Paleopathology Origins of spondyloarthropathy in Perissodactyla

B.M. Rothschild1,2,3, D.R. Prothero4, C. Rothschild1

Bruce M. Rothschild, MD; Donald R. ABSTRACT s e n t ation of in the fo s s i l Prothero, PhD; Christine Rothschild, RN. Objective record (4-10), it seemed appropriate to 1Arthritis Center of Northeast Ohio, Spondyloarthropathy has clearly been investigate an order in which contem- 2Northeastern Ohio Universities College documented as not limited in occur - p o ra ry occurrence has been noted. of Medicine, Youngstown, Ohio, USA; rence to humans. Transmammalian in Notation of the frequent occurrence of 3 Carnegie Museum of Natural History, nat u r e, it is of interest to understand the e ro s ive art h ritis in extant peri s s o- Pittsburgh, Pennsylvania; 4Department of Geology, Occidental College, Los Angeles, antiquity, and perhaps the origins, of dactyls stimulated study of its nature California, USA. this disorder in groups suffic i e n t - and antiquity. Please address correspondence and ly represented in the skeletal record. reprint requests to: Bruce M. Rothschild, Methods Materials and methods MD, Arthritis Center of Northeast Ohio, Fossil and recent skeletons of perisso - Survey was performed of modern and 5500 Market Street, Youngstown, OH dactylae from North America were sys - fossil holdings of the Acade- 44512, USA. E-mail: [email protected] tematically examined to determine the my of Nat u ral Sciences of Philadel- Received on September 5, 2000; accepted occurrence and population frequency p h i a , Pe n n s y l vania; A m e rican Muse- in revised form on May 28, 2001. of spondyloarthropathy. um of Natural History, New York City, Clin Exp Rheumatol 2001; 19: 628-632. Results New York; British Museum of Natural © Copyright CLINICAL AND S p o n dy l o a rt h ro p at hy was the most H i s t o ry, L o n d o n , United Kingdom; EXPERIMENTAL RHEUMATOLOGY 2001. common form of arthritis recognized Carnegie Museum of Natural History, in the extant and fossil records. Com - P i t t s bu rg h , Pe n n s y l vania; Cleve l a n d Key words: Spondyloarthropathy, mon in extinct families such as Bron - Museum of Nat u ral History, O h i o ; reactive arthritis, , , t o t h e riidae and Chalicotheri i d a e, a Canadian Museum of Civ i l i z at i o n , animal model of arthritis. progressive increase in the frequency Hull, Ontario, Canada; Denver Muse- of spondyloarthropathy was observed um of Natural History, Colorado; Field through geologic time in and Museum of Natural History, Chicago, Rhinocerotidae. Illinois; Florida Museum of Nat u ra l Conclusion History, University of Florida, Gaines- Erosive arthritis of the spondyloarth - ville; Institute Royale de Sciences Nat- ropathy variety is now documented as urale, Brussels, Belgium; Los Angeles not only persisting in Perissodactyla, County Museum of Nat u ral History, but as actually increasing significantly C a l i fo rnia; Museum of Comparat ive in fre q u e n cy (3-6 fold). Given the Z o o l ogy, C a m b ri d ge, M a s s a ch u s e t t s ; u nusual evo l u t i o n a ry penetrance of M i ch i gan State Museum of Nat u ra l this “disease,” the possibility must be H i s t o ry, East Lansing, M i ch i ga n ; c o n s i d e red that its persistence pro - Museum of Vertebrate Zoology, Berke- vides evidence for some unknown ben - ley, California; Museo Distoria Natu- efit to the affected host. rale-Sezione di Zoologia “La Specola” Universita di Firenza; National Muse- Introduction um of Canada, Ottawa, Ontario, Cana- Assessment of the antiquity and impli- da; National Museum of Kenya, Nairo- cations of the erosive arthritis spondy- bi, Kenya; National Museum of Natur- loarthropathy has been limited by the al History, Wa s h i n g t o n , D.C.; Roya l ava i l able anthro p o l ogic re c o rd (1). O n t a rio Museum, To ro n t o , O n t a ri o , Contrary to arthritis caused by direct Canada; Rykshius Museum van Natur- bacterial invasion (2,3), the prolonged lyhe Histori e, L e i d e n , N e t h e rl a n d s ; existence of other fo rms of ero s ive San Diego Museum of Natural History, arthritis has only recently been recog- C a l i fo rnia; Simon Fraser Unive rs i t y, n i zed (3). As spondy l o a rt h ro p at hy is Vancouver, British Columbia, Canada; not limited in occurrence to humans South Carolina State Museum, Charle- (4-8), an alternative approach was con- ston; Unive rsity of A ri zo n a , Tu c s o n , s i d e re d. Given the substantial rep re- Arizona; University of Kansas Muse-

