HARVARD UNIVERSITY

LIBRARY

OF THE

Museum of Comparative Zoology

P.J-1.

BULLETINS

OF

AMERICAN PALEONTOLOGY

VOL. XXXIII

1949- 195]

MUS. COMP. ZOOL LiBRAfiY

)CT 1 4 1952

HARVARD UNIVERSITY

Paleontological Research Institution

Ithaca, New York

U. S. A.

,-5

0

MOS. COMP. ZOOl. LIBRARY

OCT X 4 1352

HARVARD UNIVERSITY

CONTENTS OF VOLUME XXXIII

Bulletin No. Plates Pages

134. Polychaete Annelids from the Devonian of Parana,

Brazil

By Frederico Waldemar Lange 1-16 1-102

135. Review of Anticlimax, with New Tertiary Species

(Gastropoda, Vitrinellidae)

By Henry A. Pilsbry and Axel A. Olsson 17-20 103-124

136. The Living: Cypraeidae of the Western Hemisphere

By William Marcus Ingram 21-24 125-178

137. Some New Species of Carboniferous Crinoids

By Harrell L. Strimple 25-29 179-218

138. Preliminary Notes on Ocala Bivalves

By Gilbert D. Harris 30-42 219-272*

139. Miocene Stratigraphy and Paleontology of South-

western Ecuador

By Jay Glenn Marks 43-51 271-432*

Author Check List I-XVI

Bull. Amer. Paleont., v. 1-33; Palaeont. Amer., nos. 1-24

*Note: Page numbers 271 and 272 are repeated in Bulletins numbers 138 and 139.

PAGE

LINE

27

I

107

36

256

22

308

33

375

2S

401

4

ERRATA

Read Aglaurides for Agaurides

Read locklini for lockini

Read vicksburglana for -vuksburgisna

Read Born'ta (Temhlornia) triangulata (Anderson and Martin) for Bornia (Temhlornia) triangulata Keen

Insert Genus POTAMroES Brongniart, 1810 below Family CERITHIIDAE

Read cruziana for cruiziana

^

BULLETINS

LfSRASV

JUL 18 l%9

Url?VEaSiTY

OF

AMERICAN

PALEONTOLOGY

VOL. XXXIII

NUMBBR 134-

1949

PALKONTDLOtiJCAL HesEAKCU InsTITLTIOJ

Ithaca, iVnTv York U. S. A.

MUS. COMP. ZOOL LtSRARY

JUL 18 \%^

HARVARD UNIVERSITY

BULLETINS OF

AMERICAN PALEONTOLOGY

Vbl. 33

"p

Z 1 - i^

No. 13^

POLYCH^TE ANNELIDS PKOM THE DEVONIAN OF PARANA', BRAZIL

By Fredekico Waldemar Lange

Museu i'aiauaense, Ouritiba, Parana

June 1 1, 1949

Paleontological Research Institution

Ithaea, New York

U. S. A.

f^' \ZOOL

JUL 18 19^9

TABLE OF CONTENTS

as>5('inblai>c's

'CIlslS .11 itll-C

Alistract

Jiiti'odiietioii

Aekuowledginciits

Scolecodoiits

Articulate ja\ Fossil analysis .

Pauliiiites

Pauliiiitt^s jiaraii, Articulate jaw ap

Asseiublaycs

Dental elements

Mandiltles

Conipai'ison Occurrence Dimensions

Type

Carriers

Paleontologic comparison

Neontologn- comparison .....

Occurrence

Type

Right forceps

Paleontologic comparison

Neontologic comparison

Left forceps

Paleontologic eomparison

Neontologic comjiarison

Right dental plate

Paleontologic (•:imparison ....

Neontologic comparison

Left dental jdate

Paleontologic comparison

Neontologic c-omparison

Unpaired i)iece

Paleontologic comi)arison

Neontologic comjjarison

Right i)aragnath

Paleontologic comparison

Neontologic comparison .'.

Left paragnatli

Paleontologic comparison

Summary of characters

Variation

Variability of the shape (Plate 15)

Variability of denticulation (Plate 1:? and 14)

Occurrence

T>

ypes

Discussion

Taxonomy

Comprehensive comparison

18

b 6 1-2 14 15 16 17 ■25 25 25 26 27 27

27 28 28 28 28 30 31 32 33 33 33 34 35 36 37 37 37 38 39 39 40 40 40 41 41 41 42 43 46 47 48 48 52

Table of Contents Concluded

Neontologic comparison 58

Conclusion 62

Methods employed 63

Appendix 64

The Devonian of Parana 64

Historical 64

Stratigraphy 66

Bibliograpliy ." 69

Plates 73

LIBRARY

JUL IS 19^9 umm

. UNIVERSITY ^

POLYCH^TE Annelids fr6m the devonian

OF PARANA, BRAZIL

By

Frederico Waldemar Lange

Museu Pai'uiKU'usi', (_'uritiV)a, Paiani'i

Abstract

Several aiticulatc aimeliil Jaw apparatuses are described from the Lower Devonian Ponta Grossa shale of the State of Parana, Brazil. Some are complete with all the maxillary plates and the mandibles preserved in their natural position. The mouthi)arts comprise one pair of ventral mandibles and the dorsal maxillary assemblage consisting of one asymmetrical pair, each of carriers, forceps, dental i)lates and paragnaths, plus one unpaired piece. There is considerable individual variation in the buccal structures; tills may be partially attributable to ontogenetic variation in succeeding moults. Other variations are clearly accicLental during fossilization. Many hundreds of dissociated seolecodonts are found in the same shales. Because of a complete intergradation between the various assemblages and between their individual parts and the dissociated seolecodonts, all have been referred to the same species, whicli neontologic and paleontologie comparisons establish as new and requiring a new genus, for which the binoiuial Paulinites paranaensis Lange was created in 1947. The fossils are assignable to the polychaste superfamily Eunicea, but no previous family seems adequate to accomodate them ; wherefore the new family Paulinitidae Lange, 1947.

