. 1953. Isobuthus kralupensis Thorell and Lindstrém (partim). Pe- trunkevitch, p. 20. The holotype (No. 579; NMP Inv. 827) comprises only fragments, but these are fortunately of consider- able diagnostic value. They consist of a complete chela of the pedipalp, parts of combs, fragments of what possibly is the third or fourth leg and small parts of an abdominal plate. The well-preserved pedipalp comprises a very long hand with an incurved fixed finger that is distinctly falcate at the terminus. The cultrate edge is bounded by small, evenly-spaced, and very short setae. The free finger, equally as slender as the fixed, is curved forward and crosses the anterior falcate terminal part of the opposing finger in a scissorlike manner. The cultrate edge is bordered with the same type of minute setae as the immovable finger. No other ornamentation oc- curs. Hand width is 5.8 mm, hand length is 12.3 mm and free finger length is 16.4 mm (see Text-figs. 89A, GC): The combs are poorly and incompletely preserved, but they reveal 27 teeth on the large piece and 12 on Text-figure 89.—Paraisobuthus frici, n. sp. Type specimens in the Sternberg collection, National Museum, Prague, Czechoslovakia. From the Upper Carboniferous (Westphalian B-C), Radnice Member, Lower Gray Formation, Kralupy Hill, Cervena Hirka, Czechoslovakia. See foldout inside front cover for explanation of abbreviations. A. Holotype (NMP Inv. 827). Showing part of the ventral surface, pectine, and the enormous pedipalp. B. Paratype (NMP Inv. 826). C. Specimen (NMP GH 1973 (73)). Here the free finger of the pedipalp has been distorted and bent backward. 204 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the small. Fulcra are present and raised areoles are found on the rachis. Probably about 50 teeth occurred on each pectine. The teeth are the long type. The abdominal plates are distinctly lobosternous and have a well-developed doublure. The leg, which seems to belong to the fourth or last pair, shows tibial spurs, probably double, as well as uneven basitarsal spurs, one of which is greatly elon- gated (see Text-fig. 89A). The tarsus is long, with a serrated edge on the posterior side and a pointed an- terior terminal end. The posttarsus and claws are too poorly preserved for description. A long tibia (?) and femur (?) are also present, possibly of the third pair of legs. The cauda, fragments of two legs, and the free ramus of the pedipalp occur in a specimen which is registered as Inv. 826 in the collections of the National Museum in Prague, Czechoskovakia. It is from the same locality and horizon as the holotype, and also was previously identified by Fri¢ as /sobuthus kralupensis Thorell and Lindstrém, according to the label on the specimen (see Text-fig. 89B). The two leg fragments, quite likely parts of the last two pairs of legs, reveal that the basitarsal and tarsal spines were well developed; the ungues and posttarsus were small and developed without supplementary ser- rations or denticles. The tarsus in each leg is very long. The cauda is preserved laterally in its entirety, at- tached to fragments of the sixth and seventh tergites of the mesosoma. It is very thick, with most of the tergites completely smooth. The shortness of the distal tergites indicates a female. The telson is a smooth bul- bous vesicle with the usual curved aculeus. The pedipalp, which was the main factor in deter- mining the identification of this specimen as a species of Eobuthus, reveals only the entire fixed finger. It occurs with a crack that was at first mistaken for the junction of the hand with the free finger. However, the recurving of the finger shows that it is the fixed finger or part of the fifth joint. The finger is recurved, cultrate, and with a small notch at the extremity. Type information.—The holotype and paratype, preserved in shaly sandstone, come from Kralupy Hill, Cervena Hirka, which lies 25 to 30 km north of Prague, Czechoslovakia, from the Upper Carboniferous Rad- nice Member of the so-called Lower Gray Formation, Westphalian B/C (Ivo Chlupac, pers. commun., Dec. 3, 1970). The holotype is registered in the collection of the National Museum, Prague, as Sternberg Collec- tion No. 579; NMP Inv. 827. The paratype is registered as NMP Inv. 826. Derivatio nominis. —Named in honor of Dr. Anton Fric (Fritsch), to whom we owe most of our knowledge of Czechoslovakian scorpions. Remarks. —The differences between Paraisobuthus frici, n. sp., and P. prantli from Rakovnik, are nu- merous. The hand of the latter is much shorter, with much longer fingers, and with a groove running the length of the fingers. The denticles at the end of the fingers of P. prantli are not present in P. frici. The double row of large fulcra in P. prantli differs from the small fulcra of P. frici. The abdominal plates of P. prantli do not have as rounded a configuration as in P. frici. Another specimen, also from Kralupy Hill, and reg- istered in the National Museum collections at Prague, Czechoslovakia, as GH 1973 (73) is also referred to this species. It consists of a nearly complete chela of the pedipalp, but the free finger is not incurving, al- though it is obvious that this finger has been distorted, as no scorpion could have a chela in which the fingers could not possibly close. The row of setae on the cul- trate edge, however, is present and agrees otherwise with the holotype specimen (see Text-fig. 89C). Paraisobuthus duobicarinatus, new species Plates 16-18; Text-figure 90 1911. Eobuthus holti (2) Pocock, pp. 15-16, pl. I, figs. 2a, b. 1913. Isobuthus holti (Pocock). Petrunkevitch (partim), p. 22. The holotype comprises a single specimen, pre- served in an ironstone concretion, that reveals both dorsal and ventral sides of a robust scorpion. Almost the entire preabdomen is present, and, although tele- scoped, is remarkably preserved. The entire fourth walking leg is also preserved in an uncrushed condi- tion. A part of the fourth abdominal plate was lifted out of the specimen, exposing the entire comb and, more important, part of the details of the gills and gill chamber. The seventh tergite is preserved apart as an external mold, showing some degree of original inflation and both dorsal and ventral sides. The species is therefore well founded, requires a name, as it can easily be rec- ognized again, and is referred on the basis of the combs to the genus Paraisobuthus. The tergites have been slightly telescoped, and the second to the seventh inclusive are preserved. They Text-figure 90.—Paraisobuthus duobicarinatus, n. sp. Holotype, GSM 30245 (ventral part) and 30246 (dorsal part). From the Upper Carboniferous, Coal Measures (lower Anthracosia similis—Anthra- conaia pulchra Zone), Shipley (Digby) Clay Pit, Kimberley, Not- tinghamshire, England. See foldout inside front cover for explanation of abbreviations. A. Shows the ventral half of the holotype. The deep lobation of the fourth abdominal plate (AP4) is notable; also the excellently- preserved right pectine. B. Shows the dorsal half of the holotype. C. Right posterior claw of leg IV of the holotype. D. Right pectine. E. Detail of the center of the right pectine, showing the areoles on the pectinal teeth. FossiL SCORPIONIDA: KJELLESVIG-WAERING 1mm 206 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 reveal a row of pustules on the posterior edge, and scattered pustules in the epimeral parts. Each is an- teriorly bounded by a strong transverse ridge (see Text- fig. 9OB). The dorsal side of the seventh tergite (Text-fig. 90B) has two median carinae, each of which is surmounted with very coarse pustules. There is another carina on each side, also surmounted by very coarse pustules. Fine to very coarse pustules cover the center and lateral areas, apparently not in a uniform manner or discern- ible pattern. The tergite undoubtedly has a strong an- terior transverse ridge, and the base also seems to be bounded by a transverse ridge. The underside (Text-fig. 90A) shows that the ab- dominal plates are lobosternous, long, and divided by a deep cleft in the posterior median part. All are bound- ed anteriorly by a strong transverse ridge. The ventral part of the seventh tergite is nearly smooth except for two median carinae, close-set, each surmounted by medium-sized pustules. Part of the abdominal plates covered most of the pectines, but a crack in the rock permitted the lifting of a small slice of rock containing the right abdominal plate, revealing the entire pectine. It fortunately re- vealed loose fragments of cuticular matter representing the inside of the abdominal plate and the body wall. These skins were lifted off the specimen and embedded in Canada Balsam. Much to my surprise, the pieces showed shiny dark-brown cuticle (see Pl. 16, fig. 1; Pl. 17) along with white, thick, spongy tissue (see Pl. 16, figs. 2, 3; Pl. 18), which contrasted dramatically with the sclerotized parts. Covering the spongy tissue on one side were abundant cone spines such as charac- terize the gill tracts of eurypterids (see Wills, 1965; Waterston, 1975, p. 241). This is the first direct evi- dence of gills in lobostern scorpions, although it has certainly been known that such would be the case. The white spongy tissue that composes the gill tract shows no discernible structure (see Pl. 18, fig. 3). Cone spines (see Pl. 16, figs. 1-3), however, are dark brown, occur only on one face, the outer face of the body wall, and show some striations. There are some much small- er spines (see Pl. 17, fig. 1), which are also considered to belong to the gill tracts, as the form is much like that of the cone spines, rather than to the gill chamber (see Waterston, 1975, p. 250, fig. 3). It is considered, therefore, that the lobostern abdom- inal plate is the same as that which was revealed by Wills (1965) and especially by Waterston (1975). This reveals a large gill chamber with the gill tract being part of the body wall and with the abdominal plate overlapping. Waterston (1975, p. 252) suggests with good reason that the water entered the gill chamber laterally and was expelled through the posterior slit, as happens in Limulus. This is probably why the spines present on the posterior slit, bordering the posterior of the abdominal plate and doublure (see Text-figs. SOF, G), point to the posterior. If the spines pointed ante- riorly, they would slow the passage of the water. In- asmuch as they point posteriorly, the spines do not impede the passage of the water outward and further- more serve to keep extraneous material out of the gill chamber. The pectines (see Text-figs. 90A, D) are beautifully preserved, showing a segmented anterior lamella pos- sibly of three or four joints; the distal one disappears before the actual distal end of the entire comb is reached. The middle lamella is composed of irregularly-sized areoles, rounded, that range in shape from circular to irregular-elliptical. The areoles decrease in size toward the distal end. The inner part of the margin retains a very large, elliptical, basal pectinal lobe. This structure, present also in Paraisobuthus prantli, is known in much reduced form in some species of 7itvus, where it de- notes the female. It is probable that this specimen is a female, as the preabdomen is very wide. Fulcra, sub- triangular and large, are well developed and separate the teeth of the combs. The latter are long, and ap- proximately 30 occur on each comb. Areoles, although scarce, occur on some of the teeth of the combs (see Text-fig. 90D). The fourth walking leg is preserved in its entirety except for the posttarsus (see Text-figs. 90OA, B). The coxae (III, IV) are long and flaring. The first joint is short, followed by a long second joint. The third joint is robust, greatly expanded and with a strong dorsal carina. The fourth joint is long and dorsally carinated. This joint carries a long, thick spine at the distal end that has secondary spines on at least the inner part. This species possibly has double spines, although only one was noted, but the other is covered, if present. The fifth joint or basitarsus is also long with a pair of rather long spines. The tarsus or sixth joint is long and carries the paired claws or ungues. The claws are of the long type, falcate at the end, and with a single row of sharp spines on the ventral side. The spines are perpendicular to the shaft of the ungues, thus assuring the greatest traction against the substrate. Type information. —Holotype is from the L. Moysey Collection (GSM 30245, 30246, 30247 and 30248), from the Upper Carboniferous, Coal Measures (lower Anthracosia similis—Anthraconaia pulchra Zone), at Shipley (Digby) Clay Pit, Kimberley, Nottingham- shire, England. The slides of the gills are also filed with the specimen. Derivatio nominis. —duo (L.) = two +. bicarinatus (L.) = double-keeled. FossiL SCORPIONIDA: KJELLESVIG- WAERING 207 Remarks. — Pocock (1911, p. 15) “‘provisionally” re- ferred this scorpion to Eobuthus holti Pocock, but the presence of the type of keels on the venter of the sev- enth tergite is sufficient to refer this scorpion to a sep- arate species. In E. holti, the keels are convergent; in the new species described above, they are close-set and parallel. Paraisobuthus duobicarinatus differs from the type species, P. prantli, in having very well-developed ca- rinae on the dorsum and venter of the seventh tergite that are lacking in the Bohemian form. This difference, as followed in the modern treatment of living scorpion classification, is sufficient to refer them to separate genera, and this may be the case here when more is known of each. I have referred the species nomen duo- bicarinatus to Paraisobuthus because the large rounded sclerite of the pectines is also present in Paraisobuthus prantli. Paraisobuthus virginiae, new species Text-figure 91 1913. Eoctonus miniatus Petrunkevitch (partim), pp. 51-52, pl. II, fig. 15. 1953. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 15. Included as a paratype of Eoctonus miniatus Pe- trunkevitch was a small scorpion whose only similarity to the holotype was the small size. It is, however, a lobostern scorpion, whereas E. miniatus is a holostern. In addition to this major difference, the coxosternal arrangement is very different, for example, the third and fourth pairs of coxae abut the sternum in E. mi- niatus, whereas in the new species, the fourth pair abuts the genital opercula. Petrunkevitch (1953, p. 15) ap- parently arrived at the conclusion that the specimen did not represent Eoctonus miniatus Petrunkevitch and the problem was dispensed with by the statement, ““The paratype, No. 132, proved to be a distorted specimen of Alloscorpius granulosus (Petrunkevitch) and there is no need to consider it further here.” The specimen is preserved in two parts, showing both dorsal and ventral surfaces in nearly complete preservation. As in other Mazon Creek materials, it is preserved in an ironstone concretion (see Text-figs. 91A, B). Apart from the underside, the characteristics of the carapace (e.g., outline, strong cephalic shield, shape of the eye node and eyes, and the presence of three carinae on the ventral side and two on the dorsal Text-figure 91.—Paraisobuthus virginiae, n. sp. Holotype, YPM 132, part and counterpart. (This is a paratype of Eoctonus miniatus Petrunkevitch.) From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. A. Counterpart, showing lobate abdominal plates and characteristic carinate postabdomen. B. Dorsal surface, showing complete carapace and nearly complete slender pedipalp. 208 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 side of the cauda), clearly show that this specimen is not referable to Eoctonus miniatus Petrunkevitch. The dorsal side is largely flattened, but the left lateral eye, and in particular, the median eye node with the two eyes, are exceptionally well preserved. The median eye node is scutelliform, with two relatively small round eyes separated by a space slightly larger than the width of the eyes. The anterior glossate area of the carapace is well preserved and occurs in front of the median eyes. The cephalic part of the carapace has been almost completely flattened, but definite traces are present on both sides. The lateral eyes occur on the same level as the median eyes, and are small, elliptical and protrud- ing. The preabdomen is preserved in its entirety, but some telescoping of the tergites occurs. Compression has obliterated most of the anterior transverse ridges, but they can be discerned in rubber casts. Little or- namentation was noted, as the texture of the rock on this particular nodule is too coarse for good preser- vation. The tergites increase in length posteriorly to reach greatest length at the seventh tergite. Greatest width of the preabdomen occurs at the anterior of the fourth tergite. The pedipalp is preserved nearly whole and reveals a strong chela, the complete length of which is probably greater than that of the femur and tibia together. Over- all, the pedipalp is quite slender, and in life would reach past the second caudal segment. Both the femur and tibia are narrow, with the femur being considerably longer. The hand is long and slender, and the finger, not complete, is unusually long and bent. It is consid- ered to be the fixed finger; the other finger is not pres- ent. The coxosternal region clearly reveals a Paraisobu- thoidea with two pairs of coxae in front of the sternum, both with well-developed maxillary lobes, reaching to the anterior end, the third pair abutting the sternum and the fourth pair abutting the opercular plates. The sternum is not well preserved in outline, but is large and pentagonal. The opercular plates are large, elon- gate, rounded or ellipso-triangular, considerably longer than wide. Of particular interest is the presence of the abdom- inal plates. They are lobosternous, but each shows a very noticeable large, rounded area in the center of each half, occupying most of the inner surface of the plate. The last abdominal plate shows these rounded areas in excellent preservation. These areas are actually depressions in the plates, occurring as shallow pouches and are considered to be the gill chambers. The last preabdominal tergite appears to have two faint crests on the ventral side. The dorsal side of this tergite has two crests, at a diagonal to each other, that are better seen in the latex casts. The cauda is preserved in its entirety and is known from both dorsal and ventral sides. Each tergite in- creases in length slightly to the posterior. The ventral side reveals three prominent crests on each tergite, whereas the dorsal shows only two. The telson or sting- er is large, rather elongated, and bent as in other scor- pions. It definitely appears to have two crests on the ventral part of the vesicle, a structure common in many living scorpions. Measurements (in mm) of YPM 132.— Prosoma: Width at base: 3.4 Length: 3.4 Palpus (chela) length: 4.0 Pedipalp: Femur length: 2.5 Tibia length: Pa Length of tergites: 1 0.8 2 0.8 (telescoped?) 3 1.0 4 2 5 1.2 (telescoped?) 6 1.3 7 2.0 8 1.6 9 1.8 10 2:1 11 22 12 25 Telson length: 2.5 (incomplete) Vesicle width: 0.8 Total body length, excluding chelicerae: 26.0 (estimated) Abdomen length: 8.3 Caudal length: 12.8 (incomplete) Type information. — Pennsylvanian, lower part of the Francis Creek Shale, at Mazon Creek, Illinois. Holo- type is YPM 132, part and counterpart. Derivatio nominis.—Named in honor of my wife Virginia, who has contributed much to this mono- graph, and to whom fell the more onerous tasks of this work. Remarks. —Petrunkevitch (1913, p. 52) states that the underside was so deformed that ‘tno details of structure can be made out’’. This is incorrect. Although there are better-preserved scorpions, the preservation is quite good, and many details of major importance can clearly be discerned (see Text-fig. 91A). In com- paring the lobosternous scorpions occurring in the Ma- zon Creek biota, namely Eoscorpius carbonarius, Kronoscorpio danielsi, and Telmatoscorpio brevipec- tus, only the first mentioned, FE. carbonarius, might be confused with this new species. The main difference between Eoscorpius and Paraisobuthus is in the max- FossiIL SCORPIONIDA: KJELLESVIG-WAERING 209 illary lobes of the first two pairs of coxae, which reach to the anterior end or to the same level in Paraiso- buthus, in contrast to the maxillary lobes of Eoscorpius, in which the second pair only reaches the midsection of the first pair. In Paraisobuthus, the second pair has squeezed the first pair away from the midline, and the first pair of lobes is narrow, whereas in Eoscorpius the first pair is greatly expanded or spatulate. The same differences apply to the maxillary lobes in Kronoscorpio danielsi; the sterna are also totally un- alike. The large pentagonal sternum of Paraisobuthus is sufficient to separate it from Telmatoscorpio with its small barrel-shaped sternum. On the species level, if it is necessary to list some differences, details of the seventh tergite, which nearly always is an easy character for quick identification, will separate all three listed lobosternous scorpions in the Mazon Creek biota. X Text-figure 92.—Leioscorpio pseudobuthiformis, n. gen., n. sp. Ho- lotype, BM(NH) In.22832 (one of Pocock’s subsequent study suite of Buthiscorpius buthiformis (Pocock), designated as a paratype by Petrunkevitch, 1953). From the Upper Carboniferous, Upper Coal Measures, Etruria Marl Group, Coseley, South Staffordshire Coal- field, England. See foldout inside front cover for explanation of ab- breviations. A. Complete specimen. B. Lobosternous abdominal plate (AP4), stippled area restored. Genus LEIOSCORPIO, new genus Paraisobuthidae with very narrow maxillary lobes of the first and second pairs of coxae; dorsal side smooth without any granulation and with seventh tergite with a single median carina. Derivatio nominis. —leios (Gr.) = smooth, bald. Type species.—Leioscorpio pseudobuthiformis, n. gen., n. sp. Geological range. —Upper Carboniferous. Remarks. —No confusion between Leioscorpio and Pseudobuthiscorpius should exist. Although both occur in the same bed, at the same locality, the former is mainly known from the dorsal side, whereas the latter, at least for the important parts, is known from the ventral, but the maxillary lobes are totally different. In Leioscorpio, the maxillary lobes are very narrow, which, along with the lobosternous abdominal plates, rele- gated it to the Paraisobuthidae, although we have no knowledge of the eyes, median or lateral, since the anterior carapace is unknown. The very wide maxillary lobes of the second pair of coxae in Pseudobuthiscor- plus, as well as the protolobosternous abdominal plates, indicate that this genus belongs to a different family, the Pseudobuthiscorpiidae. The complete seventh tergite is preserved in both genera, but the central carina of Leioscorpio is not present in the Pseudobuthiscorpiidae. Pocock (1911, pp. 26-27, fig. 8) mistook the maxillary lobes of the first two pairs of coxae for the chelicerae. Leioscorpio pseudobuthiformis, new species Text-figure 92 1911. Anthracoscorpio buthiformis Pocock (partim), pp. 26-27, fig. 8. 1953. Buthiscorpius buthiformis (Pocock). Petrunkevitch (partim), p. 32. The specimen comprises part of the dorsum of the carapace with only remnants of the pedipalp and legs, except for the fourth leg of left side, which is fairly completely preserved. The preabdomen is complete, but only the anterior caudal elements are present. Orig- inal convexity is preserved. This specimen was one of the four paratypes of Buthiscorpius buthiformis (Po- cock) and like the others, it has been referred to other genera and species. The designation of paratypes was made by Petrunkevitch (1953, p. 32), not by Pocock (1911, pp. 26-28), who merely referred these speci- mens to his previously described scorpion: In addition to this specimen I have several examples from Dudley which, in the absence of satisfactory proof of their distinctness from each other, and from the type, I provisionally refer to this species. The “‘type”’ referred to was the specimen considered the holotype of Buthiscorpius buthiformis (Pocock), 210 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 which is from different beds and from Sparth Bottoms, rather than Dudley (Coseley). One of the important features of this specimen is that the ends of the first and second pairs of coxae are preserved. These are very narrow, of equal thickness, and with the second pair of maxillary lobes reaching the anterior margin to crowd the first pair from the midline. The carapace is preserved only as a fragment, but shows a slight median triangular sulcus. No or- namentation is present. The postabdomen is robust, with the tergites in- creasing in size posteriorly; each is stoutly constructed with a heavy, transverse ridge along the anterior edge. No ornamentation is present on these tergites. The seventh tergite is roughly triangular, long, with a no- ticeable narrow median ridge. The lower part of the preabdomen is broken away to reveal an underlying fourth abdominal plate. This is of the normal lobostern type, being well notched or bilobed at midsection (see restoration, Text-fig. 92B). The caudal segments (T8, T9 and T10) are stoutly constructed, about as long as wide, and also devoid of ornamentation. All are fragmentary. The fourth walking leg is partly preserved and is long, stout and extends posteriorly to the end of the preabdomen, as in living scorpions. Measurements are given by Pocock (1911, p. 27) and will not be repeated here. Type information.—A single specimen, well pre- served, but incomplete, preserved in dorsal aspect in an ironstone concretion collected by Dr. Wheelton Hind from the Carboniferous, Upper Coal Measures, Etruria Marl Group, Ten Foot Ironstone bed at Coseley, South Staffordshire Coalfield, England; BM(NH) In.22832. Derivatio nominis.—pseudo = (Gr.) false + buthi- formis, the species under which this specimen had pre- viously been classified. Remarks. —The differences between this species and the only other lobostern from the Coseley area with which it might be confused, are discussed under “‘Re- marks”’ concerning the generic description. Family TELMATOSCORPIONIDAE, new family Paraisobuthoidea with small barrel-shaped sternum; carapace with small compound lateral eyes, large me- dian eyes anteriorly located. Type genus. —Telmatoscorpio, n. gen. Genus TELMATOSCORPIO, new genus Telmatoscorpionidae with square carapace, anterior not greatly glossate; genital operculum larger than ster- num; pedipalp with deep groove, studded with row of elongate pits (setal sites). Derivatio nominis. —telmatos (Gr.) = swamp + scorpio. Type species. —Telmatoscorpio brevipectus, n. gen., Nn. sp. Geological range. — Pennsylvanian. Telmatoscorpio brevipectus, new species Text-figures 93, 112F 1913. Eoscorpius granulosus Petrunkevitch (partim), pp. 45-46, pl. II, figs. 11, 12. 1949. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 153. 1953. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 29. 1955. Alloscorpius granulosus (Petrunkevitch). Petrunkevitch, p. 73. 1962. Alloscorpius granulosus (Petrunkevitch). Dubinin, p. 429. The holotype (YPM 129) originally was a paratype of Eoscorpius granulosus Petrunkevitch, which has been referred to the synonymy of Eoscorpius carbonarius Meek and Worthen. The specimen consists of two parts of an ironstone concretion that had never been cleaned of the calcite crystals that covered nearly all of the essential details reported here. The specimen is pre- served showing the dorsal side on one half of the con- cretion and the ventral side on the other. Both sides are preserved from the inside. Preservation, after re- moval of the calcite, is unusually good. The dorsal side has been greatly flattened. The carapace is subquadrate with a short anterior glossate process (see Text-fig. 93B). Two subelliptical (probably round in life) median eyes are placed on an elevated, heart-shaped ocellar node, directly behind the glossate process. From the lines present, it appears that an elevated cephalic area was well delineated in the form ofa shield. Facetted lateral eyes, insufficiently well-preserved to warrant description, are present, and presumably are small and not unlike those in Eoscor- pius. A prominent raised posterior marginal rim oc- curs. The dorsal side of the preabdomen consists of the usual seven tergites, all of which increase in length posteriorly. All are bounded anteriorly by a prominent transverse ridge, and all are fringed along the posterior by a single row of small tubercles (see Text-fig. 93B). Indeed, the dorsal side of the carapace and preabdo- men greatly resembles that of Eoscorpius carbonarius, for which it could easily be mistaken if only the dorsal side were preserved. However, the tubercles are far more numerous in the latter than in 7. brevipectus. The last tergite of the preabdomen (T7) is very dif- ferent from that of Eoscorpius carbonarius. It is bound- ed anteriorly by a wide, raised transverse ridge, where- as the rest of the tergite is sparsely covered with tubercles, all minute. None appears aligned in rows or crests. Eoscorpius carbonarius has two well-defined rows of punctae on the corresponding tergite. FossIL SCORPIONIDA: KJELLESVIG-WAERING 211 The sternum consists of a barrel-shaped, or roughly rounded, hexagonal plate (see Text-fig. 93A). This is very small and only the third pair of coxae abuts it. The fourth pair abuts against the large opercular plate. The second coxae occur directly in front of the ster- num, have well-developed narrow maxillary lobes that extend anteriorly as in living scorpions. The first coxae are not well preserved in the region of the maxillary lobes, but the latter are clearly developed, narrow in shape, and abut the maxillary lobes of the second pair. They are not expanded as in the Eoscorpiidae, but appear to be much as in living scorpions such as the Buthidae, Scorpionidae and Diplocentridae, etc. The legs are preserved only in fragments, but enough are present to permit certain important interpretations. The chelicerae are badly smashed and, other than in- dicating their presence, little can be described. The denticles are present, but these are too poorly preserved for description. Three joints were noted, but four joints probably occur. The pedipalp is preserved from the dorsal side, along with the ventral side of the opisthosoma, but only the tibia, hand and fixed finger remain (see Text-fig. 93A). The entire structure reveals a powerful pedipalp (see “‘Measurements’’). The tibia is stoutly constructed, but too poorly preserved to reveal any ornamentation. The hand is poorly preserved, but enough is present to show that it is wide and not very long. The fixed finger is curved backwards, which indicates that the free finger, of which only the base is present, would be curved forward to fit against the immovable finger. No den- ticles occur along the inner edge of the finger, therefore, the inner edges are cultrate. A deep narrow groove is present, parallel to the shaft of the free finger and close to the inner edge. Several deep, elongate small pits, probably sites for bristles, occur on the inner edge of the groove. The walking legs are not well preserved, but show that the tarsus was unusually long, had two very large, curved claws, and a posttarsus, or heel, which was elongated, although preservation of the heel was not complete. It suggests a structure much as Wills found Text-figure 93.— Telmatoscorpio brevipectus, n. gen., n. sp. Holotype, YPM 129. (This is a paratype of Eoscorpius granulosus Petrunkevitch.) From the Upper Carboniferous (Pennsylvanian), Carbondale Formation, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. A. Ventral aspect. B. Dorsal aspect. 