xk3

MEMOIRS

OF THE

CAENEG-IE MUSEUM.

VOL. II. NO. 3.

FOSSIL AVIAN REMAINS FROM ARM1SSAN. By C. R. Eastman.

The history of gallinaceous birds is traceable as far back as the Middle Eocene, the earliest known genus being found in the Green River limestone of Wyoming. Tins is known by a single, nearly perfect skeleton, and seems to be closely related to the existing Ortalis. From the Upper Eocene of the Paris Basin have been de- scribed two species of Palssortyx, which comprise small pheasant-like birds, with the rostrum longer and less curved than in modern forms. Certain fragments occurring in the Quercy Phosphorites (Upper Eocene), and the calcareous marl of Vaucluse, supposed to be of equivalent age, are likewise referred to Pahcoriyx, and several species are also known from the Miocene of Southern France. The remarkable and highly specialized Hoaetzin, or Opisthocomm, of South America, whose habits and certain points of structure suggest considerable affinity with the Cuculi, is rep- resented by a closely similar genus (FHlholomis) in the Phospborites of Quercy.

The fresh-water limestone of Armissan, near Narbonne (Aude), by some geolo- gists regarded as of Upper Eocene, by others as of Oligocene age, has furnished re- mains of a single genus (Taoperdiz), which agrees somewhat closely with modern pbeasants ; and Phasianw itself occurs rather abundantly in the Middle and Upper Miocene, and also in the Lower Pliocene of Pikermi, where it is accompanied by Gallus. A species of fossil turkey (Meleagris antiqims) is known from the White River Oligocene of Colorado, the same genus being also represented in the Pleisto- cene of New Jersey, Pennsylvania and probably elsewhere in tins country. Under the name of Palseoperdix, three species of fossil pbeasants have been described from the Middle Miocene of Sansan (Gers). Gallus and Phasianus appear to be tbe only known Pliocene genera, but from the Pleistocene and cavern deposits of various

132 MEMOIRS OF THE CARNEGIE MUSEUM

parts of the world a considerable number of gallinaceous birds have been brought to light, the more important of which are enumerated in the following list.

Table Showing Geological Distribution of Gallinaceous Birds.

Middle Eocene. Galliuuloides wyomingensis Eastman. Green River limestone ; Uinta County,

Wyoming.

Upper Eocene. Palaeortyx hoffmanni (Gervais). Gypsum of Paris Basin, Montmartre.

" blanchardi Milne-Edwards. Gypsum of Paris Basin, Montmartre.

" sp. indet. Gypsum of Paris Basin, Montmartre. Filholornis paradoxa Milne-Edwards. Quercy Phosphorites.

" gravis " " " "

" debilis " " " "

Taoperdiz p>cssieli (Gervais). Lacustrine limestone ; Armissan (Aude).

" keltica Eastman. " " " "

(Jligocene. Meleagris antiquus Marsh. White River Beds ; Colorado.

Lower Miocene. Paheoiiyx cayluxensis Lydekker. Phosphorites of Cay lux (Tarn-et-Garonne).

" (jallica Milne-Edwards. Lacustrine deposits of Allier, and Phosphorites

of Cay lux. " brcvipes Milne-EdWards. Lacustrine deposits of Allier. " (V) phasianoides " " " "

" media " " " " {dcscrip. insuf.).

Middle Miocene. Palseortyx edwardsi Deperet. Saint-Alban-de-Roche (Isere). Pahmpcrdix longipes Milne-Edwards. Sansan (Gers). " prisca " " "

" (?) sansaniensis " " "

Phasianus altus " " and Saint-Alban-de-Roche (Isere).

" medius " Sansan (Gers).

" desnoycrsi " Orleannais.

Upper Miocene. Phasianus alius Milne-Edwards. Lacustrine deposits ; < Eningen, Switzerland.

EASTMAN: FOSSIL AVIAN REMAINS FROM ARMISSAN 133

Lower Pliocene.

l'lumanus archiaci Gaudry. Pikermi (Attica). G alius sesculapii " " "

Upper PI locate. Gallus bravardi Gervais. Arde, near Issoire (Puy-de-D6me).