INTRODUCTION

This paper is the Ent^lish versicjn (jf a paper pubHshed in Portuguese in Brazil in 1947 {Anelidcus poUquctas dos jolhelhus Devonianos do Parana, Arquivos do Museu Paranaense, vol. 6, art. 5, pp. 161-230, pis. 17-32, Curitiba, Parana, Brazil, Septem- ber, 1947). The principal observations of the original work are herein contained and in addition a short note concerning the Devonian occurrence in the State of F^arana.

Bulletin 134

ACKNOWLEDC^MENTS

1 he writer is happy to acknowledge his gratitude to Dr. kenneth !•.. L aster, of the L'niversity of Cincinnati, who not only SL:';'geste(l and encouraged the translation of this paper, but also kindly helped in the revision of the I'Lngiish text, and greatly as- ?;isted in seeing the paper through the press. The writer is also grateful to the Board of Directors of the Aluseu Paranaense, oi L"uritil)a. Parana, who authorized the publication of this paper and li:i\e jjrovided financial assistance toward the cost of the illustra- tions. To the Palaeontolcjgical Research Institution, the writer is indel)ted for the privileges of ])ublication and to Professor (i. D. W.arris and Dr. Katherine V. Palmer fcjr no small labor in seeing the bulletin through the press. Further thanks are also due Dr. Ernesto Marcus and Dr. Paulo Sawaya. of the Faculdade de Filosofia, Ciencias e Letras of the Iniversity of Sao Paulo, for valuable bibliogra])hic assistance; to Dr. 1{. R. Eller. of the Carne- gie Museum. Pittsburg, for his important scolecodont papers; and to Rudolf B. Lange and Carlos (iofferje, of the Museu Paran- aense, for their kind loan of Recent annelids from the Brazilian littoral. Consideralile credit also goes to Rev. Ricliard Wagner, C.SS.R., for his painstaking revision of the English text.

SCOLECODONTS

Beginning in the early Paleozoic, the worms occupv a signif- icant place in ancient faunas. Not only were they present, and often in great numbers, but, what is more striking, the worms of certain F'aleozoic occurrences are ]iracticall\- indistinguisha1)le from modern representatives. This is especially true with re- s]iect to their hard parts. This perserverance of form and struct- ure, which allowed them to attain modern times without any apparent "necessitN "' for great modification, would seem to indi- cate that we treat of a liomogeneous group of organisms, well adapted to its environment and aliundanth alile to cope with the succession of competitors that appeared in the course of time.

The general lack of hard parts makes the worms poor material for fossilization. Despite this, several ver) fine records of essen-

Brazilian Devonian Polych.t:tes : Lange

tiall}- complete fossil worms are known. Most famous, of course, are the beautiful specimens discovered by Walcott in the (."anadian Middle Cambrian (Burgess i'assj. l-'Jiler's hitiiicites from the ^'olnhofen ( lurassic) lithographic limestone of Bavaria is also another such rarit\ ; likewise the uniipie specimen described by Carlotta J. Maury U9-7) ^n^m the Itarare (Carboniferous) varvites from Anitapolis, State of Santa Catharina, Brazil, as Oliveirania santacatharinco .

I'"ossil trails which are generally attributed to worms are con- siderabh- more abundant in several cf)untries. They are best known in the Paleozoic. The writer (1942) described such a \esmiform trail fr(jm the l'"urnas sandstone (basal Devonian) of Parana, which incidentally ha])pens to be the ijldest fossil in the state so far. The most famous Brazilian trails, probably largely of worms, are those of the Itarare (Carboniferous) varxites of Santa Catharina, Parana, and Sao Paulo. These are ver\- abun- dant in several places.

^ addition to the foregoing t>pes of fossil worms, the minute j 1WS of annelids occur at many horizons, beginning with the earl\- Paleozoic. Such remains were first described in connection with the problematic conodonts ; however, their true nature was soon recognized and the term "scolecodont" coined for them by Croneis and Scott in I033 -^^ for the conodonts. the\- are still proble- matical. Although they were assigned to the worms bv Zittel and Rohon (1886) after a careful com|)arative stud}-, thev are usualh' considered b\- modern authors as attributable to fishes. The\- are, however, considered b\- other current workers as possibl\- wholl\ or partiall}' remains of annelids, crustaceans, gastropods, arach- nids, etc.

On the other hand, the very .close similarit\' between the jaws of Kecent Polychaeta and the scolecodonts leaxes no room for doubt a'^ to the nature of the latter. The writer has taken the occasion to reverify this identity of form, both from a survey of the litera- ture and from dissections of modern polych?etes from the coast of Paran.-'i and Santa Catharina.