628 Origins of spondyloarthropathy in Perissodactyla / B.M. Rothschild et al.

Fig. 1. Spondyloarthropathy in Pleistocene and modern . (A) Lateral view of fused vertebrae. Full fusion noted on right with early calcification (arrow) in anulus fibrosis on left. (B) “En face” view of sacroiliac joint. Erosions marked by arrow. um of Nat u ral History, L aw re n c e, Results o c c u rrence cannot be eva l u ated ep i- Kansas; University of Michigan, Ann C rat e r- s h aped holes with smooth, demiologically, and so will be listed Arbor; Unive rsity of Neb ra s k a , L i n- rounded edges in the iliac component solely for completeness. Characteristic coln, Nebraska; University of Pretoria, of or fusion through the sacroiliac joint erosive lesions with reactive new bone South Africa; University of Witterswa- identified spondyloarthropathy in Per- formation were found as isolated phe- tersrand,Johannesburg, South Africa; issodactyla (Fi g. 1). The presence of nomenon in metacarpals of thre e and Yale Peabody Museum, New Ha- s y n d e s m o p hytes or sacroiliac joint Miocene Rhinocerotidae (UF 59699, ven, Connecticut. involvement was used as the numerator UF 59700 and UF 59702), metapodials The diagnosis of spondyloarthropathy in determining the frequency of spon- (UNSM 1568-46 and UNSM 1869-46) was made on the basis of specifi c dyloarthropathy. Given the incomplete and scapula (UNSM 7196-80) of sacroiliac (SI) or zygapophyseal (ZA) state of the fossil record, the presence Te l e o c e ra s , and calcanei of Pliocene joint erosions or fusion or ve rt eb ra l of sacroiliac joints themselves was uti- R h i n o c e rotidae not identified as to b ri d ging in the fo rm of syndesmo- l i zed as the denominat o r. A l t h o u g h species (UNSM 676-40 and UNSM phytes (1-12). The skeletal remains of spondyloarthropathy can also produce 3012-49). fossil and recent mammals were sub- erosions in other joints (1-12), isolated Shoulder and knee involvement were jected to visual ex a m i n ation of all articular regions, to identify all occur- rences of articular and peri - a rt i c u l a r bony alterations throughout each ske- leton, to specify the types of bony al- t e rations at each occurre n c e, and to map the distribution of occurrences in e a ch skeleton. Lesions we re distin- guished from art i fact as prev i o u s ly described (3). In the event of disagree- ment as to whether a lesion represented an erosion or artifact, for the purposes of this study it was treated as artifact. The incomplete nature of preservation of skeletal fossil remains can compro- mise the assessment of the population frequency of spondyloarthropathy. For this study the sacroiliac joint was there- for e considered “sentinel.”The ratio of affected to preserved sacroiliac joints was utilized to estimate the population Fig. 2. Phylogenetic distribution of spondyloarthropathy in Perissodactyla by epoch. frequency of spondyloarthropathy.