212 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 in Eoscorpius pulcher (Petrunkevitch). Double tarsal spurs are present, at least on the third pair of legs (see Text-figs. 93A, B). Fine punctae occur as a single row on, at least, the fourth coxa, trochanter and tarsus. This row of punc- tations is probably present on all the legs, as it has been seen on the two extremes of the leg (e.g., the coxa and tarsus). A small fragment of the matrix, above the tarsus of the second leg, was taken off the specimen and dissolved in acid. This revealed the presence of numerous short brown setae, which undoubtedly oc- curred in the row of punctae. The opercular plates are very large, nearly as large as the sternum. They are elongate, and sublacrimiform, with their greatest axis parallel to the body. The large pectines (see Text-fig. 93A) are preserved in their entirety, except where part of the teeth are covered by the overlying abdominal plate. These re- markable structures appear to have a very wide basal lamella that comprises at least three segments. The distal end is covered by an overlying tergite and may continue underneath. The middle lamellae are com- posed of a great number of rounded sclerites, which are coarsest or largest toward the anterior middle part and diminish in size toward the posterior and distal parts. Fulcra are well developed. The teeth are nu- merous, elongated, and at least 16 can be counted be- fore they are covered by the overlying tergite and first abdominal plate. It should be repeated that the scor- pion is seen from the inside of the ventral side at this point. The pectinal plate is largely covered by the pec- tines, but of interest is the presence of a diamond- shaped organ that may be the median organ of the pectinal plate, and that has been noted on other fossil scorpions. The first abdominal plate is very poorly preserved and only a small part is present. The second and third abdominal plates are very well-preserved and show that they are deeply bilobed, lobosternous, and with a deep cleft in the middle. No ornamentation is present. Type information. —The holotype (YPM 1239) is from the Pennsylvanian, Carbondale Formation, lower Francis Creek Shale at Mazon Creek, Grundy County, Illinois. Derivatio nominis. —brevi (L.) = short + pectus (L.) = shield. Remarks. —The specimen was covered with a thin layer of drusy calcite crystals that obstructed all the details reported here. This layer of crystals was re- moved with HCl, which revealed the beautifully pre- served structures described above. Measurements (in mm) of the holotype (YPM 129).— length width Carapace: 6.8 8.2 at base Median eyes: 0.7 0.55 Tergite: 1 1.4 2 LiF 3 2.0 4 2:35 5 3.0 6 322. ih 4.7 8 5.8 Pedipalp: Femur: 9.0 DES Hand: 2.9 Fixed finger (incomplete): 7.4 Sternum: 1.0 1.2 Family SCOLOPOSCORPIONIDAE, new family Paraisobuthoidea (or Isobuthoidea) with carapace without ‘“‘compound”’ lateral eyes. Type genus. —Scoloposcorpio, n. gen. Remarks. —To all appearances, the large Eoscor- pius-like cephalic cheeks should have relegated this form to the Eoscorpiidae or Mazoniidae. However, the lack of compound lateral eyes precludes the possibility of this form belonging to the Eoscorpiidae, and the presence of lobosternous abdominal plates disposes of the Mazoniidae. It is referred to the Paraisobuthoidea with considerable misgivings, but with our present knowledge, this is the only superfamily that can ac- commodate this family even tentatively. This is based on the lobosternous abdominal plates, which are not excessively bilobate. Genus SCOLOPOSCORPIO, new genus Scoloposcorpionidae with quadrate carapace, me- dian eyes located in the middle of the anterior half of the carapace on a lacrimiform eye node. A deep median sulcus divides the carapace into two well-developed inflated cephalic cheeks. Derivatio nominis. —scolopo (Gr.) = anything point- ed + scorpio. Type species. —Scoloposcorpio cramondensis, n. gen., Nn. sp. Geological range. —Lower Carboniferous. Remarks. — Although the superfamily designation is still not conclusive, there seems little doubt that this family and genus are necessary as they seem to occupy the same taxonomic position among the Lobosternina as the Mazoniidae does in the Holosternina. FossiL SCORPIONIDA: KJELLESVIG- WAERING Pls} Scoloposcorpio cramondensis, new species Text-figure 94 1883. Eoscorpius tuberculatus Peach (partim), pp. 398-400, pl. 23, figs. 8b, c. The holotype (GSE 2173) consists of two detached parts preserved together on the same bedding plane and not more than 20 mm apart, in black micaceous, slightly arenaceous shale. The parts preserved are the carapace and half of a lobosternous abdominal plate, very likely the second (or first) gill-bearing appendage. Another half of an abdominal plate, in the same col- lection, is listed as GSE 5862. The holotype carapace was figured by Peach (1883, pl. 23, fig. 8b). The carapace is much like that in Eoscorpius except that the lack of ““compound” lateral eyes indicates a different family, either in the Isobuthoidea or the Para- isobuthoidea. When the coxosternal area is known, the matter will be settled (see Text-fig. 94A). The carapace is subquadrate, straight along the base, with parallel sides, rounded at the anterolateral angles and with a pointed glossate process in the median an- terior margin. The basal margin is bounded by a thick, raised rim that occurs on the lateral margins and con- Text-figure 94.—Scoloposcorpio cramondensis, n. gen.,n. sp. From the lower Carboniferous (Lower Viséan), Cramond, near Edinburgh, Scotland. See foldout inside front cover for explanation of abbre- viations. A. Holotype (GSE 2173). Prosomal carapace. B. Holotype (GSE 2173). Left half of deeply lobate abdominal plate. C. Paratype (GSE 5862). Right half of a lobate abdominal plate. tinues on to the basal fourth of said margins. Two well-developed, raised, cephalic cheeks sur- round a deep sulcus, and are separated from one another by a deep sulcus. The raised lacrimiform eye node is located well within the anterior margin, in the middle of the anterior half of the carapace. The median eyes are round, and located on the anterior of the lacrimi- form eye node. There are no lateral compound eyes and enough of the carapace is preserved to assure that they would have been preserved if present. The carapace measures 7.1 mm in length and 8.2 mm in width at the base. The associated abdominal plate (see Text-fig. 94B) is a typical lobostern, and seems to be, from its shape, the left half of the second or first gill-bearing abdom- inal plate, which undoubtedly had a median sulcus separating the two lobes, and a wide doublure. The plates are covered with very minute pustules. The plate half measures 6.7 mm in length and 9.3 mm in width. Another abdominal plate (paratype (GSE 5862), see Text-fig. 94C), but of a smaller individual, is also in- cluded in this species. The abdominal plate is probably the second or third, and reveals an entire half of a typical lobostern plate, with the deep median groove and thick doublure. It measures 6.0 mm in width. Type information. —Lower Carboniferous, Calcifer- ous Sandstone Series, Lower Oil Shale Group (lower Viséan), shore, west of Eagle Rock, Cramond, near Edinburgh, Scotland. Derivatio nominis.—Named after Cramond, near Edinburgh, Scotland, the type locality of the holotype. Remarks. — Peach (1883, pl. 23, figs. 8b, c), included under the name of Eoscorpius tuberculatus two cara- paces from Cramond. The subject of 8b is now the holotype of Scoloposcorpio cramondensis, whereas the other may also be referred to the same species, although my identification of the latter is based on Peach’s figure since the original has not been found. The holotype of Eoscorpius tuberculatus Peach was made the type species of the genus Benniescorpio by Wills (1960, p. 322), who also correctly noted the dis- crepancy in Peach (1883), who included several un- related specimens in his species. Benniescorpio tuber- culatus (Peach) is from the Coal Measures at Blair Point, near Dysart, Fife, Scotland, whereas the other specimens were from the Lower Carboniferous Cal- ciferous Sandstone. The species here described, and figured by Peach (1883, figs. 8b, c) represent two of the specimens that had been misidentified. Petrunkevitch (1953, p. 29) accepted Peach’s determination as listed in his synonymy of the species, and neither questioned the determination nor noted the great difference in 214 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 stratigraphic age involved between the specimens.*° Genus BENNIESCORPIO*’ Wills, 1960 (?) Scoloposcorpionidae with quadrate carapace, large median eyes located on a heart-shaped eye node very near the front of the shield, no compound lateral eyes. Type species. —Eoscorpius tuberculatus Peach. Geological range. —Carboniferous. Benniescorpio tuberculatus (Peach, 1883) 1883. Eoscorpius tuberculatus Peach, p. 398, pl. 23, figs. 8, 8a, 8d, 8e. 1885. Centromachus tuberculatus (Peach). Thorell and Lindstrém, Dp: 29: 1911. Archaeoctonus tuberculatus (Peach). Pocock, p. 19. 1913. Archaeoctonus tuberculatus (Peach). Petrunkevitch, p. 34. 1953. Alloscorpius tuberculatus (Peach). Petrunkevitch, p. 29, figs. 27, 28. 1960. Benniescorpio tuberculatus (Peach). Wills, pp. 322-327, pl. 56, fig. 5; pl. 57; text-figs. 28, 29. 1962. Benniescorpio tuberculatus (Peach). Dubinin, p. 429, fig. 1230. Holotype. —GSE 9675 (or 456) and GSE 9676 (or 457). The specimen consists of part and counterpart preserved in gray shale with plant remains, collected from the Coal Measures at Blair Point, near Dysart, Fife, Scotland. For description and discussion see Wills (1960). Family OPSIEOBUTHIDAE, new family Paraisobuthoidea with short, wide, lacrimiform ster- num, with notch at posterior end. The maxillary lobes of the first pair of coxae occur above the maxillary lobes of the second pair and are expanded anteriorly to form flat, spatulate lobes; prepectinal plate greatly developed and with protolobosternous abdominal plates. Type genus. — Opsieobuthus, n. gen. Remarks. — Although it is much too soon to postu- late phylogeny in the fossil scorpions, it is of impor- tance to note that this family seems to have developed directly from the Eobuthidae by the elongation of the second pair of maxillary lobes, and the inflation of the first pair of maxillary lobes, which has been squeezed out of the midline. It differs from the Telmatoscor- 3° It seems highly unlikely that the two parts preserved on the holotype slab (GSE 2173, Text-figs. 94A, B), belong to the same individual, because the complete abdominal plate would have been more than twice as wide as the carapace, which would mean an animal with a preabdomen resembling the eurypterid Carcinosoma. A:S:G, 7 Kjellesvig-Waering never got around to assigning Benniescorpio tuberculatus (Peach) to any family. Such notes as were found were written before he had worked out the ventral structures of Mazonia woodiana, and no longer apply, since M. woodiana is a holostern and B. tuberculatus is a lobostern. A.S.C. pionidae in the shape of the sternum, the shape of the first pair of maxillary lobes (spatulate in the Opsieo- buthidae and narrow in the Telmatoscorpionidae), the lack of prepectinal plates in the Telmatoscorpionidae and the deeply-bilobate abdominal plates of the Tel- matoscorpionidae as compared with the protolobos- ternous plates of Opsieobuthidae. Genus OPSIEOBUTHUS, new genus Opsieobuthidae with pectines having well-devel- oped, bandlike pectinal plate. Derivatio nominis. —opsi (Gr.) = late + Eobuthus = a genus of Carboniferous scorpion. Type species. —Eobuthus pottsvillensis Moore. Geological range. —Upper Carboniferous of In- diana. Opsieobuthus pottsvillensis (Moore, 1923) Text-figures 31A, 95, 112E 1923. Eobuthus pottsvillensis Moore, p. 131, pl. 2, figs. 1-4. 1949. Eobuthus pottsvillensis Moore. Petrunkevitch, p. 137. 1953. Isobuthus pottsvillensis (Moore). Petrunkevitch, p. 22. The holotype comprises a nearly complete scorpion, preserved in greenish-gray shale as a ventral impres- sion, and associated with numerous plants. The scor- pion measures 40.0 mm overall, with a prosoma and preabdomen 20.5 mm in length, and a cauda (including the stinger) 19.5 mm long. The original colors are pre- served, consisting of shiny, bright, brown cuticle throughout, covering the mesosoma and coxal areas, and becoming darker brown along the pedipalps and the extremities of the cauda. Preservation is unusually good and all details are readily discernible (see Text- fig. 9SA). The coxosternal region (see Text-fig. 95B) is pre- served in its entirety. The maxillary lobes of the first pair of coxae above those of the second pair are greatly inflated anteriorly, spatulate, rounded at the outer mar- gins, and do not meet at midsection, but are actually behind and above the thin maxillary lobes of the sec- ond pair of coxae. The very thin lobes of the second pair of coxae meet at midsection; they are elongated and do not project so far anteriorly as do the maxillary lobes of the first pair. The third and fourth pairs of coxae are triangular and elongate, the third abutting against the sternum and the fourth against the opercula. The sternum is wide, lacrimiform, and retains a small notch at the posterior, has rounded sides, and the an- terior is pointed. It measures 2.0 mm in length and 1.7 mm in width. The genital operculum consists of two large ellipsoidal plates that fit the base and part of the sides of the sternum. The pectinal plate is trape- zoidal, very large, long, and divided at the median part posteriorly (see Text-fig. 31A). The pectinal plate is FossiL SCORPIONIDA: KJELLESVIG- WAERING INS composed of two narrow thin plates, each of which is joined by a suture at midsection, where the articulation with the pectines occurs. A narrow transverse ridge borders each plate. Should there ever exist any doubt that the pectinal plate (and pectines) represent a so- mite, the presence of the transverse ridge along the paired pectinal plates should dispel that notion. The large pectines are attached to the middle section of these plates, and are preserved in excellent condition, showing approximately three or four joints along the anterior lamella. The intermediate lamella is undoubt- edly perliform, with numerous small rounded sclerites, and attached to the posterior side of each pectine are Text-figure 95.—Opsieobuthus pottsvillensis (Moore). Holotype, FMNH (UC) 37984. From the Upper Carboniferous (Pennsylva- nian), Pottsville Series, Strip Clay pit, near Owen Co. line, east of Clay City, IN. See foldout inside front cover for explanation of abbreviations. A. Complete specimen. Note the protolobosternous abdominal plates. B. Details of the coxosternal and pectinal areas. approximately 20 long teeth. These teeth, in shape, are the type common in living forms such as the Buthidae. Small fulcra are definitely present. On the preabdomen, all abdominal plates are pres- ent, but the fourth has been pushed under the third. The plates are rounded at the genal angles and have a slight emargination at the middle. These lobosternous plates are therefore of the type referred to here as pro- tolobosternous, and are considered to be derived from the holosternous plates. On the right side of the scor- pion are large oval impressions, interpreted as gill chambers, that have been covered by the abdominal plates. There is no ornamentation of any kind on the three plates. The last preabdominal segment is broadly triangular and with a triangular area in the posterior center, composed of two converging crests surmounted by coarse tubercles. The cauda is almost complete, but details of the ventral crests are not very discernible in the last three tergites. The first caudal segment shows a half-moon of tubercles on the anterior side with three crests run- ning the length of the segment. These are surmounted by tubercles. The second caudal segment is not so well preserved, but seems to have three crests also. The third caudal segment is not well preserved, but at least two crests are visible. The fourth and fifth segments are too poorly preserved for description. The telson is well preserved, showing a long aculeus and a bulbous vesicle without the presence of a subaculear tooth. The aculeus is stout, long and curved. The entire telson measures 5.0 mm in greatest length and 2.5 mm in greatest width. The appendages are not so well preserved, but some structures are discernible. The pedipalps are strongly developed, having a massive chela. The femur is con- siderably longer than the tibia and the palp shows only the hand and most of the fixed finger. The fixed finger has been broken at the end, but it is long, nearly straight, and at the end it curls into a hook. There are no den- ticles along the edge. Little can be said about the walk- ing legs except that there seem to be double basitarsal and double tarsal spurs on each. The two claws are long and curved. Of particular interest, however, is the very wide tibia developed on each leg. The basitarsus on one leg has many setae on the ventral surface, and, although part of the setae seem to be in linear series, it is not possible to be sure. Type information. —Pennsylvanian, Pottsville Se- ries, in strip pit in Clay County, near the Owen County line, east of Clay City, Indiana. Holotype, FMNH (UC) 37984. Remarks. — Very few specimens are preserved as well as this one. The relationship of the underside of the abdomen is especially well shown in the remarkable 216 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 preservation. Much has been added to our knowledge of the pectinal plate and the prepectinal plate, which commonly are not well preserved (see Text-fig. 31 A). Moore’s description (1923, p. 131) is unusually good, although details of the above-mentioned plates were not noted. The oral chamber in Opsieobuthus is the same as that in modern scorpions, namely, the first two pairs of maxillary lobes form the floor of the oral chamber and the coxae of the pedipalps, the lateral sides of said chamber. This scorpion, as advanced as it appears, is nevertheless a lobostern. Thus it is seen that funda- mental structures, which presumably are attributed to terrestrial existence, were developed before the respi- ratory organs changed from gills to lungs. This idea was first expressed by Laurie (1899, p. 576) in con- nection with the description of the Silurian species Dolichophonus loudonensis, and seems to be borne out by this beautifully-preserved specimen (see also Storm- er, 1954). It is, of course, possible and probable that this scorpion lived much like gill-bearing crabs that move from water to air along the roots of trees, in the swamp environments in which both lived. The water, which gave the required moisture, was carried in ample gill chambers or pouches of the scorpions. They were amphibious, and able to eat outside of the water be- cause of the development of the oral chamber. Superfamily LOBOARCHAEOCTONOIDEA, new superfamily Lobosternina with first pair of coxae without max- illary lobes and meeting in front of the sternum; other three pairs of coxae abutting the sternum. Legs tubi- form. Type family. —Loboarchaeoctonidae, new family. Remarks. —Loboarchaeoctonoidea is essentially the lobosternous “‘counterpart’”’ of the holosternous Ar- chaeoctonoidea. Family LOBOARCHAEOCTONIDAE, new family Loboarchaeoctonoidea with very large, pentagonal, longer than wide sternum. Median eyes placed well forward on the carapace; compound lateral eyes pres- ent. Type genus. — Loboarchaeoctonus, n. gen. Genus LOBOARCHAEOCTONUS, new genus Loboarchaeoctonidae with lanceolate carapace, very short legs and well-defined median carina on the me- sosoma. Derivatio nominis. —lobos (Gr.) = a rounded projec- tion, lobed + Archaeoctonus = a scorpion genus. Type species. —Loboarchaeoctonus squamosus, 0. gen., n. sp. Geological range. —Lower Carboniferous. Loboarchaeoctonus squamosus, new species Text-figures 96, 111J 1949. Archaeoctonus tuberculatus (Peach). Petrunkevitch (partim), pp. 138-139, figs. 138, 175. 1949. Centromachus euglyptus (Peach). Petrunkevitch (partim), p. 141, fig. 136. 1962. Centromachus euglyptus (Peach). Dubinin, p. 428, fig. 1242. Specimen I.—The holotype, BM(NH) 1.988, con- sists of a small scorpion, preserved mainly in dorsal aspect, exhibiting the carapace and nearly all of the opisthosoma as well as one well-preserved pedipalp, part of the chelicera and several walking legs. Little can be noted of the ventral side, but some impression of the coxosternal region occurs through the carapace, so that this important area can be discerned in detail. The imprint of a lobosternous abdominal plate is also present. The holotype is preserved in black shale as- sociated on the same bedding plane with what appear to be carbonaceous plant remains. Most of the speci- men is composed of the original cuticle although many patches have now peeled off. When restudying this specimen after some ten years, it was found that peeling had continued and the specimen was quite deteriorat- ed. Nevertheless, all details reported here can be seen on the specimen, under alcohol, and in a dry state under a very low-angled light. The carapace is broadly lanceolate. This type of car- apace looks much like that of the eurypterid Hugh- milleria. The lateral margins are nearly parallel, round- ed at the anterolateral angles and the anterior. The base is straight and very likely a ridge occurred at the base. The median ocelli are small and are located anteriorly on the carapace. They appear to be elliptical in shape, probably round in life, and are placed on a broad, lacrimiform ocellar node. The lateral eyes are com- pound, reniform, and located at the anterolateral an- gles. They apparently are slightly intramarginal. The preabdomen is composed of the usual seven tergites, each of which increases in length posteriorly. A prominent median ridge is present on at least the second to seventh preabdominal tergites. The seventh tergite also has two well-developed carinae on either side of the median ridge. Only four caudal tergites were preserved. These show a great lengthening in the last two tergites preserved, and probably indicate that the fifth caudal segment, which is not preserved, would be longer. Four definite carinae occur on the first segment and very likely on the second segment. Only three dorsal carinae occur on the last two segments that are preserved. Large setal openings occur on the tergites and some of them occur at the crest of the carinae. One of the surprising features is the enormous pedi- palp and another is the short length of the walking legs. The entire pedipalp is preserved, with the trochanter FossIL SCORPIONIDA: KJELLESVIG- WAERING Dili developed as a massive segment. The femur is longer than the tibia and both are massively constructed. The hand is quite wide, quadrate, and the free finger, also a massive structure, curves inward. Large squamose raised markings occur on all joints of the pedipalp, and the curved movable finger retains several setal open- ings. Preservation was not sufficient to determine any particular pattern. The walking legs, of which only the last two are preserved nearly whole, show that all were short and tubular. The longest leg, which is the fourth, barely extends to the middle of the sixth tergite. Squa- mose markings are also present along the legs. Two massive spurs occur, at least on the last leg, that would correspond to the basitarsal spurs. No tibial spurs seem to be preserved. The termination of each leg consists of two claws that are not curved but seem to be short and straight. Nothing was seen of the posttarsus, al- though it must be assumed that it was present. The left chelicera is poorly preserved and reveals only that the edges of the chelae had small teeth. The chelae are curving but not hook-like. The coxosternal arrangement is noted as impressions left on the area of the carapace. The squamose structure covering the coxae and sternum was responsible for the erroneous determination made by Petrunkevitch, who considered the carapace to be highly granular, as shown in his drawing of this specimen (Petrunkevitch, 1949, fig. 138). The first pair of coxae meets at mid- section, does not have any development of maxillary lobes, and is covered with squamose ornamentation. C Text-figure 96.— Loboarchaeoctonus squamosus, n. gen., n. sp. From the Lower Carboniferous (lower Viséan), Glencartholm Volcanic Beds, Eskdale, near Langholm, Dumfriesshire, Scotland. See foldout inside front cover for explanation of abbreviations. A. Holotype (BM(NH) I.988: a paratype of Archaeoctonus tuberculatus (Peach) of Petrunkevitch, 1949). Complete specimen. B. Holotype (BM(NB) I.988). Ventral organization of the coxosternal area as impressed through the carapace. C. Paratype (BM(NH) In.12848). A pedipalp. 218 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The last three pairs abut against a large, longer-than- broad, pentagonal sternum. The pentagonal sternum also seems to be ornamented with squamose markings along the frontal part (see Text-fig. 96B). There are two oviform opercular plates. This is best seen under very oblique lighting. Nothing can be noted concerning the rest of the underside except to show that in the region of the fifth and sixth tergites, the gill pouch of the fourth or fifth abdominal plate was partly impressed through the sixth tergite, and also is present on the lateral part at the right hand side of the mesosoma. Although the abdominal plates are not preserved completely, enough is present to state definitely that this scorpion is lo- bosternous. Measurements (in mm) of the holotype (BM(NH) I.988).— length width Carapace: 345 3.3 (at base) Preabdomen: 6.8 4.3 (greatest) Postabdomen: Tergite 8: V5 Tergite 9: 1.7 Tergite 10: 251 Tergite 11: Zell Pedipalp: Femur: 2.2 Tibia: 1.4 Hand: 1.6 15 Free finger: 3.0 Estimated overall: 18.0 Type information. —Glencartholm on the River Esk (Eskdale), 3'2 miles SSW of Langholm, Dumfriesshire, Scotland. From the Lower Carboniferous (lower Vi- sean) Glencartholm Volcanic Beds, Upper Border Group. Derivatio nominis. —squamosus (L.) = scaly. Remarks. —After studying the holotype specimen (BM(NH) 1.988), my results differ almost totally from those of Petrunkevitch (1949, p. 138, fig. 138). The differences are so great that comparison or detailing them would be superfluous. It is best to refer to Pe- trunkevitch’s figure 138 and compare it with Text- figures 96A and B here. Specimen IT.—A\so from the same locality and ho- rizon is a single, well-preserved, nearly complete pedi- palp (BM(NH) In.12848, In.12849; see Text-fig. 96C). Petrunkevitch (1949, p. 141) identified this specimen as Centromachus euglyptus (Peach), because *“The shape of the hand and its fingers conforms fairly well with the figure given by Peach for his type.” I disagree with the determination, as the ornamentation is entirely different; the shape of the fingers and the hand is dif- ferent, and C. eug/yptus has fingers with denticles that have a serrated appearance, as stated by Peach and verified here, whereas the fingers of BM(NH) In.12848 are entirely cultrate. Apart from this, the fixed finger is recurving to fit the curved falcate finger along its entire length. In C. euglyptus, the two fingers are op- posable-falcate and do not join along their length. This pedipalp is referred to Loboarchaeoctonus squamosus because the shape of each joint is identical, the inner edges of the chela are recurving and cultrate, and the ornamentation is the same. The ornamentation is very interesting because it 1s identical to that of the “‘scorpionid type” eurypterids (the Carcinosomidae, Megalograptidae, Mixopteridae and others), in which the opisthosoma had a shape very much like that of the scorpions, and also had a tubiform cauda capable of overhead thrusting like the scorpions. In some of these eurypterids, the telson was developed as a curved, downwardly-bent spine, much like the scorpions, also capable of overhead thrusting. The ornamentation comprises long, rounded scales, raised at the apex and slanting into the rest of the integument. The raised lateral walls are very finely striated. These scales are always black, contrasting with the dark-brown color of the rest of the integument. They are commonly setaceous. The rounded ‘“‘bead- like’ tubercles, which Petrunkevitch (1949, p. 141, fig. 136) reported, are not tubercles, but are secondary for- mations and a matter of preservation, as shown by Stormer (1963, pp. 37-38) in his study of Giganto- scorpio willsi from the same beds and having the same preservation. The chela, including the hand and fingers, measures 10.0 mm in length. It represents a much larger scorpion than the holotype. Superfamily PSEUDOBUTHISCORPIOIDEA, new superfamily Lobosternina with the first pair of coxae abutting the second pair, which in turn meets at the midline; both pairs anterior to the sternum; last two pairs abut against the sternum. Type family. —Pseudobuthiscorpiidae, new family. Remarks. —The Pseudobuthiscorpioidea differ from the Isobuthoidea, both of which are Lobosternina, in having the last pair of coxae abut against the sternum, as in living scorpions, and not against the opercular plates. It seems possible that the Pseudobuthiscor- pioidea represent a stage in development out of the Anthracochaeriloidea, which differs only in having protolobosternous abdominal plates. Family PSEUDOBUTHISCORPIIDAE, new family Pseudobuthiscorpioidea with fully-developed max- illary lobes of the first and second pairs of coxae. The maxillary lobes of the second pair of coxae extend forward to be in line with the anterior of the first pair FossiIL SCORPIONIDA: KJELLESVIG-WAERING 219 of maxillary lobes. Sternum pentagonal, very large and concave at the posterior margin. Type genus. — Pseudobuthiscorpius, n. gen. Genus PSEUDOBUTHISCORPIUS, new genus Pseudobuthiscorpiidae with maxillary lobes devel- oped so that both reach the anterior margin, and with the first pair wide but not flaring. All coxae are very short (see Text-fig. 112J). Derivatio nominis.—pseudo (Gr.) = false + Buthi- scorpius = a genus of fossil scorpions. CMI Text-figure 97.— Pseudobuthiscorpius labiosus, n. gen., n. sp. Ho- lotype, BM(NH) 1.1555, part and counterpart. From the Upper Car- boniferous, Upper Coal Measures (Modiolaris similis and pulchra Zones), near Coseley, Staffordshire, England. Some parts of the coxae are taken from both halves of the type and from rubber casts. All tergites are dorsal, but the protolobosternous abdominal plates are preserved beneath them. See foldout inside front cover for expla- nation of abbreviations. Type species. —Pseudobuthiscorpius labiosus, n. gen., Nn. sp. Geological range. —Upper Carboniferous. Remarks. — There is no other scorpion with the char- acters of this genus, and although the dorsal surface is unknown, it cannot be mistaken for any other. It ap- parently is one of the rarer scorpions at the Coseley locality. This specimen (BM(NH) 1.1555) was erro- neously included in Anthracoscorpio buthiformis Po- cock (Pocock, 1911); Petrunkevitch (1953) thought it represented the underside of Buthiscorpius buthiformis (Pocock) and later, in 1955, having referred the spec- imen to Buthiscorpius, which he included under the family Eoscorpiidae, he used this specimen as the sub- ject of a drawing purportedly showing the coxosternal arrangement of Eoscorpius (Petrunkevitch, 1955, p. 73, fig. 40(1)). In reality, the underside of Eoscorpius was then unknown; it is known now and it is a very different scorpion. At the time, the underside of Buthiscorpius was un- known, as the holotype of Buthiscorpius buthiformis (Pocock) was preserved as a dorsal impression. Now that the Mazon Creek fauna has furnished a remark- ably well-preserved specimen of Buthiscorpius (B. le- mayi) it is clear that the well-known specimen BM(NH) 1.1555 does not belong with Buthiscorpius, as that ge- nus is a Holosternina, whereas Pseudobuthiscorpius is a Lobosternina. Pseudobuthiscorpius labiosus, new species Text-figures 97, 112J 1911. Anthracoscorpio buthiformis Pocock (partim), pl. 1, figs. 2 and 2a. 1913. Eoscorpius buthiformis (Pocock). Petrunkevitch (partim), p. 35: 1949. Eoscorpius buthiformis (Pocock). Petrunkevitch (partim), p. 153. 