Pander (1856) was apjxirently the first to call attention to the scolecodonts (Silurian oi Russia) without recognizing their true nature. Among the older works on the subject, undoubtedly the

Bulletin 134

most important are those of Hinde (1879. 1880, 1882, 1896), who described th€ ,copious material whicli he had collected from the Paleozoic of North America, Great Britain, and Scandinavia. Having verified the structural similarity to the Recent worms, he created many of the generic names used for the scolecodonts and established the custom of coining names for the fossils derived from the su]:)posed modern derivatives or analogues, as for ex- ample, AraheUitcs from .-Irabclla, Ociioiiitcs from Ocnona, and Glyceritcs from Glycera.

Until recentl}' no great importance was attached t(j these fossils ; but of late their study lias been greatly augmented, especially in the United States, with the discover}- that the}- are good horizon markers. Due to their minuteness they are found intact in even the smallest well-cores and thus serve as excellent micro-guide fossils. The chief modern studies on scolecodonts are therefore tlie works of the North American paleontologists, E. R. Eller, C. L. Staufifer, and others.

The great stratigraphic value of the scolecodonts, like the conodonts, lies in the fact that the} are characteristicall}- Paleozoic, and have proved useful where certain other commonl} employed micro-guide fossils, such as the Foraminifera, are scarce or even unknown. Incidentall} , it is both interesting and perplexing to note the ap]iarent absence of the scolecodonts from Mesozoic and Cenozoic sediments. This is especially curious, considering the Paleozoic abundance of the group and the great array of modern marine polychaetes. This great hiatus in the fossil record is apparentl}- bridged so far onl} b}' the unique worm imprint from the Solnhofen Jurassic.

This lack of post-Paleozoic scolecodonts can hardly be attributed to Iftck of search. Because of economic implications, these are the ver}' strata which have been subjected to the most rigorous exam- ination b}- micropaleontologi*ts. Considering the vast array of ostracodes and Foraminifera found in these beds, it seems most unlikely that scolecodonts, perchance associated with these fos- sils, should have escaped notice.

The scolecodont material described in this paper was col- lected by the writer in the Ponta Grossa (Lower Devonian) shale at various localities in the State of Parana, Brazil., It represents several year's work, for the Parana scolecodonts are never very

Brazilian Devonian Polycii^tes: Lange

abundant and do not occur at specilic horizons. They were, thereto! e. more or less accif'entally encountered in connection with general fossil collecting in the slialc. In the end, over a thousand "isolated scolecodonts were amassed ; of these several hundred \,ere removed from the matrix for detailed stud}-.

The hrst notice of the discover}- of scolecodonts in Brazil was gi\en in a note by the writer appearing in a paper by Paulino Franco de Carvalho (1911 ). where also some of the nevvl\- discov- ere ! specimens were illustrated. Although they were again men- tioned briefly b}- the writer (1943) in a paper describing new in- \'Citebrates from the I'onta (irossa shale, their detailed stud\' and •■'escription hafl to await the assembling of the necessary neon- tclogic and paleontologic literature.

The scolecodonts are irregularly scattered throughout the Ponta Grossa shale. They usually occur as dissociated minute plates of a uniform black color and high gloss and present -a most lietcrogeneous array of shapes. Their uniformity of color is in cC'i trast to the modern polycluete mouthparts where the portions imbedded in the tissue are somewdiat paler in color, translucent, and horn\- in aspect. Our material varies from a fraction of a mdllimeter to a maximum of 2.32 mm. in length. Its composition appears normal. This was given by Croneis (in Twenhofel and Shrock. 1935'! for the scolecodonts in general as about 50% vola- tile matt-er and about 1.^9^ silicon dioxide.

Due to their small size and black color, the scolecodonts are difficult to observe on the dark-colored Ponta Grossa shale. This 's especially pronounced on fresh exposures. In somewhat weathered material, wdiich takes on a light grayish tone, the tiny fc'ssils are nK)re conspicuous I ntil the e\e becomes accustomed, they are easily overlooked, ddiis probably is the explanation for t'xiv lia\ing h^een overlooked for so long among the well-knowm -nd ,-d)n!"i'ant n^acro-fr ssil arrav of the Ponta Grossa shale com- prised of brachiopods, pelecypods, trilobites, etc. Once discovered, however, tliey have been verified in nearly all the principal expo- sure-. of the sliale, from the Caniii Kiver in the south to T.-'.'^uariai\a, near the northern limit of the Parana Devonian outcrop

10 Bulletin 134 10

Tlie hnest specimens were found in a thin layer of light gra\', clayey shale at Santa Cruz, in the district of Palmeira. The\' are more abundant elsewhere in the Parana Devonian but have never thus far been fcnind so satisfactorily preserved. While they are often \ery abundant in the dark arenaceous shales, this medium usuall} yields crushed specimens which can onl}' be ex- tracted with great difficulty. Articulate specimens are very rare.

The Recent i)ol}chjete superfamil\ h'unicea is especiall\' inter- esting for comparison with the fossils. In this group the jaw apparatus is located in the distal part of an eversible pharynx. The various buccal plates are so attached to the pharynx wall as to function effectivel\- in seizing food when the pharynx is everted through the mouth. The phar_\nx is an elongate, pos- teriorly directed pouch (when at rest > and its "inner" wall is covered with a chitinous cuticle which is coetaneous with the buccal lining. At certain places this chitinous lining becomes th.icker and forms salient folds. The consolidation of these sal- iences results in the various i)haryngeal jaw plates.