629 Origins of spondyloarthropathy in Perissodactyla / B.M. Rothschild et al.

Table I. Phylogenetic distribution of spondyloarthropathy in Perissodactyla by epoch. noted in one extant rhinocerotid, Dic- eros bicornis (NMNH 271189). Reac- Genus Miocene Pliocene Pleistocene Holocene tive bone changes were noted in meta- Rhinocerotoidea 7* 1/22* 13/95 1/5 28/87 podial III of a Miocene leo- Amynodontid nensis (UF 164393) and in an unnum- Metamynodon 2 b e red USNM Miocene Pe ra c e ra s. Hyracodontid Fusion of the navicular-ectocuneiform Forstercooperia 1 Hyrachyus 6 in a Pleistocene horse (LACM 8599) Hyracodon 8 from California is compatible with a Rhinocerotid 1/12 13/87 1 1/5 28/87 diagnosis of spondyloarthropathy. Aceratherium 1 S p o n dy l o a rt h ro p at hy was noted in Aphelops 2/5 Eocene Pe rissodactyla (Tables I, I I ) , Ceratotherium 7/24 s p e c i fi c a l ly Bro n t o t h e riidae (13%). Diceratherium 17 While insufficient nu m b e rs of Bro n- Dicerorhinus 2/9 Diceros 8/30 totheriidae were available to assess its Menoceras 25 presence in the Oligocene, 7% of Ol i - Peraceras 1/1 gocene Chalicotheriidae wer e affec t ed. Rhinoceros 11/24 Spondyloarthropathy was not observed Subhyracodon 1/11 in Eocene Rhinocero t i d a e, a l t h o u g h Teleoceras 9/39 Trigonias 1 the sample size was small. Such le- 3/23 8 sions were present in 8.3% of Oligo- Dolichorhinus 2/4 cene Rhinocerotidae, specifically Sub- Megacerops 1/2 hyracodon NMNH 3919 (Tables I, II). Helaletidae 2 Well represented in Miocene rhinocer- Chalicotheriidae 1/14 otidae (13.8%), the Pe ra c e ra s, Ap h- Moropus 1/13 elops and Teleoceras were specifically Tylocephalonyx 1 a ffe c t e d. D i c e rat h e ri u m and M e n o- Tapiridae 1 3 13/65 c e ra s appear to have been spare d. Tapirus 13/65 While insufficient Pliocene examples Equidae 9 37 1/137 1/49 9/304 13/163 were available to assess its frequency, 2 3 20% of Pleistocene Rhinocero t i d a e 1 were affected, contrasted with 32% of 5 asinus 3/18 recent Rhinocerotidae (Fig. 2). burchelli 1/53 Spondyloarthropathy was observed in caballus ~ 4/104 2/22 20% of extant Tapiridae (Tables I, II). fevus przwalskii 2/13 Tapiridae are insufficiently represented grevyi 1/18 hemionus/kiang 1 1/20 in the examined fossil record to assess neogaeum 1 its frequency in the evolutionary histo- pacificus 3 ry of the group. scotti 1 The first notation of spondyloarthropa- simplicidens 1 zebra 3/19 thy in the fossil record of the Equidae 4 (Tables I, II) was in the Miocene, in a 2 specimen of P l i o h i p p u s ( A M N H 1 71123). That individual represents less 8 than 1% of the Miocene Equidae that Kalobatippus 2 were evaluated. Two percent of Plio- 24 cene and 3% of Pleistocene Equidae 19 1 were affected, compared with 8% of 1 1 the extant genus Equus (Fig. 2). 2 E i g h t y - s even percent of the affe c t e d 1 i n d ividuals had axial joint disease. Parahippus 15 Sixty-three percent had sacroiliac joint Plesippus 2 erosions; 13%, sacroiliac joint fusion; 1/11 19 2 3 8%, syndesmophytes (anulus fibrosus calcification) (Fig. 1); 5%, zygapophy- *Fractions indicate number affected compared to number evaluated; freestanding numbers indicate seal joint disease; and 6%, costoverte- numbers evaluated, as none were affected. bral erosions or fusion. Peripheral joint