Pleistocene. Tetrao tetrix Linn. Kent's Hole Tavern, near Torquay, Devonshire.

" urogallus Linn. Forestbed (?) of Norfolk. Lagopus alius (Gmelin). Cavern desposits near Montauban (Tarn-et-Garonne).

" mutus (Montin). " " " " "

Francolinus prictus ( Jardine and Selby). Cavern deposits ; Madras, India. Coturnix novic-zeab indite Quoy and Gaimard. Superficial deposits; New Zealand. Tympanuchus pallidicinctus Ridgway. " " Oregon.

Pediocsetcs phasianellus Linn. " " "

" nanus Shufeldt. " " "

" lucasii " " " "

Polxotctrix gilli " " " "

Meleagris superbus Cope. " " New Jersey.

" ecler Marsh. " " "

" sp. indet. Caves of Pennsylvania. Gallus sp. Rare in European hone caverns. Perdu- sp. " " " "

It will he seen from the foregoing that although gallinaceous birds are plentiful in the late Tertiary, with the exception of the Pliocene, our knowledge of them in the Eocene is confined to but four genera. Two of these, futhermore, are known by a solitary individual each ; and as Eocene bird remains are seldom well pre- served, it is clear that the discovery of fairly complete skeletons from this horizon is a matter of considerable interest and importance. The Carnegie Museum is there- fore to be congratulated upon having recently acquired two tolerable skeletons of Eocene birds from the lacustrine limestone of Armissan, near Narhonne (Aude), the existence of which has not hitherto been made known.

These specimens form part of the famous Payet Collection, presented by Mr. Andrew Carnegie to the Pittsburgh Museum in 1903; and through the unfailing courtesy of Dr. W. J. Holland, Director of the Museum, the present writer has been generously entrusted with their description. For this privilege he desires to express here his sincere thanks.

134 MEMOIRS OF THE CARNEGIE MUSEUM

It has not been possible, however, to offer descriptions of both of these speci- mens in the following pages, owing to the fact that only one of them (fortunately the better preserved of the two) has not been sufficiently disengaged from the matrix to permit of its extended study. The specimen about to be described bears the Carnegie Museum catalogue number 2023, and belongs unquestionably to the genus Taoperdix, founded by Milne-Edwards1 upon a unique individual from Armissan, which had been previously described by Gervais under the name of Tetrao pessieli* Although agreeing with the type in its general characters, the Carnegie example differs from it in its greatly reduced length of wing, and in the disproportion of its limb bones; hence it may properly be regarded as constituting a distinct species. It may be fittingly designated as Taoperdix keltica, in commemoration of the name be- stowed by Aristotle upon the inhabitants of the country near Narbonne.8

Taoperdix keltica, sp. nov. (Plates XIII. -XIV.) Founded upon the crushed skeleton of a bird having approximately the size of a ruffed grouse {Bonasa wmbellus), and differing from the type species of Taoperdix in the relative proportions of its limb bones, especially its much reduced humerus ; also with shorter mandible. Upper Eocene ; Armissan.

Although the skeleton is considerably dismembered and confused, both in the type of this species and in that of T. pessieti, as may be seen from a comparison of the plates, this circumstance must be regarded as rather fortunate than otherwise, since it permits of a more precise examination of the several parts. It is also fortu- nate that these two type-specimens should supplement each other in important respects. For our knowledge of the cranium we must depend solely upon the speci- men belonging to the Carnegie Museum, although the mandible is present in both. Most of the limb bones, too, are better preserved in the new than in the older known species; but the latter, on the other hand, alone exhibits the furculum, sternum and pelvis in satisfactory manner. In the accompanying restoration of T. keltica, parts which are wanting or not clearly recognizable in the actual fossil are represented in outline after analogy with the type species or with recent pheasants, as the case may be. That is to say, when an epiphysis or articular condyle has become crushed or otherwise obliterated in the fossil before us, these parts are restored according to the

1 Oiseaux Fossiles de la France, Vol. I[., p. 225, pi. 127, Paris, 1871.

2 Comptes Kendus, Vol. LIV. (1861), p. 890.

3 Prior to the Roman conquest of Gaul, the whole of that country bore the name of Keltica ; but after it* occupation by the Romans the southern provinces were distinguished from the rest of Keltica by coufening upon them the name of Qallia Nnrbonensis. An interesting description of the Narbonnaise is given by Strabo in the fourth book of his Geog- raphy, and further accounts of this region are to be found in the well known History of Polybius. Both of these ancient writers mention among the wonders of this country the occurrence of so-called " subterranean fish " or "dug mullets," which the inhabitants obtained by digging to a depth of two or three feet in marshy districts.