1953. Buthiscorpius buthiformis (Pocock). Petrunkevitch (partim), p. 32, figs. 34 and 45 (labelled Euscorpius). The holotype (BM(NH) I.1555) comprises a small ironstone concretion with a nearly perfect impression of the ventral side of the coxosternal region and the inside of the dorsal side of the mesosoma. The coxo- sternal region is shown in great detail. The sternum is very large with an overall pentagonal outline, but deeply invaginated, roughly in a triangular depression in the posterior. This deep inverted trian- gular area is very noticeable in the posterior part, and a narrow low ridge occurs transversely along the an- terior part of it. The posterior margin of the sternum is concave and the lateral margins flare slightly pos- teriorly. The coxae of the first two pairs of legs have well- developed maxillary lobes and occur in the same ar- rangement as that in living scorpions, namely, the sec- No NO Oo ond pair meets at midsection and squeezes the large maxillary lobes of the first pair to the side. These max- illary lobes are thereby fully developed. The third and fourth pairs of coxae abut the sternum. The opercular plates are tear-shaped, the bulbous parts of which op- pose each other, and are considerably wider than long. The legs were not preserved except for one joint of the fourth walking leg. The dorsal side of the mesosoma shows that the tergites increase in size posteriorly. There are no carinae on any of the tergites, including the seventh tergite. Small scattered granules on the sides of the tergites increase in size toward the posterior of each tergite. A single row of pustules probably is pres- ent at the posterior of each tergite. The large abdominal plates, the fourth and fifth at least, were impressed on the tergites, and can best be described as protolobosternal, that is, the central pos- terior part is emarginate, but not deeply-notched as in typical lobosternal forms. Type information. —The holotype (BM(NH) 1.1555) consists of two halves of an ironstone concretion col- lected by W. Egginton from the concretions in the shales 10 ft above the Thick Coal, Upper Coal Measures (Mo- diolaris similis and pulchra Zones) near Coseley, in the South Staffordshire Coalfield, England. Derivatio nominis. —labiosus (L.) = large-lipped. Remarks.—No other scorpion combines all the characters that set this apart from others on the su- perfamily, family and generic levels: therefore, com- parisons are unnecessary. However, some scorpions occurring in the same bed may be confused with this species. It differs from Coseleyscorpio lanceolatus Kjel- lesvig- Waering (see above) in that the coxae of the first pair of legs in C. /anceolatus are very narrow in contrast to the “thick-lipped”’ first pair of coxae of Pseudo- buthiscorpius labiosus. C. lanceolatus, furthermore, 1s a holostern instead of a protolobostern. With regard to the lobosternous Leioscorpio pseudobuthiformis, the same important differences are apparent in the first pair of maxillary lobes. These are major differences that cannot under any circumstances be explained by either factors of preservation or of variation. Family PETALOSCORPIONIDAE, new family Pseudobuthiscorpioidea with first two pairs of coxae meeting at the midline, with the second pair extending less than halfway the length of the first pair. Sternum pentagonal. Type genus. —Petaloscorpio, n. gen. Remarks. —The family differs from the Waterston- iidae in the shape of the sternum, and in having the second pair of coxae not reaching so far forward nor being so well-developed as in the Waterstoniidae. It also differs from the Pseudobuthiscorpiidae in the cox- PALAEONTOGRAPHICA AMERICANA, NUMBER 55 osternal region, for in the Pseudobuthiscorpiidae the second pair of coxae has squeezed the first pair away from the midline and the maxillary lobes no longer abut one another as they do in the Petaloscorpiidae. Genus PETALOSCORPIO, new genus Petaloscorpionidae with large pectines that lack dif- ferentiation of the basal and middle lamellae and have a rachis with numerous rounded sclerites, all of which are unusually homogeneous in size. Derivatio nominis.—petalos (Gr.) = broad, flat + scorpio. Type species. —Petaloscorpio bureaui, n. gen., n. sp. Geological range. — Pennsylvanian. Remarks. —The unusual perliform comb, with an- terior or basal lamellae undifferentiated, may well be a family character rather than generic. Petaloscorpio bureaui, new species Text-figures 98, 1121, 113C1 One specimen of a well-preserved scorpion showing mainly the dorsal side of the mesosoma and anterior tergite of the metasoma with impression of the coxo- sternal region faintly imprinted; preserved in greenish- gray shale with numerous plants on the same horizon. The specimen is quite well-preserved with small de- tails such as the combs and the ornamentation readily visible, and although there is no doubt concerning its uniqueness on the species level, the type of abdominal plates and the shape of the sternum are not entirely certain. Therefore, on the generic level, there is some doubt. The coxosternal region is only faintly preserved. The sternum is probably pentagonal and the last two pairs of coxae (CII and CIV) abut against it. The maxillary lobes of the second pair seem to extend midway up the length of the maxillary lobes of the first pair, and the first pair of maxillary lobes are broadly spatulate, as indicated by the preserved basal part, and meet at the midline. Rather coarse, scalelike ornamentation covers the sternum and the second pair of coxae (CII). The genital opercular plates are very poorly preserved and appar- ently are longer than broad. The left comb is preserved in its entirety and all details are revealed. There is no anterior lamella and the overall width of the comb is very great, and flaring anteriorly. The teeth decrease in size to less than one- Text-figure 98.—Petaloscorpio bureaui, n. gen., n. sp. Holotype (UL 1092). From the Upper Carboniferous (Pennsylvanian), Escu- minac Formation at Bernard’s Lot, Miguasha, Bonaventure Co., Quebec, Canada. See foldout inside front cover for explanation of abbreviations. A. Complete specimen. B. Reconstruction of the coxosternal area. civ FossiL SCORPIONIDA: KJELLESVIG- WAERING tN tO 222) PALAEONTOGRAPHICA AMERICANA, NUMBER 55 half their length from the center to the distal end. There are approximately 50 teeth (48 counted). Fulcra are present but greatly reduced. The rachis is composed of nearly rounded, small areoles, notable for their over- all small size as compared with those in all other scor- pions. The tergites increase in size posteriorly, and the greatest width probably was reached at the fifth or sixth tergites. All are ornamented in the posterior part by punctae and by a row of semilunar scales at the base. The seventh tergite retains two central carinae, each of which is surmounted by elongated or linear pustules. Two similar carinae occur on each side of the central ones. The lateral edge of the seventh tergite is appar- ently serrated. Numerous pustules occur between the carinae. Only the first caudal segment was preserved, and this shows the same pustules as the preceding tergite, and carinae, which are too poorly preserved for deter- mination. Only one small fragment of an abdominal plate, which seems to be the third, was preserved, and is of the lobosternous type. Measurements (in mm) of the holotype (UL 1092).— length width Tergites: 1 2.6 (partly covered) 2 3.6 (partly covered) 3 4.5 (partly covered) 4 5.6 (partly covered) 5 6.3 (partly covered) 6 6.3 18.0 {/ 9.4 16.3 (anterior) 9.2 (posterior) 8 9.0 6.4 Pectines: 15.0 5.5 at midsection Overall: 140.0 (estimated) Type information. —Collected from Zone 5 of Penn- sylvanian Escuminac Formation at Bernard’s Lot, Miguasha, Bonaventure County, Quebec, Canada (lo- cality also known in literature as being in Scaumenac or Escuminac Bay) by René Bureau in 1964, and reg- istered as No. 1092 in the collection of the University of Laval, Quebec City, Canada. Derivatio nominis. —The species is named in honor of the collector, René Bureau. Remarks. —Comparison with other scorpions is su- perfluous as the comb is sufficiently different to dis- tinguish it and to warrant description as a new species. Family WATERSTONIIDAE, new family Pseudobuthiscorpioidea with lacrimiform sternum and well-developed maxillary lobes of the first two pairs of coxae, but with the second pair reaching only to the lower two-thirds of the maxillary lobes of the first pair; third and fourth pairs abutting the sternum. No lateral eyes present. Type genus. — Waterstonia, n. gen. Remarks. —The family combines rather primitive, deeply-bilobate abdominal plates with highly-ad- vanced coxosternal characters, which, interestingly enough, are almost identical with those of the Lower Devonian species Branchioscorpio richardsoni. 1 sus- pect that this scorpion is amphibious and could carry water in the gill chambers in order to live on land. The well-formed oral chamber is considered an adaptation for living on land, although not necessarily perma- nently. Genus WATERSTONIA, new genus Waterstoniidae with elongated, anteriorly-rounded carapace with two large, anteriorly-located elliptical median eyes, no lateral eyes. Tarsus of first pair of legs greatly elongated. Derivatio nominis. —Named in honor of Charles D. Waterston whose eurypterid work, particularly on the abdominal plates and gills, is exceptional and has great bearing on our knowledge of scorpions, and who kindly furnished the holotype specimen for description. Type species. —Waterstonia airdriensis, n. gen., Nn. sp. Geological range. —Upper Carboniferous. Waterstonia airdriensis, new species Text-figures 99, 112G The holotype comprises part and counterpart of a medium-sized specimen that is fragmentary, but in which all the parts important for taxonomic purposes are well preserved. It occurs in micaceous gray shale, and it was possible to expose part of the structures that were embedded therein. Thus it has been possible to reveal many of the important parts that make this specimen unique. The original cuticle is perfectly preserved, some of which would have peeled off the specimen had it not been coated with a preservative. Enough of the carapace is preserved to show that it was very long, rounded on the anterolateral corners and along the anterior margin. The base is nearly straight, with an elevated ridge along the basal part. It is covered with coarse pustules, mainly over the central part and posterior parts. Two large elliptical eyes (prob- ably round in life) are located on the anterior part of the carapace. These are the median eyes and there is no trace of compound lateral eyes. Nearly all parts of the taxonomically-important coxosternal region are known and the arrangement is of particular interest. The sternum is badly crushed, but enough of the outline is left, I believe, to be able FossiL SCORPIONIDA: KJELLESVIG- WAERING 223 to reconstruct it. It is lacrimiform, rounded on the posterior side and pointed at the anterior end. It is small and would extend into the junction of the second coxae. The third and fourth coxal pairs abut against the sternum. The first pair of coxae has very large and wide max- illary lobes. They are broadly club-shaped or spatulate and nearly flat along the anterior edge. They abut each other along the anterior third of the maxillary lobe, thus the lower, outer part of the oral chamber was composed only of the maxillary lobes of the first pair of coxae. They are rounded on the outer, lateral part and constricted where they adjoin the rest of the coxae, resulting in a pair of spatulate organs. The second pair of coxae also has well-developed maxillary lobes, but these are pointed and long and occur within the pos- terior two-thirds of the first maxillary lobes. The third and fourth pairs of coxae are rounded at the inner part 4 2 be ; ao 43 Ney ‘ ~ 711127) \ 712(Ac?) Text-figure 99.— Waterstonia airdriensis, n. gen., n. sp. Holotype (RSM 1957.1.4996). From the Lower Carboniferous, Lower Coal Measures (Westphalian A), Whiterigg Colliery, Airdrie, Lanarkshire, Scotland. See foldout inside front cover for explanation of abbreviations. A. Showing the disposition of prosomal ventral features and proximal preabdomen. B. Counterpart. C. End of the basitarsus of the first leg (I), showing the two basitarsal spurs and minor spines adjoining the end of the tarsus. D. Probably the first leg, showing very large distal spurs of the sixth joint and analogous structure on the seventh joint. E. Third leg, distally folded back on itself. NO No os and abut the sternum. The only ornamentation noted was a single row of granules along the posterior edge of the fourth coxae. The opercular plates are also lac- rimiform with a flat area where they adjoin each other along the midsection (see Text-figs. 99A, B). The pectines are of particular interest and show that the anterior or first lamella consists of at least four segments. These are covered with irregularly-shaped, rounded sclerites. The larger sclerites occur toward the inner part of the pectines. The middle lamella is quite wide and broad and is covered with small rounded sclerites. One or two larger sclerites were noted, but most of them are small. The fulcra are well developed and are made up of rounded small sclerites. The den- ticles of the pectines are long and narrow. Thirty of these teeth have been counted, but, undoubtedly, there is room for 45 to 50 teeth on each pectine (see Text- figs. 99A, B). The chelicerae are only partially preserved, but they appear to be rather wide, although probably crushed. However, the hand of the chelicera, forming most of the third segment, seems to be unusually wide. The two chela are armed with teeth, but it was not possible to determine their arrangement. Little can be noted concerning the pedipalps, as only fragments were preserved. However, they appear to have been stout, and it was noted that small denticles occurred along the edge of the free fingers. The ends of the immovable and free fingers seem to be rounded and without falcate terminations. The first walking leg is preserved almost in its en- tirety. This is mainly slender with a particularly long tarsus (see Text-figs. 99A, D). Basitarsal spurs occur (see Text-fig. 99C), one of which is a flat short spur, whereas the other is a longer, pointed spur. Another spur occurs in the same intersegmental membrane. The tarsus is particularly long and tubular, and the claws are straight and quite long (see Text-fig. 99A). These are covered with setal openings, revealing that they were rather hirsute. The posttarsus is pointed, with a round, ball-like internal area that shows definite deep and linear muscle scars (see Text-fig. 99D). Setal open- ings are also present on the tarsus. The tarsus, as well as the basitarsus, is also covered in part with setal openings. The second leg is largely unknown except for the coxae. The third leg (see Text-figs. 99B, E) is perhaps the most interesting of all, as it shows that the tibia was armed on the inner side with a very serrate edge. The tarsus is greatly reduced from what it is on the first leg. The ungues are rather straight, falcate and with spines on the venter. The fourth walking leg reveals a very long coxa, followed by a short trochanter. Only parts of the femur, tibia and patella are present. PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The dorsal side of the mesosoma is practically un- known, but fragments of the tergites do occur, and show that they were bounded on the anterior edge by a prominent transverse ridge with a row of setal open- ings. The underside, however, is preserved and shows that the scorpion was a typical lobostern with very wide, deeply-bilobate abdominal plates that were bounded at midsection by a suture. The indentation or cleft at the posterior middle of the two lobes is very deep. The cauda is represented only by fragments of three segments and, other than to show that this was a rather thick tail, nothing more can be described. Measurements (in mm) of the holotype (RSM 1957.1.4996).— 12.8 (estimated) 10.0 (estimated) Length of prosoma: Width at base: Median eye length: 2.0 width: 1.6 Estimated total length: 125.0 Type information.—The holotype (RSM 1957.1. 4996) is in the collections of the Royal Scottish Mu- seum, Edinburgh, Scotland, and is from the “Fern Bed”’, from which many fine plant specimens were described by R. Kidston, and which occurs in Whiterigg Colliery, Airdrie, Lanarkshire, Scotland. The fern bed lies 20 ft above the Upper Drumgray Coal. This horizon is in the communis Zone of the Ammanian (Lower Coal Measures or Westphalian A) (Waterston, pers. com- mun., August 19, 1968). Derivatio nominis. —Species name is taken from the town of Airdrie, in Lanarkshire, Scotland. Remarks. —The original coloration was preserved in this specimen and it showed that the entire test was dark-brown in color and that it apparently must have been a rather hirsute animal. Waterstonia (?) brachistodactyla, new species Text-figure 100 The holotype consists of a single specimen, pre- served in black shale in unusually good condition. Al- though only part of the pedipalp is preserved, enough is present so that its recognition will be easily assured if found again. The skin is totally without ornamen- tation, presenting a bald appearance, which is unique. Only part of the hand and all of the fixed finger of the pedipalp are preserved. The parts preserved show that this scorpion likely reached a length of approximately 20 cm. The fixed finger is very slender, unusually short, cul- trate, tapering and with a very slight swelling at the base along the cultrate edge; it tapers to a point rather abruptly. It is straight along the outer edge. Type information. —The holotype label reads “From FossiIL SCORPIONIDA: KJELLESVIG- WAERING a shale a few fathoms above main clay ironstone, Crookhill-Beith’’. According to personal communi- cation with C. D. Waterston, it comes from ‘1 to 3 fathoms” above the “Main Clay Ironstone’’, or the Dalry Clayband Ironstone, near the Johnstone Shell- Bed horizon—that is, the Limestone Coal Group of the Scottish Carboniferous succession, Namurian E,, from Crookhill near Beith, North Ayrshire, Scotland. The specimen is registered as RSM 1978.5.1 in the Royal Scottish Museum [Craig Collection], Edin- burgh, Scotland. Derivatio nominis.—brachistos (Gr.) = shortest + dactylos (Gr.) = finger. Remarks. —The new species cannot be compared with Waterstonia airdriensis from the same general locality because the pedipalp is not known in the latter. However, because W. (?) brachistodactyla occurs in a different horizon, and lacks the ornamentation present in W. airdriensis, it is considered different. Infraorder BILOBOSTERNINA, new infraorder Branchioscorpionina with five pairs of abdominal plates consisting of two completely-separated, rounded lobes, without doublures, below the body wall, which may have become slightly sclerotized and developed into sternites. Remarks. —The Bilobosternina seem to reveal the direction of evolution of the early scorpions. The ab- dominal plates are joined to the body wall only at the inner anterior part, and serve only to cover the gills. These scorpions are intermediate between the lobo- sterns and the living forms, and the plates are in the process of being reduced completely. Superfamily BRANCHIOSCORPIONOIDEA, new superfamily Bilobosternina with first two pairs of coxae having well-developed maxillary lobes meeting at the midline in front of the sternum; last two pairs of coxae abutting the sternum. Text-figure 100.— Waterstonia (?) brachistodactyla, n. sp. Holo- type (RSM 1978.5.1) [in the Craig Collection]. Incomplete hand and finger of the pedipalp. From the Upper Carboniferous, Coal Mea- sures, Crookhill-Beith, North Ayrshire, Scotland. NO i) Nn Type family. —Branchioscorpionidae, n. fam. Remarks. —The coxosternal area is remarkably ad- vanced for a superfamily known to date only from the Silurian and Devonian. Family BRANCHIOSCORPIONIDAE, new family Branchioscorpionoidea of small size with a pointed oval-shaped sternum; orthostern sternites with the sug- gestion of a median line, above the paired abdominal plate lobes. Type genus. —Branchioscorpio, n. gen. Remarks. —This is an easily recognizable family, dif- fering from all others of its age in the shape of the sternum and arrangement of the coxae, but, oddly enough, almost identical with the Upper Carboniferous genus Waterstonia in its coxosternal arrangement. The Silurian family Proscorpiidae Scudder, 1886, as emended by Kjellesvig-Waering (1966), bears some affinities to this new family. It has a sternum that is somewhat similar to the Branchioscorpionidae, though proportionately much larger; however, the arrange- ment of the coxae is different. In the Proscorpiidae the first pair of coxae lacks maxillary lobes and meets in front of the sternum, but all four pairs of coxae abut the sternum. Apart from this, the abdominal plates of the Branchioscorpionidae are of the bilobostern type, whereas those of the Proscorpiidae are holostern in shape. Genus BRANCHIOSCORPIO, new genus Branchioscorpionidae with operculum composed of two large subquadrate sclerites immediately posterior to the sternum; pectines large, unjointed, composed of one large plate, and basally bordered by a narrow la- mella to which are attached eight rounded teeth; preab- domen covered with large, thickened scales, and each sternite bordered posteriorly by a well-developed transverse ridge; legs much like those of present-day scorpions; large rounded gills (at least in ventral aspect) open to the outside by a large slit located at the anterior outer margin. Derivatio nominis.—branchia (Gr.) = the gills of fishes + scorpio. Type species. — Branchioscorpio richardsoni, n. gen., Nn. sp. Geological range. —Lower Devonian of Wyoming. 226 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Remarks. —Little is to be gained by a comparison of this genus with any other, as it is unique.** Branchioscorpio richardsoni, new species Text-figures 101, 111F The species is based on a single specimen (FMNH (PE) 6176a,b) collected by Eugene S. Richardson, Jr. and Robert H. Denison from the Field Museum of Natural History collecting site in Cottonwood Canyon, Wyoming, in 1962. The holotype is preserved so that the cuticle retains the original colors which, as in some modern scorpions, consist of an overall light-brown, darkening toward the ends of the pedipalp. The preservation can hardly be surpassed, as the original cuticle is present, however, in a compressed condition. It is a scorpion of about the same size as the medium-sized species of the pres- ent day, based on the size of the organs present (the carapace and most of the cauda are missing). Only the ventral side is known. The coxa (first joint) of the pedipalp, seen in ventral aspect, comprises an elongate, roughly triangular joint (P1), about twice as long as wide. It has been displaced, along with the rest of the pedipalp, to the left side of the mouth. It is rounded outward and has a small triangular spine at the inner anterior corner. Length of the coxa is 2.3 mm and 1.1 mm in greatest width. The surface of the joint is covered with crowded, very mi- nute, triangular, thornlike spinules very similar to those that Wills reported for Mesophonus (Wills, 1947, p. 84) on coxae 1 and 2. The spinules are all of the same size and all point toward the inner edge of the joint. They differ from those of Mesophonus in being ar- ranged so that they point away from the oral opening rather than being haphazardly distributed. The spi- nules toward the side of the trochanter, however, are longer and narrower. They are important because they have been considered to be stridulating*’ organs by Wills and others. The color of the first joint is light-brown, and be- 38 The most difficult part of preparing this manuscript for publi- cation has been the attempt to reconcile the revised illustrations of Branchioscorpio richardsoni with the only description of the genus and species available, dating more than 12 years earlier, before Kjel- lesvig-Waering had completed his world-wide study of the available types and had formulated his theories of fossil scorpion classification. Among his papers was a jubilant note that he had just completed the revision of the figures of B. richardsoni in January, 1979, in line with his final ideas, but that he had had to stop short of making a reconstruction of Branchioscorpio and writing up the new parts. Shortly thereafter he went to the hospital, and it was too late. Scat- tered throughout the completed parts of the manuscript were com- ments and comparisons with B. richardsoni to suggest what he thought the species was really like. A.S.C. °° These “organs” were eliminated in the final version of Text-fig. 101A. A.S.C. cause of the minute spinules, absorbs light, giving it a duller appearance than the other joints. One feature of considerable interest about the coxa of the pedipalp is the presence of a row of small tri- angular teeth, forming a gnathobase, as in eurypterids, along the posterior part of the coxa, and in life, on each side of the passage to the mouth.*° The complete trochanter (P2) of the pedipalp (see Text-fig. 101A) is preserved, turned so that the dorsal side is placed anteriorly. As such it is a triangular struc- ture, with the dorsal side rounded, but without any ridges such as occur in the trochanters of many Recent scorpions, which almost appear as if two joints had been fused. In B. richardsoni the entire structure is smooth, except for two or three very small granules along the dorsal (anterior) part. In color the trochanter is light-brown and shiny. Along the dorsal side it mea- sures 2.3 mm in length. The femur (P3) of the pedipalp, preserved flattened, with ventral side showing as in life (see Text-fig. 101A), is a long, possibly cylindrical, structure, and quite mas- sive in relation to the rest of the pedipalp. This joint is diagonal, and is ornamented with very small granules in no apparent pattern. Some of these could be setal sites or trichobothria, but this was not definitely es- tablished under high magnification. The color is light- brown, but becoming slightly darker toward the distal end. The femur measures 4.6 mm in greatest length and 1.7 mm in width at midsection. The tibia (P4) of the pedipalp (see Text-fig. 101A) is much shorter than the femur and, although not en- tirely preserved, enough is present for a suitable de- scription. It is preserved from the ventral side, and is thicker than the femur and only about half again as long as wide. In color it is shiny light-brown; the or- namentation comprises semilunar mucrones, coarser than those on the femur, but scattered about in no discernible pattern, except that they are mainly con- centrated in the posterior part of the podomere. The 4° This is another detail omitted in the final figure. A.S.C. Text-figure 101.—Branchioscorpio richardsoni, n. gen., n. sp. Ho- lotype, FMNH (PE) 6175a,b). From the Lower Devonian, Beartooth Butte Formation, of Cottonwood Canyon, Big Horn Mts., WY. See foldout inside front cover for explanation of abbreviations. A. The ventral surface exposed in the holotype (dorsum unknown). This shows both sternites*! (body wall features) (St2-4) and abdom- inal plates (AP1-5), which cover the gills. B. Reverse of holotype slab (fig. 101A): only the few parts identifiable on the surface are drawn. C. The prepectinal plate (ppp) and its median organ (MO). D. A portion of the free ramus of the pedipalp. Note fine denticu- lation (serration) on the cutting edge. E. A composite of the part and counterpart of the pedipalp, showing the serrations on both cutting edges of the dactyls. F. Showing the slitlike anterior gill opening (in black) of the second abdominal plate (AP2), left side. G. The slit- like anterior opening of the third gill, right side (AP3). H. Right leg IV, showing condyle and socket. FossiIL SCORPIONIDA: KJELLESVIG-WAERING 227 *! Tt might be argued that the structures St2—4 are internal surfaces of tergites, but Kjellesvig-Waering believed otherwise. His interpretation here is crucial to his concept that abdominal plates and sternites are non-homologous structures, which at full development, to the exclusion of one or the other, performed analogous function. In response to a question from Dr. Rolfe concerning the sternite-tergite problem, Kjellesvig- Waering (written commun., date unknown) replied: ““You may wonder how I could detect [sic] the sternites (St) from the tergites. Easy—two ways: Ist the ornamentation is pointing (the U) ventrally and 2nd the transverse ridges are bowed ventrally (trr)”. A.S.C. i) NO oo tibia measures 2.6 mm in greatest length and 1.9 mm in greatest width, which is at midsection. Almost the entire chela is preserved except for the distal end of the fingers (see Text-figs. 101A, E). It also is preserved from the ventral side and is unusually long and narrow. The hand, in particular, is very long, mea- suring 1.7 mm in width and 4.3 mm in length through the midsection, or to the inner junction with the free finger. It is rounded at the posterior end, with the two sides parallel. The entire chela is shiny dark-brown increasing to darker shades at the distal ends. It is covered only in part by small scalelike mucrones, or granules, some of which are rounded and definitely reveal a small opening that, no doubt, represented sites for setae. Smaller round openings can also be noted, although not in any defined pattern. Some of these may represent trichobothria. The fixed ramus is nearly complete (see Text-figs. 101A, B, E), although parts of the extreme distal end are lacking. It occurs on both part and counterpart, in particular on FMNH (PE) 6175b. The fingers are long, tapering structures with an inner serrated edge that consists of a single row of triangular denticles. Both rami contain the same denticles, and are approxi- mately of the same size. The fixed ramus measures 3.9 mm along the inner edge to the junction with the free ramus. This is an incomplete measurement as the distal end is gone. The fingers are very dark-brown in color and no ornamentation was noted. Of the four walking legs, only the second and fourth are preserved almost entirely, although the basal podo- meres of all are present and reveal the important cox- osternal area, which seems to be of great diagnostic value in our classification. The coxae of all the legs are preserved complete, and show that the first two pairs join together at midsection much the same as in pres- ent-day scorpions, whereas the last two pairs abut against the sternum described above. This arrange- ment of the coxae (Text-fig. 111F), surprisingly is much like that present in Buthiscorpius, Eoscorpius and Ma- zoniscorpio among the Carboniferous scorpions, and still more surprising, as in modern scorpions. The shape of the sternum, however, is totally different from any- thing known, except in Waterstonia (see Text-fig. 112G). The second walking leg on the left side (underside) is preserved almost entirely, but parts of it are missing (see Text-fig. 101A). The trochanter (II1) is a large triangular, flaring podomere and, although not entirely preserved, enough is present to reveal its unusual size. Junction with the coxa is very clearly seen. In color it is dark-brown, and quite shiny. No ornamentation is present. The following joints (II2 to II6) are rather incomplete, although II3 and IIS5 are nearly complete. PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The latter are cylindrical and retain a few scattered scales. Joint II5, the basitarsus, is noteworthy for hav- ing a swelling at the posterior junction with the pre- vious joint. Some scales occur on the swelling, other- wise the joint does not retain any ornamentation. A single claw, undoubtedly the anterior (the posterior claw was lost in excavation), occurs in normal position. It is long and slightly curved, but not as in modern scorpions. It is important to note the complete lack of any spurs in the integumental tissue between the joints. This is also true of the fourth leg. The last leg (IV) on the right side is well preserved except for the trochanter, which is fragmented, and the distal end (see Text-figs. 101A and H). Measurements (in mm) of the last leg of the holotype (FMNH(PE) 6175a,b).