In the Eunicea the complex jaw apparatus consists of a bi- laterally arranged series of plates which are held together and articulated by muscular tissue. The apparatus is divided into two distinct f)arts, which also are characterized 1)y distinctive chemical compositi(jn. First, there is the ventral pair of mandi- bles which are distinctly calcareous. The two elongate mandi- bular shafts are generally fused along the median line, and, con- trary to the maxillary plates, onh' their anterior liord^r is pro- truded through the mouth. The other part is comprised of the dorsal buccal armature consisting of a complex of chitinous max- illar}- plates. These are mostly paired structures of character- istic forms. The maxillee are generalh- numbered from rear to front (I-IV, or more), and are denominated:

T. Forceps or pinchers : a pair of toothed or edentulous jaws, united and supported b\- a basal pair of "carriers." IL Dental plates : a generall}- denticulate pair of plates. in. Unpaired piece: a usually denticulate plate on the left side;

not always present. IV. Paragnaths : one or more pairs of minute denticulate dis- tal plates.

11 Brazilian Devonian Polych^tes : Lange ii

111 the >c\er;il represcntatixes of the l*"unicea the maxilhe of c.'ich pair are s)ninietrical ; in others there exist slight differences ;'.s to shai)e and nuini)er of denticles occurring- on the inner niar- ;;in of the two plates of a pair. The form and nnm])er of max- illary plates and their arrani^ement in the jaw apparatus are fundamental criteria in the classification of the luinicea.

I'ach of thee many mouth-part pieces corresponds to a scole- codoiit. of course. Despite the .^enerall} dispersed and detached occurrence (,f these fossils, it is not difficult to identify them as t" li'jiit or left side of tlieir orii^inal jaw apparatus, hecause the x.^'jiou-- plates are ,^enerall_\ inwardl) curved and/(jr possess denticles on the inner marf^in and a distinct cjpening or fossa for rniscle attachment on the inferior side. This last serves for 1 'etiniti\ e orientation.

in addition to the principal maxillae, there may also occur !.,;()ups of small accc ory plntes, either lateral (jr anterior. These are difficult to distin.ijuish, and easil\- confounded with broken fra.i^ments of the standard ])lates. They are of minor taxonomic si.i^nihcance.

Due to the difl'c^crt cliemical nature of the two principal [la.rts of the jaw a])p:ir:;tus (calcareous mandibles and chitinous maxilhc) usuall_\' onl_\ ore ov the other structure is preser\e(l in an\- particular fossil medium, though mold traces of the less permanent element can scmctimes be distinguished.

I'dilers (1867-70), for example, showed that only the calcar- eous mandibles w^ere actually preserved in connection with the soft-part fossils of annelids in the Solniiofen lithogra[)hic lime- stone, although it was jxissible to distinguish the molds of the associated maxilhe. The Snlnhofen conditions were exception- al, however, for in nearl_\- e\ery other .case so far known of artic- ulated mouthparts, the mandibles disa])pear completel_\- during fossilization, and onl_\- the chitinous maxilhe remain. In this the Parana Devonian shales are no exceptitjn. There, to(j, the onl} common evidence of the mandibles are impressions, often red-stained molds, i)rol)al)ly discolored by the decoin])osition of the mandibles themselves.

L'p to the jiresent time, only the Parbados and Santa Cruz lo-

12 Bulletin 134 12

calities of the Parana Devonian have yielded a few intact mandi- bles ; but even these are so poorly preserved and so brittle that they break at the slightest touch. It is thus most difficult to re- move them from the matrix for stud}-.

The chitinous maxillary plates, on the otlier hand, are usual- ly beautifully preserved, whether in articulated assemblages or in transported material. Clearly they were much more resistant both to chemical and physical factors.

Probably various factors were operative in causing the almost universal dispersal of the scolecodont mouthparts, thus making assembled jaws among the rarest finds in paleontology. Appar- ently upon the death of the worm the jaw articulation was so fragile that upon decomposition of the tissues the slightest move- ment of the water was sufficient to scatter the buccal plates. Some of them may even have floated automatically because of putrefactive gasses held in their cavities and openings. Hinde (1882) presented the interesting hypothesis that this almost uni- versal scolecodont disruption ma_\ possibl\' be attributed to scav- enging ostracodes, the shells of which he often noted in scole- codont associations. The Parana scolecodonts are similarly abundanth' associated with ostracofles, thus corroborating HindeV observations elsewhere.

Ectiysis or moulting may be another factor in the dissociate nature of scolecodonts. Heider (1922) says that in the modern genus Staiiroccplialits tlie maxill?e are moulted and replaced by new plates localized in a submaxillar alveolar fold. He also called attention (1924) to a similar observation b}' Ehlers for Eunice where the chitinous jaw apparatus had been expelled, and onl\- the cellular materials which originally filled the maxillary openings were retained, subsequently to secrete new plates. Neither of these authors could determine the exact nature of the moult, whether there is simultaneous ecdysis of the whole as- semblage, or if it is a differential moult, plate by plate. Either mode, and especially the latter, would effectively explain the commonly dissociate nature of the scolecodonts.