630 Origins of spondyloarthropathy in Perissodactyla / B.M. Rothschild et al.

Table II. Identification of spondyloarthropathy in Perissodactyla. Tel e o c e r as (UNSM 2479-76 and LACM 92016-MP, respectively) and 3rd pha- Family Individuals with Spondyloarthropathy* lanx in a Pleistocene Dawson hors e (N M C 36318 Ð third phalanx) we re Rhinocerotidae noted. Aphelops AMNH FAM 114805; UF 69944 Ceratotherium AMNH 51858; BMNH 75-2384; FMNH 29174; FMNH 125413; Discussion NMNH 164635; NMNH 164589; NMNH 164592 Sacroiliac joint erosions and/or fusion Diceros AMNH 24741; AMNH 113776; AMNH 133777; BMNH 1876.2.15.5; and syndesmophytes (Tables I, II, Fig- IRSNB 9714; NMNH 162933; NMNH 198298; NMNH271189 ure 1) in Brontotheriidae, Chalicotheri- Peraceras UNSM 14-22-637 i d a e, R h i n o c e rotidae (S u b hy ra c o d o n, Rhinoceros Pe ra c e ra s, Ap h e l o p s, Te l e o c e ra s a n d sondaicus BMNH 1948.12.20.1; NMNH269392 all extant species), Tap i ri d a e, a n d unicornis AMNH 35759; AMNH 54456; BMNH 1953.8.13.2; Equidae are pathognomonic for spon- BMNH 1961.5.10.1; FMNH 57639; IRSNB 1208; LACM 85986; dyloarthropathy (1-12). Diagnosis was NMNH 336953; UNSM-ZM13844 confidently made, as the limitations of Subhyracodon NMNH 3919 x-ray interpretation (2, 13-15) do not Teleoceras AMNH 44-1944.1; AMNH 15489; AMNH 8392; AMNH 8415; apply to direct visual examination (16). AMNH 8431; UNSM 2708-87; UNSM 2079-76; UNSM PI2240; UNSM PI2270 ( False positive x-ray interp re t at i o n s , Not identified to species - Love bone bed UF - not catalogued i n d i c ating the presence of sacro i l i a c Brontotheriidae joint erosions or fusion, relate to the Dolichorhinus AMNH 1843; CM 3147 irregular shape of the human sacroiliac Megacerops AMNH1848 joint). The only phenomenon wh i ch Chalicotheriidae may occasionally mimic the sacroiliitis Moropus AMNH 1920 of spondyloarthropathy (on direct visu- Tapiridae al examination) is infectious sacroili- Tapirus BMNH 63.7.1.1; BMNH 85.802; BMNH 85.809; LACM88943; itis (2, 3). Howeve r, i n fe c t i o n - re l at e d NMNH 11883; NMNH 14648; NMNH 15540; NMNH 218726; NMNH 252944; NMNH 261025; NMNH 270353; NMNH 281393; bone lysis and exuberant new bone for- ROM 94410 mation are quite distinctive and thus Equidae were easily excluded in this study (2, Equus LACM 8599; LACM 121054; NMC 45629; UNSM SH-40; 3). UNSM 574-48; UNSM 8053-41; UNSM 10222-39 R e c ognition of spondy l o a rt h ro p at hy, asinus BMNH 1960.11.10.4; BMNH 1981.12.9.1; IRSNB 12970 on the basis of pathognomonic sacro- burchelli CM 17826 iliitis extended back to the Eocene in caballus NMNH 15780; NMNH 267482; RMNH 123861; RMNH 145107; RMNH 29797; RMNH 400330 the Perissodactyla (Tables I, II). It was fevus przewalskii BMNH 1963.1.25.1; NMNH 303321 common in extinct families such as grevyi AMNH 82037 Brontotheriidae (13% in the Eocene) hemionus BMNH 1957.7.18.1 and Chalicotheriidae (7% in the Oli- zebra CMNH 6845; CMNH 30381; IRSNB 3974 gocene) and present in 20% of extant Pliohippus AMNH 71123 Tapiridae. A progressive increase in the frequency of spondyloarthropathy was * AMNH - American Museum of Natural History, AMNH FAM - American Museum of Natural Histo- ry, BMNH - British (London) Museum of Natural History, CM - Carnegie Museum of Natural History, observed through geologic time.This CMNH - Cleveland Museum of Natural History, FMNH -Field Museum of Natural History, IRSNB ranged from 8.3% of Oligocene Rhino- Institute Royale de Sciences Naturale, Brussels,LACM - Los Angeles County Museum,NMC National cerotidae to 15% in the Miocene, 20% Museum of Canada,NMNH - National Museum of Natural History (Smithsonian),RMNH - Rykshius in the Pleistocene and 32% in extant Museum van Naturlyhe Historie,ROM - Royal Ontario Museum, UF - University of Florida, UNSB and UNSM - University of Nebraska. species (Fig. 2). This compares to an increase in frequency in equidae from less than 1% in the Miocene to 2% in erosions and fusion were distributed to lanx) and a Pleistocene