— width at length midsection IV2: 3.6 2 IV3: 3.0 1.0 Iv4: 2.8 0.8 IVS: 1.4 0.6 IV6: 2.0 0.4 Each joint appears to be cylindrical but slightly ta- pering. All are covered with scales, pointing distally, much coarser than any noted elsewhere, and many of these scales have a small perforation, very likely rep- resenting the site of a seta. In the distal joints (IV4 to IV6) the posterior part is ornamented with a single row of serrations. Joint [V4 reveals two rows, probably one on each side of the leg. No spurs are present on any of the joints. Podomere I'V2 reveals a condyle at the distal end, and opposing this is the corresponding socket (see Text-fig. 101H). The color of the leg is light-brown, with the scales being of a darker color. The sternum is elongated, subovoid, pointed ante- riorly and not so pointed on the posterior end. A nar- row doublure occurs dorsally, and this is separated at the anterior into two separate bands (see Text-fig. 101A). The sutures of the doublure of the metasoma in the eurypterids have been described (Kjellesvig- Waering, 1961, p. 817), and it is apparent that these sutures are the same or quite similar in the sternum of the scorpions. No ornamentation occurs on the ster- num and the color is uniform light-brown, with a dark- er shade on the doublure. The sternum is 1.8 mm in length and 1.2 mm in width. The genital operculum comprises two large subquad- rate plates, one of which is preserved completely, whereas the other is mainly covered by the coxa of the last leg. The cuticle of the opercula is more dark-brown in color than the other parts of this scorpion, except the pedipalps. The opercula measure 1.1 mm in di- FossiL SCORPIONIDA: KJELLESVIG- WAERING ameter, and are devoid of any ornamentation. The pectines are preserved in their entirety and are quite different from those known from any scorpion in the Upper Paleozoic. (The only other scorpion from the earlier Paleozoic that shows the pectines, very poorly-preserved, is Waeringoscorpio hefteri Stormer, 1970.) Each is composed of a large unjointed plate, nearly triangular in shape, with the sides rounded. Along the basal edge is a very narrow, also unjointed, lamella to which are attached eight elliptical and very wide teeth. It appears certain that each comb is composed of eight teeth. These teeth show minute perforations along the outer edge that apparently are the peg organs. These can be seen under high magnification on the second tooth of the right comb. It is interesting to note that the neonates of the living species Megacormus granosus Gervais of Mexico (Hoffmann, 1931, fig. 14) reveal a basal unjointed rachis that recalls the comb of Branchioscorpio richardsoni. Jointing of this part occurs in later stadia. The pectines measure 2.2 mm in length and 2.1 mm in width. The combs are attached to a narrow base that is the first “‘sternite” or anchor to the combs (ppp in Text- fig. 101A). Point of attachment seems to be at a notch- like lobe at both ends. In the center, however, (see MO in Text-figs. 101A, C) is a short clublike organ. It was not possible to note any joints, but the end appears to be divided. This organ resembles in many respects the short female (Type B) organ of the eurypterids, in par- ticular that of Megalograptus ohioensis Caster and Kjellesvig-Waering (1964, fig. 19). The underside of the mesosoma is the most inter- esting part of the scorpion as the separate, rounded abdominal plates seem to underlie large, rounded or ovoid gills. The abdominal plates are of the split or bilobostern type. The mesosoma appears to have been thrust foward, perhaps due to ecdysis, so that abdom- inal plates and the corresponding gills overlap one another. The abdominal plates can be seen in the pos- terior part of the mesosoma. The first pair apparently had been mainly lost before deposition, as only the right posterior abdominal plate is present. Following this, and covering the undisturbed gills (AP2), are three long, squarish sternites. On the counterpart (FMNH (PE) 6175b), the relationship of the gill to the outlines of the plates is well shown. This reveals that the ab- dominal plates are separated by a considerable distance from each other (see Text-fig. 101A). The gills are almost surely much more complicated than they appear in the specimen. They are large, ovoid, cuticular membranes and are not attached to the ab- dominal plate, but must be attached to the venter of the scorpion. It has been noted (see Text-figs. 101A, F, G) that at the outer anterior end of the gills is a large No N Ke) opening, bounded on the outer lip by a single series of serrations, which cannot be referred to any known scor- pionid structure. By inference, it appears that this must also be part of the gills, perhaps an opening, or outlet of the gills; it occurs in the same place in relation to two different gills and retains the same structure. There is little doubt that other parts of the gills have been flattened out and that all that can be seen now is the ovoid part and the opening (outlet?). This is a problem that must be settled with additional material. The ovoid part of the gills, which is the ventral part, reveals a dendritic pattern, probably cuticular rein- forcement, such as is found in the lungs of some scor- pions, rather than vascular markings, and is similar in many respects to the arborescent ridges on the gills of the eurypterids (see Moore, 1941, pp. 64-66). The sternites are bordered by a well-developed an- terior transverse ridge and the ornamentation com- prises wide, semilunar, thick scales scattered in no ap- parent pattern. The metasoma is known only from the anterior two tergites. The first, which forms the last tergite of the preabdomen, is roughly pentagonal, and is seen from the inside of the dorsal side. It is covered with the same type of thick scales as the rest of the dorsum, and is surrounded by a thick doublure, but no crests are de- veloped. The first tergite of the cauda is nearly twice as long as wide, but nevertheless is wide enough to suggest a well-developed or strong cauda. It is surmounted on the venter by two crests, each of which is covered with small scales as also are the sides, suggesting lateral crests. This tergite measures 4.1 mm in length and 2.2 mm in width at midsection. Type information.—The holotype (FMNH (PE) 6175a,b) comes from the Beartooth Butte Formation, light-gray, argillaceous, dolomite (dololutite) of Lower Devonian age, associated with eurypterids, plants, and fishes, from Cottonwood Canyon, Big Horn Moun- tains, WY, U.S.A. Derivatio nominis.—The specimen was named in honor of Dr. Eugene S. Richardson, Jr., invertebrate paleontologist at the Field Museum, who first called my attention to the Cottonwood Canyon scorpions. Family DOLICHOPHONIDAE Petrunkevitch, 1953 (emend.) Branchioscorpionoidea with carapace having com- pound lateral eyes and small median eyes placed on a small intramarginal eye node. Type genus. —Dolichophonus Petrunkevitch, 1949. Remarks. — Discovery of the bilobosternous abdom- inal plate means that the family can no longer retain 230 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 the genus Proscorpius, as advanced by Petrunkevitch (1955, p. 70), as the latter is a holostern. Genus DOLICHOPHONUS Petrunkevitch, 1949 (emend.) Carapace subrectangular, tapering anteriorly, very long and strongly emarginate on the anterior margin; compound lateral eyes at the anterolateral corners, pro- truding. Type species. —Palaeophonus loudonensis Laurie, 1899. Geological range. —Middle Silurian. Remarks. —Generically there is no scorpion that can be mistaken for this one, which is the oldest known. It occurs, as do all Silurian and Lower Devonian scor- pions, associated with eurypterids. Dolichophonus loudonensis (Laurie, 1899) Text-figure 102 1899. Palaeophonus loudonensis Laurie, p. 576, pl. 1, fig. 1. 1904. Palaeophonus loudonensis Laurie. Frié, pp. 64-65, text-fig. 80. 1913. Palaeophonus loudonensis Laurie. Petrunkevitch, p. 32. 1949. Dolichophonus loudonensis (Laurie). Petrunkevitch, p. 132, fig. 171. 1953. Dolichophonus loudonensis (Laurie). Petrunkevitch, pp. 11- 12, figs. 7-10, 117. 1955. Dolichophonus loudonensis (Laurie). Petrunkevitch, p. 70, fig. 38(4). 1962. Dolichophonus loudonensis (Laurie). Dubinin, p. 425, fig. 1224. Type information. —Holotype, Middle Silurian, Wenlock, at Gutterford Burn, Pentland Hills, Scotland, registered as RSM 1897.32.196 [Hardie Collection]. Preserved in yellow-brown, shaly sandstone, the fossil retains considerable convexity or relief. The holotype consists of only one piece, preserved as a mold (counterpart) of most of the upper surface of a scorpion that barely reached an estimated 80 mm in total length from the anterior of the carapace to the end of the stinger (not present). Most of the twelfth tergite is not preserved, but this was mistaken for the telson (stinger) by Petrunkevitch (1953, fig. 7). The specimen also retains small fragments of abdominal plates that are of great importance for taxonomic pur- poses. The holotype was studied in Edinburgh, both in a dry state and (particularly) under alcohol, which revealed several important morphological features hitherto unknown. An excellent rubber cast made by Charles D. Waterston of the Royal Scottish Museum revealed further details. The drawing made by the writ- er on Text-figure 102 reveals all the details that had previously been overlooked. The carapace is very long, rectangular, slightly tapering, but definitely strongly- bowed inward. Two small median eyes, on a low small teardrop-shaped eye node occur on the anterior central part. The eye node can best be seen on the rubber cast. Large, bulbous, compound eyes undoubtedly occur. Under alcohol, the left eye (in life) was easily discern- ible, although facets were not preserved in the coarse matrix. As noted in other primitive scorpions, small median eyes “always” are accompanied by large, com- pound lateral eyes, and vice versa. It came as a distinct surprise that the small median eyes had not been pre- viously noted as, even in the dry state, they are well shown. This is the oldest scorpion known, and there- fore of considerable taxonomic importance. The appendages are mainly unknown except for the counterpart sedan} Text-figure 102.—Dolichophonus loudonensis (Laurie). Drawn from the holotype RSM 1897.12.196 [Hardie collection], under alcohol and a rubber cast thereof, made by Charles Waterston. From the Middle Silurian, Wenlock, at Gutterford Burn, Pentland Hills, Scot- land. See foldout inside front cover for explanation of abbreviations. FossIL SCORPIONIDA: KJELLESVIG- WAERING Dill right pedipalp (note that the figure is of the counter- part), which is not notable for any peculiarity and will not be described further as the drawing reveals all that can be said about it. No denticles were noted and the edges apparently are cultrate. A small fragment of the first leg of the left side is preserved. The chelicerae are well preserved and are falcate; small denticles were noted on the fixed ramus of the right chelicera, but they probably represent only a small part of the teeth on the edges of these chelate append- ages. The mesosoma likewise is normal, with strong, anterior transverse ridges not noted by Petrunkevitch (1953, fig. 3). Each tergite increases in length. The over- all slenderness of the mesosoma may indicate a male, but not necessarily so, as unimpregnated or younger females are also slender in mesosomal aspect. The last tergite of the preabdomen (first metasomal) definitely shows two rounded ridges on the side, and two ridges rather close together on the midsection. The cauda, unlike that shown by Petrunkevitch (1953, fig. 7), is incomplete and, instead of being thin, is actually rather thick and wide. Only four caudal tergites and a small piece of the fifth are present. No telson, despite Petrunkevitch, is present. By far the most important morphological feature found is the presence of a full half of an abdominal plate of the bilobosternous type. This plate resembles those of Branchioscorpio richardsoni, n. gen., n. sp. The plate is sufficiently well-preserved to show that it had to hang free and independent of the corresponding half on the other side. The so-called “hidden first tergite”, upon which the suborder Protoscorpiones Petrunkevitch, 1953 (p. 5) was based is, of course, fictitious (as was stated before by Kjellesvig-Waering, 1966, p. 361, in regard to Pro- scorpius osborni (Whitfield)), and merely represents a swelling at the back of the carapace, known also in many other scorpions, fossil and living. The nature of the rather coarse rock (very sandy) precluded the preservation of such details as setae, ommatidia, etc. Inasmuch as there are so many differences with re- spect to the structure of this scorpion from that given here and what Petrunkevitch (1953, pp. 11-12) pos- tulated, it is necessary to give new measurements. Remarks. —As stated above in the description, 1m- portant taxobases such as the median eyes, the median eye tubercle, the compound, or aggregated, lateral eyes and, in particular, the small fragment of a biloboster- nous abdominal plate were found for the first time, and therefore, because of these characters, it was nec- essary to emend the family and generic descriptions as given above. My study disagrees almost entirely with that of Petrunkevitch (1953, pp. 11-12), as do the cor- responding drawings. The right pedipalp is not present, and the small segment present is likely the second joint of the first leg. Transverse ridges are present, as are well-developed carinae on the seventh to eleventh ter- gites (twelfth tergite only partially preserved). Petrunk- evitch (1953, p. 11) states and shows in his figure 7 that the cauda is supposed to be very narrow. Laurie (1899, p. 576, pl. 1, fig. 1) also shows a very narrow cauda. Both of these authors mistook the median ca- rina for the the edge of the caudal segments. These carinae can be clearly seen in Petrunkevitch’s photo- graph (1953, fig. 117), as well as the actual thick cauda extending beyond the two carinae. The photograph also reveals the impossibility of there being more than the four full segments and a fragment of the fifth (T12) instead of the five plus a stinger as shown by Petrun- kevitch (1953, fig. 7). Measurements (in mm) of the holotype (RSM 1897.12.196).— Carapace length: 10.5 (at midsection) Width at base: 10.0 Width at anterior: 7.5 Opisthosoma: length Tergite No. 1: 2.4 Tergite No. 2: S22 Tergite No. 3: 3.4 Tergite No. 4: 3.6 Tergite No. 5: 4.0 Tergite No. 6: 4.0 Tergite No. 7: 5.6 Tergite No. 8: 4.5 Tergite No. 9: 7.0 Tergite No. 10: 7.0 Tergite No. 11: 7.0 Tergite No. 12: (incomplete) Suborder NEOSCORPIONINA Thorell and Lindstrém, 1885 (emend.) (nom. transl. et correct. ex Neoscorpil Thorell and Lindstr6m, 1885) Terrestrial scorpions breathing through book-lungs, sternites perforated by stigmata. Type family (herein designated).—Scorpionidae Leach, 1815. Geological range. —Carboniferous to Recent. Infraorder ORTHOSTERNINA Pocock, 1911 (nom. transl. and restricted ex Orthosterni Pocock, 1911) Neoscorpionina with undivided, stigma-bearing, rectilinear true sternites, without doublures, that char- acterize all living scorpions. Geological range. —Carboniferous to Recent. 232 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Superfamily SCORPIONOIDEA Leach, 1815 (nom. transl. Petrunkevitch, 1955 ex ‘“‘family”’ Scorpionides Leach, 1815) Orthosternina with first coxae meeting at median line or wedged behind second pair; second pair of coxae meeting in front of sternum; third and fourth pairs of coxae abutting sternum. Sternum pentagonal or rang- ing from very wide to triangular-pentagonal. Geological range. —Carboniferous to Recent. Family PALAEOPISTHACANTHIDAE, new family Scorpionoidea with well-developed prepectinal seg- ment (sternite No. 1). Very large sternum, with max- illary lobes of first pair of coxae broadly spatulate. Sternites with small round stigmata. Type genus. —Palaeopisthacanthus Petrunkevitch, 1913. Geological range. —Upper Carboniferous of North America and England. Remarks. —There are numerous structures that re- semble those of some of the living families. The large sternum is a structure that is more common in older fossil forms. However, the development of spatulate maxillary lobes of the first pair of coxae is quite similar to that in the Chaerilidae and some of the Chactidae. I consider the type of stigma to be of family value, and the round stigma of the new family is found in some living genera, such as the chactid Broteochactas and in the Chaerilidae. It is not the purpose of this study to revise the living families, but there is no doubt that, as they stand, there is little meaning to the separation. The presence of the well-developed prepectinal lobes is, of course, a major difference from the other Or- thosternina. One genus is recognized: Palaeopisth- acanthus Petrunkevitch, 1913. Another genus, Comp- soscorpius Petrunkevitch, 1949, is also referred to this family, although the first maxillary lobes of this genus are not spatulate, but narrow; very likely it should go into a different family. Nevertheless, the coxosternal area of Compsoscorpius is mainly unknown and, until specimens revealing the venter are known, it seems preferable to maintain Compsoscorpius within the family Palaeopisthacanthidae. The three simple lateral eyes certainly indicate the terrestrial habitat of this genus. Genus PALAEOPISTHACANTHUS Petrunkevitch, 1913 (new definition) Palaeopisthacanthidae having carapace with strong- ly-cuspidate anterior margin. Type species. —Palaeopisthacanthus schucherti Pe- trunkevitch, 1913. Geological range. — Pennsylvanian of Illinois. Palaeopisthacanthus schucherti Petrunkevitch, 1913 Text-figures 103, 104, 111G 1913. Palaeopisthacanthus schucherti Petrunkevitch, pp. 48-49, pl. 2, figs. 8-9; text-figs. 11-12. 1949. Palaeopisthacanthus schucherti Petrunkevitch. Petrunke- vitch, p. 154. 1953. Palaeopisthacanthus schucherti Petrunkevitch. Petrunke- vitch, p. 33. 1955. Palaeopisthacanthus schucherti Petrunkevitch. Petrunke- vitch, p. 74, fig. 43(6). 1962. Palaeopisthacanthus schucherti Petrunkevitch. Dubinin, p. 431, figs. 1231, 1255a, b. 1966. Palaeopisthacanthus schucherti Petrunkevitch. Vogel and Durden, pp. 655-658, pl. 81, figs. 1-3; text-figs. 1-2. Petrunkevitch, in his original description and in sev- eral other reports (see above), did not notice the im- portant stigmata, which are perfectly preserved. These were reported in an outstanding paper by Vogel and Durden (1966, p. 655). To date, the stigmata reported by Vogel and Durden remain the only ones known in a Paleozoic scorpion. Other reports of stigmata by Fri¢ (1904, pp. 69-71) regarding Microlabis sternbergii Corda, and in Palaeophonus nuncius Thorell and Lind- strém (1885, p. 14) have been fully checked by me and are entirely erroneous. Petrunkevitch (1953, pp. 20 and 24) also studied the same specimens and agreed that there were no stigmata. It is now possible to go further and state that it would have been impossible for these forms to have had stigmata, as these would have to occur on the face of the abdominal plate and not the sternite, inasmuch as the two scorpions in question are meristosternous and lobosternous respectively. The holotype and only known specimen is preserved in two parts, showing the interior of the dorsal and ventral sides respectively on each half of the nodule. Preservation is remarkably good, showing fine details such as the lateral eyes and punctations throughout the scorpion. It also revealed that the skin on parts of the pedipalp is finely punctate, with larger, limbated setal openings scattered throughout, which indicate trich- obothria. The preservation is so perfect that an attempt is made to detail the position of these trichobothria (see Text-figs. 104C—H) on the pedipalp. The carapace is trapezoidal, wider than long, with greatest width posteriorly, and the anterior edge glos- sate at the central part. There are three lateral eyes at the anterior corners, and the median eyes are on an elongate mound approximately in the middle of the anterior half of the carapace. The median eyes are cir- cular, with strong interocular ridges. A raised marginal rim is developed on the lateral edges and posterior part of the carapace. Fine punctations occur along the an- terior edge and on the lateral parts of the carapace; otherwise it is smooth. It is very similar to the carapace in some of the living scorpions, although the anterior FOssIL SCORPIONIDA: KJELLESVIG-WAERING 233 is cuspidate. It is not, however, produced forward to the extent that it is in scorpions such as Mazonia, Gigantoscorpio, Eoscorpius, etc., which are considered aquatic and in which the glossate anterior was appar- ently a structure used for digging in the mud. The dorsal tergites are seen from the inside, each bounded on the anterior by a heavy marginal rim and on the lateral sides by a doublure. Each tergite increases in length posteriorly and is ornamented with numerous punctations that are concentrated on the central and lateral margins. The chelicerae are robust, showing that the free finger has four inner teeth and approximately three smaller rudimentary teeth or ridges on the inner side. This type of chelicera would seem to correspond more closely to those of the Chaerilidae than any other family. The opposing fixed ramus comprises four teeth with a long terminal tooth that fits between the two terminal teeth of the fixed ramus (see Text-fig. 103E). Again, the fixed ramus seems to correspond closely to the arrangement in the genus Chaerilus (see Vachon, 1963, p. 162). The pedipalps are robustly developed and show all of the joints except the coxae. The trochanter is deeply sculptured with coarse granulations. This is followed by the femur, which is robust and barely granulated ventrally. On the dorsal side, a linear series of granules occurs at the central part. The tibia is ornamented with a linear series of granules on the dorsal side and with a row of granules across the ventral face. The hand is preserved in its entirety, showing it to be robust, but with long curved fingers. Apparently two or three ridges may have been present on the hand, but these are not sufficiently preserved for description as they are flat- tened. The fingers are curved, and at least the free finger /f ij Ventral Dorsal Ch4 Cha Ch4 Dorsal Text-figure 103.—Palaeopisthacanthus schucherti Petrunkevitch. Holotype, YPM 140. From the Upper Carboniferous (Pennsylvanian), Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. See foldout inside front cover for explanation of abbreviations. A. Showing the dorsal surface. B. Ventral aspect. C. Left chelicera, drawn from the specimen immersed in alcohol. All four joints of the appendage are shown. D. The anterior carapace and glossate process, as well as the lateral eyes, are particularly well shown. E. Cheliceral detail, showing restored fixed finger, making use of parts and counterparts of the holotype. 234 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 has a well-developed keel running alone the outer face. The inner edges of both fingers are without denticles of any kind. The location of the granules and limbated tricho- bothria are detailed on the pedipalp joints; ventral and dorsal are shown on Text-figures 104C-H. In com- paring these with modern forms, the reader is cau- tioned that compression has occurred, but they are not entirely flattened. Parts of the walking legs are preserved, but not suf- ficiently so for description. Heavy punctations occur on the ventral side of the femur of the third walking leg. The venter is preserved so as to reveal the inside; however, the combs are covered by the sternites, and only their outlines are discernible (see Text-figs. 103B, 104A-B). The coxosternal region is clearly revealed and shows that the sternum is large and pentagonal. The anterior is not clear, but it definitely is triangular, giving a pen- tagonal shape to the sternum. In the center is an ellip- tical area that was sunk inward in life, undoubtedly for the attachment of muscles. A narrow doublure can be seen on the lateral margin. The first and second pairs of coxae occur exactly as in Recent scorpions; that is, maxillary lobes of the first two pairs of coxae are well developed, and the second pair meets at the midsection with the first pair immediately behind the maxillary lobes of the second pair of coxae. The third and fourth pairs abut against the sides of the pentagonal sternum. The first pair is spatulate, again as in the Chaerilidae. The genital operculum consists of two lacrimiform plates with the blunt ends opposite one another. Vogel and Durden (1966) report two rounded areas within the genital opercula, but the rounded, circular struc- tures are more probably areas of muscle attachment and are not part of the actual sclerite, which is distinctly lacrimiform. Of particular importance is the presence of a well-developed bilobate prepectinal plate. The large anvil-shaped pectinal plate, to which the pectines are attached has been jammed partly under the prepectinal plate. The relationship and normal position can be seen in Text-figures 104A-B. The combs are almost completely covered, but the outline of the periphery can be noted, and on the left comb one of the teeth has been impressed through the sternite. The fulcra are large. The combs are revealed as long, wide structures, and it is estimated that there must have been more than ten teeth on each pectine. There is evidence of small, rounded, areoles or scler- ites, such as occur in the middle lamella of the comb in Recent scorpions and in such fossil scorpions as Eoscorptus. The four sternites are preserved. These are true ster- nites, without doublures, and show that they are or- thosternous and with round stigmata. All eight stig- mata have been seen, although two of the anterior sternites are not so well-preserved as those following. The round stigmata recall some of the living Chactidae, such as Broteochactas, and the Chaerilidae. The last preabdominal tergite does not have ridges ventrally and is ornamented with a series of punctations along the dorsal side. The postabdomen is not preserved. Measurements have been given by Petrunkevitch (1913) and Vogel and Durden (1966). Type information. —Pennsylvanian, lower Francis Creek Shale, Mazon Creek, Grundy County, Illinois. Remarks. —Vogel and Durden (1966, p. 657) as- sume that the holotype specimen is exuviae from ec- dysis, but this does not seem to be the case. The pres- ervation is very different from that of other Mazon Creek scorpions, and it appears to have been a dried specimen that had secondarily been swept into the water and later covered by sediments. This is based on the interesting fact that the ventral side of the scorpion is completely concave, fitting against the convex dorsal surface. The split of the nodule occurred through the integumental skin on the side, thus the dorsal side is on the side of one nodule, preserved from the inside, and the venter on the other, also preserved from the inside. Recent scorpions, when dried, also become con- cave on the ventral surface and become pressed or bowed upward against the convex dorsal surface. In the holotype, the venter is bowed outward, but this apparent convex surface is actually a concave surface as it is seen from the inside, and is not the actual ventral surface. The dorsal surface accordingly is concave, but, inasmuch as this is the concave inner surface, it is actually the inside of a dorsal surface. There is no question that this scorpion is a pulmo- nate, terrestrial form, which definitely belongs in a separate family from the living forms. The presence of spatulate maxillary lobes in the first pair of coxae, the type of chelicerae and stigmata seem to be significant Text-figure 104.—Palaeopisthacanthus schucherti Petrunkevitch. Holotype, YPM 140. From the Upper Carboniferous (Pennsylva- nian), Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. Figures C through H are drawn mainly from latex casts of the ho- lotype. Trichobothria are represented by circles, pustules by black dots. See foldout inside front cover for explanation of abbreviations. A. Showing the posterior coxosternal region and opercular and pectinal areas from the inside. B. A reconstruction of the coxosternal and pectinal areas; no restoration involved. C. Dorsal view of the femur of the right pedipalp. D. Ventral view of the femur of the right pedipalp. E. Dorsal view of tibia of right pedipalp. F. Ventral view of tibia of right pedipalp. No trichobothria are present over most of this surface. G. Dorsal view of hand and part of the fingers of night pedipalp. H. Ventral view of hand and part of fingers of right pedi- palp. FosstL SCORPIONIDA: KJELLESVIG-WAERING 236 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 in showing the possibility of this scorpion being a fore- runner of the Chaerilidae (see Text-figs. 103B and 104B). Genus COMPSOSCORPIUS Petrunkevitch, 1949 Palaeopisthacanthidae having carapace with round- ed anterior margin. Type species. —Compsoscorpius elegans Petrunke- vitch, 1949. Geological range. —Upper Carboniferous (Coal Measures) of England. Remarks. —The holotype of Compsoscorpius elon- gatus Petrunkevitch, 1949, is a male of Compsoscor- pius elegans Petrunkevitch, 1949. The holotype spec- imen of Compsoscorpius elegans is a female. Typhlopisthacanthus anglicus Petrunkevitch, 1949, is merely a very poorly-preserved specimen of Compso- scorpius elegans. All three specimens, therefore, are synonymous and were described in the same publi- cation by Petrunkevitch (1949). As first reviewer, I hereby select the genus and species of Compsoscorpius elegans Petrunkevitch to be the correct taxon to use for this scorpion. The presence of three lateral eyes, indeed, of simple lateral eyes, whether one or five, is a character that occurs only on pulmonate or land-living scorpions. Therefore, although the underside of Compsoscorpius is unknown, I do not hesitate to consider it an Or- thosternina. Compsoscorpius elegans Petrunkevitch, 1949 Text-figures 105-107 1911. Anthracoscorpio buthiformis Pocock (partim), p. 26, fig. 7. 1949. Typhlopisthacanthus anglicus Petrunkevitch, p. 145, figs. 143, 182. 1949. Compsoscorpius elegans Petrunkevitch, pp. 149-150, figs. 152- 154, 183-185. 1949. Compsoscorpius elongatus Petrunkevitch, pp. 150-151, figs. 147-150, 186-188. 1953. Compsoscorpius elegans Petrunkevitch. Petrunkevitch, pp. 32— 33. 1953. Compsoscorpius elongatus Petrunkevitch. Petrunkevitch, p. 33: 1953. Typhlopisthacanthus anglicus Petrunkevitch. Petrunkevitch, p. 34. 1955. Compsoscorpius elegans Petrunkevitch. Petrunkevitch, p. 75, fig. 44(1). 1962. Compsoscorpius elegans Petrunkevitch. Dubinin, p. 431, figs. 1235, 1253. Petrunkevitch (1949, pp. 144-145) erected the genus Typhlopisthacanthus, with T. mazonensis (Petrun- kevitch) as type species, for a rather poorly-preserved scorpion from the Mazon Creek area of Illinois. The genus was purported to be an eyeless scorpion with a very small tail, somewhat resembling a gigantic pseudoscorpion, or scorpions such as the living Hor- murus, which are characterized by having small tails. He further referred a new species from the British Coal Measures to his new genus (7. anglicus). It has been shown in the review description of the holotype of 7. mazonensis, given above, that the eyeless feature and the small tail were fictitious and that, furthermore, the holotype was nothing more than a poorly-preserved specimen of a female of Eoscorpius carbonarius Meek and Worthen and, therefore, the genus 7yphlopisth- acanthus Petrunkevitch, 1949, was an invalid genus, being a junior synonym of the genus Eoscorpius Meek and Worthen, 1868. 