ARTICULATE JAW ASSEMBLAGESi

Di«ssociated scolecodonts are the rule wherever they are known.

13 Brazilian Devoxian Polych^tes : Lange 13

L'p to the present time, only tour Paleozoic and one Mesozoic discoveries of articulated polych^ete jaw assemblages have been described. All of these are incom[)lete, either showing a jjartial displacement of the elements, loss of plates, or sei'ious damage. L'nder the head of [)aleontologic comparisons in the latter part ot this i)aper. there api)ears a brief descrii)tion of these live as- semblages. There it can Ije seen that the fragmentary nature 01 all precluded complete analysis and definite biologic assignment.

Considering these tacts, the Parana discovery of several es- sentially intact iaw assemblages is the more remarkable. The_v are apparently unic|ue and appear to ha\e considerable paleonto- Idgic significance both from the standpoint of the polychcctes and : Iso for the problematical seolecodonts.

This Brazilian Devonian articulated material comes from Santa Cruz, Parana. The assemblages were encountered spor- adically among many detached scolecodonts. .Vlthough the}' have lieen intensively sought over a period of \ears. onl}' 12 articu- h'ted sets have sf) far been recovered. C?)f these, seven are pre- served in their natural unaltered position ; the remaining five have been somewhat shifted and also lack certain of the maxil- la r\- plates. In two of the first -^even even the ventral mandibles are preservetl in their natural position. The simultaneous oc- currence of both mandibles and maxillary plates in these assem- blages would seem to indicate that the animals had been interred before decomposition commenced thus forestalling dispersal of the niouthparts. A few additional scattered mandibles have turned up at Santa Cruz. That they should occur in dissociation from the maxillae is much to be expected considering the fact that they are not intimately associated in the pharynx with the dorsal maxillary plates.

The Santa Cruz locality has also yielded so far in addition : two pairs of articulated forceps with their corresponding car- riers ; two isolated forceps, also with carriers ; two pairs of mandi- bles ; three pairs of carriers, joined in their life position ; and al-

1 The term "assemblage"' is employeil liei'e for tlie articulated jaws comprising the buccal armature of the i)olyclia^tes in tiie same manner as already used in conodont studies bv Scott (.Journal of Pnlcontcdofjv, vol. 8, p. 448, 1934).

14 Bulletin 134 14

S(t mail}- Imiidrcd isolated scolccodonts, representing;' all the [)late ff'rnis found in the assemblages. This array of intact material thus makes Santa Ci'iiz the most si.ynihcant localit} as yet known for sc(jleC(A'ont studies.

FOSSIL ANALYSIS

As will he demonstrated in the discussion at the end of the sjiecies desciiption, no close similarity exists hetween the articu- late jaws from the L>e\'onian of I'arana and the ti\e pre\ious discoveries. Moreoxer it will I)e l)rought out that the isolated scolecodonts so far descriljed do not constitute a sound basis for classification since their real biolo,i;ic affinities cannot be deter- mined.

Conseqr.entl} . the identilication and taxonomy of our new fos- sil assembla_':;cs has had i)erforce to be based primarily on neon- t;;lo;;"ic CMiipaiisons. Amont;- the i)ertip.ent moc'ern worms the sh,a])e and disposition of the maxillary plates in the articulated .i[<paiatus, a.-, enibled in their life position, pro\ide all the neccs- sar}-. and traditional, elements for classification. Soft i)arts are unessential for this.

The ."general organization of the Parana articulated I'aws indi- cates alii^nment with the modern superfamily I'Lunicea. Herein the number and arrans^ement of the maxillary plates show ii'reat- er similarit}- to the modern families Onuphida.^ and Eunicidai tlian to any others. However, the denticulate forceps of the I'arana material precludes assignment to either of these, the most si.ijnilicant trait of bcjth being edentulous inner forcep mar- gins. Lven more impressive differences exist between the fos- sils and other existing families. Wherefore the establishment ( 1947) of a new familx' to accommodate the new Parana ma- terial, the detailed description of which follows:

Phylum ANNELIDA

Class CHAETOPODA

Order POLYCHAETA

Superfamily EUNICEA

Family PAULINITIDAE Lange, l'J47-

Type of the family, Paiilinilcs paranacnsis Lange, 1947.

- Date of (ji-ig'iiial l'ui'tu>;iK'st' ])UlirK';itioii oL' t'::c t-niiti/nt dl' tiity jiapiM' jr. tlu> Arqiiivos do Musovi I-'aiaiiaea.sc. C'uiitii)a, Brazil, ISri>ti'iiiln'r. li>17.

15 Brazilian Devonian Polych.^tes : Lange 15

For the diagnosis ot the new laniil}- 12 articuhite jaws wei'e a\ailahle. I'he complete assemblages show two ventral manihhlcs and seven dcjisal maxillai'v i)lates groujied on the l)asal jiair of carriers. The i)rincipal traits of the family I'aulinitidic are the lollow ino- :

I. Mandibles inarticulate; shafts inwardlx- curved. J. Alaxilhe in asymmetrical pairs; disposed around the lorceps when withdrawn.

3. Carriers short, smooth, slender, with curved margins; arti- culated at a swelling of the anterior border; without ventral nx'dian piece.

4. Forcei)S as\ mmetrical ; denticulated along the entire inner margin ; with a large anterior hook ; right forcep with a detachable basal plate.