Sheri d a n the Pliocene, 3% in the Pleistocene, the shoulders, elbows, wrists, forefeet, County, Florida horse (UNSB 6747), and 8% in extant (Fig. 2). knees, ankles and hindfeet. f ra c t u res in an Eocene Pa l a e o s yo p s Among the Miocene Rhinocerotidaet O s t e o a rt h ritis was ex t re m e ly ra re in (AMNH 129392 Ð foot) and a Pleisto- Pe ra c e ra s, Ap h e l o p s, and Te l e o c e ra s the fossil record. Minimal osteophytes cene horse (UF 136490 Ð metapodial we re affe c t e d, while D i c e rat h e ri u m were found in one Miocene Aphelops with secondary infection), a metapodi- and M e n o c e ra s appear to have been (UF 69977). Isolated exostoses in Mio- al stress fracture in a Miocene Perac- spared (unaffected). While most of the cene Parahippus leonensis (UF 163928 eras (UNSM 8001.82) and scapula and affected horses were not identified to Ð tarsal and UF 168370 Ð mid-III pha- metapodial infection in two Miocene s p e c i e s , re a c t ive bone ch a n ges we re

631 Origins of spondyloarthropathy in Perissodactyla / B.M. Rothschild et al. noted in metapodial III of Parahippus Conclusion Semin Arthritis Rheum 1992; 21: 306-16. lemesis (UF164393) and probably rep- While spondy l o a rt h ro p at hy is cl a s s i- 6. ROTHSCHILD BM, WANG X-M, CIFELLI R: Spondyloarthropathy in ursidae: A sexually resent an example of the peri p h e ra l fied as a disease, perhaps (analogous to t ransmitted disease ? N atl Geographic Res arthritis that often occurs in spondylo- malaria-protective effects of sickle cell 1993; 9: 382-284. arthropathy. anemia), is its persistence evidence for 7. ROTHSCHILD BM, WANG X-M, SHOSHANI J: Perhaps the most impressive aspect of some unknown host benefi t ? Rhino- Spondyloarthropathy in proboscideans. J Zoo Wildlife Med 1994; 25: 360-6. Perissodactyl evolution is the size pro- cerotidae and Equidae are represented 8. ROTH S C H I L D BM , ROTH S C H I L D C: No laugh- gression. While spondy l o a rt h ro p at hy by extant fauna (e.g., Rhinoceros) and i n g m at t e r : S p o n dy l o a rt h ro p at hy in hya e- was common the Perissodactyl giants are amenable to chrono-epidemiologic nidae. J Zoo Wildlife Med 1994; 25: 259-63. ( B ro n t o t h e ridae and Chalicotheri d a e ) , a n a lysis of the disease impact. Such 9. ROTHSCHILD BM, WOODS R J: A rt h ritis in New World monkeys: Osteoarthritis, calcium they left no known descendants. This is analysis may allow greater understand- pyrophosphate deposition disease and spon- not so for horses and rhinos. Early ing of the contemporary distribution of dyloarthropathy. Intl J Primatol 1993; 14:61- horses were quite small and early rhi- the disease. Perhaps further study will 78. nos relatively smaller than their mod- allow identification of the nature of that 10. BJORKENGREN AG, SARTORIS DJ , SHERMIS S, RESNICK D: Patterns of paravertebral ossi- ern descendants. theoretical benefit. fication in the prehistoric sabertoothed cat. While this might suggest a relationship Amer J Roentgenol 1987; 148: 779-82. or at least a direct correlation of body Acknowledgements 11. MCEWEN C, DITATA D, LINGG D: Ankylos- mass with spondyloarthropathy occur- We wish to express our appreciation to ing spondylitis and spondylitis accompanying Drs. Laura Abraczinkas, Jon Alexan- ulcerative colitis, regional enteritis, psoriasis, rence, examination of the full record of and Reiter’s disease: A comparat ive study. this almost pan-mammalian disease der, Roy Carlson, Ken Carpenter, Ame- Arthritis Rheum 1971; 14: 291-318. (17) suggests otherwise.The frequency o l e re Coeri o m o n , Melissa Mora l e s 12. G O F TON JP, BENNETT PH, SMYTHE HA, of spondyloarthropathy is actually in- Cogan, George Culver,Ted Daeschler, DECKER JL: Sacroiliitis and ankylosing spon- Mary Dawson, Francis Desmedt, Lize dylitis in North A m e rican Indians. A n n distinguishable among monkeys, from Rheum Dis 1972; 31: 474-81. the squirrel-sized Callithrix to howler De Wet Bro n n e r, Judith Ege r, M a rc 13. D U R BACK