7. anglicus Petrunkevitch also is based on imaginary characters and, as shown in the redescription and new figures, is nothing more than a very poorly-preserved specimen of Compsoscorpius elegans Petrunkevitch and, judging from its inflated state and sparse ornamentation, is a female of that species. It is also shown below that Compsoscorpius elongatus Petrunkevitch is a junior synonym of C. ele- gans Petrunkevitch, the latter being a female, whereas the former is a male. Both specimens, as well as the holotype of Typhlopisthacanthus anglicus, are from the same horizon and locality, namely, the Coseley area of Staffordshire, England. As stated under ““Remarks” concerning the genus Compsoscorpius, Compsoscor- pius elegans is the correct name for this species. The description of the holotype of Compsoscorpius elegans Petrunkevitch given by Petrunkevitch (1949, p. 149) is mainly correct, although his figure is lacking in important details that would possibly have resulted in revealing that C. elegans and C. elongatus are syn- onymous. The two differ only in the presence of coarser pustules on the holotype of C. elongatus, and in the overall more slender aspect of the latter. Both of these traits can definitely be attributed to sex, as the holotype of C. elongatus is a male, a determination based on the great length of the twelfth tergite, the coarser or- namentation and the more slender aspect. On the other hand, the more robust aspect of the holotype of C. elegans and the finer pustules reveal that this is a fe- male. These are good secondary characters for sexual determination. The three lateral eyes show that this scorpion was probably a land dweller that breathed through book-lungs, as present-day scorpions do, and is very likely related to the American species Palae- opisthacanthus schucherti Petrunkevitch, which has been shown to have stigmata and is a typical Orthos- ternina. Text-figure 105.—Compsoscorpius elegans Petrunkevitch. Holo- type (BM(NH) 1.7883), part and counterpart. Female. From the Carboniferous, Upper Coal Measures, Coseley, Staffordshire, En- gland. See foldout inside front cover for explanation of abbrevia- tions. A. External dorsal aspect. B. Internal dorsal aspect, except for ventral aspect of the right pedipalp. 238 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 The marginal rim, shape of the carapace, shape of the median eyes and eye node, the presence of three lateral eyes on a side, the pattern of ornamentation, the details of the peculiar, large tubercle on the last preabdominal tergite, as well as the coarse pustulation surmounting the caudal carinae (note that the tail is turned in the holotype of C. elongatus), are identical in C. elegans and C. elongatus, so that the writer does not hesitate to refer them to the same species. The holotype of Typhlopisthacanthus anglicus Pe- trunkevitch is so distorted and poorly preserved that there is considerable doubt as to the propriety of its ever having been described. Petrunkevitch’s descrip- tion is erroneous as well as the figure given (1949, pp. 145-146, figs. 143, 182). The figures given here (Text- figs. 107A, B) show the part and counterpart; it should be noted that the specimen is preserved in a curved position, therefore the figure has been “rolled over” to include all on a single plane. The carapace shows little, but has a median division or sulcus, which seems to separate it into halves. The eyes are not preserved, but would occur in the indicated ocellar mound. It is reck- less to state that there were no eyes; there are none on the specimen, but that is entirely due to preservation. The tergites are greatly telescoped together, making the carapace appear to be huge, as noted by Petrunkevitch (1949, p. 145). However, of diagnostic value is the presence of two maxillary lobes, belonging to the first and second coxae and indicating a scorpion much like the living ones. Both narrow maxillary lobes reach the anterior to form a well-developed oral chamber as in modern scorpions. It would be safe to assume that the third and fourth coxae abut a sternum that very likely is pentagonal. This would give a coxosternal arrange- ment much as in present-day scorpions, but also like some present in the Pennsylvanian. The legs are pre- served only as very small fragments. The presence of a small tail such as occurs in some present-day scorpions (Hormurus) is entirely illusory. There are no caudal segments preserved and much less the telson. The scorpion is identical to Compsoscorpius elegans Petrunkevitch, a determination which is based on the identification above of C. elongatus as the male and C. elegans as the female of a single species. The or- namentation on the carapace of the holotype of 7)- phlopisthacanthus anglicus is identical to that of the male of C. elegans (holotype of C. elongatus), includ- ing the row of larger punctae on the base of the cara- Text-figure 106.—Compsoscorpius elegans Petrunkevitch. Para- type, BM(NH) In.15862 (holotype of C. elongatus). Male. From the Carboniferous, Upper Coal Measures, Coseley, Staffordshire, En- gland. See foldout inside front cover for explanation of abbrevia- tions. counterpart Tel FossiL SCORPIONIDA: KJELLESVIG- WAERING 239 pace; the punctations along the lower half of each ter- gite and the punctae (trichobothria?) developed into two rows on the ventral part of the femur of the pedipalp are also identical. Type information. —All three specimens are from the Carboniferous, from the same horizon of concre- tions about ten feet above the Thick or Ten-foot Coal, Upper Coal Measures, Coseley, Staffordshire, England. All specimens are in the British Museum (Natural His- tory): holotype, BM(NH) 1.7883; paratypes, BM(NH) In.15862 and In.31261. Family SCORPIONIDAE Leach, 1815 Scorpionoidea with third and fourth pairs of coxae grown together along lines of contact; single spur in intersegmental membrane between metatarsus and tar- sus of first and second pairs of legs, no such spur on third and fourth pairs; three pairs of lateral eyes. Genus MIOSCORPIO, new genus Scorpionidae having carapace rounded and protrud- ing in front; sternum deeply divided by a median groove. Derivatio nominis. —Scorpion from the Miocene. Type species. —Scorpio zeuneri Hadzi, 1931. Geological range. — Miocene. Remarks. —This is the only known Miocene scor- pion genus. It differs from the genus Scorpio in lacking the deep anterior cleft that results in the bilobed an- terior margin so characteristic of that genus. The Mio- cene form retains the produced anterior that recalls the older fossil scorpions and that is still retained by the living Brachistosternus, a Bothriuridae of the northern Text-figure 107.—Compsoscorpius elegans Petrunkevitch. Para- type (BM(NH) In.31261), part and counterpart (holotype of 7y- Phlopisthacanthus anglicus Petrunkevitch). From the Carboniferous, Upper Coal Measures, Coseley, Staffordshire, England. See foldout inside front cover for explanation of abbreviations. A. External dorsal aspect. B. Internal dorsal aspect. and western part of the South American continent, and by the troglobite Typhlochactas of Mexico, a Chac- tidae. Mioscorpio zeuneri (Hadzi, 1931) 1931. Scorpio zeuneri Hadzi, pp. 134-148, figs. 1-8. Type information.—The holotype and paratypes come from the Upper Miocene Sprudelsinter at Bot- tingen, Swabian Alps, Germany, and are on deposit in the Staatliches Museum fiir Naturkunde (SMN), Stutt- gart, West Germany. Dr. Jovan Hadzi kindly sent me the original casts used by him in his description of this interesting species. Later, in Nov. 1970, I was able to study the original specimens, which had been kindly sent to the Institut fiir Palaontologie der Rhein, at the Friedrich-Wilhelm University at Bonn by Dr. M. Warth of the Staatliches Museum fiir Naturkunde in Stuttgart, where the types are deposited. The scorpions are preserved as hollow external molds in the limestone, which apparently is a travertine deposit. It is necessary therefore to make casts before details are readily revealed. New silicone casts did not reveal any details not previously seen in the casts made by Hadzi; however, there are a number of important details that have been determined by a review of the original specimens. Hadzi shows the first and second pairs of coxae meet- ing in front of the sternum, without the development of maxillary lobes in the second pair of coxae (HadZi, 1931, fig. 7a). This would result in a coxosternal ar- rangement more primitive than most of the Carbon- iferous scorpions and, indeed, would preclude the scor- pion in question from being in the family Scorpionidae Leach, 1815, or even in the superfamily Scorpionoidea Leach, 1815. I disagree with Hadzi’s excellent description of the arrangement of the coxae, at least in respect to the coxal maxillary lobes. My review clearly shows well- developed maxillary lobes in both the first and second coxae, and the latter meet at midsection and are squeezed in between the first coxae and their maxillary lobes. Thus they are identical to those of living scor- pions. The carapace is preserved on one specimen (object 88) and, although I was unable to determine the type of eyes, the outline of the carapace is clearly preserved. This is straight at the base, rounded at the anterolateral angles, and with a distinct protrusion along the anterior margin. The carapace is probably longer than wide, although the entire width has not been seen. It mea- sures 5.2 mm in length. As stated above, this shape is an important taxonomic difference from that present in the genus Scorpio, which has a bilobed anterior. The surface of the carapace does not reveal any deep sulci, 240 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 although the basal margin, towards the sides, seems inflated. This would certainly be indicative of the Scor- pionidae. The sternum is small, deeply cleft at midsection, at least for the posterior three-fourths of the length. The base seems to be straight and the overall shape is pen- tagonal with parallel sides. The coxae of the last two pairs of legs abut the sides of the sternum. I was unable to determine the type of stigmata*’ present, but there is no doubt that they are present and that the scorpion is an Orthosternina. GENERA INCERTAE SEDIS Genus TITANOSCORPIO, new genus Very large scorpions with pedipalp having very short hand, both fingers curved inward, concave along the inner edge, and not joining along the edges. Edge armed with single (?) row of small denticles, at least on the distal part. The distal ends of both fingers hooklike. Tibia very short. Type species. —Titanoscorpio douglassi, n. gen., n. sp. Geological range.—Carboniferous of the Mazon Creek area of Illinois. Remarks. — Although it is not possible to determine the family to which this unusual scorpion should be delegated, there is no doubt that on the generic level itis distinct. Unique morphological structures are found in the two long, hooklike terminations of the pedipalp fingers, as well as the opposable concavity of both fin- gers. In other scorpions, the fixed finger fits against the free one. The presence of very small denticles along the edge of the fingers, although well-known in various living scorpions, is an uncommon character in Pal- aeozoic scorpions. It would appear that the pedipalp would have a scissorlike action in closing, rather than that of nearly all scorpions where pressure is caused by the fitting of the curved or concave free finger against the convex fixed finger. The size of this scorpion rivals that of Gigantoscor- pio willsi Stormer and Eoscorpius carbonarius Meek and Worthen (see above, description of a new and very large specimen of the chela of a pedipalp, FMNH (PE) 32083, No. H-19 in the Herdina collection). A scorpion with pedipalps of this size would probably reach 30 cm in body length. The short tibia suggests that this was a burrowing scorpion, although the size of the scorpion indicates a water-dweller, as is the case with nearly all of the Mazon Creek scorpions. #2? Hadzi (1931, p. 139) claims to have noted traces of a slitlike stigma on one of the sternites of object 86. A.S.C. Titanoscorpio douglassi, new species Text-figure 108 DLD Specimen I.—Holotype: Specimen in the per- sonal collection of David Lyon Douglass, Western Springs, Illinois,** consisting of three parts of an iron- stone concretion showing ventral and dorsal sides of the chela and attached femur of the pedipalp, as well as a small piece of the inner filling representing the distal part of the free finger. Type information. —Pennsylvanian, Francis Creek Shale of the Mazon Creek area, Illinois, collected by David Lyon Douglass, 1966. The chela of the pedipalp comprises a wide, nearly quadrate hand, without any discernible ornamenta- tion. The free finger is curved inward, convex rather than concave as in other Pennsylvanian scorpions, and is decidedly hooklike. The edge of the free finger con- tains at least one row of small, rather blunt denticles and presumably the same occur on the fixed finger, as in other scorpions. No pitting or other ornamentation was noted on the fingers. The femur is very short, stout and subrectangular in general shape. Measurements (in mm) of DLD unnumbered speci- men (holotype: Specimen I).— length width Hand: (center) 17.5 (ventral) 18.5 22.5 (dorsal) Free finger: 27.8 (inside) 5.5 (middle) Immovable finger: 31.2 (inside) 5.3 (middle) Tibia: (center) 20.2 RD! Greatest length of chela: 52.0 Text-figure 108.— 7itanoscorpio douglassi, n. gen., n. sp. From the Upper Carboniferous (Pennsylvanian), lower Carbondale Forma- tion, Lower Francis Creek Shale, Mazon Creek, Grundy Co., IL. Specimens represented by figures A to G are in the personal collection of David Lyon Douglass (DLD), Western Springs, Illinois.*? See foldout inside front cover for explanation of abbreviations. A-E. Details of the left pedipalp, paratype DLD Specimen III). Labelled ‘Bartowski Collection, write Gene Richardson” on the back of the drawings. A. Inner side showing joints P2-6. B. Outer side, segments P4-6. C. Detail of P4 of figure A, showing the pustular organization. D. Enlargement of the end of the chela, inner side (fig. A), to show details of the fine linear pustulation and large pustules on the fixed finger. E. Detail, tip of segments P5-6, outer side. F. Dorsal view of the terminus of the pedipalp, holotype (Specimen 1). G. Ventral view of the same. H. Paratype (RC Specimen II) in the private collection of Richard Cramer, Forest Park, IL 60130 U.S.A. Third segment of the pedipalp (P3). 43 IT spoke with Mr. Douglass, then of 95563 Highway 101, Ya- chats, OR 97498, U.S.A., on April 23, 1985. He said that a more permanent address would be: 1312 Chicago Avenue, Evanston, IL 60201, U.S.A. P.R.H. FossiIL SCORPIONIDA: KJELLESVIG- WAERING 241 skin removed left inner inclined downward 1cm left outer broken away inclined downward 1mm 1cm 242 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Derivatio nominis. —The species has been named in honor of the collector. RC Specimen II. —Paratype: Specimen in the col- lection of Richard X. Cramer (1435 Marengo Ave., Forest Park, IL 60130, U.S.A.), collected August, 1975, in Pit 11, Will County, Illinois. The paratype consists of the right humerus (P3) of the pedipalp. It is very stout, short, with only one superior carina on the posterior dorsal side. Socket for articulation of the brachium is present midway at the distal end. A raised area is present at the conjunction with the trochanter (see Text-fig. 108H), which would be overlain by the latter joint. The length of humerus is 21.8 mm, and the width is 20.0 mm. The paratype is smaller than the holotype, although still a formidable scorpion about 20 cm long. Little is left of the original skin. DLD Specimen III. —Paratype: Specimen illustrat- ed in Text-figures 108A-—E, labelled on the back of the drawing “Bartowski Collection, write Gene Richard- son” and accredited to the private collection of David Lyon Douglass in the figure description.*4 Genus WATTISONIA Wills, 1960 Small scorpion with last preabdominal tergite having two low carinae. Corresponding ventral plate larger, with large median carina and two low lateral crests. Type species. — Wattisonia coseleyensis Wills, 1960. Geological range. —Upper Carboniferous. Wattisonia coseleyensis Wills, 1960*° 1960. Wattisonia coseleyensis Wills, pp. 319-322, pl. 56, figs. 2-4; text-figs. 26, 27. Genus PALAEOMACHUS Pocock, 1911 Hand broad and oval, its width greatly exceeding that of the brachium, its length exceeding that of the fingers, which are short and in contact when closed. The length of the immovable digit about equals the width of the hand. Type species. —Eoscorpius anglicus Woodward, 1876. Geological range. —Upper Carboniferous. #4 No text was found about the specimen, and on the back of the drawing it was originally assigned to Eoscorpius carbonarius. A.S.C. 45 Kjellesvig-Waering left no text on this species. The holotype, BU 722A,B, was collected by Mr. J. T. Wattison of Shrewsbury, almost certainly from the Ten-foot Measures above the Thick Coal, Coseley, South Staffordshire. It consists of two half-nodules with a small scorpion with normal, rather hairy tergites. The seventh ven- tral tergite is thickly coated by posteriorly-directed setae attached to small hair-bases. All other organs unknown. Nothing could be added to Wills’ description. A.S.C. Palaeomachus anglicus (Woodward, 1876) Text-figure 109 1876. Eoscorpius anglicus Woodward, pp. 57-59, pl. VIII, fig. 3. 1911. Palaeomachus anglicus (Woodward). Pocock, pp. 16-17, fig. 2a. 1911. Palaeomachus anglicus (Woodward). Bather, p. 673, fig. 1913. Palaeomachus anglicus (Woodward). Petrunkevitch, pp. 33- 34. 1953. Palaeomachus anglicus (Woodward). Petrunkevitch, p. 38, fig. 132. 1955. Palaeomachus anglicus (Woodward). Petrunkevitch, p. 79. 1962. Palaeomachus anglicus (Woodward). Dubinin, p. 433, fig. 1259. Holotype. —BM(NH) I.944a and b. Upper Carbon- iferous Coal Measures, Skegby New Colliery, near Mansfield, Nottinghamshire, England. Two _ pieces showing part and counterpart of a pedipalp lacking its coxa, but otherwise complete.*® SCORPIONIDA, gen. et sp. indet. Stormer, 1969 1960. Eurypterid or scorpion. Stormer, p. 90, text-fig. 2. 1969. Scorpionida gen. et sp. indet., Stormer, p. 30, pl. 2, fig. 8. The specimen consists of a well-preserved cauda, consisting of the last four segments and telson. The description is adequate, and all that can be added is that the long, twelfth tergite denotes the specimen is a male. Type information.—From the Lower Devonian Klerfer beds of Willwerath, Eifel, Germany. Text-figure 109.—Palaeomachus anglicus (Woodward). Holotype, BM(NH) I.944a,b. From the Upper Carboniferous, Coal Measures, Skegby New Colliery, near Mansfield, Nottinghamshire, England. See foldout inside front cover for explanation of abbreviations. A. Fragmental pedipalp, counterpart. B. Fragmental pedipalp, part. C. Curious and unique interlocking of the base of the free dactylus (P6) and dactylus (P5) with a special sheath boss on P5 to accom- modate the blade boss of P6, and cross-section to show the inter- locking bosses. 46 Other than the description of the text-figures, the author left no notes on this specimen. The dimensions are given by Petrunkevitch (1953, p. 38). The curious interlocking system of the base of the free dactylus (P6) and the immovable finger (5) with a special sheath boss on PS is diagrammed in Text-figure 109C. A.S.C. FossiL SCORPIONIDA: KJELLESVIG-WAERING 243 P6 P6 P5 P5 P5 eg ae p2 B P3 SS pa P5 P6 P5 . P2 = Ps Qo eet counterpart Text-figure 110.—Coxosternal areas of the Holosternina (contin- ued on Text-fig. 111). A. Praearcturidae—Praearcturus gigas Woodward. B. Mesopho- nidae— Mesophonus perornatus Wills. *7C. Anthracoscorpionidae — Allobuthus macrostethus, n. gen., n. sp. D. Anthracoscorpionidae— Anthracoscorpio juvenis KuSta. E. Phoxiscorpionidae— Phoxiscorpio peachi, n. gen., n. sp. F. Buthiscorpiidae— Buthiscorpius buthiformis (Pocock). G. Eoctonidae— Eoctonus miniatus Petrunkevitch. H. Gi- gantoscorpionidae— Gigantoscorpio willsi Stormer. I. Heloscorpion- idae— Heloscorpio sutcliffei (Woodward). J. Centromachidae— Cen- tromachus euglyptus (Peach). K. Archaeoctonidae— Archaeoctonus glaber (Peach) (after Stormer, 1963). L. Spongiophonidae— Spon- giophonus pustulosus Wills. M. Stenoscorpionidae— Stenoscorpio gracilis (Wills). N. Acanthoscorpionidae—Acanthoscorpio mucro- natus, n. gen., n. sp. O. Allopalaeophonidae— Allopalaeophonus ca- ledonicus (Hunter). P. Mazoniidae— Mazonia woodiana Meek and Worthen. 47 In the original figure description C was labelled ““Anthraco- buthidae— Anthracobuthus’’. We can only guess what he was refer- ring to. A.S.C. Text-figure 111.—Coxosternal areas. A-E. Holosternina. A. Proscorpiidae— Proscorpius osborni (Whitfield). B. Waeringo- scorpionidae — Waeringoscorpio hefteri Stormer. C. Labriscorpion- idae— Labriscorpio alliedensis Leary (Kjellesvig- Waering’s version). D. Labriscorpionidae— Labriscorpio alliedensis Leary (Leary’s ver- sion). E. Palaeoscorpiidae — Palaeoscorpius devonicus Lehmann. F. Bilobosternina— Branchioscorpionidae— Branchioscorpio — richard- soni, n. gen., n. sp. G. Orthosternina— Palaeopisthacanthidae— Pa- laeopisthacanthus schucherti Petrunkevitch. H. Meristosternina— Palaeobuthidae—Palaeobuthus distinctus Petrunkevitch. I. Meris- tosternina—Cyclophthalmidae—Cyclophthalmus senior Corda. J. Lobosternina— Loboarchaeoctonidae— Loboarchaeoctonus squa- mosusS, N. gen., N. sp. Text-figure 112.—Coxosternal areas of the Lobosternina (see also Text-figure 111J). A. Paraisobuthidae— Paraisobuthus prantli, n. gen., n. sp. B. Eo- scorpiidae—Eoscorpius carbonarius Meek and Worthen. C. Eobu- thidae—Eobuthus rakovnicensis Frié. D. Isobuthidae—Jsobuthus kralupensis (Thorell and Lindstrém). E. Opsieobuthidae— Opsieo- buthus pottsvillensis (Moore). F. Telmatoscorpionidae— Te/mato- scorpio brevipectus, n. gen., n. sp. G. Waterstoniidae— Waterstonia airdriensis, n. gen., n. sp. H. Kronoscorpionidae— Kronoscorpio dan- ielsi (Petrunkevitch). I. Petaloscorpionidae— Petaloscorpio bureaui, n. gen., n. sp. J. Pseudobuthiscorpiidae — Pseudobuthiscorpius labio- sus, n. gen., n. sp. K. Pareobuthidae— Pareobuthus salopiensis Wills. L. Anthracochaerilidae—Anthracochaerilus palustris, n. gen., n. sp. Text-figure 113.— Morphological stages, but not necessarily in or- der of phylogenetic development, of the floor of the oral tube. The more primitive are at the bottom of the page, progressing upward to the completely-developed oral tube at the top of the page. A. A group of holosternous scorpions: 1. Centromachidae— Cen- tromachus euglyptus (Peach). 2. Mazoniidae—Mazonia woodiana Meek and Worthen. 3. Mesophonidae — Mesophonus perornatus Wills. 4. Heloscorpionidae— Heloscorpio sutcliffei (Woodward). B. Another group of holosternous scorpions: 1. Archaeoctonidae— Archaeoctonus glaber (Peach). 2. Gigantoscorpionidae—Giganto- scorpio willsi Stormer. 3. Anthracoscorpionidae— fam) s..2..- OR ae ty te 29,30 eurypteroides, Archaeophonus ..............00cseeeeeeeeeeeeees 50-52,196 euglyptus, Centromachus .... 22,78,84,86—87,94,192,218,243,244,247,253 E}OSCONDIUS) se sccu sca os cscs asiereesoascsoentse2 ise conceesseseeacessercesss 86,253 Europhthalmus Petrunkevitch, 1949 ................ 162,163,178,253 longimanus Petrunkevitch, 1949 ....... 18,177,178,179,180,253 unyptenGameceers.---eeeceeee see 21,23,24,26-27,31,48,96,144,196 IB Uny teri ae terse ice tic secsctoc econ conc econ cosas Besce sels seascesces 144,166 BunypterdarZone vs cscscaccccecsecsececsescestsccest seven tescu ss Saesieescces: 53 Eurypterina Eurypterus EDOSSI DVI DV ZINGCA SAI Lela eecte res coc eceseoree cece eceesereees cree tee 253 Euscorpionina (Euscorpiones) Petrunkevitch, 1949 .............. 253 WEEUS CONDIUS Wee eeeee ee tonte deen 115 buthiformis (Pocock) 253 GVOMULION etic cece ass seen sce ve wee conc cn seeviebes < seencswotan ieee 24-25,27-28 extermallmorphology jesse seececsceeeveo tec sisese ene see tee sean os oa-aine sec aeece F exuviae, ecdysial CV OS apts ee sects ce soe nee Sonus cules Suse dazeeeeds ese seuss eueserts Inolochroali(Compound) hcecsescreteete ccoeessesccetaterceceoresercestese 13 schizochroali(aggrepate)mrescssesecsesscceseeceseeseceecestaseeseecreenes 13 me SCXUAILGIMOLDNISMigs eenssccersscasecosececocseercor cece ccessuseesces 79,94 size relationships of median and lateral ............. 14-15,52,144 Reistmantelia) BriGs W904 teee..ceaseeeco-ceecedcs-eess [S15 1525756) 2511 OV MALIN TIC ILO OA srmenaceserns one e ic sesccc rere seeseien ete ce 151,156-157 Field Museum of Natural History [FMNH] ..................... 254,F fixative 149 Florida State University [FSU] 3,86,F fossil scorpions, characteristics and nature of .................. 11-21 France FAUMAanNCewBasin ere arcs ser teee wes sor ewater stle sees wove ce secalost srenaaee ce 186 INOrthtofaViOSPeSinewsscesccssctcsesccstecnecctes -cemecseesess st aceeashemescce 30 Francke (pers. commun.) ... 15,19 Fri¢ (1873) 129,134,153,154,160,203 Fri¢ (1902) 129 Brici(i904) yess sescseeseese 39,57,107,129,134,142,151,152,153,154, 156,158,160,163,198,202,203,232 ECD O77) reeesee reese ne reese oopieacen ai see aceon fens: tuasesiereatstes 39 frici, Paraisobuthus 203-204,206 U1] Cel eee TR er Re A MR TN Sic BOA e aN 74 al (UO TIM) eee sae ances conte ooteenc tes ccs secoscuvacevcuesneceesceeseeee 30 Garnettiidae Dubinin, 1962 113,163,251,B (GarneltiusyAssem Dla ge merersen cero crcer corte oer eea eee 115 Garnettius Petrunkevitch, 1953 ................... 16,17,173,162,163, 166,189,251 hungerfordi (Elias, 1936) ....................0...- 12,15,714-120,196 (QA)IETOS, poceenbecunpepsnecconeeraescs 116,120 Generavincertae!Sedisi- sees ene eee eee 126,240-242,252 GeologicallSurveyof'Ganada\[GS@]) <...2.-..-....-0.-2--2-se2-000- 255,F Geological Survey Museum, London [GSM] .................... 255,F Geralinura ? sutcliffei Woodward, 1907 ..................... 86,88,253 Gerardikiolmi Collection eeetestereesce ase cresees ce enesens seeee ae eeee nes 13 (GLIA! nb andesbocuseebatbeaEade0082 ObaaEOnETEECE OBE Nob EDOSSeCEoe HOH ECOBEDEEEEO 14 Gigantoscorpio Stormer, 1963 ......... 15,17,18,20, 76, 158,233,250 willsi Stormer, 1963 ............. 19,22,31,75,76-79, 125,189,192, 199,218,243,244,247 Gigantoscorpionidae, n. fam. .................. 76,243,244,247,250,B Gigantoscorpionoidea, n. superfam. .........................-- 75,250,B gigas, Praearcturus 78, 121-125,243,244 pillsyeee ss. 2.8222. 5-8, 16-18 ...... 23-24,26,28-32,34,35,95-97, 101,126-128,157,206,229 ODEMIN GS ieeess eee ei etie es 33,82-83,157,226,229 pouchesior chambers) cvce-2---- -se-ceceseeeesesee- 31,96-97,139,206 glaber, Archaeoctonus .................... 21,53,66-68, 70,192,243,244,247 CENLFOINACRUS Ree Ane eS ere e RTOS TS SOaE ee ee wet 66,253 Eoscorpius 66,68,253 pnathobase: si2escceseessesscssscscse ste sceuesee se 16,48,226 Gdttingen University Museum [GUM] ........................6655 256,F GrabawlGl928) ee orca: seceoe no roc eieen oe eee oa oaeecaeeae eet ta toateseeea 81 Grabaurand: Shimeri(19i10)iesceccsvasceseseetteess cossereecese-cerscroreere 81 gracilis, IM CSODRONUSY covs.cseceseseesiatocvessevsnes oss cesses 32,34,74,80,94,254 IS LENOSCOMDIO Ness ses cence eane scene seaee ese noe ees 34,73, 74,75,90,243,244 granulosus, Alloscorpius ........ 100,163, 168-170, 172,175,176,207,210,253 IE OSCOMPIUSI x seceseateesteeee eae 135,139-140,163,176,210,211,253 (GOS DUS Rance c ean cons ox tate SeS Leu Nos ET e SOR aS USSD ae SRO RT 60 TT QdOGENES encase ctr cacik leaake noes bias Fone cou Nea iat aeie a Hae tiet 115,167 troglodytes\(Reters) ie src.cctsac cece ste sss sc cee rete sek reese eee ose ees 22 iHadruroidesilunatusi(XOCh)hecceses: eeseentenreecerste se eesneee cee 243,248 Hadz(OSI) esta Mineets- cate neon hee ce centte terete 12,239,240 lan dlarschi(l 9 OG) ise sesccascottes.ceecsssvosseessoacercootuscssutezecseaseess 81 Harrington (1959)! sess occcccacesesessesecdesetecoscesesecscdeceoesaseevare 13 Harrington, Moore, and Stubblefield (1959) .............0......2004 13 hefteri, Waeringoscorpio ............2..000+++ 26,31,48,52,229,243,245 IETCLOSCONPION WIRED. .ucccceses-cccsseiesse 2: << onesee cceess=s2 ears 86-88, 250 sutcliffei (Woodward, 1907) ...............66..05 88-90, 243,244,247 Heloscorpionidae, n. fam. ...................065 86,243,244,247,250,B Plenmigi(i922) ecco scotecicdecas tcc esceiwes sat coteevacasssiaeesstescess seeees 121 Herdina specimens 174,175,196 ¥esslete (i969) fesscsescecces eecscacsecteeccen cece semen eters coeeemeees 121,125 FT CLEPOMEINUS comec ova suscties oo suesoudasteuesuecce sas saawssvceeeeats 17,31,144 FreubuschiQlioG2)icesecccntcseecscrsnec ests caves teececeeteoceee teens 61 Mofimani(1'93)l)) essseccsces cate sccc ce nease seca ce atesoneeee rece nee eeene tate 229 FLOM USD Oe eo ee sae ace testes cces selene sadeeee cece see as Holm (1896) Holm (1897) Holm (1898) Holm (1899) olm=Willsitechniqueicesces.s--«sescesecosseeeeeecsecene=esesec seca 92-93 holochroal (compound) eyes, definition .........................0200 13 Holosternina, n. infraorder ... 14,15,16,21—22,24,26-28,31,33,34, 36-37,38,92,94,97,142,151,163, 212,243,244,245,247,250,B holti, 1 Ga TH (Sc ARE ao DESCRIBE E Oar DEERE CD OED C ECR CO NOR EE PAU AOINOG 162,207 (2?) Eobuthus 193,204,253 ISODULRUS rescore eee re coon ease ee ee erae ene caeere cen net ee 162,253 NEL OM FAUT US | occ oe core oe ae aoe Beek Woe ES 236,238 Hueberi(l960)) se.scseseececeresecue sca ccesscteccsessctceacsesseneceosolueras 126 Aucbert wii DNOSCOMPIOme.sceecese a eeter ees eee OSLO J 6 ON CLUE TICE (esecesoeciorcie cca ce ocd 1a S0COUC IEE HCC O DOSER OCS CEO 31,216 Dantksit!Salter 26. cx ssnccet cc -cccwcn ec eacarsae see seceemunt rece 12,182,183 hungerfordi, (GORNEHIUS cree sa cerece se see eT er I ac ene soe 12,15, 714-120,196 Mazonia 113,114 Hunter (S86) Pecos secon ercc esses te aee cadens tetiosttob basen a saeaeeetens 56,57 unten 8 88) aseccccscesscasecciacc. conc ctoscedvacscdceesssces-reseorsedeeacees 57 RUNtCErIPRAIACOPNONUS messes eee scen cesses cee tance seen teaeeeeetennan 57,254 Hydroscomplidaemmptamlgres-sesssseeteee eeeriseessseccscesseeoss 63,250,B Ely rOSCONPUS Tis BEN sa eeree sesso acct cease ences nets ccede= seers 63,106,250 GENISONT MASP cose sie cse sees eec esse teeter SP 12,63-66,125 282 PALAEONTOGRAPHICA AMERICANA, NUMBER 55 Illinois MazoniGreek /Aréal i. «ccc seccrssoee teste oeese ett esncceescnecteens 109,163 Grund yA CON eeeeee cere eee eecse neces 95,136,138,164,169,171,173, 196,207,211,233,240 Pit 1, Peabody Coal Co., Grundy Co.—Will Co. line ....... 174 Pit 6, Peabody Coal Co. near Braidwood ..................... 104 Pit 11, Peabody Coal Co., Will Co.-Kankakee Co. line .. 81, 98,194,242 Strip mines, Will Co.—Grundy Co. line .....................6: 101 illinois: State Museum [SMe sccsce-cesesestecesncrassmecece ect setmaee 254,F Indiana Glay Gity4 Owens COs se ccteccasecosencecseattecsceocesn ese 86,87,215 infans A, Mesophonus 32,74,254 infans D, Mesophonus 32,74,254 infans E, Mesophonus 32,80,254 LIL LQEUSYREI OSCORDIUS meet attnec cee cae ce eciec cer seces saeen sare ose e rene eres 253 ATA AO LOST See tec os roars ree ee eee eee ea cee men atne eeeueaitee 35-37 Inst. fiir Palaontologie, Friedrich Wilhelm Univeristat, Bomar [ESWW) ers aes oes ese sk ace sadecnaene ssceeaceveasavedes tees 256,F Institute of Geological Sciences, Edinburgh [GSE] ............ 255,F mntestinal:canal, preservatlONn .....--2.c-.e<0e0.e-2e eee: 3) fee 61,90 ISCHNUTUSIOCHIOPUSPINOGHE ce sece ace cge dees soe ceees te tases scout den uscstcest 22 Isobuthidae Petrunkevitch, 1913 ....... 133,151, 752-153,156,158, 162,198,243,246,247,B Isobuthoidea Petrunkevitch, 1913 ..... 135, 151-152, 158,162,190, 192,193,198,212,213,218,251,B Tsobuthus Fric¢, 1904 ...............++ 19,29,151,152, 153, 156,192,251 HolteMPOcoGks GW ll), cc.ssccesteeesscivccstseeese toe teswaecee. secs 162,253 kralupensis (Thorell and Lindstr6m, 1885) .. 152,/53-154,160, 168,192,203,204,243,246,247 ORNALUSIETICT 904) presse. aaseroneceea cearseanee sess sce cece aa ses 156,253 pottsvillensisi(NioOres 11923) ccsueceuvecesee 17,1325133)242:252 anglicus| (Woodward! 1187116) e2seasecceecseerenees sees ceecee eee 242-243 sp» betrunkevitch1949) 22... 