5. Dental plates asymmetrical; denticulate; with a shank on the outer mar'^in ; plates smaller than the forceps.

6. Un])aired jnece denticulate; located on the left side of the a[>paratus.

7. Taragnaths asymmetrical ; denticulate.

As will ])e brought out in tlie hnal discussion, these characters recall the families ()nuj)hid;c and luinicicUe of the nioclern fauna; !io\\e\er, certain distincti\e cliaracteristics do not allow assign- ment to either of these families.

Genus I'AULIMTES Lange, 1947

(lenot}i)e. Puiiliiiites paraiiaciisis Lange, 1947. Lower L'evonian Ponta Grossa shale, of Santa Cruz, State oi I'arana, brazil.

Since this monot} pic genus shares the species traits, described below , onl\- the principal generic features are given at this place.

I'roboscidial armature ccmsisting of one pair of mandibles, seven maxillary plates and a pair (jf carriers. The arrangement is as follows :

On the ventral region one pair (jf long, conical mandibles which are not directly joined or articulated ; these have elongate frontal l)ieces .connected obliquely with the shafts which are long and narrow and taper to an acute, inwardly curved, posterior extremity.

Articulated maxillar} pieces on the dorsal side consisting of : two short posterior slender carriers, without ventral median piece,

Bulletin 134 16

and with the inner margins of their shafts free and incurved; jo!i:e I onl\ at a shght overlapping of the thickened anterior hor- de:" ; upon the cariiers the other maxillary pieces are arranged in a semicircle. The asymmetrical forceps are falcate and end aiiterioilv in a stout fang or hook; a variahle number of small, b:ck\;::rd directed denticles extend along the whole length of the i:i:er niaigin; a small, oblong basal plate fills an angular bight of the poste ior margin of the right forcep. Beneath the forceps, and nearl}- entirel}- covered by them, are found two smaller, ir- e"ularl\ ''crtate and asymmetrical dental plates of subtriangular shape, and with a medium-sized shank on their outer margin. L'nder the left dental plate occurs an elongate and subtriangular ■mpaired piece, with irregular denticles on the inner margin. Two small, inegularl)' oblong and asymmetrical paragnaths are dis- posed 1 bliqiel . at the anterior region of the articulated jaws; their minute c'enticles point outwards In the withdrawn appara- t::5, the maxillary plates are disposed around the forceps.

The ai7i:dties of Paiilluifcs to other genera are taken up at the end of the species description.

The name f^aiilinifcs is giv'en in honor of Dr. I'aulino France i-V Carvalho wlio first gave notice (1941) of the writer's discov- e ■}" of these fossils.

Paulinites paranaensis Lange, 1947 Plates 1-1.t; text figs. 9-11

T!iis analysis is based on the u aforementioned articulated j:.w apparatuses, two of which it will be recalled we:e complete, even to the presence of the mandibles. In addition man_\- hundred isolated scolecodonts from the same Devonian shale of Parana, all presenting the same characteristics as the various parts of the assembagcs, have been assigned to this same species.

Tlie articulate iaws preserve the same position seen in tlie buccal armature of the Recent FAinicea. Some of the fossil assemblages present the dorsal and others the ventral aspect in the matrix. Because of the overlapping nature of the undisturbed plates, it has been necessary to remove some of the jaws from two of the assemblages in order to expose underlying plates.

First, let us look to the articulated apparatuses ; then to the different corresponding pieces.

Brazilian Devonian Polych^tes: Lange 17

AETICULATE JAW APPAKATUSEJS

The complete articulated jaws, when preserved in their natural position and observed from the dorsal side, present the following arianL;ement :

At thi.' posterior region of the assemblage two sxmmetrical carriers are disposed side by side, in nearl}- parallel lines, joined oiil}- at an overlapping of the thickened anterior border. This frontal swelling of the carriers hts perfectly in a depression of the posterior margin of the forceps and acts as a base for their flexible articulation.

In this manner, supported by the carriers, the forceps occur in a ])alf-opened position, converging at the base in a \' ; the preser- vation of the forceps in this position is accidental, for being articulate, they could open freely up to a certain point and close completel\- in order to seize prey.

L'nder the forceps, and following the same orientation, occur the dental plates. The posterior margin of these plates ends upon the upper heavy border which surrounds the fossa of the forceps. Since the dental plates are smaller than the forceps, they are nearly entirely covered by the latter, so that onl}- a few denticles and a small part of their frontal margin remain visible.

The unpaired piece occurs under the left dental plate, ending posteriori)- at the point where the inner border of the fossa of the dental plate forms a bight ; the unpaired piece is completely concealed by the dental plate and can be disclosed for examination only b\ the removal of the left maxillse I and IL

The small paragnaths are disposed at the anterior end of the articulated jaws, with which the\' are not directly connected. Since in the living animal the\- occup}- a somewhat oblique and twisted position, it happens that in the fossil assemblages they are generall)- preserved with their denticulate margin turned out- wardl}-, in contrast to the other maxillary plates, where the denti- cles occur on the inner oj)posed margins.

In two assemblages of Pauliiiifcs the writer further observed the occurrence of a small plate, provided with only one terminal denticle, situated over the paragnaths. Due to its minute size and isolated occurrence, it has not been possible to ascertain if

1 8 Bulletin 134

this plate represents an auxiliary parao^nath (maxilla V), which occurs in some Recent i^enera, or if it is only a fragment of another plate, accidently placed in this position. The elucida- tion of its true nature must he postponed until a more nearly [perfect specimen is availahle.