MA, S C H U M ACHER HR, E D E L- or cap u chin monkeys (18). The fre- Frank, Wolfgang Fuchs, Eugene Gaf- S T E I N G: A b n o rmalities of the sacro i l i a c f n ey, Linda Gord o n , John Hey n i n g, joints (SIJ) in diffuse idiopathic ske l e t a l q u e n cy of spondy l o a rt h ro p at hy in hyperostosis (DISH):Demonstration by com- chimpanzees (28%), gorillas (20%) and Thor Holmes, D ave Ja n i ge r, C h e ri puterized tomography (CT). Arthritis Rheum orangutans (17%) also does not (19-21) Jo n e s , James Kitch ny, Tom Lab e d z , 1988; 31: S102. support a body mass relationship. The M e ave Leakey, B ryn Mader, L a rry 14. DIHLMANN W: Diagnostic Radiology of the M a rt i n , Ti m o t hy Mat s o n , C a n d a c e S a c roiliac Jo i n t s. Stuttga rt , G e o rg Th i e n e frequency in bears (equivalent in polar Verlag, 1980. and sloth bears) is independent of M c C a ff rey, B ruce McFa dd e n , C h u ck 15. RESNICK D, N I WAYAMA G, GEORGEN T G: size/mass (22). The frequency in Euro- M e r b s , Nina Mudida, Guy Musser, Degenerative disease of the sacroiliac joint. pean bison is actually signifi c a n t ly Marta Pagesi, Darlene Patterson, Bob Invest Radiol 1975; 10: 608-21. P u rdy, A l b e rt Saunders , Kevin Sey- 16. ROT H S C H I L D B M , P OT E AT G B, W I L L I A M S greater than in the much larger Ameri- E, CRAWFORD WL: Inflammatory sacroiliac can bison (17). The evolutionary ad- mour, Sally Shelton, Bill Stanley, Hans joint pat h o l ogy : E va l u ation of ra d i o l ogi c va n t age of spondy l o a rt h ro p at hy still Dieter Sues, Duane Schlitter and assessment techniques. Clin Exp Rheumatol remains elusive, though quite impres- Richard Thorington for their assistance 1994; 12: 267-74. in facilitating our examination of the 17. ROTHSCHILD BM, ROTHSCHILD C: Spondy- sive. A structural support contribution l o a rt h ro p at hy as a transmammalian pheno- to mass effect, however, seems unlike- collections they manage and to Charles m e n o n , rep ro d u c i ble in its manife s t at i o n s ly. Kibiy, Ogetto Mwebi, and Ezekiel Sav- across species lines. J Paleopath 2000; 11: Osteoarthritis proved to be a relatively ala for assistance in the examination 103-4. process. 18. ROTHSCHILD BM, WOODS RJ: A rt h ritis in ra re pro blem in Pe ri s s o d a c t y l a , i n New World monkeys: Osteoarthritis, calcium agreement with observations of other pyrophosphate deposition disease and spon- free ranging mammals (4-9). Exostoses References dyloarthropathy. Intl J Primatol 1993; 14:61- were also rare. The very low frequen- 1. ROTHSCHILD BM, WOODS RJ: Character of 78. p re-Columbian North A m e rican spondy l o- 19. ROTHSCHILD BM, WOODS RJ: S p o n dy l o- cies of fra c t u res and infection may arthropathy. J Rheumatol 1992; 19: 1229-35. a rt h ro p at hy in go rillas. Semin A rt h ri t i s reflect a low population occurrence or 2. RESNICK D, N I WAYAMA G: D i agnosis of Rheum 1989; 18: 267-76. s i m p ly poor surv ivo rship. A n i m a l s Bone and Joint Disord e rs. Philadelphia, 20. ROTHSCHILD BM, WOODS RJ: Reactive ero- with fractures or infections may have Saunders, 1988. sive arthritis in chimpanzees. Amer J Prima - 3. ROTHSCHILD BM, MARTIN L: Paleopatholo - tol 1991; 25: 49-56. quickly succumbed and therefore were gy: Disease in the Fossil Record. London, 21. ROTHSCHILD BM, WOODS RJ : Inflammatory not available for analysis. Stress frac- CRC Press, 1993. arthritis in Pongo. J Med Primatol 1996; 25: t u res appear to be ex t re m e ly ra re in 4. ROTHSCHILD BM, WOODS RJ: Spondyloarth- 414-8. f re e - ra n ging Pe ri s s o d a c t y l a , for they ro p at hy in go rillas. Semin A rt h ritis Rheum 22. ROTHSCHILD BM, WANG X-M, CIFELLI R: 1989; 18: 267-76. Spondyloarthropathy in Ursidae: A sexually were observed in only one individual Ð 5. ROTHSCHILD BM, WOODS R J: S p o n dy l o- t ransmitted disease ? N ational Geograp h i c in a Miocene Peraceras. arthropathy as an Old World phenomenon. Research and Exploration 1993; 9: 382-284.

632