5.2c.csecseaenacseesssnesie- see nen = 132,254 Paleontological Institute, Academy of Sciences, Leningrad, NES: SeREATEASS J civ ex cetieceassoaecee aces ac tsueeseawcsceaetecemecesaet 256,F Palaeophonidae Thorell and Lindstrém, 1885 ... 18,60, /42,251,B Palaeophonoidea Thorell and Lindstr6m, 1885 ..... 18,/42,251,B Palaeophonus Thorell and Lindstrém, 1884 ..... 14,17,28,66, 142, 144,146,251 caledonicus Hunter, 1886 ................. De dict 56,57,148,254 RURLCrINPOCOCKSE1.9 Ollrenesascce soe tone scens eeeencece weeeee ees 57,254 (2) lightbodyi Kjellesvig-Waering, 1954 ..................00...000. 149 loudonensisaurien 899s 2.7298 -secose: act eces reese. oinenee 230,254 nuncius Thorell and Lindstrém, 1884 ............ ere 29,56, 142-149,232 OsbDornuWihitheldsaS8S° gecssssceetecvetenceeeseees seceeecasesses 39,254 SDE CACH IS GS: ere ee. oa nesee cess eats ceieSaners oe te neaee nc aeee eee oseenenees 57 Palaeopisthacanthidae, n. fam. .......... 35,232,236,243,245,252,B Palaeopisthacanthus Petrunkevitch, 1913 ...... 12,15,17,19,29,30, 31,90,162,163,232,252 mazonensis Petrunkeyitchs, 19113) 2 osce.cqe..c22e=ceseeee see 163,254 schucherti Petrunkevitch, 1913 ...... 29,31,96,232-236,243,245 Palaeoscorpiidae Lehmann, 1944 .................. 61,243,245,250,B Palaeoscorpioidea Lehmann, 1944 ....................:000000 61,250,B Palaeoscorpionidae Lehmann, W944 0.0.5.2 ..2..cs een — yl Tiphoscorpionidae Cyclophthalmidae Palaeobuthidae Palaeophonidae Anthracocheirilidae Isobuthidae Eobuthidae Eoscorpiidae Pareobuthidae Kronoscorpionidae Paralsobuthidae Telmatoscorpionidae Scoloposcorpionidae Opsieobuthidae Loboarchaeoctonidae | Pseudobuthiscorpiidae | Petaloscorpionidae Waterstoniidae Branchioscorpionidae Dolichophonidae Palaeopisthacanthidae Scorpionidae No LCE Large LCE Carapace unknown Coxosternal area Coxosternal area Short thick legs A unknown LCE unknown Slender legs ending ending in spine-like ? BN } in 2 long claws posttarsus with Long cylindrical legs > Deeply lobosternous AP a SY x hy vary. short/unguea Loe No LCE Protolobosternous AP | Very short legs Round stigmata pec usualy LCE unkr Noicomiee Carapac: known CEICEEES Wahi) Carapace unknown 3 pairs of LE Sipaira OuLE apace un short and slender esi al = Text-figure 5.—Key to the classification of the Scorpionida, show- ing the relationship of the higher taxa, infraorders through families {prepared by A.S.C.]. See foldout inside front cover for explanation of abbreviations. I = Se _ a a NN se NN PREPARATION OF MANUSCRIPTS Palaeontographica Americana currently appears irregularly, on an average of about one monograph each year. This series is a publication outlet for significant longer paleontological monographs for which high quality photographic illustra- tions and the large quarto format are a requisite. Manuscripts submitted for publication in this monograph series must be typewritten, and double-spaced throughout (including direct quotations and ref- erences). All manuscripts should contain a table of contents, lists of text-figures and (or) tables, and a short, informative abstract that includes names of all new taxa. Format should follow that of recent numbers in the series. All measurements must be stated in the metric system, alone or in addition to the English system equivalent. The maximum dimensions for photographic plates are 178 mm 229mm (7” x 9”; outlined on this page). Single-page text-figures should be drafted for reproduction as single column (82 mm; 3'4”) or full page (178 mm; 7”) width, but arrangements can be made to publish text-figures that must be larger. Any lettering in illustrations should follow the recommendations of Collinson (1962). Authors must provide three (3) copies of the text and accompanying illus- trative material. The text and line-drawings may be reproduced xerographically, but glossy prints at publication scale must be supplied for all half-tone illustrations and photographic plates. These prints should be identified clearly on the back. All dated text-citations must be referenced. Additional references may be listed separately if their importance can be demonstrated by a short general com- ment, or individual annotations. Referenced publication titles must be spelled out in their entirety. Citations of illustrations within the monograph bear initial capitals (e.g., Plate, Text-figure), but citations of illustrations in other articles appear in lower-case letters (e.g., plate, text-figure). Original plate photomounts should have oversize cardboard backing and strong tracing paper overlays. These photomounts should be retained by the author until the manuscript has been formally accepted for publication. Explanations of text-figures should be interleaved on separate numbered pages within the text, and the approximate position of the text-figure in the text should be indicated. Explanations of plates follow the Bibliography. Authors are requested to enclose $10 with each manuscript submitted, to cover costs of postage during the review process. Collinson, J. 1962. Size of lettering for text-figures. Journal of Paleontology, vol. 36, p. 1402. Gilbert Dennison Harris (1864 - 1952) Founder of Paleontographica Americana (1916) ISBN 0-87710-401-8 Or ie » Paleontographica Americana NUMBER 56 JANUARY 8, 1988 Late Ordovician Sponges from the Malongulli Formation of central New South Wales, Australia by J. Keith Rigby and Barry D. Webby Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850 U.S.A. PALEONTOLOGICAL RESEARCH INSTITUTION Officers PRESIDENT 5 be oe ee eee ee eee eres Tenor WILLIAM P. S. VENTRESS MIGE-PRESIDENIIO: 30 eters Der ee ea es ee eo aes JAMES E. SORAUF SEGRETAR Yi. 20 i ep a ER A SE ee ON Ree Eee HENRY W. THEISEN TEREASURER GG rit ric ae Sarto ee ue ee he ere ean eye ae JAMES C. SHOWACRE ASSISTANT sDREASURERS spo cere ee ae ne eet ther ae aenoee cde meee JOHN L. CISNE TD IRECTOR is oo cose soc conte ene reat ie Bene OR oT open PETER R. HOOVER THE G ATH COUNSEL ceo kes ies eee ete ae cannes nee ersuatoeh HENRY W. THEISEN Trustees Bruce M. BELL (to 6/30/90) WILLIAM A. OLIVER, JR. (to 6/30/89) RICHARD E. ByrD (to 6/30/89) EDWARD B. PICOU, JR. (to 6/30/89) JOHN L. CISNE (to 6/30/88) JAMES C. SHOWACRE (to 6/30/90) J. THoMAS DuTRO, JR. (to 6/30/90) JAMES E. SORAUF (to 6/30/88) Harry A. LEFFINGWELL (to 6/30/90) HENRY W. THEISEN (to 6/30/89) RoserRT M. LINSLEY (to 6/30/89) RAYMOND VAN HouTtTTE (to 6/30/88) CATHRYN NEWTON (to 6/30/88) WILLIAM P. S. VENTRESS (to 6/30/90) A. D. WARREN, JR. (to 6/30/88) BULLETINS OF AMERICAN PALEONTOLOGY and PALAEONTOGRAPHICA AMERICANA PETERS RoE OO VERS oacd Scere Oh eae ee AR Socablatatr ov ortaretccene sete owe hate EDITOR Reviewers for this issue WILLARD D. HARTMAN and CARL W. STOCK A list of titles in both series, and available numbers and volumes may be had on request. Volumes 1-23 of Bulletins of American Paleontology have been reprinted by Kraus Reprint Corporation, Route 100, Millwood, New York 10546 USA. Volume 1 of Palaeontographica Americana has been reprinted by Johnson Reprint Corporation, 111 Fifth Ave., New York, NY 10003 USA. Subscriptions to Bulletins of American Paleontology may be started at any time, by volume or year. Current price is US $30.00 per volume. Numbers of Paleontographica Americana are priced individually, and are invoiced separately on request. for additional information, write or call: Paleontological Research Institution 1259 Trumansburg Road Ithaca, NY 14850 USA (607) 273-6623 The Paleontological Research Institution acknowledges with special thanks the contributions of the following individuals and institutions PATRONS ($1000 or more at the discretion of the contributor) JAMES E. ALLEN (1967) AMERICAN OIL Company (1976) ATLANTIC RICHFIELD COMPANY (1978) CHRISTINA L. BALK (1970, 1982, 1983) HANs M. Bo ttt (1984) RuTH G. BROWNE (1986) Mr. & Mrs. KENNETH E. CASTER (1967) CHEVRON O1L Company (1978, 1982) EXXON COMPANY (1977 TO DATE) Lots 8. FOGELSANGER (1966) GULF OIL CoRPORATION (1978) MERRILL W. HAAs (1975) ROBERT C. HOERLE (1974-1977) RICHARD I. JOHNSON (1967, 1986) J. M. McDona.p FounbaTION (1972, 1978) Mosit Ort CorPoRATION (1977 TO DATE) SAMUEL T. PEEs (1981) RICHARD E. Petit (1983) RoBeRT A. PoHOwsky (1982) Texaco, INC. (1978, 1982, 1987) UNION OIL OF CALIFORNIA (1982 TO DATE) UNITED STATES STEEL FOUNDATION (1976) CHARLES G. VENTRESS (1983 TO DATE) CHRISTINE C. WAKELEY (1976-1984) NorMaAN E. WEISBoRD (1983) INDUSTRIAL SUBSCRIBERS (1988) (3300 per annum) EXXON PRODUCTION RESEARCH COMPANY Mosit EXPLORATION AND PRODUCING SERVICES SHELL DEVELOPMENT COMPANY (continued overleaf) R. TUCKER ABBOTT JAMES E. ALLEN ELizABETH A. BALCELLS-BALDWIN CHRISTINA L. BALK Bruce M. BELL Rospert A. BLACK Hans BOL. DaviD JOHN BOTTJER RUTH G. BROWNE J. DAviD BUKRY Lyte D. CAMPBELL JOHN L. CARTER ANNELIESE S. CASTER KENNETH E. CASTER JOHN E. DUPONT ARTHUR N. DUSENBURY, JR. R. H. FLOWER Lots S. FOGELSANGER A. EUGENE FRITSCHE ERNesT H. GILMOUR MERRILL W. HAAS ANITA G. HARRIS STEVEN M. HERRICK RosBert C. HOERLE F. D. HOLLAND, JR. RICHARD I. JOHNSON Davip B. JONES PETER JUNG Davip GARRETT KERR CeciL H. KINDLE Mary E. KINDLE WILLIAM F. Kose, II LIFE MEMBERS ($200) Jiri Kriz THORWALD KRUCKOW RALPH L. LANGENHEIM, JR. Harry A. LEFFINGWELL Ecpert G. LEIGH, JR. GERARD A. LENHARD Louie N. MARINCOVICH DONALD R. MOORE SAKAE O’HARA SAMUEL T. PEES RICHARD E. PETIT Rospert A. POHOWSKY JOHN PoseTA, JR. JOHN K. PoPE ANTHONY RESO ARTHUR W. ROCKER WALTER E. SAGE, III JOHN B. SAUNDERS JUDITH SCHIEBOUT MirIAM W. SCHRINER EDWARD S. SLAGLE Davip H. STANSBERY CHARLES G. VENTRESS EmiLy H. VOKES HAROLD E. VOKES CHRISTINE C. WAKELEY THOMAS R. WALLER NORMAN E. WEISBORD RALPH H. WILLOUGHBY ARMOUR C. WINSLOW Victor A. ZULLO Membership dues, subscriptions, and contributions are all important sources of funding, and allow the Paleontological Research Institution to continue its existing programs and services. The P.R.I. publishes two series of respected paleontological monographs, Bulletins of American Paleontology and Palaeontographica Americana, that give authors a relatively inexpensive outlet for the publication of significant longer manuscripts. In addition, it reprints rare but important older works from the pa- leontological literature. The P.R.I. headquarters in Ithaca, New York, houses a collection of inver- tebrate type and figured specimens, among the five largest in North America; an extensive collection of well-documented and curated fossil specimens that can form the basis for significant future pa- leontologic research; and a comprehensive paleontological research library. The P.R.I. wants to grow, so that it can make additional services available to professional paleontologists, and maintain its position as a leader in providing Resources for Paleontologic Research. The Paleontological Research Institution is a non-profit, non-private corporation, and all contri- butions are U.S. income tax deductible. For more information on P.R.I. programs, memberships, or subscriptions to P.R.I. publications, call or write: Peter R. Hoover Director Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850 U.S.A. 607-273-6623 Paleontographica Americana NUMBER 56 JANUARY 8, 1988 Late Ordovician Sponges from the Malongulli Formation of central New South Wales, Australia by J. Keith Rigby and Barry D. Webby Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850 U.S.A. Library of Congress Card Number: 87-62934 Printed in the United States of America Allen Press, Inc. Lawrence, KS 66044 U.S.A. AbStractircmtns att otcaus fetacsnialace ar. Sate ae Introduction . ety Stratigraphic Relationships Geological Setting and Paleoenvironments Paleogeography. . . Paleozoogeography Acknowledgments . Systematic Paleontology Introduction SVDOMY INES teem pe een sir rerteee eke aie Terminology, Measurements, and Abbreviations .. . Class Demospongia Order Lithistida Suborder Rhizomorina Family Haplistidae . Genus Haplistion ..... Genus Warrigalia Genus Taplowia....... 5 Genusiewinia sea ce. sen e Brats Genus Boonderooia .... Suborder Megamorina Family Saccospongiidae ..... Genus Clie/denospongia Suborder Orchocladina Family Anthaspidellidae Genus 4rchaeoscyphia .. Genus Aulocopium. Genus Perissocoelia ... . Genus Hudsonospongia Genus Patellispongia... GenusrPsarodictyumirns. aes ce Genus Amplaspongia .. Genus Malongullospongia Genus Pseudopalmatohindia ... . Genus Dunhillia . . Genus Yarrowigahia ...... Genus Gleesonia . Genus Vandonia CONTENTS Suborder Tricranocladina Family Hindiidae Genus Hindia . Genus Belubulaspongia. . . Genus Palmatohindia Genus 4rborohindia .. Genus Mamelohindia .... Genus Fenestrospongia . Suborder Sphaerocladina Family Astylospongiidae Genus Astylostroma. . Class Hexactinellida Subclass Amphidiscophora Order Reticulosa Superfamily Dictyospongioidea Family Dictyospongiidae Genus Tiddalickia . . Subclass Hexasterophora Order Lyssacinosa Family Pyruspongiidae Genus Wongaspongia ... Order Amphidiscosa Family Pelicaspongiidae Genus Walliospongia .. Genus Liscombispongia . Genus Wareembaia Genus Kalimnospongia. . root tuft . Class Calcarea Order Sphinctozoa Suborder Porata Family Angullongiidae Genus Nibiconia Genus Belubulaia References Cited Platesa.. Index 61 63 65 70 12. 74 76 77 719) 82 84 85 87 89 LIST OF ILLUSTRATIONS Text-figure Page 1. Index map to the Cliefden Caves area of central New South Wales ........... 00. c cece eee eee eect eee terete eens 6 2. Generalized stratigraphic section of Upper Ordovician rocks in the Cliefden Caves area... 2.02.02 eee eee 7 3. Stratigraphic sections across the Malongulli Formation ................0 2 ces e eect ee eee ee tee e eee een nent ante n snes nee ees 8 4. Exposures of the Malongulli Formation, showing limestone breccia deposits at Coppermine Creek, Sugarloaf Creek, and Glee- ea} OE SK Oe) Cea age ca eee or A Roe en GG oo naa en en eo ae OR em mrirat SOs COS AOMar as 008,6.0.0-0 9 5. Camera lucida drawings of spicules of Haplistion regularis, nN. Sp... 2.0... 0-660 -5- Mole ate Bags wise Ox ED eee Sede a eee 17 6s (Camera lucida drawing of spicules.of Warrigalia elliptica; DASPs 2 .c1.< ci. eeaie es oe cies 2s = aiajals Sinise ne cid eim ayes Gin oie o ein oe Sele oles 19 J. Sketch of a heloclone-like spicule from Lewinia complanaia; USP... 2.26.0 c cece cee ee seen beet te see eee seen ween eee 25 SeeSketchesiof.spicules.of Cliefdenosponerailarn ima: te Spey te eiotas era yctejctcsevalets sea ale s10/e/ ae, ofa) c)svore\shoetvel ly @ eiate aneis oy Siete tie stehebohe, ofet eke st teeyare 28 9. Generalized vertical cross-section through Archaeoscyphia minganensis (Billings) ........ 2.2... 55-50-0022 29 10. Spicules and generalized vertical section of Hudsonospongia cf. H. cyclostoma Raymond and Okulitch, 1940 ................5.5 35 lie Sketches; of spicules:of Patellisponeiataustralisg Ws SDs cicestacs nite sq /s.c 2 ols ccrese ote sie/aie epalsel= us =cyeveraselig) saya 6 <8 ote else > Clete yanen iateres tte 38 leeSketchestopdendroclones Of Amp aspongiar Dil ba, TA SPs creyexsie estes oe) te) oxic viays) er ove) sper euss shateahecsial e/a sAterey te te esi ete] chal ayatanater teveteteteheuel 42 13. Sketches of dendroclones and terminations of dendroclone fragments of Amplaspongia magnipora, N. SP... 2... 6-2 es 43 14. Sketches:of spicules of Malongullospongia:delicatula, ND. SPp. 2.0... c ce uc cece ee eee ye Sea stein aise ele ets ce eee ooo nuance 45 15. Diagrams of species of Dunhillia, n. gen., showing growth forms and relationships of canals and ostia and spicules.............. 48 16. Vertical cross-section along the axis of Yarrowigahia brassicata, 0. SP..... 2.000. cence tence eee tee eee n cence nsec eteenees 56 17. Spicules and beam-like development in Gleesonia porosa, n. sp. .......-..- Nera gates GEN aioe a Oe eee 58 [Si ;Sketches oftspiculesiofi:Belubulaspongia) gigantea; ./SPire orai< ares. Department of Geology & Geophysics, University of Sydney, New South Wales 2006, Australia. A total of 34 genera (26 new) are described and subdivided into 44 species (39 being new). The lithistid demosponges are the most diverse, and include rep- resentatives of the following suborders— Rhizomorina, Megamorina, Orchocladina, Tricranocladina and Sphaerocladina. Of these, the Rhizomorina has five genera (four new) subdivided into seven new species, the Megamorina includes one new genus and species, the Orchocladina, 13 genera (eight new) subdivided into 17 species (14 new), the Tricranocladina, six gen- era (five new) subdivided into nine species (eight new), and the Sphaerocladina, one new genus and species. Additionally, there are six new genera (seven new species) of hexactinellid sponges and two silicified members of the sphinctozoans (one a new genus and species). A wide range of isolated spicules also occurs 6 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 SPONGE UNIT 3b SPONGE UNIT 3a “CLIEFDEN CAVES” —_ “KALIMNA” #7 a yp xia DJ : . “- AR? o> ~~ ™ thi s “ de eihias Cig me of et PLATE 10 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 7 eee . rit sf << Figure AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 10 l= Seb erissocoeliainabra vuGWESPECLeS| weer eee eae eee eines pee re ee ae aie Se NaS Holotype (AMu. F66808): Gleesons Creek lower breccia. ike ty Side view shows the upward-expanded, obconical to mushroom-shaped form of the species, with a dense dermal layer of fused spicules over the annulate stalk. Upper part of the sponge has several oscula and subtangential radiating canals characteristic of the species. (x 2). . Basal view shows relationships of the dense, wrinkled, dermal layer to the prominently-radiating skeleton of the endosome. Circular canals are cross-sections of the series that is normal to the mounded summit. ( * 2). . Details of canals and skeletal structures around them. One of the oscular openings shows the almost septate margin of the spongocoel produced by the stacked radial series of canals. The principal skeletal structure is essentially normal to the oscular surface. Horizontal dendroclones connect prominent trabs. (x 8). Broken surface shows vertical section through part of the endosomal skeleton. Prominent rod-like structures are trabs produced by confluent tips of rung-like dendroclones. Radiating canals show in cross-section as stacked series of circular openings. Canals parallel to the trabs show as interruptions in the regular skeletal net. (6). Photomicrograph shows moderately-large dendroclones as horizontal elements in floors of some of the circular canals. Den- droclones have long, smooth shafts and Y-shaped terminations that merge to produce the trabs, which are here essentially at right angles to the upper surface. (22). 6-8. Hudsonospongia cf. H. cyclostoma Raymond and Okulitch, 1940 2.0.0.2. eee sf voyapsie sieiss Hypotype (AMu. F66812): Gleesons Creek lower breccia. 6. Vo Side view of lower part of the conical-cylindrical sponge shows prominent trabs connected by small dendroclones in the upward- and outward-radiating net of the sponge and seen along the margins as essentially cross-sections of the ladder-like series. ( x 2). Basal view shows the radiate nature of the skeleton and the rounded base of the sponge, which lacks a prominent dermal layer. Beginnings of the radiating canals are evident in the upper part of the sponge where the skeleton has been partially broken away. (x 2). Photomicrograph shows skeletal detail of the upper part of the sponge, in which prominent rod-like elements are subvertical trabs connected in ladder-like series by subhorizontal dendroclones. Regularity of the skeleton shows best in the upper right, where the ladder-like series are seen somewhat diagonally. These same elements, where seen at right angles to the trabs, appear like the regular net in the lower part of the photograph. Canals appear as interruptions in regularity of the net. (<6). 107 Page 32 108 Figure PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 1|1 l=9" ‘Hudsonospongia cl... cyclostoma Raymond and Okulitchs 194002 a csc ei ciate ie eieiseeeie cisierrsiaeieiscie aaa teers 1, 6-8. Hypotype (AMu. F66811): Gleesons Creek lower breccia. HE > Side view shows general uniform texture of the exterior and, where broken, the upward-expanding skeletal structure and a narrow moderately-shallow spongocoel, into which open large ostia of excurrent canals. (1.5). . Photomicrograph of upper gastral surface shows irregular sweeping trabs and cross-connecting, rung-like dendroclones, somewhat thickened around excurrent ostia. (<8). . Photomicrograph of the exterior and outer endosome shows impervious dermal layer, which is only locally developed, formed by expanded upper or dermal ends of trabs. In general, details are obscured in heavy silicification. (8). . Diagonal view into the endosomal part of the skeleton shows well-organized trabs, as major elements, cross-connected by long-shafted, smooth dendroclones. A coring oxea shows against the canal in the lower right (arrow). (8). . Hypotype (AMu. F66810): Gleesons Creek upper breccia. Vertical view of summit and into shallow, narrow spongocoel shows coarse, vertical axial canals and somewhat finer radial canals in the moderately-uniform anthaspidellid skeleton. (2). . Basal view shows the general radiate nature of the skeleton and somewhat annulate growth form of the lower part of the sponge with ill-defined dermal layer. (1.5). . Side view of nearly complete specimen capped by a shallow spongocoel. Skeletal structure is typically radiate, pierced by radiating canals. (x 1.5). . Photomicrograph of skeleton with large vertical trabs cross-connected by long, smooth-shafted dendroclones. A single coring oxea (arrow) is exposed in the lower right and in broken trabs in the right center. (* 9). . Enlarged dermal view of the skeletal structure. Elongate elements are dendroclones whose expanded tips produce nodose- appearing porous trabs seen in cross-section. (x 9). PLATE 11 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 Py pW ing eeg, \ Ti Saat PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 12 Figure =e Rerissocoeliaihaprasnew.SPeCleSinm.= seiieieisieisel sai siciocicie citicicroie ieleiccsie se 2 ls cies Hae suse AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 12 1-4. Paratype (AMu. F66809): Gleesons Creek upper breccia bed. ile Rw Vertical view into broken summit shows axial canals in limited oscular development characteristic of the species. A dense dermal layer encloses the canaliculate endosome. (* 2.5). Side view shows obconical form and annulate, stalked, lower part of the sponge encased in a dense dermal layer. (* 2). Basal view shows annulate dense dermal layer. Growth was somewhat irregular. (* 2). Photomicrograph shows impervious dermal layer composed in large part of fused, distal ray tips of endosomal dendro- clones. Small rhizoclones or subtangent dendroclones may be present but obscured. (x 9). 5, 6. Paratype (AMu. F66814): Sugarloaf Creek breccia. 7-9. Patellispongia australis, new species .... 5: Diagonal view from above shows the dense dermal layer around the canaliculate endosomal skeleton, pierced by coarse, subvertical canals at the base of oscular pits. Flared base suggests that the specimen was cemented to a solid substrate. (x9). . Fractured upper surface shows skeletal structure of robust, distinctly Y- or X-shaped dendroclones, particularly well- shown in the upper left by large axial canals. Axial canals are separated by single, ladder-like, series of spicules. Large, straight monaxial fragments are foreign. (x 27). 7. Paratype (AMu. F66817): Gleesons Creek upper breccia. Diagonal view of relatively small specimen shows upward-expanding irregular funnel-shape with a smooth gastral surface around saucer-like spongocoel and moderately-smooth, dermal layer. (* 2). Holotype (AMu. F66815): Coppermine Creek breccia. 8, 9. 8. Side view shows thick walls and rounded oscular rim around the broad, open spongocoel. Irregularly-branching subvertical canals mark the dermal margin. Circular ostia are radial canals that pierce the walls. (x 1.8). . Vertical view into partially matrix-filled, saucer-like spongocoel, shows the thick walls and the smooth gastral surface and characteristic radiate, uniform nature of the skeleton in the upper center. Pestle-like Palmatohindia cylindrica, n. sp. 1s the large sponge that overlies Pate/lispongia. (1.5). 109 Page 110 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 13 Figure 1=9" Patellispongia australis. new: SPECIES: 21.4 eee cies ariel sco Riu Ge bet Shoe CIa SEER CRETE rE Te eee 1, 2. Holotype (AMu. F66815): Coppermine Creek breccia. 1. Side view shows the general form of the species, canaliculate outer surface, thick walls, and smooth gastral surface on the saucer-like spongocoel, which is partially filled with matrix and part of Palmatohindia cylindrica, n. sp. (* 1). 2. Detail of the outer surface of the sponge and the upper oscular rim shows the subtangential canals; open, endosomal net at the oscular rim and the somewhat more dense dermal layer on the ridges between the canals. (5). 3-9. Paratype (AMu. F66816): Coppermine Creek breccia. 3. Longitudinal cross-section shows surface of pinnation near the dense, gastral margin, from which trabs arch upward and outward toward the dermal margin on the right. Numerous oxeas show as coring spicules in the trabs. Coarse, irregularly- oriented fragments are foreign. Radial canals arch upward from ostia at the dermal margin. (* 6). 4. Gastral view shows regular, smooth, moderately dense gastral layer, which is complete in middle and upper parts, and relationship to the upward- and cutward-radiating endosomal skeleton. (x 2.5). 5. Dermal view shows dense dermal layer where some tangential canals are roofed and connect to the exterior through mounded circular pores, or where only partially-roofed, as linear grooves that lead up toward the oscular margin. (x 2.5). 6. Photomicrograph of dense dermal layer which appears minutely nodose, although details are obscured in the silicification. Low rims surround the incurrent ostia. (= 8). 7. Photomicrograph of the dense gastral layer, perforated by small, excurrent ostia. Nodes on the surface are produced by globular expansions of ray tips of subtangential dendroclones, although details of individual spicules are obscured in the silicification. (= 18). 8. Photomicrograph of the upper part of the dermal surface shows partially-roofed subdermal canals characteristic of the species, and the nodular exterior. (8). 9. Photomicrograph shows skeletal details in a transverse section through the wall. Numerous, long oxeas core trabs that are formed by fusion of tips of dendroclones that wrap finger-like around the oxeas. The surface of pinnation extends through the central part of the photograph. The dense gastral layer shows at the left margin. Y- and X-shaped dendroclones are common throughout the skeleton. (* 1). PLATE 13 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 14 CANA, NUMBER 56 PALAEONTOGRAPHICA AMERI Rit sar’ > » Pea r Lbtid p Figure AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 14 l= Gee SALOdICLYHINICrASSUIT SNC WASDCCIESIA ERE eee een ones ae ence ein aeiecinteie nie PROCESO S TCE Holotype (AMu. F66818): Coppermine Creek breccia. Ne 2 General gastral view of the fragmentary holotype shows uniform distribution of excurrent openings where the gastral layer is complete and, where incomplete, the uniform radial nature of the endosomal layer. (x 0.6). . View across a broken surface with the gastral layer in the foreground and various sections of the endosomal layer in middle and upper parts. Irregularity of the gastral layer in the lower part contrasts to well-oriented trabs of the interior of the endosomal net. Upper section is cut normal to the trabs and shows the distinct canals. (* 2). . Photomicrograph of the gastral layer shows characteristic, almost confused appearance of the layer and the irregular, excurrent openings. (8). . Dermal layer shows uniform net with trabs oriented essentially normal to the dermal surface, cross-connected by rung-like dendroclones. Circular incurrent canals are approximately parallel to trabs in this part of the wall. (x8). . Vertical cross-section through the wall longitudinal to the structure. The surface of pinnation occurs toward the base. Strongly- curved trabs meet the dermal surface essentially at right angles. Coring oxeas show as trab continuations in the left center and lower right of the figure. (x 8). . Cross-section through the middle of the wall shows uniform structure, pierced by circular radial canals. Section is vertical- longitudinal, parallel to the dermal surface. Canals in the interior may be lined by ladder-like rows of dendroclones. (* 8). 111 U3 92 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 15 Figure 1-4. Amplaspongia bulba, new species ....... 2.6.0.6 ee ea PO rt era oR arin Arie eri nce 0.6 ob 1, 2. Holotype (AMu. F66819): Coppermine Creek breccia. 1. Vertical view of the mounded gastral surface of the large specimen shows clusters of excurrent canals only moderately developed in an otherwise uniform skeleton pierced by isolated circular canals. (< 0.5). 2. Enlarged view of the summit center shows skeletal structure of trabs, as circular areas, cross-connected by long, thin, dendroclones. Circular canals roughly parallel trabs and are uniformly spread, except where locally clustered as in the lower left and upper right. These clusters are exaggerated by convergence of discontinuous, only irregularly-developed, horizontal excurrent canals. Long monaxons on the surface are foreign. (* 2). 3, 4. Paratype (AMu. F66820): Coppermine Creek breccia. 3. Photomicrograph of vertical section shows continuous, regular trabs and uniform, long-shafted dendroclones; skeleton disrupted by two moderately-large, vertical canals and a few discontinuous horizontal canals, as for example, at the left margin. (x8). 4. Side view shows arched upper surface of the sponge and uniform, upward-radiating structure of the skeleton in a vertical longitudinal section. Principle canals are interruptions parallel to the trabs, but a few circular canals occur throughout the body of the sponge. (1.5). 55,6; Am plasponRia MAGNIPONA, NEWISPECIES:.« sracus sta csavcrssertacastorsiee trie ecerassner evar, a3 0 oke/ neat ehcke ete ralete oeste demede aete eke coevereekcre Rekecses cat aero eee Holotype (AMu. F66821): Coppermine Creek breccia. 5. Vertical view of the summit of the massive sponge shows large, unclustered excurrent canals and lack of spongocoel and differentiated gastral layer. (= 0.8). 6. Enlarged vertical view of the gastral surface shows large excurrent canals separated by thin tracts or ladder-like series of dendroclones, with long smooth shafts, that combine to produce trabs which are oriented normal to the gastral surface. (<7). PLATE 15 MBER 56 J PALAEONTOGRAPHICA AMERICANA, Nt WN tou ; aia we ye “hw \ ia &é 4 ighiee * ae a! as S re : ; PLATE 16 6 JMBER ) PALAEONTOGRAPHICA AMERICANA, NL AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 16 Figure I PAM PIASPORLLGLINGALTIIP OLA GD EWASDECLES amr ery etree erence rte rere are as oes oie eed eL Sore e ere syevs tie ehslis\e oera e]evey sus! sveyait rereyonevernfaverslaieecevers Holotype (AMu. F66821): Coppermine Creek breccia. Side view of the large specimen shows general upward- and outward- radiating nature of the skeleton, pierced by moderately straight canals parallel to the trabs. The sponge has overgrown crinozoan debris (center) and other sponges such as Hindia Duncan, 1879, at the base. (= 0.8). me VialongullospongiaydelicatulanewiSPeCieSimmt- ltteers steele = ceases el esas fe efaieaveiiehelele eyseeyeiel s syelieresers shagaeetey ais Ca) ite tte cuit rate Holotype (AMu. F66822): Gleesons Creek lower breccia. 2. Vertical view onto the dermal surface shows clearly-defined trabs, oriented normal to the surface, cross-connected by long smooth, dendroclones or X- and Y-shaped dendroclones, like those in canals near the center. Major canals are parallel to trabs. (<15). 3. Side view of the exterior of the sponge that is covered on the left by the dermal layer. Endosomal canals immediately below the dermal layer show on the right and near the base. (* 1.25). 