The arrangement oi the different maxilla? in the assemhlage may l)e ohserved more easil\- when viewed from the ventral side, as ma}' be seen in the corresponding figures.

The mandibles occur on the ventral side, beneath the maxillar}' plates. In one of the articulated assemblages they lie as an in- verted V, with the mandibular shafts separated posteriori}, and the nearest contact being near the frontal plates which are, how- ever, discrete. This separation of the shafts is accidental and may be due to the pressure of the overlxing plates during the process of fossilization.

In the second complete jaw apparatus (No. P. 104) one man- dible is somewhat displaced toward tlie right ; in the other con- jugated pairs of mandililes which were found dissociated from the apparatus ^the shafts lie nearl\- ]iarallel, converging slightl\' posteriorly.

There follows a detailed description of the several articulate assemblages of Pmdinitcs poronaensis. The numbers preceded by a P. are from the catalogue of the Museu F^aranaense, where, with two exceptions, the material on which this paper is based is stored. The specimens here given as P. 106 have laeen deposited in the United States National Museum, Washington, D. C, and P. 107 is deposited in the Museum of the Division of Geolog\ and Mineralogy of the F'ederal Government in Rio de Janeiro.

Assemblage P. 101. Holotype Plate 1, figs. 1-10

Articulated jaw apparatus, preserved in its natural position, the jaws showing the dorsal aspect (fig. 2), consisting of one pair of forceps, one pair of dental plates, one unpaired piece, and one pair of paragnaths. The pair of carriers, preserved on tlie oppo- site piece of the matrix, shows the ventral aspect (fig. 4). From this assemblage the forceps and the left dental plate were later removed to disclose the other underlying plates (fig. i).

The principal features of tlie different parts of this assemblage are as follows :

19 Brazilian Dfa'onian Polycit^tes: Lange 19

Right forcep (figs. 9, 10). Timer margin with 6 small frontal denticles under the hook. f(jllo\ved hy 1 1 normal, hackvvard directed denticles, which de- crease in size posteriori), terminating upon the hasal projection; l);isal plate i)reser\ed in its natural position. Length of the right forcep, 1.72 mm.

Left force]) (figs. 5, 6). Inner margin with () frontal denticles followed hy 11 normal, Ijackward pointed denticles which decrease in size ])ostcriorly ; h.isal region of the inner margin smooth ; the upper horder of tlie fossa of this specimen is somewhat crushed. Length of the left forcep. 1.68 mm.

Right dental plate (fig. i). Outer lateral margin with a stout shrmk directed ol)liquely hackward ; inner margin with i frontal denticle -|- I tooth -|- I intermediate denticle followed hy ij normal, po.steriorly decreas- ing denticles. Length of the right dental jilate. 1.30 mm.

Left dental plate (figs. 7, 8). ( )uter margin with medium-sized shank ; inner margin with 2 frontal denticles followed hy 12 normal (jnes which decrease in size toward the posterior. Length of left dental plate, 1.42 mm.

Lhipaired piece (fig. i). Plate slightlx' twisted due to compression; anterior margin cu.rved, ending in an external heak, inner margin with 2 frontal denticles -|- i tooth followed hy 5 normal, slightly inclined denti- cles decreasing in size |)osteriorl\-. Length of unpaired piece, C.60 mm.

Right paragnath (fig.i).

Denticulated margin with 12 denticles, the anterior ones jiointed, the posterior rounded and with a weak median furrow. Length, 0.72 mm.

Left paragnath (fig. I), Plate somewhat twisted and partially covered, only 4 denticles heing visihle. Length of the twisted plate, 0.39 mm.

Carrier (figs. 3, 4). Conjugated pair, with lateral part of the frontal thickening of the right carrier overlap])ing an equal ])art of the left one. The carriers have been preserved in their natural position, on the base

20 Bulletin 134 20

of the cast of the articulate jaws, o.n the counterpart of the matrix. Length, 0.94 mm.

Assemblage P. 103.— Symplerotype ■• Plate 3, fig. 2

Articulated jaw apparatus showing the ventral aspect, with the niandihles preserved ventrally in their natural position and cover- ing the maxilhe, so that the denticulation of the different plates could not be observed. Mandibles.

Pair conjugated at the anterior region and separated posterior- ly, showing the under side with, frontal plates. Length, 2.10 mm. Maxillae.

Covered by the mandibles, only a part of the forceps being visible, the right one with basal plate, the dental plates, a small part of the unpaired piece, the paragnaths, the right one being somewhat displaced and showing the dorsal side, with 11 denticles, and the right carrier, broken at the base. Since the frontal margins of the different plates were covered b}- the mandibles, it was not possible to measure the entire length of the maxillse.