EASTMAN: FOSSIL AVIAN REMAINS FROM ARMISSAN 135

usual conditions in modern Phasianidx. All of the bones shown in Plates 1 and 2 are precisely as they occur in nature.

Passing now from these general considerations to an account of the several parts of the skeleton, we note that the head is of average size, rather abruptly truncated behind, with a maximum width of 2.2 cm. across the occiput, the orbits being rather posteriorly situated. The mandible is slender, without apparent downward curva- ture in front, approximately 3 cm. in length. The sternum and pelvis, not clearly indicated in the Carnegie example, but well shown in the type species, are regarded by Lydekker1 as suggesting affinity with Nwftidia and Meleagris. It is an open ques- tion, however, whether the lateral aspect of the sternum has not become widened through mechanical compression of the matrix ; at any rate the published figure creates a suspicion that such has been the case. A small portion only of the sacrum is preserved in the Carnegie specimen, the furcula is wanting, and the coracoids not distinctly reognizable. The last-named elements in T. pessieti are stout and straight, without subclavicular process, and with broad sternal facets. In both species the scapula is long and sabre-shaped, without being expanded posteriorly.

Important points of comparison are furnished by the limb bones, which are excellently preserved. It is evident that the species under discussion possessed but feeble powers of flight, the humerus being notably shorter than in T. pessieti and most living pheasants. This bone exhibits a slight double lateral curvature, and has a head of moderate size. The remaining wing bones are proportionally abbreviate, the brachium being of about equal length with the humerus. Nearly all of t lie bones of the man us in both wings are well preserved, and the same is true of the pes, The femur is relatively stout, of medium length, and with a slight forward curva- ture, the head being concealed. The tibio-tarsus is a stout bone, its length having a ratio of 1.3 as compared with the femur, and 1.7 as compared with the tarso-metatar- sus. The latter element is without spurs, so far as indications show, and appears to have been more or less flattened from front to back. The trochleas are of moderate size and widely separated. The phalangeals are slender and of medium length.

The relative length of the principal limb bones in this and other species is ex- hibited in this table, the tarso-metatarsus being taken as a standard of comparsion.

Species.	Tarso-metatarsus.	Tibio-tarsus.	Femur.	Humerus.
	100 100 100	174 172 172	130 133 124	143
	121
	141

The table given below is intended to facilitate comparison of the actual measure- ments of various bones of the skeleton in both species of Taoperdix. As there exists

1 " Catalogue of Fossil Birds iu the British Museum," p. 139 (Loudon, 1891).

136

MEMOIRS OF THE CARNEGIE MUSEUM

in some instances a discrepancy between the measurements given in the text of Milne-Edwards' work and those indicated in the plates, we have arranged the data obtained fron both sources in parallel columns.

Table of Comparative Measurements.

Taoperdic keltica.	Taoperdix, ja-ssieti.
Name of Parts.	Length.	Data from Text.	Data from Figure.
	3.3 cm. 5.7 4.4 3.7 3.6 3.3 2.0 (Incomplete) 3.0 (Wanting)	3.5 cm. 6.2 4.6 5.0 4.8 (Wanting) 2.5 3.4	3.7 cm.
	5.0 5.2 and 4.S 4.8
Ulna
	2.5
	3.8

No indications of plumage accompany either of the Armissan skeletons, although isolated feathers, and even egg-casts, are not uncommon in the lacustrine deposits of the south of France and in Switzerland. For references to the literature concern- ing detached feathers, one may consult an article on " Fossil Plumage," in the Sep- tember number of the American Naturalist for 1904.