4. Vertical cross-section through the massive hemispherical sponge shows the upward-radiating skeletal system interrupted by large canals characteristic of the interior. The two sets of horizontal canals, essentially at right angles to each other, and vertical canals produce a moderately-rectangular canal system in the interior. One set of horizontal canals is seen best near the base of the sponge and the other, at right angles in the middle and upper part of the sponge. (1.5). 5. View down onto the gastral surface shows irregular horizontal canals as vermiform impressions, with circular ostia of the vertical series in ridges between. (1.5). 6. Photomicrograph of skeletal detail shows relationships of trabs to one of the large horizontal canals. X- and Y-shaped den- droclones are essentially horizontal and combine with coring oxeas to produce the sweeping, delicate, skeleton. (x 13). 7. Photomicrograph of vertical section through the middle of the sponge shows curved tracts of trabs and ladder-like series of dendroclones, bending around some canals but interrupted and discontinuous at other canals. Variations in shapes of dendro- clones show well in canal margins where spicule tips are isolated. Principal canals are at right angles to the cut surface (7). 13 114 Figure lineMalongullospongiaidelicatulamewiSpecieS te eee eee eee ee eee eee Holotype (AMu. F66822): Gleesons Creek lower breccia. Photomicrograph shows interruptions of delicate skeleton by horizontal canals, essentially in the plane of the section. Somewhat smaller vertical canals show near the center of the figure (x 7). to PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 17 - ePseudopalmatohindiaidigitatay NewiSPeCies: eee eee eee eee eee eerie en ee Holotype (AMu. F66823): Coppermine Creek breccia. 5 Nodular and irregular thickened dermal layer of expanded dendroclones and trabs, which are oriented normal to the surface. Dermal layer pierced by small, incurrent canals. (x 9). . Irregular, broken surface of digitate sponge shows large excurrent canals along axes of digitations or palmate blades; canals subparallel to trabs in the interior of the specimen. Direction of growth, indicated by expansion of the skeletal structure, is toward the bottom. Fragments of other anthaspidellids and spherical specimens of Hindia Duncan, 1879, occur above the dark, V-shaped matrix fill between digitations. (1). . Reverse of figure 3 shows large, excurrent canals essentially in the middle of the palmate blade along the left and upper surface. A long wrinkle or digitation extends diagonally down toward the lower right between several spherical Hindia, and shows characteristic pores of the dermal layer, right of the large Hindia at the base. The specimen is in growth position. (* 1). . Detail of a tip of one digitation shows the relatively-massive, nodose layer, right center, and the more open endosomal part of the skeleton. Large excurrent canals are in the center along the axis of the digitation. These openings are the same as those on the upper right in figure 4. (3.5). . Three layers of the skeleton are differentiated. The lower, knobby dermal layer contrasts to the more open-textured endosomal layer. A somewhat thickened gastral layer forms the wall of the excurrent canal in the upper left. The dendroclone-based nature of the skeleton is evident in the endosomal layer. (9). . Vertical section shows upward- and outward-radiating net of thin dendroclones and delicate trabs. These elements are thickened in the gastral layer around an excurrent canal, the vertical tube. (8). PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 17 PLATE 18 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 Figure AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 18 PES CUAOPAIMALONINGIQIAISILALA INC WASDCCIOSH RA ee eer elects oie ee ea aerate ener ane eee seeiels ayieeioeaies cueterneseronanriererane Holotype (AMu. F66823): Coppermine Creek breccia. Photomicrograph shows relatively rouusis smooth-shafted, dendroclone tips that unite to form the trabs, which are arched upward and outward. (x 16). Aa 2, JOT ATA EE WEP elstOley a gaencasecconssuecoubsoosoons oon Dh er aantna Hehe ea ceeteg tose Merits rT ter or ree Siete nyevnys os We 3516; ty 2), Os Paratype (AMu. F66829): Coppermine Creek breccia. 2. Side view shows general tubular growth form of the species. Low, mound-like short exaulos-like tubes or labropores are incurrent ostia. Uniform skeleton shows where broken in the upper part. Horizontal ring-like shelves of rhi- zoclones show on the gastral surface. (= 4). 11. Photomicrograph of dermal surface shows characteristic small, tile-like, rhizoclones as dermal elements and forming rims around incurrent ostia. Dermal rhizoclones may extend across two rows of ragged endosomal dendroclone tips, which form more narrowly-spaced, less regularly-spaced elements in the lower part of the figure. ( = 20). Paratype (AMu. F66828): Coppermine Creek breccia. Juvenile tip and conico-cylindrical early stage of the species. (* 3). Paratype (AMu. F66832): Coppermine Creek breccia. Exaggerated annulate growth with dense dermal spiculation and only a few isolated incurrent ostia. (6). Paratype (AMu. F66826): Coppermine Creek breccia. 5. Enlarged view shows the very regular nature of the skeleton and stacked, tile-like rhizoclones that form the dermal layer, in vertical rows opposite tips of radially-arranged dendroclones. Stacks of spicules show spacing of trabs in the interior beneath the dermal layer. Dermal layer pierced by low, exaulos-like, incurrent canals. ( 8). 6. Transverse cross-section shows relationship of incurrent canal to exaulos-like outer ostia and the horizontal, con- centric midwall canal. Thin dermal layer clearly defined. Trabs, seen in cross-section, are normal to the transverse section. (= 10). Holotype (AMu. F66824): Coppermine Creek breccia. 7. Diagonal view of the nearly-complete upper end of the sponge, shows rounded oscular margin and prominent shelf- like ring of a cluster of rhizoclones on the gastral surface of the spongocoel. (x 8). 9. Side view shows small, rectangular, tile-like rhizoclones of the dermal layer in vertical rows, deflected up onto the margins of the low, exaulos-like incurrent canals. Ends of expanded endosomal dendroclones show as ragged surfaces below the rhizoclone dermal spicules. (15). 10. Side view shows conico-cylindrical form of the species with somewhat distant incurrent canals with low labropore rims or short exaulos-like tubes. Weakly-annulate and vertically-striate appearance is typical. (x 4). . Paratype (AMu. F66831): Side view shows somewhat exaggerated annulate growth form with exaulos-like tubes as incurrent openings. (4). Paratype (AMu. F66825): Coppermine Creek breccia. Vertical section shows regularity of the endosomal skeleton of parallel trabs with rung-like dendroclones moderately widely-spaced. Shelf-like rings on the gastral surface are made largely of rhizoclones. Cross-sections of horizontal ring canals interrupt the skeleton at midwall, in the middle part of the section. (* 15). Wis) 48 116 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 19 Figure 1=32 Dunhilliaitubulaxnew SpecieSts..4c seme cee a ee ee ee REALE ERE eee EE ee eee 1. Paratype (AMu. F66833): Coppermine Creek breccia. Vertical view of natural cross-section shows general proportions of the wall, spongocoel, and relationships of the horizontal midwall canal to other canals and incurrent ostia, like that at the right. (12). . Paratype (AMu. F66825): Coppermine Creek breccia. Broken cross-section shows the horizontal shelf-like rings on the gastral surface and well-organized skeleton of the endosomal layer. (4). 3. Paratype (AMu. F66827): Coppermine Creek breccia. View shows relationships of incurrent tubes that pierce the dense dermal layer and the subcylindrical annular growth characteristic of the species. (6). 4214. Dunhilliavapertura) New SPeCleSun acme cece ee eek eee ee CO OnE eR ere RE eae eRe ee ne 4-7. Paratype (AMu. F66837): Coppermine Creek breccia. 4. Side view of small twig-like specimen shows moderately-dense dermal layer pierced by widely-spaced ostia, in the lower part, and more closely-spaced ostia above. (* 4). 5. Enlarged view of the upper part of the paratype shows uniform endosomal skeleton and vertically-elongate incurrent ostia. Dermal rhizoclones show best in the lower part where they overlap rows of expanded radial ends of endosomal dendroclones. (= 10). 6. Natural horizontal section of upper end of paratype shows incurrent canals and only moderately-defined, horizontal midwall canal, and cross-sections of nearly vertical trabs cross-connected by horizontal dendroclones. Gastral layer ill-defined but dermal layer prominent. (x 12). 7. Natural diagonal, near basal, cross-section shows relationships of horizontal midwall canal to incurrent ostium and well-defined gastral and dermal layers. (= 12). 8, 9. Holotype (AMu. F66836): Coppermine Creek breccia. 8. Enlarged side view of the upper end showing well-organized vertical rows of expanded ends of radially-oriented dendroclones, overlain in middle and upper parts by irregular and tile-like elliptical to rectangular rhizoclones. Elongate ostia are of two canals, stacked vertically, that connect to different interior ring canals. (= 15). 9. Side view shows general distribution of the ostia and general uniform structure of the sponge. (x8). 10. Paratype (AMu. F66844): Coppermine Creek breccia. Side view shows paired incurrent canals are clearly separated immediately beneath the dense dermal layer. (x 4). 11. Paratype (AMu. F66838): Coppermine Creek breccia. Side view shows distortion of regularity of rhizoclone dermal spicules around ostia with a low rim. Rhizoclones are rectangular elements that overlap beyond the expanded dendroclone ends, best shown near the upper ostium, and below that and to the right. (x 20). 12, 13. Paratype (AMu. F66840): Coppermine Creek breccia. 12. Side view shows relatively rare ostium with four incurrent canals leading from one rimmed opening. Dermal layer dense but obscure. (x8). 13. Natural cross-section shows relationship of the incurrent openings to somewhat discontinuous midwall canal, dense dermal layer and uniform endosomal skeleton. (x 10). 14. Paratype (AMu. F66842): Coppermine Creek breccia. Vertical cross-section shows organization of the endosomal layer, the gastral surface on the tubular spongocoel and lack of horizontal shelf-like rings typical of Dunhillia tubula, but with excurrent ostia arranged in horizontal layers. (15). i) PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 19 PN pcg & ee pet Ji a oe Bee: PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 20 AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 20 Figure Aen O ll OM DUMALIAICHIDLAld sNews SDCGCICS ey eae ately eliotn ce Tee eee cesses bere eee 1-3. Holotype (AMu. F66845): Coppermine Creek breccia. 1, 2. Side views show general subcylindrical form with dense dermal wall interrupted by irregular to elliptical pore fields of clustered incurrent canals. Broken upper end shows characteristic uniform endosomal skeleton and part of central cylindrical spongocoel. Dermal layer somewhat obscured. (x 4). 3. Enlarged view shows relationships of pore fields to curved outer endosomal trabs that diverge around the lower edge but begin abruptly at the upper edge of the pore field. Massive rhizoclone-based skeleton of the pore field shows between circular to elliptical ostia. (x 10). 4. Paratype (AMu. F66848): Coppermine Creek breccia. View shows unrimmed diagonal pore field. Regular endosomal trabs show well below the pore field. (x 10). 8. Paratype (AMu. F66849): Coppermine Creek breccia. Natural horizontal section shows relationships of the canal system, grouped incurrent canals, and a moderately continuous, though narrow, horizontal midwall canal as it interrupts the regular endosomal net. Expanded dermal spicules show around the periphery. (* 10). 9, 10. Paratype (AMu. F66846): Coppermine Creek breccia. 9. Natural vertical section shows the fairly-regular endosomal skeleton and gastral layer of the tubular spongocoel where rhizoclones occur but do not form prominent horizontal shelf-like rings. (* 16). 10. Skeleton composed of irregular small rhizoclones in matted felt-like layer shows between incurrent ostia. Reg- ularity of the endosomal layer shows below. ( 20). Se GtnULILiPOrata MUCW.SDCCIES mt eer TTT eo re elec eae ear RE eee dein ace Holotype (AMu. F66850): Coppermine Creek breccia. 5. Diagonal view of upper surface shows vertical and horizontal cross-sections of regular endosomal wall. Dense dermal layer pierced by numerous ostia shows in the lower right. Comparatively large excurrent ostia of the gastral layer show around the tubular central spongocoel. (= 10). 6. Horizontal section of broken base shows well-organized endosomal layer perforated by incurrent canals, with small canals that interconnect with the moderately well-defined, though small, midwall horizontal canal. Connections to excurrent canals are less obvious. The dense dermal layer contrasts sharply to the ill-defined gastral layer of the cylindrical spongocoel. ( 10). 7. Side view of the lower part of the holotype shows numerous, fairly uniformly-distributed, incurrent ostia, each with a minor rim and each perforating the well-organized but somewhat curvilinear trab- and dendroclone-based skeleton. (x10). 117 118 Figure PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 21 1=65 (Dunhilliaimultiporata NeweSPeCieS. 2 ajac cei crocs «+ cies = eee stors ols sistas shel als s'aseratal «pelt fete otetnveneyer ofeiedeyefepet veri ser: ore tetoneloh stele conc ren tae 1. Holotype (AMu. F66850): Coppermine Creek breccia. Side view shows general growth form and proportions of canals, skeletal structures and the spongocoel. (6). . Paratype (AMu. F66856): Coppermine Creek breccia. 2. Side view shows well-preserved dermal layer with vertical rows of rectangular rhizoclones and outer ends of dendroclones interrupted by widely-spaced incurrent ostia. (10). 3. Side view of the complete fragment. (4). 5. Diagonal natural cross-section of lower end shows principal radial structure of both incurrent and excurrent canals. The horizontal midwall canal shows in partial section in the lower part of the broken surface. Numerous ostia pierce the lower dermal layer and moderately large excurrent ostia show in the spongocoel wall. (10). . Paratype (AMu. F66852): Coppermine Creek breccia. Natural horizontal section shows general skeletal structure of the endosome, ill-defined gastral layer and more prominent dermal layer, the latter interrupted by numerous circular incurrent canals in the outer part of the sponge. Horizontal midwall canal ill-defined. (15). . Paratype (AMu. F66853): Weathered end shows prominent vertical trabs and horizontal rung-like dendroclones in well- organized skeleton around tubular spongocoel. (18). 1=9:, Yarrowieahia brassicatasMew/SPeCleS} s. a. « Malongullospongia’ delicatula; mewsSpeCieS. 3. 442-1 ee rs = easel ies derail ei eee tee ee ee ee eee eee Holotype (AMu. F66822): Gleesons Creek lower breccia. 1. SEM photomicrograph shows trabs cored largely by subparallel, moderately loosely packed oxeas; trabs cross-connected by smooth, long-shafted, Y- and X-shaped dendroclones whose branching tips grasp oxeas. (50). 2. SEM photomicrograph shows articulation of dendroclone cladome ray tips with coring oxeas of the trabs and the porous trabs of the silicified specimen. (= 200). . Amplasponsias bulba,, DEW SPECIES so. 5 sox ace setens cle Srerasialaro sed eo epee pate PS TIS vals eaves nite Shey Ventage ces Mata vone pe Pevev one Ricie me Maree yee I ete ene 3. Paratype (AMu. F66820): Coppermine Creek breccia. Long, smooth-shafted dendroclones with short cladomes articulate with adjacent spicules to mutually produce trabs that are the vertical porous elements. (50). 4. Holotype (AMu. F66819): Coppermine Creek breccia. SEM photomicrograph shows characteristic long, smooth axes of den- droclones whose mutually-interdigitating ray tips produce the trabs. (x 50). . Perissocoelia habra, new species ........ MA TON head esrecoe Boe saitsles Pon cP REO AUSF SR AOR Oe PET EET OORT To Holotype (AMu. F66808): Gleesons Creek lower breccia. 5. SEM photomicrograph shows somewhat irregular orientation of spicules and bifid, finger-like, cladome termination of one of the long, smooth-shafted dendroclones typical of the species. (= 100). 6. Fused ray tips of numerous dendroclones. The most distinctly-outlined dendroclone is in the lower right where the long shaft is subvertical and the prominent Y-shaped cladome surrounds the small pore. The less-expanded brachyome tip is lost in fusion with adjacent spicules. The open texture and canals are characteristic of the form. (x 100). ; \GleesOnia POrOSA: NEW SPECIES: » veces oe er oases Shei Seo ee Ee SCLIN eae. eee enor OUT OLSEN PO Re SEATS ee VANS eae Paratype (AMu. F66859): Gleesons Creek lower breccia. SEM photomicrograph shows dendroclones, in the center, with arborescent complexly-branching ray tips that combine to produce the prominent webs, at the bottom and top of the photograph. These are among the coarsest spicules known in the family. (x 25). 25 PLATE 6 MBER 5 J Nt PALAEONTOGRAPHICA AMERICANA, PLATE 26 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 26 Figure I-10 wiindra sphaeroidalis Duncan, 1879)... ......---.- s02-22 22 eesees - ae es Be EUROS Bie eins 1, 2. Hypotype (AMu. F66866): Coppermine Creek breccia. 1. Moderate-sized specimen shows spherical growth form and distribution of two sizes of canals in the skeleton. ( 3). 2. Photomicrograph of the surface shows skeleton of tricranoclads with prominent centrum from which radiate three equally-spaced rays. Stacks of spicules surround coarse, presumably excurrent, and fine, presumably incurrent, canals. Distal surface of centrum and of radiating rays sculptured with low nodes. (= 15). 3-5. Hypotype (AMu. F66867): Sugarloaf Creek breccia. 3. Moderately large spherical sponge with well-developed canals. (2.5). 4. Broken radial surface shows stacked canals and tricranoclones in which each centrum is oriented distally (upward) and the three cladome rays oriented proximally (downward). Tight fusion of cladome tips to those of proximally-adjacent centra obscures the character of individual spicules. Coarse canals are excurrent and outlined by several spicule series. Smaller canals are surrounded by fewer spicules and are considered as incurrent. (x 15). 5. Tangential view of sponge exterior shows spicule series around various canals. (8). 6. Hypotype (AMu. F66864): Coppermine Creek breccia. Moderate-sized spherical specimen shows characteristic construction of the skeleton pierced by uniformly-spaced radial canals. (* 4). 7. Hypotype (AMu. F66863): Coppermine Creek breccia. Small specimen shows small canals of two sizes still retaining proportions similar to those of larger specimens. (= 4). 8. Hypotype (AMu. F66862): Coppermine Creek breccia. Tiny specimen shows characteristic skeletal structure and canals. (<4). 9. Hypotype (AMu. F66865): Gleesons Creek upper breccia. Spherical sponge with adhering spiculitic matrix. Natural fracture through the center provides a cross-section, illustrated in Plate 27, figure 3. (= 2.5). 10. Hypotype (AMu. F66868): Sugarloaf Creek breccia. Largest specimen of species recovered retains relative proportions of small to large canals. Fracture at the lower left shows typical radiating skeletal structure. (= 1.5). 124 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 27 Figure l=35-Hindiasphaeroidalis:Muncan a8) Ore ete hee oe ee IRE Toso re ee 1, 2. Hypotype (AMu. F66867): Sugarloaf Creek breccia. 1. Photomicrograph of the center of the sponge shows resorbed (?) opening at the center into which project the inner tips of oxeas that occur in some of the radial canals. Sizes of spicules and canals increase away from the center. (x 7). 2. Natural cross-section shows open, possibly resorbed, central area and the radiating canal and skeletal structure characteristic of the species. ( x 3). 3. Hypotype (AMu. F66865): Gleesons Creek upper breccia. Prominent oxea tips converge into the probably-resorbed central opening in the center of the sponge, which shows characteristic radiating skeletal and canal structure. Matrix adheres to the lower right surface. (= 5). 4-6", Belubulaspongiaysiganteasnew Species) ene eee eee Eee eee eee eee eee ene 4, 5. Holotype (AMu. F66869): Coppermine Creek breccia. 4. Photomicrograph of dermal surface shows undifferentiated tricranoclones around uniform ostia of the incurrent canal system. Brachyomes are oriented distally and the three cladomes extend proximally into the skeleton. (8). 5. Side view shows annulate conico-cylindrical to obconical growth form of the species with a complete rounded oscular margin but broken base. Arrow points to top. (* 2). 6. Paratype (AMu. F66870): Coppermine Creek breccia. Moderately coarse, nodose distal surfaces show on the tricranoclone spicules. (= 18). PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 27 Une Fig tA oC. A oie Bs 3 Fd 4 PLATE 28 56 NUMBER ALAEONTOGRAPHICA AMERICANA P Eee i SN Seba xe < AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 28 Figure Sem ClUDULASPONLIA CISANted eWhSDECIES) Winans Neca ieee icine ciel ccs eee Sle oes s = SER te yaa ysuaysoare aye aasterenode evens suas Messnalces 1. Paratype (AMu. F66870): Coppermine Creek breccia. Basal view shows obconical form, simple cylindrical spongocoel, uniform endosomal skeletal structure and moderately-thickened spicules of the dermal layer. (x 2.5). 2, 3. Holotype (AMu. F66869): Coppermine Creek breccia. 2. SEM photomicrograph shows dense silicification of tricranoclones around uniformly-circular incurrent ostia. ( = 40). 3. Diagonal view of weathered silicified surface shows sculptured distal surfaces of tricranoclone cladome rays and bifid or otherwise sculptured brachyomes that project distally (arrows). (18). AS SMeLalInaloninalaimultlpOra ene WaSPeCleSn ry scee ene ito rae eee Ge eels mice eaverc iene a) erate) ee ae aloceietane ee eel eveieyers ersteevieas neice 4-6. Holotype (AMu. F66871): Coppermine Creek breccia. 4. Broken horizontal surface shows clustered large excurrent canals in the irregularly-bladed sponge with moderately-uniform, although locally-aligned, endosomal skeletal structure and the thin dermal layer. Some excurrent canals have a thin gastral layer. Outer part of the endosome is perforated by skeletal pores between clads of stacked spicules. (* 2). 5. Vertical section through an irregular part of the holotype shows parallel, large excurrent canals. The dense dermal layer shows in the curved surface on the right. (0.75). 6. Broken horizontal section shows the cluster of major canals in the interior of the wrinkled blade, the uniform endosomal skeleton and the dense dermal net. The spherical sponge at the bottom is a specimen of Hindia Duncan, 1879. (0.5). 7. Paratype (AMu. F66873): Coppermine Creek breccia. Horizontal section of a relatively small specimen shows the undulating bladed structure of the sponge, and size and spacing of excurrent canals along the midline. Largest openings are those which initiated lowest in the sponge. ( * 3). 8. Paratype (AMu. F66874): Sugarloaf Creek breccia. Side view shows obconical form of some immature sponges, with the uniform ostia of incurrent canals along the margin and aligned large excurrent canals on the summit. (= 3). Figure PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 29 1=72 Palmatohindia:multipora new; SpeGleS). «ooh ke oe Oe eke ee eee sks OO ot ee Eee ee rae Paratype (AMu. F66875): Coppermine Creek breccia. View of serpentine- obladed: sponge, from above, shows rows of coarse excurrent openings, moderately-uniform endosomal skeleton and smooth dermal layer perforated by small incurrent ostia. (< 0.5). bo IG > By . Holotype (AMu. F66871): Coppermine Creek breccia. Diagonal side view shows characteristic uniform ostia in the dense dermal layer and the upward-pinnate spicule stacks in cross-section at the right. Two large excurrent openings are subparallel to the broken surface. Arched stacks of tri- cranoclones show particularly well along the left side of the transverse section. (* 2). View from above shows the coarse excurrent openings at midblade, in the lower part, and the dense dermal layer in the upper right. Small canals between stacked series of tricranoclones characterize the endosome, except in the outermost part where incurrent canals are better differentiated. (x 8). Vertical longitudinal section shows upward-radiating structure of stacked tricranoclones and the fairly well-organized canals parallel to those tracts inside the dense dermal layer. Large openings in the upper right are excurrent canals in the middle of the blade. (5). Vertical longitudinal section shows relationships of coarse exterior excurrent canals to the skeletal structure and the moderately smooth, dermal layer. The rounded summit is unbroken. Paratype (AMu. F66874): Sugarloaf Creek breccia. Numerous circular excurrent openings in a complete rounded summit of an obconical sponge, an early stage of growth. (* 3). . Paratype (AMu. F66876): Coppermine Creek breccia. Subprismatic to circular incurrent ostia outlined by rays of tangential tricranoclones form small “‘calicles’’ characteristic of the genus. (= 10). PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 29 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 30 AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 30 Figure A Sea lnalOnindlainullipOLa ele waSDeEClCS Waa eee EERE ere er eee eer ee een eee eee 1, 2. Paratype (AMu. F66877): Coppermine Creek breccia. 1. Horizontal transverse section shows central excurrent canals and porous endosomal part of the skeleton made of tricranoclones, with sculptured distal surfaces on cladome rays and moderately well-defined brachyomes, in the lower right. (= 10). 2. Photomicrograph shows tricranoclones with sculptured distal surfaces of cladome rays and nodose or swollen central brachyomes, in profile along the upper central part and from above in the right center. (x 20). 3, 4. Holotype (AMu. F66871): Coppermine Creek breccia. 3. SEM photomicrograph, distal view of most of two tricranoclones with sculptured distal surfaces and swollen nodes at centers of ray junctions, perhaps best seen in the spicule on the right. Articulation of tips of cladomes perhaps shows best in the lower center, near the foreign oxea fragment. (= 100). 4. SEM photomicrograph shows moderately-uniform ostia between diverging tricranoclone rays. Distal surface of spicule in the lower center shows weak sculpture. Most others merely show complex fusion of spicules. (75). 8. Paratype (AMu. F66873): Coppermine Creek breccia. Gastral surface of a large excurrent opening shows moderately- uniform spacing of tricranoclones and lack of differentiated incurrent canals through the endosome into the large tube. (8). PEO ML AUNALORINGIAl (2) sf AV OSU ADC WASDCCICS er peet rte Pera Vay ister vavele el everetarernics cc veiavevereralste assur ehchersiseatarcvaicielavetel sha stiensvsistatonecs/aiescrs 5-7. Paratype (AMu. F66880): Coppermine Creek breccia. 5. Diagonally-broken upper surface shows uniform texture of the tricranoclone-based skeleton and the ill-defined gastral layer. Uniform dermal layer shows in cross-section on the left, where perforated diaphrams at the base of the ‘‘calicles” are the uniform thin layers inside row of canals at the dermal margin. (* 10). 6. Transverse basal section shows uniform skeletal pattern and cylindrical spongocoel which extends throughout the sponge. (8). 7. Side view shows weakly annulate subcylindrical form with dermal surface marked by prominent subprismatic “calicles” of the dermal layer. ( = 5). 9, 10. Holotype (AMu. F66879): Coppermine Creek breccia. 9. Photomicrograph of dermal layer shows prismatic “‘calicles’’ outlined by vertically-elongate cladome rays of dermal spicules. Base of the “‘calicles” defined by a prominent level of thin perforate diaphrams, particularly well-shown in lower right center and lower left. (15). 10. Horizontal transverse section of obconical sponge with cylindrical spongocoel, uniform endosomal layer and moderately-dense dermal layer. (x 5). 128 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 31 Figure Ie Palmatohindial(?):favOsa ANCWsSPCCIES' 755 4. Photomicrograph shows moderate preservation with light dots as centra of the uniformly-spaced sphaeroclones. Larger spicules over the surface and at the base are foreign. (8). . Photomicrograph of skeletal structure of an area in the upper right of figure 2 shows rosette-appearing nodular centra of sphaeroclones and delicate radiating rays that interconnect from centrum to centrum. (25). General view of the massive arched-laminar structure of the sponge. Figures 2 and 3 are enlargements of areas in the lower center near the prominent thin fractures. (2). 5-7. Paratype (AMu. F66894): Sugarloaf Creek breccia. D: 6. Photomicrograph of uniform delicate skeleton, with sphaeroclone centra cross-connected by radiating rays and interrupted only locally by small canals. (= 30). SEM photomicrograph shows relatively porous delicate skeleton made of tiny sphaeroclones. Triangular skeletal pores occur between the rays that fuse the structure together. Small canals are circular interruptions in the uniform net. (= 100). . SEM photomicrograph of moderately well-silicified area, with a nodose centrum near the center of the photograph and radiating rays of that and adjacent spicules united with other centra to produce the characteristic skeleton. These are the smallest sphaeroclones known in Early Paleozoic sponges and their uniform size throughout the skeleton is distinctive. (< 200). 133 Page 76 134 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 37 Figure 1,2 Tiddalicktatquadrata: Mew: SPCClES =... cernnie olaele tats cease eeate eo) sy eioies ood ane tah bole) Syegelane keener nen Earners Reet eee reteset tee ee Holotype (AMu. F66895): Sugarloaf Creek breccia. 1. Photomicrograph of the rectangularly-arranged skeleton made of straps of monaxons or rhabdodiactines but with large stauracts at ray junctions. Dermal layer, if present, is made of irregular hexactines in intervening quadrules. (* 8). 2. Most of the specimen shows relationships of spicule straps to irregular spiculitic matrix into which the straps are impressed. (<2). 3=75 Wongaspongia minor New: SPCGleS=..m sesicca ne clere tere en ae etree ere eee eee eres aes ee ee ot tt 3, 4. Paratype (AMu. F66898): Sugarloaf Creek breccia. 3. Dermal view shows large incurrent ostia with canals that extend approximately to midwall but end blindly there. Skeleton of irregular tracts of normal, irregularly-oriented hexactines. Spicules line foreign burrow in the upper part. (* 3). 4. Photomicrograph shows detail of dermal layer and outer part of the endosome made of relatively-robust normal hexactines that are irregularly-oriented. (= 10). 5-7. Holotype (AMu. F66897): Sugarloaf Creek breccia. 5. Side view of collapsed thin wall of cup-shaped sponge with smooth exterior perforated by incurrent canals that extend to midwall. Gastral surface shows in the upper right beyond the broken margin, above the small specimen of Hindia Duncan, 1879. Excurrent openings slightly larger than incurrent openings in diplorhysal canal systems. Base and oscular margin not preserved. (* 1.5). 6. Photomicrograph of the central part of the exterior shows irregular orientation of coarse normal hexactines, in the smooth dermal and outer endosomal part of the skeleton, associated with somewhat finer spicules in the interior of the smooth wall. (<8). 7. View from above of the collapsed thin-walled sponge. Dark groove represents the spongocoel. Radial incurrent and excurrent canals show near the gastral and dermal margin. Prominent round vertical midwall canals show near the center of the sponge wall. Spherical specimen of Hindia is in matrix attached to the smooth exterior of the sponge. (1). PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 37 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 38 AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 38 Figure 1-5. Wongaspongia minor, new species .. . Te Se eA CR RE ee 1-3. Paratype (AMu. F66898): Sugarloaf Creek breccia. 