Assemblage P. 102 Plate 2, tigs. 1-4

Articulated jaw apparatus, preserved in its natural position and

sliowing the dorsal aspect, consisting of a pair of carriers, a pair

■-■ kSiiiee in the holotype the mandibles iiad uol been preserved, the writer had to select another fossil j;:w apparatus lor tlie description of tne con- jugated mandibles; being in doubt about tne designation of this comple- mental type, Uie writer submitted the question to Kev. Jesus Moure, C. M. F., Director of the Section of Zoology of the Museu Paranaense, from whom he received the following explanation :

"I made a careful research to find a designation for the second specimen used to complete the diagnosis of the holoty^je which lacks some pieces, and based my stuily especially on the classical work of H. T. Fernald : On Type Nomenclature (Ann. Entom. Soc. Amer. 32 (4) : 689-702, December 1939). After running through the long list of 108 names employed for the designation of types of a species, 1 did not find one available for your case. Though the term ' allotypus ' ' has been use<i to designate a com- plemental ty])e, tiiis designation can no longer be employeil in its broader sense, because totlay nearly all authors use allotype to designate a specimen used as type for the desciiption of the opposite sex from tiiat of tae holo- type. Even Muttkowski, who proposed the term allotype in 1910, later substantiated the definition of this term and writes in 19.H8 : "By Allotype I meant to designate a specimen of the o^jposite sex regardless of date taken, place taken, or author describing it'. Since in your case we have to deal with the lack of pieces and not with difference of sex, it seems evident that a new name has to be coined to designate a complemental type of the holotype, and I would propose the term Symplerotype, which signifies exactly complemental type."

21 Brazilian Df.vonian Polych.^.tf.s : L.anof. 21

of forceps, a pair of dental i)lates, an unpaired piece, and the rij^ht ])aragnath. When the piece of shale was si)lit ()i>en, on one side remained the maxillary assemhla^e ( 1 '. 102, tii^. i) , on the otlier ine corresponding cast with the ])air of carriers and the anterioi' hrcjken part of the left force]) ( 1'. 102 a. tig. 2). The two forcei)s and the left dental ])late were later removed to expose the other maxillc'e, preser\ed in their natural ])osition (tig. 4). During the preparation, the hroken hasal portion of the left forcej) was lost, and the hasal ]date detached from the right forcep.

Tlie characteristic features of the different maxilL'c are the tdlhtwing :

Right forcep.

L'nder the hook 8 frontal denticles, followed hy S normal denti- cles, which decrease in size ])osteriorly, hasal plate detached dur- ing the preparation. Length, 1.64 mm. Left forcep.

With hroken basal portion, originall\- with 6 frontal denticles f(jllowed hy 8 normal ones on the whole extension of the inner margin. Length, i.('/)mm. Right dental ]ilate.

Anterior margin ending laterally in a stout shank ; inner margin with 2 frontal denticles + i tooth -{- i intermediate denticle followed hy i i normal ones, all ]:)ointed ha,ckward -Am] decreasing in size posteriorly. Length. 1.22 mm. Left dental plate.

C Hiter lateral margin with medium-sized shank; inner margin with 3 frontal denticles under the anterior tooth, followed hv 10 normal denticles. Length. 1.30 mm. Lnpaired piece.

I'^rontal margin slightly cur\ed. ending in an external heak : inner margin with 2 frontal denticles -[- i tooth -\- i intermediate denticle followed hy 5 normal ones. Length, 0.8] mm. Riglit i)aragnath.

Denticulated margin with 13 rf)unded denticles. Length. 0.61 mm.

22 Bulletin 134 22

Carriers.

Conjugated pair attached to the cast (fig. 3), overlapping at the anterior thickening. Length, 0.78 mm.

Asscmb.age P. 104 Plate 3, tig. 4

Complete articulate jaw apparatus preserved in its natural l)osition and sliowing the \entral as'pect, in which the maxillre are partially covered h}- the left niandible, so that it is not possible t(. measure the length of the maxillarx' plates or to count their denticles; this apparatus presents the following composition:

( )n the ventral i-egion the ])air of mandibles showing t\ye under side, the left mandible covering the jaw assemblage, the right one si mewhat displaced laterally.

On the base cf the maxillary assemblage occurs the pair of somewhat overlapped and displaced carriers, followed b}- the pair of forceps, the pair of dental plates, the unpaired piece and the two i^aragnaths. Since the frontal ])arts of all the maxillary plates are covered, it has not been possible to take their measure- ment. Length of the mandibles, 1-75 mm.

Assemblage P. lO.'j Plate 3, fig. 3

Articulate jaw apparatus, complete with all the maxillre, pre- served in natural position and showing the ventral aspect. Ex- cepting the ])aragnaths, the other maxillft" occur with their inner margins turned downwards so that the denticles are covered.

The dimensions of the different pieces are as follows:

Carriers _. 0.92 mm.

Forceps ^ 2.05 mm.

Right dental plate 1.50 mm.

Left dental plate 1.56 mm.

L^npaired piece ^_ ._ 0.83 mm.

Right paragnath 0.89 mm.

Left paragnath - 0.61 mm.

The right paragnath is turned inwardly and partially covered by the other plates, so that only 6 denticles are visible ; the left l)aragnath is preserved in its natural position, only slightly turned to the posterior, and has 9 denticles, of which the first one is broken.

23 Brazilian Devonian Polych^tes: Lance 23

Assemblage P. 10« Plate 4, fiK- ^

Deposited with the I'liited States National Museum, Washins^- t ,n, IX C.

Small articulate jaw ai)]iaratus jtreserved in its natural pcjsi- tion .and showin;^" the dorsal surface, in which t'le l'orce])s cover the other maxilke. The comitosition of tliis assemhla^e is the