The following brief notice concerning the nature of the deposits at Aix and Armissan, together with their principal fossil contents, is sufficiently interesting to be reproduced in its entirety. It is taken from the second volume (pp. 551, 552) of Milne-Edwards' important work on the " Fossil Birds of France " :

" Les marnes d'Aix (Bouches-du-Rhone), si riches en restes de Reptiles, de Poissons et d'lnsectes, en empreintes de feuilles, etc., n'ont pas encore fourni d'osse- ments d'Oiseaux, mais on y a trove des oeufs et des plumes admirablement conserves, dont quelques echantillons font partie du musee de Marseille. Ces marnes parais- sent s'etre deposees a la meme epoque que le gypse des environs de Paris. En 1836, M. Coquand y a trouve des restes de Paleotherium, d'apres lesquels il etablit ce parallelisme qui ne fut pas adopte par Dufrenoy, mais que la plupart des geologues actuels s'accordent a admettre.

" C'est probablement aussi a la meme epoque que se sont deposes les calcaires lacustres d' Armissan (Aude), dans lesquels M. P. Gervais a fait connaitre l'existence d'un (iallinace, le Tetrao Pessieti, que j'ai etudie recemment et range dans un genre nouveau intermediaire entre les Perdrix et les Paons, et que j'ai [re-]nomme Tao- perdix Pessieti. Ces calcaires, toujours accompagnes de gypse, de rognons de soufre et de silex pyromaque, contiennent des debris de Paleotherium, d'Anoplotherium, des Reptiles, des Poissons d'eau douce, des Insectes, et une riche flore de vegetaux qui paraissent avoir peri sous l'inrluence des causes qui out agi a Aix."

EASTMAN : FOSSIL AVIAN REMAINS FROM AKMISSAN 137

SUPPLEMENTARY NoTE.

Since the above description was set in type, the task of working out the second bird skeleton (Cat. No. 2022), referred to above on page 134, has been skilfully per- formed by Mr. 0. A. Peterson, one of the preparators of the Carnegie Museum. In its original condition many of the bones were only partially visible, and portions of them had been injured by the process of sawing or grinding to which the slab had been subjected prior to its acquisition by the Museum. As the result of Mr. Peter- son's manipulation all of the preserved portions have been clearly exposed, thus permitting a far more accurate knowledge of details than could otherwise have been obtained. Fortunate as is tins circumstance, it renders all the more conspicuous the fact that many characteristic portions, such as the articular extremities of several of the limb-bones, have been irretrievably lost through sawing the slab in two.1 Notwithstanding the considerable injury done to the remains, they still furnish us with an important and very precious example of Eocene bird remains. The ex- treme rarity of complete individuals, not more than four or five being known from the Armissan deposits, increases the value of every item of information that is obtainable from any one of them. In the case of the present specimen, the loss of the bead deprives us of perhaps the most significant information, concerning which we would have eagerly sought enlightenment. It remains for us to make as search- ing an inquiry as is possible from a decapitated body.

Fortunately we are able to satisfy ourselves in regard to a number of particulars with approximate accuracy. The conformation of the pelvis, typically gallinaceous, and the form and proportions of the limb-bones, all point to a very close relation- ship between this skeleton and the type species of Taoperdix. Indeed, it appears impossible to doubt that the present specimen and the type of T. pessieti are specifi- cally identical, the differences between them falling well within the limits of indi- vidual variation. This conclusion is supported, in our judgment, by the following table of measurements, in which it has not been considered necessary to repeat the measurements of T. keltica, owing to their relatively greater disproportion. The same characters which enable us to discriminate between that species and the type compel us to distinguish between it and the second Carnegie skeleton now under discussion. A protographic reproduction of the new specimen (No. 2022), of slightly less than the natural size, is given in Plate XV. of this Memoir. In Plate XVI. several characteristic bones are shown as they occur in nature, except that in the case

1 The specimen is evidently a piece of pavement slab, which was only discovered to contain fossil remains after it had been sawn. Editor.

138

MEMOIRS OF THE CARNEGIE MUSEUM

of the limb-bones, some of the articular extremities that have been injured are here partially restored.