1. Interior or gastral view shows large excurrent canals and intervening spicule tracts. Excurrent canals end at midwall or slightly beyond in diplorhysal system. Skeletal texture is somewhat finer than on the exterior. ( 3). 2. Photomicrograph of the gastral margin shows a mixture of fine and coarse spicules of the gastral and inner part of the endosomal skeleton. (x 10). 3. Spicules line foreign burrow, on the exterior, defined by small spicules in contrast to the moderately-coarse hexactines of the dermal layer. (= 10) 4, 5. Holotype (AMu. F66897) is in the upper half, and Wongaspongia major n. sp., holotype (AMu. F66899) and paratype (AMu. F66900), are the fragments in the lower half of the figures. Sugarloaf Creek breccia. 4. Diagonal view of fragments of Wongaspongia major with relatively coarse canals, below, in comparison to fine canals in Wongaspongia minor, at the top. (0.75). 5. View along walls of both species of Wongaspongia, n. gen., showing contrast in canal sizes and wall thicknesses of the two species. Of the two fragments of Wongaspongia major in the lower half of the figure, the paratype (AMu. F66900) is above and to the left of the holotype (AMu. F66899). (x1). 4-7. Wongaspongia major, new species ....................- RPS SS eG AAG Lae HOP ACCS Read NM TERS R TS pena et cs Set ease aN s ol cig Pete ee Tec ae STS 4, 5. [see above] 6, 7. Paratype (AMu. F66901): Coppermine Creek breccia. 6. General view of the gastral interior shows large excurrent openings and fine, irregularly-oriented spicules of the gastral and inner part of the endosomal layer. ( = 2). 7. Photomicrograph shows delicate, irregularly-oriented spicules on the gastral surface and weak bundling of some of the spicules with long rays. (8). 1135 Page 80 136 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 39 Figure 1-6. sivas a PIAJOTTICW: SPECIES: < oycae arc eve oie Citic ce etter ers ys eee toe ev eveerete eletan tere otspege elope = easiest en . Paratype (AMu. F66902): rperiear Creek breccia. Tufts of major ececarele! long-rayed spicules are characteristic of the inner endosomal and gastral layer, producing an unusual, perhaps, pathologic node. (4). 2-5. Holotee (AMu. F66899): Sugarloaf Creek breccia. 2. Enlarged part of the dermal layer shows coarse incurrent canals and irregularly-oriented normal hexactines of the dermal layer. Lower part obscured in matrix. (* 4). 3. Photomicrograph shows characteristic, coarse, smooth-rayed hexactines of the dermal layer in irregular orientation on the moderately-smooth wall. (10). 4. Enlarged gastral surface shows large excurrent canals and finer spicules of the gastral and inner endosomal layer. Excurrent canals, like incurrent ones, end blindly in a diplorhysal system where tips of incurrent openings occur in spaces between excurrent ones, with communication laterally through the open net. (* 4). 5. Photomicrograph of the gastral surface shows the relatively-delicate spicules and their irregular orientation. (* 10). 6. Paratype (AMu. F66901): Coppermine Creek breccia. Fragment shows coarse canals and intervening relatively-coarse irregular spicules of the dermal net (* 2). 1. Walliospongia: gracilis New SPCCieS) «wc 3. co spe sois = Sis oc or are 9 aye trsarane et egos opacais of es segs euelcge eneteye saejener sued spite feds) 4) felt tos 4. 5. Holotype (AMu. F66905): Coppermine Gree breccia. Enlarged side view shows outer fused dictyonine-like dermal layer and the less regularly-oriented principal endosomal net made of long-rayed hexactines. (* 7). Paratype (AMu. F66908): Coppermine Creek breccia. Interior dermal layer shows the transition from the well-organized, vertically-oriented, outer part of the dermal dictyonine net into the thin, irregular subdermal layer, with attached bits of the principal endosomal net of more regular hexactines inside the somewhat vesiculate-appearing subdermal layer. Buttress-like structures form margins of canals that are subvertical and radiate upward and outward from the endosomal layer. (= 10). Paratype (AMu. F66911): Coppermine Creek breccia. Interior of the dictyonine dermal layer shows prominent vertical ori- entation of major rays in the fused structure, with a web-like transition zone. Annulate appearance is related to shingled layering of spicules in the outer endosome and inner dermal layer. Shingling produces lateral connections between vertical canals near the exterior of the sponge. ( * 10). . Paratype (AMu. F66907): Coppermine Creek breccia. Fragments show the strong alignment of major vertical rays in the inner part of the dermal layer, in the lower part, and the irregular, vesicular-appearing transition layer, in the upper part of the fragment. (= 10). PLATE 41 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 COR Re ee vote PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 42 L'a SMILE SSAA S S Se a) : AS: Rib \ ¥: ¢c \ (ie AIX / ENA if Figure AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 42 1=6: Wareembaia concentrica, new species ........22..... 0220000 ee sss i 5 oa Enns A Re eae GAs 1-5. Holotype (AMu. F66905): Coppermine Creek breccia. ne tu Side view shows somewhat-flared base and upper cylindrical form, made of alternating laminae of open- and closed- texture in an upward- and outward-expanding growth pattern. Fragment of a dense dermal net shows above the flared base and contrasts with the more open endosomal layer. (x 2.5). . Side view shows relationship of the upward- and outward-flaring dermal layer to the canal system that surrounds the endosomal part of the skeleton. The organized dictyonine dermal layer contrasts sharply to irregular spicules in the endosomal net. ( x 2.5). . Vertical view down into the central spongocoel. The endosomal skeleton is the alternating layers of dense and open texture. (2.5). . Side view of the lower part of the sponge shows the radiate irregular skeleton of the base, the nodose dense dictyonine dermal layer and the irregular spicules of the endosomal layer. (x 6). . Vertical view of the basal attachment surface shows its radiating, irregularly-fibrous, crudely-dictyonine net and the hexactine-based skeleton above that attachment surface. (6). 6. Paratype (AMu. F66908): Coppermine Creek breccia. Photomicrograph of the exterior of the dermal layer with long, vertical rays ornamented by subspherical nodes. Vertical rays laterally-fused at regular intervals to produce an almost ladder-like webbing and solidly-fused structure. Spicules generally oriented upward and outward, which produces the alternating bands of nodose and smooth vertical rays. (= 10). 139 Page 86 140 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 EXPLANATION OF PLATE 43 Figure 149. Wareem baila: concentrica® NeW: SPECLES ier fever-,ccpats siete ree eee a hee eae Lee Se nee ey keke ee cetera eee take e eet to eats ogateted ateieeen ste ee eee 2. Paratype (AMu. F66910): Coppermine Creek breccia. 1. Exterior of the dermal layer shows the en echelon packing of spicules capped by nodes on the dermal surface and cross- braced and fused to lateral spicules by horizontal rays or synapticulae. Diameters of nodes increase as diameters of spicules increase vertically in the thin solidly-fused layer. (* 10). Photomicrograph of the central part of the specimen, illustrated in figure 1, shows dermal nodes and taper of the vertical rays with en echelon replacement. All are fused laterally by horizontal elements. Layer perforated by pores with essentially the same diameters as the nodes. The structure is a solidly-fused dictyonine net. (x 30). 3. Paratype (AMu. F66909): Coppermine Creek breccia. View shows attachment base with obvious hexactine derivation and radiating buttress-like elements of the basal part of the endosome with a transition into the dermal layer. (* 10). 4. Paratype (AMu. F66913): Gleesons Creek lower breccia. Side view shows flared attachment base and the irregular, concen- aoe ag an endosomal skeleton in the cylindrical part of the sponge. The dense dermal layer is ill-defined in this view. x 3). 5. root ee ee er eee tn Pe ee RR Or ria ak oA RSE emt Contin OORT eoaand oop eosoandb ben pagonaT i noosc Figured specimen (AMu. F66914): Coppermine Creek breccia. Aligned monaxons and occasional eines with associated loose oxyhexactines. (x 5). 6575 (Kalimnospongia pertiusa) new: SPECieS:.. <<. aj<.2 :-cs.csereietete hese eee reel oe sasae states vatedes ars ol ovclodny ronelcheteneyete enero kehekereuei er elat- eet cheteney ote neReRoastens Paratype (AMu. F66916): Coppermine Creek breccia. 6. Dermal view of the fragment with vesicular dermal layer and distinctive large pore subdivided by fin-like blades. (* 6). 7. Endosomal view shows outer part of the endosomal skeleton fused with synapticulae between rays of the coarse hexactines, rays of which extend as buttressing elements beneath the blades that subdivide the distinctive large pore. (x 6). tN PLATE 43 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 PLATE 44 56 -AEONTOGRAPHICA AMERICANA, NUMBER AL P, A > aN Lae ° nore a AUSTRALIAN ORDOVICIAN SPONGES: RIGBY AND WEBBY EXPLANATION OF PLATE 44 Figure 1—S eK QIIUINNOSPONg Ia pertuSsid New SDCCIESre elise ieee erie eisiacrresecisiel sires see ares clean ie aye Roc ht Gra ea tee tre Ae SES Holotype (AMu. F66915): Coppermine Creek breccia. 1. General view of polymict block of associated sponges. The holotype of K. pertusa, n. sp., includes coarse-textured skeletal elements in the right center, as well as the dense, laterally-connected, thin-walled frond immediately below the open-bladed pore and the upper thin part of the wall, which is nearly on end. Spherical specimens of Hindia Duncan, 1879, tubular Dunhillia, n. gen., and somewhat coarser Pa/matohindia, n. gen., are associated fragments. (= 1.5). . Photomicrograph of lower part of figure 1 shows the dictyonine outer layer and the irregular endosomal layer of the thin- walled part of the sponge. These layers are strongly reminiscent of cylindrical Wareembaia, n. gen. (x 10). 3. Photomicrograph in gastral view of one of the large pores, with the bladed partitions, and surrounding endosomal and gastral parts of the skeleton. Coarse hexactine rays are reflexed and some support the blades in the pore. The irregularly-vesicular gastral layer is obscured by the coarse endosomal hexactines, some of which are united with synapticulae into the solid structure. Spicules are also fused where they cross. (= 4). ASS He BEL DUIALIPACKHAMM ) MWEDDY{ANGURIS D281 9 SS eererepiere masters erereerectceete revel eis terete vsvele) sete evel leresedelionsl/sYelel eitie‘elavst-telalctlevoila\atienete «(eteteierehsioks Hypotype (AMu. F66921): Sugarloaf Creek breccia. 4. Fragment shows exopores in the central tube and flaring fragments of interwalls. (8). 5. The central tube junction with the interwall is marked by a ring of pores. (* 8). 6-9), INEDTOTHS CHINTLE, WEG SudalGSeaenuo doce omons GU nooeS Gon doUsEbda desu oo Ona ompO OC ooramodcoon Heo ome rcrcacen mascara 6. Paratype (AMu. F66919): Sugarloaf Creek breccia. Side view of parts of two chambers shows distinctly porous walls and necked connections between chambers; a tubular exaulos extends to the right. (= 8). 7, 8. Paratype (AMu. F66920): Sugarloaf Creek breccia. 7. View of the relatively smooth interior of a chamber and the open pore to the exaulos. One of the exopores has a distinct rim on the interior. (x 8). 8. View of the outside of the chamber and the exaulos shows porous chamber walls, but an impervious exaulos wall. (8). 9. Holotype (AMu. F66917): Sugarloaf Creek breccia. Side view of cluster of chambers shows porous wall. The somewhat- elongate spindle shape results from prominent exaules. Chambers are necked at their junction. (* 8). to 141 142 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 INDEX Note: Page numbers are in light face; plate numbers are in boldface type; principal discussion pages are in italics. Actinocoelia Finks, 1960 ... 44,56 adnata, Nibiconia ............. 5,10,13,90 GILernataPNYSOSPONGIG a:2.cscestetere ss tases cceecaeceee- se aecee tee te= ee 77 Amiad CUSHBASING cress. ores toecer are cosa tebe secuinteeeree cee ce seer e eeeee sere 7 AlriplasDONelauner Bel acces cacceretstccertne tec eecee eee eee ere eee 41 bulbasme Spi, - tse ssasceeteesoccsessesccesee 15,25 23: 5,10,41,42,44 MORNINOVGNDs SPias-csteses sass eae see re 1531655 ---: 5,10,42,43,44 amplexicaulis pauperatus, OrthOgraptus .....0..00c0cecvee eee eeee eee eee 9 AMu. [Australian Museum, Sydney, New South Wales, Australia] 16,17-19,21-25,27-29,31,32,34,36-39,41-47,50,52, 54-59,61,63-65,67-77,79,81,82,84,85,87,89,91 AngullongMufie tone Nee seanechaiee etka eee eer 44.76 MICAS SPs tence derevcersd cess ogesc dodvecsaccsseeevtees 36... 5,10,76 Aulocopium Oswald B47. ce.ccssdecescs-os:scsese- Secs -seceeees 14,31,34 aurantium Oswald, 1847 ................. 8:9" se. SLOSS 37532 compositum’ ConwentzZ,01 905) <,.ccs.cece-eeescescusseer-eeeeeseeere 31532 aurantium, AulocopiuM .........0......6.04 SiON es. 5, 0511531532 AUStralasiat so sccciscxsccorone soe Se ee ceoe cae Suses one veenanh OBS nee sete ee OE EEE SE 5 Australia ... 5,13,15,16,27,31,32,36,37,39-41,63,83 Central iets. c.cceaccstce cose -sverasecescneusnmatoennsstascetncweneeteteehonce cers 7 CASEIN oensctectestccoceascte: stcedieav ese nse cacsscawe nassaneucet eases teeeece 13 New: Southi Wales: tnccshonsuwsecsscnessene caoseemeeenesee ses 6,10,61,65,90 CAN GULLON Gy - senna tats cates on Laces rele oveed estan decinigenaaanseneetorers 14 SIBOONGEROO! Saaacdec ances sesnmaceneascueseasrnSeeeeaeeeee 6,14,25,47,90 “GurrajOneyR ark?” ce iccecnges iisesenetaeasswsaeesesaevecat aeaeeasceeses 63 soICAlIM MA ches cats rece ccees oeech ets socneueote eos ve sae neta stnets 6,87 SsluiscOmbe)POGIS) i iaz saGs.crsasse2snavsteteneaseete sceceecn ease tenw seer 84 SV ANGON iecncttcccatonscaseetdeieecntstees dene sewer ececea es Someeeemete vee 59 SOM BA taaecenisstss sresdseutevecse vet st teen nesdoecsetecesetemden sveenore 79 Bathurst erccc sore acccte sce Seo Pon one eae nee reece Soe ae ne ese 6 BelubulavRiver cs: sce:-0.00esees. cee 5,6,10,11,21,47,56,57,63,87 Bayne vines tee re eeee ee Bee fae tetra Ie a ate es Renee ae 6 Canowindra—Mandurama road .................:2ceseeeeeeeeeeeeee 82 Central aye sstece ett ae oe at rete we teas cor arate commen 7,12,16 Gliefdem Caves: arcatsscscect ee sseesecece eee 6,7,12,14,16,27,30,87 Main! @liefdeni Gave. Caves Urack secrscccesetesssssceesscceae 13 Nibicon Cave (CL-43) .......... Tiddalick Cave (CL-29) Wareemba Cave (CL-9) Warrigal Cave (CL-28) Yarrowigahi Cave (CLE-13) ci vc.cstscsccsssecscsessesccsetseacanes 56 Coppermine! @reeksee----ese- eee 6,8-10,21,25,31,32,39,47,87 OWA error eS eee ee Ree 2 Sn ene nee 6 Gleesons! Grech: sere essere eee 6,8—-12,32,57 Gunningbland! ok wccesccseenacenctassee seeae eee eee eae ene canes ees 63 Eachlan\ River: .2iesccccecos co ceo soca onsen 6 Dimestone: Creeks 5. ..:... 5,10,48,50-52 USSU LEtS OLLASI Qn neon ucc cence see sn ote c eer a eee eee eee 90 Eastont(il 960) heres sccccecccae ioc cene ease eee Seer ee ee eee 61 CASLONENSIS, MSEDlLOLMADIUS aca nacre ee seeece sacs oe ses ner neanaeteemecsetes 7 Ecclimadictyon cribratum Webby and Morris, 1976 .... Edriospongia Ulrich and Everett in Miller, 1889 .................. ClEQANS HD ICELOSVADLUS mecceeseeeecee nections oseone Sone meatessecease sees 7 Ellessand’ Wood ((lOOM)i 7 .te scons ceuedeaeesuesneaerccdeactensecse cremmease 9 elliptica, Warrigalia ..............00..6.. 224 pees 5,10,11,/8,19,20 Emibleton (973) ec seccseccseweceseosandetassoceseocs seseeetbascesaeecesessts. 13 EochaunactisiRigbyand Dixons 1979) ce.cc.c-.22s--0eeeerne ences 18,27 Eospongia Billings, 1861 57 Ethendpel (9G) ieseoscuceseescccte nce nec occis iceetan ase shanaecssacnate see 7 etherid geri Cliefdenella\ersccssecsctecteasee- <-cese te cacee sees e sence eee 7 Ethmophyllum minganensis (Billings) ..................00..00e00es088+ 29 EUraS Lathe case cssea eee aie eau ecins sess coe oee ero n cee euceteUeemececetee ect cere J EXbT (0) 6{0) qacancapen ps dnosceaQuoco ses rBSaacKanoacKanoEecca. Germany Spain pee cenecoar seo ccooeer es EX CUITENU CANAL |GENNe | eeseseneeteee te ee een eae cases ssoecceneceseoncees 15 explanata, FenestrospOngia ............+. 34:35) 0 5,10,74,75,76 IE GVISLINGLEIOWELy UO OME an te eiecseeteccen testes sesncane=seenccceecnese 11,12 favosa, Palmatohindia (?) ........--.....--+++ 80:31F 5,10,69,70 144 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 ENeSIrOSPONGIG= s, BEN acces ee cewsves eteecer reveceseseeeeee me eece tee 74,76 CxDlANAtd, NA SPs, jesseeeesecaees ese se ese = 34535) 5,10,74,75,76 fibrosa, Calamopora .... FRINdid ween in ksi(9 60) ieee ccsseseneee sence esas Fini S(O Gis sob ec nese ta cerned tence aac te anne Teen ete nee inksi@l 7) eeaecse ecco ct ce resco ce ccssct cone sles senende tee doee aden an neehoe Finks (1983a) Finks (1983b) Finksella Rigby and Dixon, 1979 flaccidus, Leptograptus ..............00004+ Flower QUO Goss ooustrcaceeeacasncussuets sisseene ore vestemence serena FOCHStE OOS iv .ctcse sees sevetwadhanerteestactanicas seisasoreessepeueueeserexnast SO LIALG sg GNAUNQCHIS) seats theo oeece ree seen ac shaneae hee een eee eee FossilMHallll Tim estoney cc aceresnaeescseduesccacdeacee tees ecntesssereoess aceasta Caves rack NOrizOn) cc.cccc e+e. cecesnee Dunhill Bluff Limestone Member Gerth (ODT) oe, Mecaceestecte sntass anceshacteeceakre sees eseseeesmise 32,34,41,44 gigantea, Belubulaspongia ................... PHPAY Weeeeee 5,10,63,64 Girty (U8 9 Si LUB OT) ease oe oe. oa scecewseteecene dats cotestg seedateesstepes 61 Girtyi(L S09) eee ese ie er ek ey eee sae ee 61 Girtyocoelia Cossmamn, U909 a. ss... sreeececce see secceee=setecsace arise 90 GICESONIG. NS GENE hac eon cnmec cere rocnstesence: ue eesetecie 17,22,57,58,60 (POTOSCESINSD tee ore vesascercn a0 ans sci D2 S23,25" wezess S110:57558359 Gleesons Creek lower breccia deposit (3a) .................. 73951963 DIO GIG 5.2 eae cr acwten sacvexeesteanswes Ne gestdvscnsexcesscsssesta cee 19,84 BIOCK 3) seecerecs chic cshess.c sceradticspavacusietasee seats sestessueteseence 37,87 block 5 block 7 DOCK ON terene ste ener cocoon dace eo eeea ties Sorcha oeeeee ern ate EGO ORAL Gleesons Creek lower breccia unit (3a) [see Gleesons Creek lower breccia deposit (3a)] Gleesons Creek lower breccia unit, block 7 [see Gleesons Creek lower breccia deposit (3a), block 7] Gleesons Creek upper breccia deposit (3b) ............... 6-11,37,63 DIGCKA3 Fo saccave cds save ectwa es veiate-tseoeis soeceseee DOCK 7 jai ears ress secacanetestertens ceases Glenister (1952) .......... Glossaryiof: GOlOSy iccscccvad-czecs eects o41e2teeeatesesisasetaiesteeesessns Goldfuss:(Wi826=11829); <2... .2-c see cencoee shoes sess oseneseeseusooasszees css Gondwanan continental margin of Tasmania ....................... 13 Goonumbla Volcanics! a. 262. 25.42. 2- Brass;-and)Harrisoni(:98'5)) \..-2..2-----.52-as+---eee< over 13,14 WN Ke) Kos a¥92) 5 Bale psoeoeasedssasucnBascicusbasu dpe SecResnAes ode renneeec ues 11-14 INMOOrS (MOO) Ieee cs cana sence usc neG eae acsessecdcnde cd ouutecanseaaeua seats? i multipora, Palmatohindia ............... 28,29,30 ...... 5,10,65,66 multiporata, Dunhillia ............... 5,10,48,52,54,55 ITAL VATS, EXO! Sconaaneesonosonene coconsnpsdonsochuosasoceod 32,41,44 National Science Foundation Grant DEB 8200860 ............... 15 WNevadocoeliatBasslers W927 Mresc: sccsccestece sec ecceeeataeseeeer ener 34,44 New Southawialesisland=arciressrc.seneeesserescesecsess orc ercre ss ee eres 13 INIDICONIGANM SON tener re nose eee eee ee eens ceeee ce ctaraeee 90 GANQLAN TAS tarcerecs eae seen E a AAS 52: 5,10,13,90 Nicholson (1872) 29 MICHOISOMUAS D RACTOLINCSieesesen eecicn cacececcncec cere eee eas cate ie atee 61 nodosa, Liscombispongia ..............0..06++ 40,41. ...... 5,10,81,84 INorthvAmerica 2. cise. sesc-vceccsseserescosciseseseseeseees 11,16,27,39,41 Ap palachiansrepiomrecsercce cere eaeee eeree cere nsee eenaeeensactestes CASLODM pees sents regs fon ses proce ect naes cen tocews North American assemblages INorthrupi(liOS9) Beeerecsscs caer cs secccncee aes tecens ones aceechne ecco noscaes oculatayPatellispongia maemmccr see 37,39 @kulitehy (935) Pest sc ter ar eer ee eee Cons a oecean ae aee 29 OLAINATAEIAD OWLO pret eter tenes ace 3°46 eee 5,10,21,25 Orthograptus amplexicaulis pauperatus Elles and Wood, 1907 .................. 9 calcaratus cf. basilicus Elles and Wood, 1907 orihoplecturns)StereOdichyurn sestres eee cee tenses. chee osculumi(defined] emescee sree eee en eee cate ee oSstimmidetined | fe arenes cy. cae seen ee ere see eee eee ene ee @swaldi(iS4yik rac secussccareete sasacrwseitsectanscsceeses: 5,11,14,31,32,34 Ott (OGTR. siatgeleaseeeseniee eae vec caiaet seees ce ote ono veR oe ease eee oo 90 Packhami(loss)i ates scence tcut paseo cochence sceene coe eae aac aseenoerenseine 13 DACKHAMIUNBEIUDUIGIAMapen: eae ce nae ee scree eee eee eae wee eee se 75033911 packhami (?), Belubulaia .............0...0...00+- Ags. G2 5,10,13,91 Paleo=Pacific Oceans fieacsscataccks. sete anadstesese case tease scencteeteaee ans 14 Palmatonindiaany gent meses cscscseree cose eee 46,47,65,74 CVLINGTICAADASD eter ccnvavee teceette senses tee Si, ene: 5,10,67,68 IMUITIPOLG! NaS shoesesee eet sstoss os: s2e8 28'29:30i nee 5,10,65,66 Palmatohindia (?) favosa, N. Sp. ..........+- 30:31 5... 5,10,69,70 Parkes Platform yerececeace cee co tee eee eee ee ee ne eer Parrishi(hi9 82) it etenccccoacnse ites cates celeste eee eee ee PAPVASATDOORINGIG terac.c cee cacverssscss sce eoeesee Pate and Bassler (1908) Patellispongia Bassler; W927 s2tcc2.-tscceseese- sees 14,37,38,39,72,76 CUSUGIISHI ESD Aetna 123135 eee 5,10,37,38-40 ClintonitBassler 927) cose ce cscs sea nee ee ane oeee 39 WAALNID OT GIBASSIET OA ne ws cove ecas. sateatscasecsestesnecsterecedes 39 INLIUNUTLED OVI EASSICTIPUO Diltten secc vest ese s senseacsc acs seus ce sree eee ener ee 39 OcUlatasBasslertO2Firss.cscnstaccecc occas eee nee eee 37,39 Pelicaspongia Rigby, 1970 .... : Percivali(lOnGy is. cco: socee ce shaw ere ash ve recs coo ot hen oe eee ae eee 7,9 Percival (97:8) ihc siceieees oe sce ste. cvesess oats. bests steceeeaeeac res 10,12,13 Perciviali(G9:79) etree racsweseccavdess acseveecsaeseu eect cacevseee eee cite 10,12 PercivalialaniGea yrs. cecte ese te ates coset ease sacs oe ecco sae 14 perdentata® Glie{denell avin. sz. sceoscscncsoateen ian s- cee cesce cae ce tea see nee 63 Perissocoeliay 0. BED. .sscvecsesesces.c-cseeeive oe deinsasstesssacusecevesesoies 32 ADV GANS SDitectn-seesee seas cee vee DAOM225) as. e- 5,10,32 pertusa, Kalimnospongia .................2+4- 43,44... 5,10,87,89 Petraia minganensis\ Billings, 1859) ........2.0-:22-c2-22cs0ss2s+es00s-: 29 Phacellopesma' Gerth: 927 co csscngeescste cccwee -3e-seeeo ee 32,34,41,44 Physospongia Hall 8 82 xoc sare cesses ee areate cee eee ee eee ene 77 GILErNAL@alli NS 82u, cesncctsscstcsse tease ceetaenentsetteeeccese sp tcceaeee Ul GAWSONISELAlMN S82). cizscesessscomeeece ete cease cee eee 77 Pickett! (U969))..ccccscsccetsvessoseestoenstaaneeee ooeeue cesarean. eset eneene sees 65 Pickett; (G82), cseceacasnteerosssce cute costesarsetoracceceec stay oesesteece tes 21 PICke tty BS) ie cacck owas one scetee soe sean secs eee omens seemecn tonemer cet Pickettiand ell ((U9S3) iis ccasn.ccoosacchocecadecneseestesee Seacostesmesesces planata, Mamelohindia ................--. DIQNUM A PSQNOGICLVUINIGs. eacmeeenee eset eeaecetecen seas te cena eeeeeet enter Plunkett Trust, Lilian and Winifred .... POTCARANOMAIOLIOSSA gevannsccecaee se des doses eee sone ee ee eee JPOLOSAs(GICESONI Gs speeee ence eeete sees DDS 25 DroterOns StCr€OdICLY UM ver.acseeccncetcevectecesoescessvoncescctuets oereste Protospongia Salter, 1864 .................. wagued Cordecea ur venstes det coNe 7 Psarodictyum Raymond and Okulitch, 1940 .................... 39,72 CFASSUMIETNS SD cenvioncencee sac nenseascuc dia see seee seo 14) 5,10,39 magnificum Raymond and Okulitch, 1940 ................... 39,40 planum Raymond and Okulitch, 1940 ...........0.........02.022 40 Pseudopalmatonindiasin Gen gaes-2.cec eoee teen eee ete 46,65 GIS GIGEN ES Dap attre sere eae ee AZ7EUS Sescces 5,10,46 QUAGT ALA NIGAGLICKIQ pewecrsctedeatcaesaseanacese ate Bie ese. 5,10,78 AINOS A MEASSOCOELIQMs miscen outers Ses ee ok Sos cae een eee 70 Rauf (S86) esses eee seten eee es Re aearerh BU. dea Seicaccabtucedeasesoes 61 146 PALAEONTOGRAPHICA AMERICANA, NUMBER 56 Raufti (US 93) se acne Sacseecccces stsee econ tveeebare reoeeereee ee eros 27,61 Reautt (S94) xc. cescsscsecosssessscnesenscectausseeeeetts pee scome ees eo OOS Raita 95)) shoes ances cc oczeettecsees oavescanaceecsaesaces 29,31,32,41,63 Raymond and Okulitch (1940) .......... 5,14,34-36,39,40,41,57,72 Read: Bays Rormationyscss.cessccesseasaete tee cesecteaeeiea senses sen Ooeess 16,27 regularis, Haplistion ............... Ar gesteet 5,10,76,17-19,21,22,25 RETO 958) eaessccses se acs sceeveaneeees Stas TT) Reid (1963) ... 29 PRETO GA Yes ke cba contae sacs ses naceaes coaaeee axeeinaesesscrenaiseate saeceeres= 86 FREIGsCU 96 Sa) hes pesca: sosceceare teehee cae ene senee an sme sden qSemtee sone mnees ee 27 REIGN (USGS D) recom. con eeeeramsastes den tases tetee uecategtass mien cnstuneee eres: 61 Rhopalocoelia Raymond and Okulitch, 1940 14 Rietschel (1968) sous» TAO) Rigby ya(9 G7) heer sceccsee seas concseneec use 79 Rigby (1970) ......... 79,82,83 Re ya (UO Tél) eee sae ts aacenine ventemereedcciceeancoactssenessensateacen ote ec 79 Rig Wa (OMAN cmesee enemas nateaceiace easton seuelne sande ecvaceeessmesnan decreas: 83 Ragbya(lO SS al eac.cacetcscteesecmet neetesscnyscceme ees ee eee Ti. Ruighyi(li983 bil ecccsesscas A stoi vecusranataces cauestsieceessontasetreae os Rig b yi (9 SA etnias achscanaccaeet con asuence siaucartece cer ceateenerec mee setae Righyi (986) weree.scersany cts sccatwanseacaetteceasaieeecesetecsce Rigby and Bayer (1971) Rigbyzands Dixon: (979) eccecreseesteeeeee ness -seeresenesae 16-18,27,57 RighviandsRotten (986) reerc.ueceergastecisarcoecesensserecescoesaress. 7,14 Rigby and'Potteri(ini prep:)) s., s.c.scsecoteesare cues sacesdeussvcsseeese= seas 16,18,21,41 ON GINGA ESD eareteeseese terest 354. ee 5,10,21,25 PasSmMan Orogen cee ics coasanoaes ese oacseckoc es eare ewes anes ceceemeenes LETS: Macquanie Volcanic'Belt 2c... -2csccseccace--s0essecetesmeanereeteeees 12 Molong High ace, willl Pasman:Sea’ --.-:.-c-c0sc-s- een LS) Tasmanide fold belt Ree ws) Termiemand! Dermiér) (956) sec. -cvace-ccs-u--ceasceneeecuevh eeeeeneee 90 Termier: Termier: and: Vachard (1977) = = Ee = = a = 2 : 2 » 2 m 2 m Z m 2 m = (ep) . — — _ n ; J1UWYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSON! a ee ee z g 2 x. ¢ ‘es WN § =z WG 5 rs = af pg = AS =| OMEN co oO Ne ae ro} xi OY, (OMS TO Se ae ES 2 2 2 EGU 2 WS 2 =e BW 8 ES z = 8 Gg 2S & = = = _ SS = 5 = a= ” * ra w = 3 7) 2 ” » oA STITUTION NOILALILSNI NVINOSHLIWS saluvudit_LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLI ee Ae He Ww 2 Rt Ww a WwW a ca a SS = nn R S aN a n = - - <” = nn as SN a A \S a at (ae ty pp = (94 = SAG < 4 WSS < c \AS&E > - 2 = ae I~ a \S = a =. YY A is z e z ie eee B z STITUTION NOILNLILSNI NWINOSHLINS S3IYVYUSIT LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLI fe 27) z n Zz ‘ 2 Zz - 27) z = = = Ws = = = = = tye i - “ = \y 3 4 2 YG.5 Nw? 2 Na = 3 WS 2 : 5 2B NN 2 2 NAG 3 TNE Oo Be fe) Hy L As ie) a5 RE BE AS ze = 2/V% 7 5 S 2 E Wo 2, = Sue 3 E ee ee ae z 1!'YVYG!I1 LIBRAR IES SMITHSONIAN INSTITUTION NOILNIILSNI_NVINOSHLINS | S3 | yVYa!I] LIBRARIES SMITHSON! & us . Ww @ us 5 i e = c S cc YS ms c < J loa P = a = NS ae = a 5 2 3 i o we S) = 2 ay ad Sj 2 2 uy STITUTION NOILALILSNI NVINOSHLINS S3I1YVYEGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLI = is = a = a = a a - w =) o 2 “oO =) ELLER o \e D = Pe) = x = a \) y = = = = ‘ <3 = > \ rex > Es > = > = Ps) = pe) = ‘ a E pe) w ot oa w — Yn . = | 2 o 2 D z Oo z Ws \}YVesl1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSN! NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSON ae ae a Z : g ge pet), 1= \\ SS 2 AN 5 bY “4z = fr AS =| “p We D Qe Kz Oo WN = OY fe) Boge DS Qk = YUG 2 KX & 2D NE fe ff J 2 ues AER” B iy g =z NS S 2 = \N SG; Ky = 2 E \. 2 (iY = YS = a : = YG, Ye = = = ° = = > = . > = > = . > | ” ) rd ” ‘. a n ead n > z STITUTION NOILNLILSNI NVINOSHLIWS S3IYWYGIT LIBRARIES SMITHSONIAN INSTITUTION | NOILNLILSNI NVINOSHL! i= eG = oO = a = x. S S @ = ee ® XK = @ = bof foo - NN ac 4 Xe a = “a. vy jf - a [J ays < 2S = a if Lf 3 < 2 BK : aS OE = = Gy + 3 [-N = fo) = Oo = ys oO — (2) “) 4 Zz ee | 4 =) rs 4 2 : 1YVYGI1T LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3JIYVYUGIT LIBRARIES SMITHSON at S = JE ; ive) = y oO —_ oo — wo — 4 = i Gly, ee) 5 Be) YX igs: a) = ee > - fo Kf Fa > \ QE > = Z ol —_ ly “és Le — > o] aN SWS —_ 2 mas bl 7 = eas eae = = eax XS me = =, we Fe m ” m ” m SW 2 m wv a) z= no = a) : = wo = STITUTION NOILALILSNI NVINOSHLINS SSIYVYEIT LIBRARIES SMITHSONIAN NVINOSHL * o z a ; aS Y z a = g = y ity = = = = < = < Xy 2 &2 : 2 Ns : : 3 e an 3 AE ? g 2 SING 2 ne 2 NE : E ANe2 : g, : ES SS = > = “ro Ys =e s YG NINOS Tr: o) SS Ol a) a) Fa! IVYEPIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSIT LIBRARIES SMITHSONIAN / 3 0) %G z NO! LI NO = S um = Lom « z f= ms | = = “0 = w NE 2) = o : S : 2 E 2 ON 5 A ; > — = = > Wes 2 es EF ¥ | 2 =a a - i = au —— m 2 m ee = n aes n wo z on z= [o7) xX: z 2) 2 . ITUTION See ee aa eee eile a AES oT SOnIAm TINS TIEUION NOILNLILSN!I NVINOSHLIWS Rn n eee xe n Zz rea: : ZN 5 a Ea GY Oo? =r : fe) Yl | AN 2 iy E Z EM 2Z =z WO 8 yy Ss t > = > = eS = = / = See 7) 2 ” oe ee 7) Bre 77) | JvYdit LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3JIYVYSIT LIBRARIES SMITHSONIAN ud ; NGUNLILSNINVINGSHEINS ° =. fe) = fe) = ro) =e = o = wo = Gy OD = o Z > We 5 : = Gly : > We = > ANE 2 = Vd CoS = > WN - = NSS"'5 2 i i ae = = XS {VUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLINS S31NVUGIT LIBRARIES | SMITHSONIAN : ze . . ‘ ty, * z < < = fy 2 z z a WN = x OY z \ = ip x fe) rT MK AW O = ir = ALY g E z E Wo" Zz: i \S = = “2 = co a 8 ITUTION NOILNLILSNI NVINOSHLINS S3IYVYE!IT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NWINOSHLIWS LIBRARIES SMITHSONIAN 7p) => w = = no _ WwW a. WwW = A if a = = = 4 4 = a = \\ < =F a= ze 4 < (os Yq SY : e = = ao 4 SN SS a = = iva = o Se, = 3 3 = S =5 Fs =! on z a) = JVYS!IT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYE!IT LIBRARIES SMITHSONIAN E 5 : j Es : : : = fs rs o o = ‘oO = wo = wo = > : : 5 2 5 2 5 > Fy > E > = > rx z E zl 7 5 : a o z a z o 2 ue = ITUTION NOILOLILSN! NVINOSHLINS S3JIYVYUSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS ne w Zz wo = es n” Zz Ge n z s. = < = (= .wW< = < @. = 5 S = S = Jez NS q z \; =| af fix = Se Bo re) = oe ORE. =f o NWS WS So) Ke Oo g 2 g ZR 2 2NX 8 GH ? 2 E 2 -™ 2 = § 2G = > = > = oS = > : = Fas ” ca (22) sg os (Tp) a ” NYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31u¥vVudIT LIBRARIES SMITHSONIAN Ss = wo = 2) = (7p) = z ne z Ww z x Wi 2 WW 7) oe ee ed = a) Ws 2) = a = = eR Vig A =H XK = 4 tht fo = x. “uy Z = a 2 ANXO = “.¢f, a < PH a < 2 WS « eC = Uh d su ee 3 Gus) = eu EW EGY: Gy) Pe rs) a ro) = 3 = ro) a “Yes 2 = z =} z Se ey FA = ITUTION NOILALILSNI_NVINOSHLINS. S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION _ NOILNLILSNI_NVINOSHLIWS a i S = = z UF 5 leg > (@) = {eo} = (o) = = — o = w = @ = 2 = Be) IN ES x0 = Meo = es) E = yee E = = ne E z = 7m NS 2 au = 2 E 2 ASN S (‘ayes ; D re D z o Zz o RARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLINS S3IUVYSIT_ LIBRARIES SMITHSONIAN = ee z < z my, = z = _— —- z . = re z Ng : 3 : 5 Ly 3 SAN 8 2 SA B ‘oD B 2 Sn S NOY 2 E Wrz. = 2 E N\O"2z SY = = SS > = = = i Seah %) wu c < x o =| z i) Ss > \= |= 7 = z < z2 fo} a x | = = a eave peye re Scer amen tener aeemrreereetaran ipawat ora ou Per aye Serato akties =a aresan re! ou)