Table of Measurements.

Type Specimen of Tiwptidix jicmicli.	Carnegie Example of Same Species.	Type Specimen of Taoperdix pessUti.	Carnegie Example of Same Species.
Name of Parts.	Data from Text.	Data from Figure.	Length.	Name of Parts.	Data from Text.	Data from Figure.	Length.
	3.5 cm. 6.2 4.6 5.0	3.7 cm. 6.3 5.0 5.2 and 4.8	4.2 cm. 6.9 5.0 5.5	Ulna	4.8 cm. 2.5 3.4	4.8 2.5 3.4 5.0	5 5
		2.8
		3.7
		5.0-1-

Explanation of Plates.

Plate XIII. Taoperdix keltiea, sp. nov. Photograph of the specimen in its natural condi- tion, x {.

Plate XIV. Taoperdix keltiea, sp. nov. Restoration of skeleton, missing parts represented in outline, x }.

Plate XV. Taoperdix pessieti (Gervais). Photograph of the second Carnegie specimen (Cat. No. '2022) in its natural condition, x \.

Plate XVI. Taoperdix pessieti (Gervais). Illustrations of the more characteristic bones of the skeleton (Cat. No. 2022), some of the articular extremities slightly restored. All figures of the natural size. 1, humerus; 2, radius and ulna; 3, meta- carpus; 4, phalanges of pes ; 5, tarso-mctata rxux ; l>, pelvis; 7, eoraeoid ; 8, femur; 9, tibio-tarsus ; 10, scapula.

Memoirs Carnegie Museum, Vol.

PLATE XIII.

TAOPEBDIX KELTICA Eastman. Photograph of the Type in the Carnegie Moseum.

Memoirs Carnegie Museum, Vol. II

Plate XIV.

Jfohwy&CcJu., dti.

Taoperoix KELT10A Eastman. Bestobation Natural Si/j:. Tun Missing Parts Supplied in Outline.

Memoirs Carnegie Museum, Vol. II.

Plate XV.

TAOPEUD1X PSSSIETl (GerVAIS). PHOTOGRAPH OK THE ObIGWAL IN THE CARNEGIE M. sk. M.

Memoirs Carnegie Museum. Vol. II.

Plate XVI.

ff

1

TA OPEBDIX PEBS1ET1 ( < i EBVAIS) .

1

MEMOIRS

OK THE

CAENEG-IE MUSEUM.

VOL. II. NO. 4.

DESCRIPTION OF NEW RODENTS AND DISCUSSION OF THE ORIGIN OF D.EMONELIX.

By 0. A. Peterson.

One of the field parties of the Carnegie Museum with the writer in charge was detailed hy the Director to collect fossils in western Nebraska and eastern Wyo- ming during the season of 1904. In the collection which was secured are some very complete remains of rodents, belonging to two new species of Steneqfiber not pre- viously described. This material fully substantiates the views foreshadowed by Professor Scott, Dr. Matthew, and others, that this genus is clearly distinct from Castor. In the following detailed description of this new material it will be seen that there are some features, especially in the cranium, which are similar to those in the genus Castor, hut these do not necessarily imply relationship. The skeleton as a whole shows osteological characters very similar to those of Aplodonfria rufa with which I have compared it.1 Oynomys ludovicianus is also used for comparison in studying the present fossil forma

The material was discovered in the Upper Miocene sandstones, or Harrison ( / hrnuntcli.r) beds, in Sioux Co., northwestern Nebraska, and also in Converse Co., Wyoming, immediately across the Nebraska-Wyoming state line, and in the same general locality. Fossils are comparatively rare in this horizon. The list of genera found is as follows: Promerycochoerus, f Mesoreodon, a Peccary, Oocydactylus, and other small species of camels not identified. The commonest fossils from this horizon are the new species of rodents described in the following pages.

I take pleasure in acknowledging the valuable assistance of Mr. Earl Douglass

1 A specimen of Aplodontia rufa (Col. U. S. Dept. Agri., No. 77975) was kindly tarnished for oomparison hy Dr. C. H. Merriam, Smithsonian Institution, Washington, D. C.

I